Natural resources and bioeconomy studies 74/2020 Role of dietary fish oil and plant oil supplements in ruminal lipid metabolism and fish oil induced milk fat depression in lactating cows Doctoral Dissertation Piia Kairenius Natural resources and bioeconomy studies 74/2020 Role of dietary fish oil and plant oil supple- ments in ruminal lipid metabolism and fish oil induced milk fat depression in lactating cows Doctoral Dissertation Piia Kairenius Academic dissertation To be presented, with the permission of the Faculty of Agriculture and Forestry of the University of Helsinki, for public examination in Auditorium 109 (B2), Forest Sciences Building (Latokartanonkaari 7) of the University of Helsinki on November 13th, 2020, at 12 o’clock. Natural Resources Institute Finland, Helsinki 2020 Supervisors: Dr. Heidi Leskinen Natural Resources Institute Finland (Luke) Production Systems, Milk production Prof. Aila Vanhatalo University of Helsinki, Finland Department of Agricultural Sciences Prof. Kevin J. Shingfield Natural Resources Institute Finland (Luke) Production Systems, Milk production Pre-reviewers: Associate Prof. Kevin Harvatine The Pennsylvania State University, USA College of Agricultural Sciences Department of Animal Science Associate Prof. Rui Bessa University of Lisbon, Portugal Faculty of Veterinary Medicine Opponent: Prof. Veerle Fievez Ghent University, Belgium Department of Animal Sciences and Aquatic Ecology Custos: Prof. Aila Vanhatalo University of Helsinki, Finland Department of Agricultural Sciences Author contact-info: Piia Kairenius Natural Resources Institute Finland (Luke) Animale, FI-31600 Jokioinen, Finland piia.kairenius@luke.fi ISBN: 978-952-380-065-6 (Print) ISBN: 978-952-380-066-3 (Online) ISSN 2342-7647 (Print) ISSN 2342-7639 (Online) URN: http://urn.fi/URN:ISBN:978-952-380-066-3 Copyright: Natural Resources Institute Finland (Luke) Author: Piia Kairenius Publisher: Natural Resources Institute Finland (Luke), Helsinki 2020 Year of publication: 2020 Cover photo: Tintti Design & Anna Lehtinen Printing house: Unigrafia Ltd., Helsinki, Finland Natural resources and bioeconomy studies 74/2020 3 Abstract The objective of the research described in this thesis was to provide new information on the ruminal biohydrogenation of long-chain n-3 polyunsaturated fatty acids (PUFA), such as 20:5n-3, 22:5n-3 and 22:6n-3, for altering bovine milk fatty acid (FA) composition, with the potential to improve human health. Emphasis was not only placed on the po- tential to increase milk fat n-3, but also to modulate ruminal lipid metabolism and to explore the mechanisms driving milk fat synthesis and its regulation in lactating cows in order to understand the mechanisms and metabolic pathways underlying the diet- induced changes in milk fat depression (MFD), milk FA composition and specific FA in- termediates and end products associated with MFD. Experiments documented in I–IV encompassed detailed investigations of ruminal (I-III) and mammary (IV) lipid metabo- lism. Experiment reported in I was conducted to build up methods for the analysis of long-chain 20- to 22-carbon FA intermediates formed during ruminal biohydrogenation of n-3 PUFA. The detailed analysis of fish oil (FO) and omasal digesta of lactating cows fed FO enabled the structure identification of 27 previously unidentified 20- to 22- carbon FA intermediates, containing at least one trans double bond. No conjugated 20- carbon FA were detected in omasal digesta. Results demonstrated that the hydrogena- tion of 20:5n-3, 22:5n-3 and 22:6n-3 in the rumen proceeds via two principal mecha- nisms that involve sequential reduction or isomerisation of cis double bonds closest to carboxyl group and provided clear evidence of extensive biohydrogenation of 20:5n-3, 22:5n-3 and 22:6n-3 in cows fed FO. Experiments documented in II-IV involved two physiological studies in which, the ef- fects of dietary FO supplements alone (II; IV) or in combination (III) with sunflower (rich in 18:2-6; SFO) or linseed (rich in 18:3n-3; LFO) oil on animal performance (II-IV), ruminal lipid metabolism (II; III), microbial ecology in the rumen (II; III) and milk fat composition (IV) were investigated in lactating cows. Dietary FO supplements increased the intakes of 20:5n-3, 22:5n-3, 22:6n-3 and total FA (II; III), whereas decreased dry matter intake (II; III). Dietary oil supplements decreased (II) or had no effect (III) on ruminal volatile FA concentrations, but FO at high amounts (II) or when supplemented with plant oils (III) promoted an increase in molar proportions of propionate (II; III) and butyrate (II) at the expense of acetate (II; III). Supplements of FO modified ruminal metabolism of 16- and 18-carbon PUFA, caus- ing increases in trans 16:1, trans 18:1 and trans 18:2 flow and a decrease in 18:0 at the omasum, and at high amounts promoted trans-10 18:1 accumulation at the expense of trans-11 18:1. Dietary FO had no substantial influence on ruminal outflow of conjugated linoleic acid (CLA). Extensive ruminal biohydrogenation of 20:5n-3, 22:5n-3 and 22:6n-3 resulted in increases in numerous 20- and 22-carbon PUFA containing at least one trans double bond at the omasum. Natural resources and bioeconomy studies 74/2020 4 Relative to FO, ruminal metabolism of 22:6n-3 was more extensive on diets contain- ing plant oils, whereas the biohydrogenation of 22:5n-3 and 20:5n-3 showed no differ- ence between FO and diets containing plant oils (III). The inhibitory effects of FO on the reduction of 18-carbon PUFA to 18:0 were influenced by the source of 18-carbon PUFA in SFO and LFO. The ruminal outflow of 18:0 was lower and accumulation of trans 18:2 and 20- to 22-carbon FA intermediates greater for LFO than SFO. Supplements of SFO and LFO caused trans-10 and trans-11 18:1 to accumulate, trans-10 18:1 being the most abundant FA intermediate in SFO. Alterations in the ruminal metabolism of FA were not associated with substantial changes in rumen protozoal counts or analysed bacterial populations known to be capa- ble of biohydrogenation (II; III), but lowered Butyrivibrio spp. numbers in response to increasing levels of FO (II). Supplements of FO decreased milk fat yield and content and increased 20:5n-3, 22:5n-3 and 22:6n-3 concentrations in milk fat (IV). Enrichment of milk long-chain n-3 PUFA was associated with decreases in 4- to 18-carbon saturated FA and several-fold increases in CLA, trans FA and PUFA concentrations. Dietary FO resulted in the appear- ance of 37 unique 20- and 22-carbon FA in milk. FO-induced MFD (up to -40.6 % reduction in milk fat synthesis) was associated with changes in the concentrations of multiple FA in milk, in particular increases in milk fat trans-10 18:1 and cis-11 18:1 concentrations, but not with changes in the amount of trans-10,cis-12 CLA in milk and omasum or estimated milk fat melting point (IV). The negative relationship between ruminal outflow of trans-10 18:1 and milk fat secretion confirmed that a shift in ruminal biohydrogenation of 18-carbon FA toward trans-10 pathway has a role in the regulation of milk fat synthesis during FO-induced MFD. A de- crease in 18:0 supply in combination with increased mammary uptake of cis-11 18:1, trans-10 18:1, and trans 20- and 22-carbon FA intermediates originating from the rumen may contribute to the reduction of milk fat observed during FO-induced MFD. The dietary supplements of FO alone or in combination with plant oils increased the ruminal outflow of FA intermediates containing at least one trans double bond and en- riched long-chain n-3 PUFA in bovine milk with associated changes in the abundance and distribution of FA. These changes may have implications for the host metabolism and the nutritional quality or ruminant-derived foods. Keywords: biohydrogenation, rumen, fish oil, plant oil, sunflower oil, linseed oil, lactating cow, polyunsaturated fatty acid, n-3 fatty acid, conjugated linoleic acid, milk fat, trans fatty acid, Butyr- ivibrio, microbial ecology, gas chromatography, mass spectrometry, silver-ion thin-layer chroma- tography Natural resources and bioeconomy studies 74/2020 5 Tiivistelmä Kala- ja kasviöljylisäysten vaikutus lypsylehmien rasva-aineenvaihduntaan pötsissä sekä kalaöljylisäyksen aiheuttama maitorasvasynteesin heikkeneminen Tämän väitöskirjatyön tavoitteena oli tuottaa uutta tutkimustietoa monityydyttymättö- mien n-3-rasvahappojen, kuten 20:5n-3, 22:5n-3 ja 22:6n-3, biohydrogenaatiosta pötsis- sä sekä mahdollisuuksista muokata maitorasvan koostumusta ihmisravitsemuksen kan- nalta suotuisammaksi. Lypsylehmien nurmisäilörehuvaltaiseen ruokintaan lisättiin kala- ja kasviöljyjä, minkä avulla selvitettiin pötsin biohydrogenaation reaktioreitteihin vaikut- tavia ravitsemuksellisia säätelymekanismeja ja pyrittiin lisäämään rasvahappojen 20:5n-3, 22:5n-3 ja 22:6n-3 pitoisuuksia maidossa. Samalla tarkasteltiin pötsissä tapah- tuvan biohydrogenaation väli- ja lopputuotteena muodostuvien rasvahappojen sekä maitorasvasynteesin heikkenemisen eli maitorasvan depression välisiä yhteyksiä. Väitös- kirja perustuu kolmeen ravitsemusfysiologiseen tutkimukseen, joiden tulokset on jul- kaistu väitöskirjan osajulkaisuissa I-IV. Ensimmäisessä tutkimuksessa, jonka tulokset on raportoitu osajulkaisussa I, tutkit- tiin kalaöljyä saaneiden lypsylehmien pötsistä satakertaan virtaavan ruokasulan sisältä- miä pitkäketjuisia 20- ja 22-hiilisiä tyydyttymättömiä rasvahappoja sekä kehitettiin ana- lyysimenetelmiä, joiden avulla määritettiin näiden rasvahappojen rakennetta, kuten kaksoissidosten paikkaa hiiliketjussa ja cis-trans rakenneisomeriaa. Erilaisia rasvahap- poanalytiikan menetelmiä yhdistämällä löydettiin jopa 27 aikaisemmin tunnistamatonta pötsin biohydrogenaation väli- ja/tai lopputuotteena muodostunutta, 20- ja 22- hiiliketjun pituista rasvahappoa, jotka sisälsivät vähintään yhden trans-kaksoissidoksen. Satakertaan virtaavasta ruokasulasta ei havaittu lainkaan yli 20 hiiltä sisältäviä rasva- happoa, joilla olisi ollut hiiliketjussaan ns. konjugoitunut rakenne. Tulokset osoittivat, että rasvahappojen 20:5n-3, 22:5n-3 ja 22:6n-3 biohydrogenaatio pötsissä on hyvin te- hokas reaktiosarja, joka alkaa pääsääntöisesti lähimpänä rasvahapon karboksyyliryhmää sijaitsevien cis-muotoisten kaksoissidosten pelkistyessä tyydyttyneeseen muotoon tai niiden muuttuessa trans-muotoon isomerisaation seurauksena. Toisen ja kolmannen tutkimuksen tulokset on julkaistu väitöskirjan osajulkaisuissa II- IV. Tutkimusten tarkoituksena oli selvittää rehuun lisätyn kalaöljyn (II; IV), kala- ja aurin- gonkukkaöljyn (III; runsaasti linolihappoa; 18:2n-6) sekä kala- ja pellavaöljyn (III; runsaas- ti α-linoleenihappoa; 18:3n-3) vaikutuksia lypsylehmien ravintoaineiden saantiin, pötsis- sä tapahtuvaan rasva-aineenvaihduntaan ja pötsimikrobien ekologiaan (II; III) sekä mai- torasvan koostumukseen ja ravintoaineiden hyväksikäyttöön maidontuotannossa (IV). Kalaöljyn lisäys nosti odotetusti kokonaisrasvahappojen sekä rasvahappojen 20:5n-3, 22:5n-3 ja 22:6n-3 saantia, mutta vähensi lehmien kokonaiskuiva-aineen syöntiä (II; III). Kalaöljyn lisäys yhdessä kasviöljyjen kanssa ei vaikuttanut pötsissä muodostuvien haih- tuvien rasvahappojen kokonaispitoisuuteen (III), mutta sen pitoisuus väheni lisättäessä lehmien rehuun pelkkää kalaöljyä (II). Kalaöljyn lisäys yksin (II) tai yhdessä kasviöljyjen kanssa (III) lisäsi pötsissä muodostuneiden propionihapon (II; III) ja voihapon mooli- osuuksia (II) vähentäen etikkahapon suhteellista määrää (II; III). Natural resources and bioeconomy studies 74/2020 6 Kalaöljyn lisäys muutti pötsin 16- ja 18-hiilisten monityydyttymättömien rasvahap- pojen biohydrogenaatiota merkitsevästi, aiheuttaen pötsistä satakertaan virtaavien trans-16:1, trans-18:1 ja trans-18:2 rasvahappojen kokonaismäärien lisääntymisen sekä rasvahapon 18:0 virtauksen vähenemisen. Suurimmalla kalaöljyannoksella rasvahapon trans-11 18:1 virtaus kääntyi laskuun ja rasvahapon trans-10 18:1 virtaus lisääntyi voi- makkaasti. Kalaöljyn lisäys ei vaikuttanut juurikaan pötsistä ulos virtaavan konjugoidun linolihapon (CLA) kokonaismäärään, mutta lisäsi lukuisten 20:5n-3, 22:5n-3 ja 22:6n-3 rasvahappojen biohydrogenaation väli- ja lopputuotteina muodostuvien, trans- kaksoissidoksen sisältävien, 20- ja 22-hiilisten rasvahappojen virtausta satakertaan. Kalaöljyn lisääminen rehuun yksin tai yhdessä kasviöljyjen kanssa ei vaikuttanut ras- vahappojen 20:5n-3 ja 22:5n-3 biohydrogenaatioon pötsissä, mutta rasvahapon 22:6n-3 pelkistyminen oli perusteellisempaa silloin kun kalaöljyä annettiin yhdessä kasviöljyjen kanssa (III). Kasviöljyille ominaiset erot 18-hiilisten tyydyttymättömien rasvahappojen (18:2n-6 ja 18:3n-3 auringonkukka- ja pellavaöljyille vastaavasti) pitoisuuksissa vaikutti- vat siihen, miten voimakkaasti kalaöljylisäys rajoitti 18-hiilisten tyydyttymättömien ras- vahappojen pelkistymistä pötsissä rasvahapoksi 18:0 sekä siihen, minkälaisia reaktioväli- tuotteita epätäydellisen biohydrogenaation seurauksena pötsissä muodostui (III). Lisät- täessä kala- ja pellavaöljyjen seosta rehuun pötsistä ulosvirtaavien trans-18:2 rasvahap- pojen ja trans-kaksoissidoksen sisältävien 20- ja 22-hiilisten rasvahappojen määrä oli suurempi ja rasvahapon 18:0 virtaus pienempi kuin lisättäessä kala- ja auringonkukkaöl- jyjen seosta. Molemmat kala- ja kasviöljyjen seokset aiheuttivat rasvahappojen trans-10 18:1 ja trans-11 18:1 kertymisen pötsinesteeseen, mutta lisättäessä kalaöljyä yhdessä auringonkukkaöljyn kanssa rasvahapon trans-10 18:1 muodostus oli runsaampaa. Muutokset pötsin monityydyttymättömien rasvahappojen biohydrogenaatiossa ei- vät vaikuttaneet pötsi- ja satakertanesteestä määritettyjen alkueläinten ja pötsin biohydrogenaatioon osallistuvien bakteeripopulaatioiden kokonaismääriin (II; III), vaikka Butyrivibrio-ryhmään kuuluvien bakteerien määrät laskivatkin kalaöljyannosta lisättäes- sä (II). Maidon rasvatuotos ja rasvapitoisuus pienenivät kalaöljyä lisättäessä (IV). Samalla maidon rasvahappokoostumus muuttui ihmisravitsemuksen kannalta suotuisammaksi. Kala- ja kasviöljyjen lisäys vähensi maitorauhasen de novo-synteesistä peräisin olevien tyydyttyneiden rasvahappojen, 4:0–12:0 ja 16:0, sekä rasvahapon 18:0 pitoisuuksia ja nosti trans-rasvahappojen ja terveyttä edistävien monityydyttymättömien rasvahappo- jen, kuten CLA, 20:5n-3, 22:5n-3 ja 22:6n-3, pitoisuutta maitorasvassa. Kalaöljyä saanei- den lehmien maidosta tunnistettiin yhteensä 37 erityistä 20- ja 22-hiilistä rasvahappoa, joiden erittyminen maitoon lisääntyi kalaöljyannosta nostettaessa (IV). Kalaöljylisäyksen aiheuttama maitorasvan depressio (jopa -40.6 %:n vähennys mai- torauhasen rasvasynteesissä) oli yhteydessä lukuisten maidon rasvahappojen pitoisuuk- sien muutoksiin (IV). Maidon trans-10 18:1 ja cis-11 18:1 rasvahappojen pitoisuudet kasvoivat maitorasvan synteesin vähentyessä, mutta ne eivät yksin selitä havaittua kala- öljyn aiheuttamaa maitorasvasynteesin heikkenemistä. Trans-10,cis-12 CLA:n osuus maidon ja satakerran ruokasulan rasvahapoista ei ollut yhteydessä maitorasvan depres- sioon. Lisäksi tässä tutkimuksessa kalaöljyn lisäys ei vaikuttanut laskennalliseen maito- Natural resources and bioeconomy studies 74/2020 7 rasvan sulamispisteeseen eikä maitorasvan synteesin heikkeneminen näin ollen ollut seuraus maitorasvan korkeammasta sulamispisteestä. Pötsistä ulos virtaavan trans-10 18:1 rasvahapon ja maitorasvan tuoton välillä oli selvä negatiivinen yhteys. Tämä vahvistaa näkemystä siitä, että maitorasvasynteesin lasku johtuu ainakin osittain kalaöljylisäyksen aiheuttamasta 18-hiilisten monityydytty- mättömien rasvahappojen biohydrogenaation etenemisestä vaihtoehtoista reaktioreittiä pitkin, jolloin pötsissä muodostuu runsaasti trans-10 18:1 rasvahappoa ja rasvahapon 18:0 satakertavirtaus sekä eritys maitoon vähenee. Kalaöljyä saaneet lehmät käyttivät rasvahapon 18:0 saannin vähentyessä enemmän rasvahappoja cis-11 18:1 ja trans-10 18:1 sekä trans-kaksoissidoksen sisältäviä 20- ja 22-hiilisiä rasvahappoja maitorasvan muodostamiseen, jolloin nämä pötsin biohydrogenaation välituotteena muodostuneet trans-rasvahapot voivat olla osatekijänä maitorasvan depressiossa (IV). Tulokset osoittivat, että kalaöljyn lisäys lypsylehmien rehuun, yksin tai yhdessä kas- viöljyjen kanssa, aiheuttivat pötsistä ulosvirtaavien pitkäketjuisten, trans- kaksoissidoksen sisältävien, rasvahappojen määrän lisääntymisen. Lisäksi kalaöljylle ominaisten terveysvaikutteisten monityydyttymättömien n-3 rasvahappojen osuus mai- torasvassa kasvoi. Kala- ja kasviöljyjen lisääminen lypsylehmien rehuun vaikuttaa siis paitsi eläimen aineenvaihduntaan myös märehtijäperäisten elintarvikkeiden ravitsemuk- selliseen koostumukseen. Avainsanat: biohydrogenaatio, pötsi, kalaöljy, kasviöljy, auringonkukkaöljy, pellavaöljy, lypsyleh- mä, monityydyttymätön rasvahappo, n-3-rasvahappo, konjugoitu linolihappo, maitorasva, trans- rasvahappo, Butyrivibrio, mikrobiekologia, kaasukromatografia, massaspektrometria, hopea-ioni- ohutlevykromatografia Natural resources and bioeconomy studies 74/2020 8 Acknowledgements I wish to express my sincere gratitude to my supervisors in lipid and animal science, Dr. Heidi Leskinen at Natural Resources Institute Finland (Luke; formerly MTT Agrifood Re- search Finland), Prof. Aila Vanhatalo at the University of Helsinki, Finland, and late Prof. Kevin J. Shingfield at Luke, for their guidance, valuable comments, constructive criticism and unfailing support, especially during the final steps of preparing the scientific publica- tions in this research. I wish to extend my special appreciation to LicSc Vesa Toivonen, and Minna Aalto at Luke for their invaluable assistance and guidance in the analysis of fatty acid composition as well as support and friendship during all these years. I thank Prof. Tuomo Varvikko, Dr. Jutta Kauppi, and MSc Ilkka Sipilä at Luke for providing excellent research facilities and skillful technical personnel at Animal Produc- tion Research in Jokioinen to assist in the completion of the animal experiments and laboratory analysis. Valuable contributions of all my co-authors, MSc Anu Ärölä, Dr. Sep- po Ahvenjärvi, MSc Timo Hurme (Luke), Prof. Pekka Huhtanen (Swedish University of Agricultural Sciences, Sweden), Docent J. Mikko Griinari (University of Helsinki, Finland), Dr. Stefan Muetzel (AgResearch Ltd., New Zealand), Dr. Delphine Paillard, and Prof. R. John Wallace (University of Aberdeen, UK), to experimental data processing, microbial ecology analysis, and scientific writing are very much appreciated. I am also grateful to Dr. Päivi Mäntysaari, Dr. Ali Reza Bayat, Prof. Marketta Rinne, and Prof. Johanna Vilkki at Luke for all the help and support in other research activities during these last few years. My warmest thanks are due to the team members of the Animal Metabolism Unit in Jokioinen, Aino Matilainen, Laila and Aaro Hakkarainen, Hannu Peltonen, Mirja Seppälä, Jani Tuomola, Outi Karesma, Helvi Kananen, and Aune Kirvelä for their assistance in conducting the animal experiments and sample preparation. My appreciation also ex- tends to entire laboratory staff at Luke Jokioinen, Mervi Mikkola, Tuija Hakala, Pia Wesin, Juha Widen, Tarja Virta, Marja Rauvola, Helmi Vaittinen, and Marketta Ratilainen under the supervision of MSc Taina Jalava for their contribution to laboratory analysis. I am greatly indebted to the referees appointed by the faculty, Associate Prof. Kevin Harvatine at Penn State University, USA, and Associate Prof. Rui Bessa at University of Lisbon, Portugal, for their careful evaluation and constructive criticism on the thesis. Financial support from the Finnish Ministry of Agriculture and Forestry, the University of Helsinki Research Funds and Luke is gratefully acknowledged. This work was supported in part by BIOCLA, an European Union 5th Framework Program Project (QLK1-2002- 023621, Production of CLA Enriched Products by Natural Means). I want to express my special thanks to my fellow PhD colleagues, past and present, Dr. Eija Valkonen, MSc Eija Talvio, MSc Laura Ventto, Dr. Anne Honkanen, Dr. Anni Halmemies-Beauchet-Filleau, Dr. Tomasz Stefański, and Dr. Erja Koivunen for their en- couragement and support during the long process of completing this work. I am also very grateful to my very good colleagues Mari Talvisilta, MSc Petra Tuunainen, MSc(Tech) Satu Ervasti, MBA Mikko Räsänen, MSc Janina Stefańska, MSc Mari Kukkola, and Dr. Elina Tampio at Luke, as well as our visiting researchers Dr. Pablo Gutiérrez Toral, Dr. Pilar de Frutos, Dr. Gonzalo Hervás (Instituto de Ganadería de Montaña, Natural resources and bioeconomy studies 74/2020 9 Spain), Dr. Kirsty Kliem (University of Reading, UK), Dr. Sylvain Lerch (Agroscope, Swit- zerland), and Dr. Mina Vazirigohar (University of Zanjan, Iran) for their fellowship and all the most memorable trips and events during these years. Finally, I want to express my warmest thanks to my support group at home farm. During this work, I have been surrounded by the enormous support of friends and fami- ly. I warmly thank them all for their encouragement, patience and support, as well as sharing the ups and downs in life with me. This research is dedicated to my closest fami- ly, and the memory of late Prof. Kevin J. Shingfield with gratitude for his inspiration, guidance, and commitment to this work and above all for his warm and everlasting friendship. Natural resources and bioeconomy studies 74/2020 10 List of original publications This thesis is based on the following publications: I Kairenius, P., Toivonen, V., and Shingfield, K. J. 2011. Identification and ruminal outflow of long-chain fatty acid biohydrogenation intermediates in cows fed diets containing fish oil. Lipids 46:587–606. https://doi.org/10.1007/s11745- 011-3561-1 II Shingfield, K. J., Kairenius, P., Ärölä, A., Paillard, D., Muetzel, S., Ahvenjärvi, S., Vanhatalo, A., Huhtanen, P., Toivonen, V., Griinari, J. M., and Wallace, R. J. 2012. Dietary fish oil supplements modify ruminal biohydrogenation, alter the flow of fatty acids at the omasum, and induce changes in the ruminal Butyrivib- rio population in lactating cows. J. Nutr. 142:1437–1448. https://doi.org/10.3945/jn.112.158576 III Kairenius, P., Leskinen, H., Toivonen, V., Muetzel, S., Ahvenjärvi, S., Vanhatalo, A., Huhtanen, P., Wallace, R. J., and Shingfield, K. J. 2018. Effect of dietary fish oil supplements alone or in combination with sunflower and linseed oil on ru- minal lipid metabolism and bacterial populations in lactating cows. J. Dairy Sci. 101:1–15. https://doi.org/10.3168/jds.2017-13776 IV Kairenius, P., Ärölä, A., Leskinen, H., Toivonen, V., Ahvenjärvi, S., Vanhatalo, A., Huhtanen, P., Hurme, T., Griinari, J. M., and Shingfield, K. J. 2015. Dietary fish oil supplements depress milk fat yield and alter milk fatty acid composition in lactating cows fed grass silage-based diets. J. Dairy Sci. 98:5653–5671. http://dx.doi.org/10.3168/jds.2015-9548 The publications are referred to in the text by their Roman numerals. The articles are reprinted with the kind permission of their respective copyright owners. All experiments were conducted at the Animal Metabolism Unit of Natural Resources Institute Finland (Luke, Jokioinen; formerly MTT Agrifood Research Finland). Natural resources and bioeconomy studies 74/2020 11 Contributions The contribution of all authors to the original articles of this thesis is described in the following table (initials of authors are listed in alphabetical order). I II III IV Planning the experiment KJS PH PK SA AV JMG KJS PK PH SA AV KJS PH SA AV JMG KJS PK PH SA Conducting the experiment KJS PK AV AÄ KJS PK SA AV KJS SA AV AÄ KJS PK SA Lipid analysis KJS PK VT AÄ KJS PK VT HL KJS PK VT AÄ HL KJS PK VT Microbial ecology analysis DP KJS RJW SM KJS RJW SM Experimental data analysis KJS PK AV AÄ DP KJS PK PH RJW AV KJS PK PH RJW SM AV AÄ KJS PK PH TH Manuscript preparation KJS PK TV AV AÄ KJS PH PK RJW TV AV HL KJS PH PK RJW AV KJS PH PK RJW AV = Aila Vanhatalo AÄ = Anu Ärölä DP = Delphine Paillard HL = Heidi Leskinen JMG = J. Mikko Griinari KJS = Kevin J. Shingfield PH = Pekka Huhtanen PK = Piia Kairenius RJW = R. John Wallace SA = Seppo Ahvenjärvi SM= Stefan Muetzel TH = Timo Hurme VT = Vesa Toivonen Natural resources and bioeconomy studies 74/2020 12 Abbreviations ACACA acetyl-CoA carboxylase Ag+ silver ion BHBA ß-hydroxybutyrate CLA conjugated linoleic acid CVD cardiovascular disease DM dry matter DMI dry matter intake DMOX dimethyloxazoline DNA deoxyribonucleic acid FAME fatty acid methyl ester FA fatty acid FID flame ionization detector FO fish oil FASN fatty acid synthase GC gas chromatography GL glycolipid HDL high-density lipoprotein HPLC high-performance liquid chromatography iNDF indigestible neutral detergent fibre LDL low-density lipoprotein LFO diet based on grass silage supplemented with 500 g/d of linseed oil and 200 g/d of fish oil LO linseed oil MFD milk fat depression mRNA messenger ribonucleic acid MS mass spectrometry MUFA monounsaturated fatty acid NEFA non-esterified fatty acid NDF neutral detergent fibre OBCFA odd- and branched chain fatty acid PCR polymerase chain reaction PL phospholipid PLS partial least squares regression PUFA polyunsaturated fatty acid SCD stearoyl-CoA desaturase, ∆-9 desaturase SFA saturated fatty acid SFO diet based on grass silage supplemented with 500 g/d of sunflower and 200 g/d of fish oil SO sunflower oil SREBP1 sterol regulatory element-binding protein TAG triacylgycerol TLC thin-layer chromatography VFA volatile fatty acid Natural resources and bioeconomy studies 74/2020 The following table describes fatty acid families (e.g. omega-numbering), systematic and trivial names, shorthand abbreviations and notations of selected (and more common) short- and long-chain fatty acids used in the text. Family1 Systematic name Trivial name Shorthand abbreviation Shorthand notation Saturated fatty acids2 ethanoic acid acetic acid 2:0 propanoic acid propionic acid 3:0 butanoic acid butyric acid 4:0 hexanoic acid caproic acid 6:0 octanoic acid caprylic acid 8:0 decanoic acid capric acid 10:0 dodecanoic acid lauric acid 12:0 tridecanoic acid 13:0 tetradecanoic acid myristic acid 14:0 pentadecanoic acid 15:0 hexadecanoic acid palmitic acid 16:0 heptadecanoic acid margaric acid 17:0 octadecanoic acid stearic acid 18:0 nonadecanoic acid 19:0 eicosanoic acid arachidic acid 20:0 heneicosanoic acid 21:0 docosanoic acid behenic acid 22:0 tricosanoic acid 23:0 tetracosanoic acid lignoceric acid 24:0 pentacosanoic acid 25:0 hexacosanoic acid cerotic acid 26:0 heptacosanoic acid heptacosylic acid 27:0 octacosanoic acid montanic acid 28:0 13 Natural resources and bioeconomy studies 74/2020 Family1 Systematic name Trivial name Shorthand abbreviation Shorthand notation Monounsaturated fatty acids2 n-7 cis-9-hexadecenoic acid palmitoleic acid cis-9 16:1 16:1n-7 n-4 all-cis-9,12-hexadecadienoic acid cis-9,cis-12 16:2 16:2n-4 n-4 all-cis-6,9,12-hexadecatrienoic acid cis-6,cis-9,cis-12 16:3 16:3n-4 n-3 all-cis-7,10,13-hexadecatrienoic acid HTA 16:3n-3 n-3 all-cis-4,7,10,13-hexadecatetraenoic acid 16:4n-3 n-1 all-cis-6,9,12,15-hexadecatetraenoic acid 16:4n-1 n-9 cis-9-octadecenoid acid oleic acid cis-9 18:1 18:1n-9 trans-9-octadecenoic acid elaidic acid trans-9 18:1 trans-10-octadecenoic acid trans-10 18:1 n-7 cis-11-octadecenoic acid cis-vaccenic acid cis-11 18:1 trans-11-octadecenoic acid vaccenic acid trans-11 18:1 Polyunsaturated 18-carbon fatty acids2 n-6 all-cis-9,12-octadecadienoic acid linoleic acid, LA cis-9,cis-12 18:2 18:2n-6 trans-9,trans-12-octadecadienoic acid linoelaidic trans-9,trans-12 18:2 n-6 all-cis-6,9,12-octadecatrienoic acid γ-linolenic acid, GLA cis-6,cis-9,cis-12 18:3 18:3n-6 n-3 all-cis-9,12,15-octadecatrienoic acid α-linolenic acid, ALA cis-9,cis-12,cis-15 18:3 18:3n-3 cis-9,trans-11,cis-15-octadecatrienoic acid cis-9,trans-11,cis-15 18:3 n-6 all-cis-3,6,9,12-octadecatetraenoic acid γ-stearidonic acid 18:4n-6 n-3 all-cis-6,9,12,15-octadecatetraenoic acid stearidonic acid, SDA 18:4n-3 Conjugated linoleic acids2 cis-9,trans-11-octadecadienoic acid rumenic acid cis-9,trans-11 CLA trans-10,cis-12- octadecadienoic acid trans-10,cis-12 CLA Polyunsaturated long-chain 20-, 21-, 22- and 24-carbon fatty acids2 n-9 cis-11-eicosenoic acid gondoic acid cis-11 20:1 20:1n-9 n-7 cis-13-eicosenoic acid paullinic acid cis-13 20:1 20:1n-7 n-6 all-cis-11,14-eicosadienoic acid cis-11,cis-14 20:2 20:2n-6 n-3 all-cis-14,17-eicosadienoic acid cis-14,cis-17 20:2 20:2n-3 14 Natural resources and bioeconomy studies 74/2020 Family1 Systematic name Trivial name Shorthand abbreviation Shorthand notation n-9 all-cis-5,8,11-eicosatrienoic acid mead acid 20:3n-9 n-6 all-cis-8,11,14-eicosatrienoic acid dihomo-γ-linolenic acid, DGLA 20:3n-6 n-3 all-cis-11,14,17-eicosatrienoic acid ETE 20:3n-3 n-6 all-cis-5,8,11,14-eicosatetraenoic acid arachidonic acid, AA 20:4n-6 n-3 all-cis-8,11,14,17-eicosatetraenoic acid ETA 20:4n-3 n-3 all-cis-5,8,11,14,17-eicosapentaenoic acid EPA 20:5n-3 n-3 all-cis-6,9,12,15,18-heneicosapentaenoic acid HPA 21:5n-3 n-9 cis-13-docosenoic acid erucic acid cis-13 22:1 22:1n-9 n-7 cis-15-docosenoic acid cis-15 22:1 22:1n-7 n-6 all-cis-13,16-docosadienoic acid 22:2n-6 n-3 all-cis-13,16,19-docosatrienoic acid 22:3n-3 n-4 all-cis-10,13,16-docosatrienoic acid 22:3n-4 n-9 all-cis-4,7,10,13-docosatetraenoic acid 22:4n-9 n-6 all-cis-7,10,13,16-docosatetraenoic acid adrenic acid 22:4n-6 n-3 all-cis-10,13,16,19-docosatetraenoic acid 22:4n-3 n-6 all-cis-4,7,10,13,16-docosapentaenoic acid osbond acid, n-6 DPA 22:5n-6 n-3 all-cis-7,10,13,16,19-docosapentaenoic acid clupanodonic acid, n-3 DPA 22:5n-3 n-3 all-cis-4,7,10,13,16,19-docosahexaenoic acid DHA 22:6n-3 n-9 cis-15-tetracosenoic acid nervonic acid cis-15 24:1 24:1n-9 n-7 cis-17-tetracosenoic acid cis-17 24:1 24:1n-7 n-6 all-cis-9,12,15,18-tetracosatetraenoic acid 24:4n-6 n-6 all-cis-6,9,12,15,18-tetracosapentaenoic acid 24:5n-6 n-3 all-cis-9,12,15,18,21-tetracosapentaenoic acid 24:5n-3 n-3 all-cis-6,9,12,15,18,21-tetracosahexaenoic acid nisinic acid 24:6n-3 1In the omega nomenclature the double bond positions are counted from the methyl end of the carbon chain. For example, the n-3 FA have a “signature” double bond in the third carbon from the methyl end of the molecule (omega-3 or n-3 FA). 2Fatty acids consist of carbon atom chains that have a methyl group (CH3) at one end and a carboxylic acid group (COOH) at the other. Two main classes of FA are saturated and unsaturated fatty acids, which are further divided to monounsaturated and polyunsaturated fatty acids. 15 Natural resources and bioeconomy studies 74/2020 16 Contents 1. Introduction ........................................................................................................ 18 2. Aims of the thesis ................................................................................................ 27 3. Materials and methods ....................................................................................... 29 3.1. Experimental animals and designs .............................................................................. 29 3.2. Experimental treatments............................................................................................. 29 3.3. Experimental measurements ...................................................................................... 30 3.4. Statistical analyses ....................................................................................................... 31 3.5. Summary of experimental designs .............................................................................. 33 4. Results and discussion ......................................................................................... 34 4.1. Analysis of long-chain fatty acids by chromatography ................................................ 34 4.2. Impact of dietary fatty acids on nutrient intake, digestibility and rumen fermentation characteristics .............................................................................................. 38 4.2.1. Fatty acids in feeds ................................................................................................... 38 4.2.1.1. Fatty acids in grass silage and concentrates ......................................................... 38 4.2.1.2. Fatty acids in oil supplements ............................................................................... 38 4.2.2. Effect on nutrient intake and digestibility................................................................ 39 4.2.3. Effect on rumen fermentation characteristics ......................................................... 39 4.3. Lipid metabolism in the rumen ................................................................................... 42 4.3.1. General changes in the omasal flow of fatty acid biohydrogenation products and saturated fatty acids .......................................................................................................... 42 4.3.2. Biohydrogenation intermediate products of 16-carbon fatty acids ........................ 43 4.3.3. Biohydrogenation intermediate products of 18-carbon fatty acids ........................ 43 4.3.4. Biohydrogenation intermediate products of 20- to 22-carbon fatty acids .............. 51 4.3.5. The extent of apparent ruminal biohydrogenation of dietary unsaturated fatty acids ................................................................................................................................... 58 4.3.6. Rumen microbial ecology ......................................................................................... 60 4.4. Lipid metabolism in the mammary gland .................................................................... 61 4.4.1. Effect of dietary fatty acids on milk production and composition ........................... 61 4.4.2. Fatty acids in milk ..................................................................................................... 62 4.4.2.1. Saturated fatty acids and cis-9 18:1 in milk .......................................................... 62 4.4.2.2. 16-carbon fatty acids in milk ................................................................................. 63 Natural resources and bioeconomy studies 74/2020 17 4.4.2.3. 18-carbon fatty acids in milk ................................................................................. 64 4.4.2.4. 20- to 22-carbon fatty acids in milk ...................................................................... 65 4.4.2.5. Transfer efficiency of long-chain n-3 fatty acids from feed to milk ...................... 66 4.4.3. Milk fat depression .................................................................................................. 67 5. Conclusions ......................................................................................................... 74 6. Future research ................................................................................................... 76 References ............................................................................................................... 78 Natural resources and bioeconomy studies 74/2020 18 1. Introduction Lipids in human nutrition. Lipids and fatty acids (FA) play an important role in human health and nutrition. Ruminant derived foods are an important and versatile source of nutrients, such as amino acids, calcium and bioactive lipids in the human diet (Givens and Shingfield, 2006; Shingfield et al., 2013). As ruminant derived foods are also rich in saturated fatty acids (SFA) there has been an increased interest in developing nutritional strategies for altering the composition of milk and meat fat to improve long-term human health (Givens and Shingfield, 2006; Shingfield et al., 2008b). Clinical studies implicate an excessive consumption of medium-chain SFA and trans FA and low intake of polyunsaturated FA (PUFA) and n-3 PUFA as risk factors for cardio- vascular disease (CVD) and other metabolic diseases (Meier et al., 2019; Mensik et al., 2003; Willet et al., 1993). It is well established that SFA, in particular myristic acid (14:0) and palmitic acid (16:0), and possibly also lauric acid (12:0), decrease the relative pro- portions of high-density lipoprotein (HDL) and increase low-density lipoprotein (LDL) cholesterol and increase the risk of CVD in humans (Hu et al., 2001; Givens, 2008). In contrast, dietary consumption of long-chain n-3 PUFA, such as α-linolenic acid (18:3n-3), eicosapentaenoic acid (20:5n-3), docosapentaenoic acid (22:5n-3), and do- cosahexaenoic acid (22:6n-3) is beneficial for human health and has been shown to in- duce protective effects against CVD (Wang et al., 2006; Harris et al., 2008; 2009; Palmquist, 2009). There is also evidence to indicate that moderate increases in the con- sumption of n-3 PUFA improves immune function, prevent certain inflammatory diseas- es and neurological disorders and may prevent certain cancers (Wang et al., 2006; Harris et al., 2009; Palmquist, 2009; Yashodhara et al., 2009; Deckelbaum and Torrejon 2012). Fatty acid 18:3n-3 must be obtained from the diet as it cannot be synthesised by hu- mans. Synthesis of 20:5n-3 and 22:6n-3 from 18:3n-3 in humans is limited (Burdge and Calder, 2005) and therefore the diet is the principal source of long-chain n-3 PUFA. Milk and dairy products as a source of lipids. The increased awareness of the asso- ciation between diet and health has led to nutritional quality becoming an important determinant of consumer food choices. Milk and dairy products are the main source of 12:0, 14:0, 16:0 and total SFA in the human diet and also make a significant contribution to trans FA intake. Lowering the concentration of these nutritionally undesirable FA and increasing the content of specific bioactive lipids recognized as having potential or puta- tive beneficial effects on human health (such as cis-9 18:1, cis-9,trans-11 conjugated linoleic acid (CLA), 18:3n-3, and n-3 PUFA) in ruminant milk, meat and dairy products offer an opportunity to improve long-term human health without requiring changes in consumer eating habits. Developing food products and dietary regimens that promote human health is also central for preventing and reducing the economic and social impact of chronic diseases. Nutritional strategies to modulate lipid metabolism and increase the total unsatu- rated FA and n-3 PUFA content of milk in lactating cows have often included supple- menting feeding rations with plant oils, oilseeds, marine lipids (e.g. fish oil (FO) and ma- Natural resources and bioeconomy studies 74/2020 19 rine algae) and rumen protected lipids (Kliem and Shingfield, 2016), and selecting differ- ent forage sources (e.g. red clover, pasture, grass hay and grass silage; Halmemies- Beauchet-Filleau et al., 2013b) and concentrate components (e.g. feeds high in rapidly fermentable carbohydrates such as cereal grains) with plant oils (Loor et al., 2004; Ventto et al., 2017) and using different forage:concentrate ratios (e.g. high or low levels of starch/forage; Kliem and Shingfield, 2016). These dietary approaches are not only used for improving the nutritional quality of milk, but also to alter ruminal or tissue lipid metabolism and achieve a desired response, such as a decrease in milk fat synthesis to explore the mechanisms driving milk fat synthesis and its regulation in ruminants. Addi- tion of lipids to the ruminant diet offers opportunities to increase the partitioning of energy towards body tissues during early lactation with the intention of reducing the extent and duration of negative energy balance in high producing dairy cows (Qin et al., 2018), as well as a model for studying the nutritional regulation of mammary lipogene- sis, particularly lipogenic gene expression (Bernard et al., 2018). Bovine milk fat. Bovine milk contains approximately 3-5 % fat depending on e.g. breed, state of lactation, management, and feeding strategies of dairy cows. Milk fat is secreted from mammary epithelial cells as milk fat droplets (milk fat globules), that con- tain mainly triacylglycerols (TAG; 96-98 %; Jensen, 2002), and smaller amounts of di- and monoacylglycerols (about 2 % of the lipid fraction), cholesterol esters (CE; less than 0.5 %), phospholipids (PL; including sphingolipids; about 1 %) and non-esterifield FA (NEFA; about 0.1 %) constitute the globule membrane (a protein rich polar lipid coat) surrounding the milk fat droplet (Jensen, 2002; Shingfield et al., 2010a; Bernard et al., 2018). Milk fat TAG are synthesised from more than 500 individual FA (Jensen, 2002; Lind- mark Månsson, 2008). Fatty acids are a wide group of compounds with different chain lengths, branching, degree of unsaturation (number of double bonds; Figure 1A), config- uration of double bonds (cis-trans; Figure 1A) and other functional and structural groups (Rodriguez-Estrada et al., 2014). In most cases, double bonds next to each other in FA are separated by a single methylene group (methylene interrupted), but some naturally occurring FA have conjugated double bond system, that is, two double bonds separated by one single bond (Figure 1B). Most natural FA have double bonds in the cis configura- tion. Trans double bonds arise from ruminal biohydrogenation or industrial processes (hydrogenation and refining processes; Shingfield et al., 2008b). A summary of FA no- menclature and shorthand notations of the most common FA used in scientific publica- tions of this thesis is presented as a separate table in the beginning of this work. The predominant SFA in milk are 14:0, 16:0 and stearic acid (18:0). These SFA ac- count for 75 % of the total FA, with a further 21 % occurring as MUFA of which the most prevalent is oleic acid (cis-9 18:1). Only 4-5 % of total FA in milk fat are PUFA, occurring mainly as linoleic acid (18:2n-6; ca. 1-3 %) and 18:3n-3 (0.5-2 %; Jensen, 2002; Lindmark Månsson, 2008). Approximately 3 % of the FA in milk are trans FA, especially ∆4-16 trans 18:1 isomers (Jensen, 2002), of which trans-11 18:1 is the most abundant (Precht and Molkentin, 1999; Shingfield et al., 2008b). Although it is still uncertain if dairy trans FA Natural resources and bioeconomy studies 74/2020 20 are as harmful as trans FA from plant-derived industrial fats and oils (Kleber et al., 2016), there is some evidence that the distribution of trans 18:1 and trans 18:2 isomers differs between ruminal and industrial trans FA (e.g. Shingfield et al., 2008b; Lock et al., 2005). Milk fat contains also a number of bioactive FA, including butyric acid (4:0), odd- and branched-chain FA (OBCFA), and cis-9,trans-11 CLA, which have been suggested to possess anti-inflammatory, anti-obesity, anti-carcinogenic, anti-mutagenic, anti- adipogenic and anti-diabetogenic properties, and have been reported to improve differ- ent biomarkers of cardiovascular health in animal models and in vitro studies with hu- man cell lines (Huth et al., 2006; Shingfield et al., 2008b; Gebauer et al., 2011). Typically, milk fat contains 2-5 % of 4:0, 2-3 % of OBCFA, and 0.3-0.6 % of cis-9,trans-11 CLA, but concentrations are subject to considerable variation. “Conjugated linoleic acid’’ is a collective term to describe a mixture of geometric and positional isomers of 18:2 containing a conjugated double bond (Figure 1B). Dairy products are the main source of CLA in the human diet, with the cis-9,trans-11 isomer accounting for 70–80 % of total CLA intake (Lawson et al., 2001), as cis-9,trans-11 is the major isomer of CLA in bovine milk (ca. 65.6–88.9 % of total CLA; Shingfield et al., 2008b). The most abundant OBCFA in ruminant milk fat are usually isomers of tridecano- ic acid (iso 13:0), tetradecanoic acid (iso 14:0), pentadecanoic acid (15:0, iso 15:0 and anteiso 15:0), hexadecanoic acid (iso 16:0) and heptadecanoic acid (17:0, iso 17:0 and anteiso 17:0) (Bauman et al., 2016; Vlaeminck et al., 2006). These originate mainly from microbial OBCFA synthesis in the rumen but also from de novo synthesis (15:0 and 17:0) in ruminant tissues, including the mammary gland (Vlaeminck et al., 2006). Milk fat is poor source of long-chain n-3 PUFA, and naturally almost devoid of n-3 very long-chain PUFA, specifically 20:5n-3 and 22:6n-3 (on average, 0.06 and 0.03 % of total FA, respectively; IV). Marine lipids, FO or marine algae in particular, have been commonly used as sources of n-3 long-chain PUFA in dairy cow rations to increase the milk fat content of these FA. The concentration of ≥ 18-carbon FA, including 18:2n-6, 18:3n-3, 20:5n-3, and 22:6n-3, in milk is dependent on the absorption of these FA from the small intestine and uptake by the mammary gland. Ruminal lipid metabolism. After ingestion, dietary plant lipids, such TAG, PL and glycolipids (GL), are released from their structural components through mastication and microbial digestive processes within the rumen, and FA are released from these lipids by the action of microbial lipases (called as lipolysis) (Harfoot and Hazlewood, 1988; Jenkins et al., 2008). Bacterial breakdown of dietary lipids is quite rapid and rather complete for most unprotected lipids (85-95 %) as observed earlier in the rumen of lactating cows fed grass hay (Bauchart et al., 1990), fresh grass (Halmemies-Beauchet-Filleau et al., 2013a) or grass silage-based diets (Halmemies-Beauchet-Filleau et al., 2013a). A number of factors related to diet and rumen environment, such as high level of PUFA intake or low rumen pH, that are known to inhibit the activity and growth of certain bacteria in the rumen, have been determined to influence the rate and extent of lipolysis in the rumen (Palmquist et al., 2005; Lourenço et al., 2010). Natural resources and bioeconomy studies 74/2020 21 A B cis-9,trans-11 CLA and trans-10,cis-12 CLA a conjugated double bond system - a pair of double bonds separated by one single bond Figure 1. Structural differences between cis and trans fatty acids in A) Dhaka, V., Gulia, N., Ahlawat, K.S. and Khatkar, B.S., 2011 J. Food Sci. Technol. 48:5:534-541, Copyright (2011) Springer Nature, and a conjugated double bond system in B) AOAC Lipid Library, 2019 cited on 27 December 2019. After lipolysis, the next step in the transformation of unsaturated FA in the rumen is biohydrogenation. During biohydrogenation, rumen microbes convert unsaturated FA to SFA via a stepwise reduction process during which cis-double bonds of unsaturated FA are first isomerised to trans FA intermediates (called as a cis-trans isomerisation) fol- lowed by hydrogenation of the double bonds in the carbon chain to yield saturated end products. Numerous, previously reviewed (Palmquist et al., 2005; Jenkins et al., 2008; Shingfield et al., 2010a; 2013), in vivo and in vitro studies have elucidated in detail the major biohydrogenation pathways of plant-derived 18:2n-6 and 18:3n-3 (Figure 2; Shing- field and Griinari, 2007). Under normal rumen conditions biohydrogenation of 18:2n-6 and 18:3n-3 is considered to proceed via isomerisation of the cis-12 double bond result- ing in the formation of conjugated 18:2 or 18:3, respectively. Conjugated FA intermedi- ates, such as cis-9,trans-11 CLA and cis-9,trans-11,cis-15 18:3, respectively, are tempo- rary metabolites and are further reduced to 18:0 as the final end product with trans-11 18:1 (vaccenic acid) as a common intermediate. The final reduction step is considered to Natural resources and bioeconomy studies 74/2020 22 be rate limiting and therefore trans 18:1 intermediates can accumulate (Harfoot and Hazlewood, 1988; Griinari and Bauman, 1999; Bauman and Griinari, 2001). Besides general biohydrogenation pathways of 18:2n-6 and 18:3n-3 described in Figure 2, numerous minor biohydrogenation pathways exist in the rumen that are de- pendent on the ruminal microbial ecosystem, resulting in the formation of a wide range of biohydrogenation FA intermediates, such as several 18:3, 18:2, CLA and trans 18:1 isomers (Shingfield et al., 2010a). Although most of these ruminal FA intermediates are further hydrogenated to 18:0, some of them escape from the rumen depending on feed- ing management, and are absorbed in the small intestine, and yet further incorporated into ruminant milk and tissue lipids. Dietary lipid supplementation in the diet of ruminants. Dietary supplements of plant oils rich in 18:2n-6 (e.g. sunflower oil; SO) (Shingfield et al., 2008a) and 18:3n-3 (e.g. linseed oil; LO) (Loor et al., 2004) or FO (Shingfield et al., 2003) are known to in- crease trans-11 18:1 formation in the rumen of lactating cows. Feeding plant-derived 18:2n-6 or 18:3n-3 together with FO further increases cis-9,trans-11 CLA concentrations in milk from lactating cows (Whitlock et al., 2002; Palmquist and Griinari, 2006; Shing- field et al., 2006a), but the enrichment varies depending on the composition of the basal diet and the amount and source of lipid supplements. The increases in cis-9,trans-11 CLA in milk in FO diets are accompanied by elevated proportions of trans 18:1 and trans 18:2 isomers (Chilliard et al., 2001; Shingfield et al., 2003; Loor et al., 2005a), indicating that ruminal biohydrogenation of 18-carbon PUFA to saturated end product, 18:0, is incom- plete (Shingfield et al., 2003; Lee et al., 2008; Shingfield et al., 2010b). Furthermore, dietary supplements of oils containing 18:2n-6 or 18:3n-3, but not FO, lower the propor- tions of 12:0, 14:0, and 16:0 in milk fat (Shingfield et al., 2013). Oil supplements can also be used to increase the ruminal outflow of long-chain n-3 FA intermediates available for absorption and incorporation into milk and meat. Several studies have shown that FO can be used to increase the 20:5n-3 and 22:6n-3 concentra- tions in ruminant milk and tissue lipids (Scollan et al., 2006; Chilliard et al., 2007; Palmquist, 2009). However, the level of enrichment of n-3 PUFA on milk and meat is limited. A number of detailed physiological studies in ruminants have indicated that 20:5n-3 and 22:6n-3 are hydrogenated extensively in the rumen (Shingfield et al., 2003; 2010; Lee et al., 2008) and disappear during incubations with rumen fluid in vitro with the extent of hydrogenation being dependent on the amount of FO supplementation (Dohme et al., 2003; AbuGhazaleh and Jenkins, 2004; Wasowska et al., 2006), but the metabolic pathways involved and the intermediates formed are not well known. Only few studies, including the experiments reported in this thesis (II; III), have indicated that the biohydrogenation of long-chain PUFA in FO or marine algae result in the formation and accumulation of numerous 20- and 22-carbon biohydrogenation intermediates con- taining at least one trans double bond in cattle (Shingfield et al., 2003; 2010; Lee et al., 2008) and in nonlactating sheep (Toral et al., 2010; 2012), but the effects of 18-carbon PUFA supply on ruminal long-chain FA metabolism and microbial communities in rumi- nants fed FO are not well established. Natural resources and bioeconomy studies 74/2020 23 Role of microbial population in ruminal biohydrogenation. The main members of the microbial population in the rumen are comprised of bacteria, bacteriophages, proto- zoa, fungi and archaea that live in a symbiotic relationship and are collectively responsi- ble for microbial fermentation in ruminants. The microbial ecology related to ruminal biohydrogenation has attracted considerable interest as PUFA are considered detri- mental to rumen microbes (Maia et al., 2010; Huws et al, 2015; Enjalbert et al., 2017). Bacteria, rather than protozoa, are thought to be responsible for ruminal biohydrogena- tion, but relatively few strains capable of biohydrogenation have been identified (Har- foot and Hazlewood, 1988; Lourenço et al., 2010). The identification of specific microbial species able to catalyse one or more reactions remains challenging, as biohydrogenation intermediates and products may also be exchanged between populations. Among cultivated bacteria, all members of the Butyrivibrio group form cis-9,trans- 11 CLA from 18:2n-6 much more rapidly than do other species, but only B. proteoclasti- cus is shown to be capable of reducing trans-11 18:1 to 18:0 (Jeyanathan et al., 2016). Previous reports highlight that the effects of FO on ruminal biohydrogenation in growing cattle are associated with changes in the ruminal bacterial community known to be ca- pable of biohydrogenation (Kim et al., 2008; Huws et al., 2010; 2011). In addition, popu- lations of specific bacteria in the rumen were found to be altered by marine algae sup- plements in lactating sheep fed diets containing SO (Toral et al., 2012), but there are no reports on the effects of FO with plant oils on the relative abundance of key biohydro- genating bacteria in cattle. Ruminal biohydrogenation of PUFA is a complex process comprising of isomerisa- tion, desaturation and hydrolysation reactions and different biohydrogenation path- ways. A different set of FA intermediates may be produced under specific feeding strat- egies, such as diets causing milk fat depression (MFD), which may shift the ruminal bio- hydrogenation towards a higher production of trans-10 18:1 at the expense of trans-11 18:1 (Figure 2; Shingfield and Griinari, 2007). A shift toward trans-10 biohydrogenation pathway in the rumen is more pronounced in cows fed diets containing relatively high concentrations of plant-derived PUFA (Griinari et al., 1998; Piperova et al., 2000; Loor et al., 2005b), high-starch, low-fibre diets (Piperova et al., 2002; Zened et al., 2013), high- concentrate diets containing plant oils (Loor et al., 2004; Ventto et al., 2017), or diets containing high marine lipids (e.g. II; Toral et al., 2016a). Milk fat synthesis in the mammary gland. Under normal dietary and physiological conditions, ca. 40 % of total FA in milk fat originates from de novo synthesis in the mammary gland, while the rest originate from the direct mammary uptake of circulating plasma FA (Chilliard et al., 2000). Substrates for mammary de novo synthesis are mainly acetate (2:0) and ß-hydroxybutyrate (BHBA) derived from fermentation of feed compo- nents in the rumen (Lock and Bauman, 2004). Substrates are used by the mammary epi- thelial cells in the presence of two key enzymes, acetyl-CoA carboxylase (ACACA) and FA synthase (FASN) (Shingfield et al., 2010a) to synthesize short- and medium-chain FA, providing all 4:0 to 12:0, most of the 14:0 (ca. 95 %), and about half (ca. 50 %) of 16:0 secreted in milk (Chilliard et al., 2000). The direct uptake of preformed FA provides re- Natural resources and bioeconomy studies 74/2020 24 maining of the 16- and all of the ≥ 18-carbon long-chain FA in milk fat (ca. 60 %; Chilliard et al., 2000). Fatty acids of 10- to 20-carbon atoms may be desaturated in the mammary epitheli- al cells by Δ9-desaturase complex (stearoyl-CoA desaturase; SCD) that is responsible for introducing a cis double bond at position Δ9 of FA. Special attention has been directed to SCD, which is the main enzyme responsible for converting SFA into MUFA (18:0 to cis- 9 18:1). The SCD activity plays an important role in ruminants because it reverses rumi- nal biohydrogenation (i.e. saturation) of dietary MUFA and PUFA in the rumen. In rumi- nants, 60 % of cis-9 18:1 (derived from 18:0), 50-56 % of cis-9 16:1 (derived from 16:0) and 90 % of cis-9 14:1 (derived from 14:0), and > 60 % of cis-9,trans-11 CLA (derived from trans-11 18:1) originate from mammary SCD activity (Bernard et al., 2018). Most of the CLA isomers appearing in milk fat are derived directly from the biohy- drogenation of diet 18:2n-6 and 18:3n-3 in the rumen (Baumgard et al., 2000; Bauman and Griinari, 2003). However, 64-97 % of cis-9,trans-11 CLA and also a majority of trans- 7,cis-9 CLA secreted in milk originates from Δ9-desaturation of trans-11 18:1 (Griinari et al., 2000; Corl et al., 2001; Mosley et al., 2006) and trans-7 18:1 (Corl et al, 2002; Piper- ova et al., 2002), respectively, in the mammary gland (Figure 2; Griinari and Bauman, 1999). Milk fat depression in lactating cows is characterised by decreases in milk fat within a few days without changes in milk yield or other milk components (e.g. Bauman and Griinari, 2003; Shingfield and Griinari, 2007). Several theories have been proposed to explain diet-induced MFD, including 1) increased formation of biohydrogenation inter- mediates that inhibit, directly or indirectly, milk fat synthesis, 2) lowered availability of 18:0 for endogenous cis-9 18:1 synthesis via SCD in the mammary gland in marine lipid- induced MFD, and 3) an increase in the supply of trans FA formed in the rumen contrib- uting to lower milk fat synthesis by increasing the milk fat melting point above body temperature, exceeding the capacity to maintain milk fat fluidity and thereby lower the rate of fat removal in mammary epithelial cells (Loor et al., 2005a; Shingfield and Gri- inari, 2007; Gama et al., 2008; Harvatine et al., 2009). Of these, the most widely accept- ed hypothesis is the biohydrogenation theory of MFD (Bauman and Griinari, 2001) which suggests that the changes in ruminal lipid metabolism increase the formation of specific FA biohydrogenation intermediates from dietary PUFA that directly inhibit milk fat syn- thesis (Bauman and Griinari, 2001; 2003). These specific FA intermediates associated with MFD often derive from the alterna- tive trans-10 biohydrogenation pathway and contain a double bond in the trans configu- ration at the 10th carbon atom from the carboxyl end. Because these biohydrogenation intermediates, e.g. trans-10 18:1, trans-10,cis-12 CLA and trans-10,cis-15 18:2 are avail- able for absorption and incorporation into milk fat, they are suggested to cause reduc- tions in milk fat content resulting in MFD (Bauman and Griinari, 2003; Shingfield and Griinari, 2007). Trans-10,cis-12 CLA formed during the isomerisation of 18:2n-6 in the rumen (Wal- lace et al., 2007) is the only intermediate shown unequivocally to lower milk fat synthe- Natural resources and bioeconomy studies 74/2020 25 sis in lactating cows (Baumgard et al., 2000; 2002; Glasser et al., 2010; Harvatine and Bauman, 2011), but it does not, in isolation, explain MFD in cows fed diets containing FO (Loor et al., 2005a; Gama et al., 2008; Toral et al., 2015). Earlier findings involving postruminal infusions of FA, such as a mixture of CLA isomers, suggest that other inter- mediates, including cis-10,trans-12 CLA (Saebø et al., 2005), trans-9,cis-11 CLA (Perfield et al., 2007), and possibly trans-10 18:1 (Shingfield et al., 2010a; Harvatine et al., 2009; Conte et al., 2018) may also inhibit milk fat synthesis in the lactating cow. However, it is probable that other, yet unidentified, biohydrogenation intermediates or metabolites or other additional mechanisms contribute to the regulation of mammary lipogenesis and MFD in cows fed diets containing marine lipids (Loor et al., 2005a; Gama et al., 2008; Shingfield and Griinari, 2007). Even though dietary FO supplements cause MFD, direct measurements of biohydro- genation intermediates formed in the rumen of lactating cows under these circumstanc- es are limited. Characterizing the changes in ruminal biohydrogenation and microbial ecology is central in understanding the mechanisms underpinning physiological re- sponses to lipid supplements containing long-chain PUFA in lactating ruminants and the impact of different nutritional strategies on composition of milk fat, which is a major source of SFA in the human diet (Eilander et al., 2015). Dairy products are a significant source of fat in the diet in most developed countries highlighting the need to under- stand how to produce milk with nutritionally more beneficial FA composition. Natural resources and bioeconomy studies 74/2020 Figure 2. Major pathways of 18:2n-6 and 18:3n-3 metabolism in the rumen and changes (dashed arrows) in biohydrogenation pathway of 18:2n-6; so called trans-10 shift that occur on diets causing milk fat depression (modified from Shingfield and Griinari, 2007 Eur. J. Lipid Sci. Technol. 109:799-816). Arrows pointing towards mammary gland indicate the important role of ∆9-desaturase (SCD) in the appearance of cis-9,trans-11 CLA in ruminant milk fat (modified from Griinari and Bauman, 1999 in Advances in Conjugated Linoleic Acid Research Vol. 1, pp. 180-200; Bauman and Griinari, 2001 Livest. Prod. Sci. 70:15-29). 26 Natural resources and bioeconomy studies 74/2020 27 2. Aims of the thesis The general aim of this thesis was to provide new information on the metabolism of 18- carbon MUFA and PUFA and long-chain 20- to 22-carbon n-3 PUFA, such as 20:5n-3 and 22:6n-3, in the rumen of lactating cows by studying the transformations occurring through supplementation of dairy cows’ diet with FO alone or with plant oils rich in 18:2n-6 or 18:3n-6. The research findings reported in this work also provided further insight into the possible causes for FO-induced MFD and varying effects of FO on milk FA composition in lactating cows by describing the mechanisms involved in the ruminal biohydrogenation, microbial ecology and mammary lipid metabolism. In addition, this research helped to explain the appearance of FA biohydrogenation products in milk (and meat) with putative biological activity relevant to the prevention of chronic diseases in humans and novel long-chain FA intermediates whose bioactivity is still unknown. For individual experiments, the objective was to build up methods for the analysis of long-chain 20- to 22-carbon FA intermediates formed during ruminal biohydrogena- tion of n-3 PUFA by combining different analytical techniques. These procedures were applied in the detailed analysis of different feed and oil supplements, omasal digesta and milk fat. The experiments were designed to investigate the effects of dietary FO supplements alone (I; II) or in combination with plant oils rich in 18:2n-6 or 18:3n-3 (III) on the ruminal and mammary lipid metabolism in order to understand the mechanisms and metabolic pathways underlying the diet-induced changes in milk fat depression (MFD), milk FA composition and specific FA intermediates and end products associated with MFD. To achieve this objective, experiments documented in I-IV encompassed de- tailed investigations of animal performance (II; III), ruminal lipid metabolism (I-III) and bacterial populations (II; III), the relationships between the flow of FA at the omasum and secretion of milk FA (IV) and changes in milk fat synthesis (IV) to provide further insight into the possible causes of FO-induced MFD, and evaluated the potential of die- tary FO supplements to enrich long-chain n-3 FA in milk (IV). All the experimental treat- ments were formulated to meet the objectives of providing new information on changes in ruminal biohydrogenation of long-chain polyenoic n-3 FA and milk fat synthesis, met- abolic pathways involved and specific FA intermediates and end products associated with MFD. Natural resources and bioeconomy studies 74/2020 28 The main hypotheses tested in this research were: i) Dietary FO supplements alone or in combination with plant oils inhibit the bio- hydrogenation of unsaturated FA in the rumen and alter the flow of specific 16- to 22-carbon FA intermediates, 20:5n-3, 22:5n-3 and 22:6n-3 at the omasum. ii) Biohydrogenation of long-chain n-3 PUFA in FO may proceed via different reduc- tion mechanism than the well-known biohydrogenation pathways of 18:2n-6 and 18:3n-3. iii) Dietary plant-derived 18-carbon PUFA lower the ruminal biohydrogenation of PUFA, including 20:5n-3, 22:5n-3 and 22:6n-3, and increase the ruminal outflow of trans-11 18:1. iv) The effect of FO supplements on microbial biohydrogenation may differ depend- ing on whether 18:2n-6 or 18:3n-3 is the main source of 18-carbon PUFA. v) Dietary FO supplements alone or in combination with plant oils modify rumen microbial communities and induce changes in the abundance of key rumen bac- terial populations known to be capable of biohydrogenation. vi) Dietary FO supplements increase the ruminal outflow of biohydrogenation in- termediates that have not been previously identified, and these intermediates are transferred into milk fat. FO supplements also enrich CLA, 20:5n-3, 22:5n-3 and 22:6n-3 concentrations in milk fat. vii) The effect of dietary FO on the flow of trans FA intermediates at the omasum, observed changes in milk FA composition and calculated milk fat melting point are related to FO-induced MFD. Natural resources and bioeconomy studies 74/2020 29 3. Materials and methods 3.1. Experimental animals and designs The experiments documented in I-IV were conducted as three separate experiments (Table 1). All the experimental procedures used are described in detail in I-IV, and only a brief outline is presented herein. Experiments were performed with multiparous Finnish Ayrshire dairy cows in late or mid-lactation fitted with rumen cannulae, except for exp. 2, in which 2 Finnish Ayshire and 2 Holstein-Friesian were used. The cows averaged 197 days in milk (SD 9.40), 589 kg live weight (SD 16), and 27 kg milk yield (SD 1.14) at the beginning of the experiments. Experiments were conducted as a balanced 4 × 4 Latin Square; except for exp. 1 that was carried out according to a design where all cows were fed grass silage-based diet with no additional lipid during the first experimental period followed by the same basal diet supplemented with 250 g/d FO during the second peri- od. The experimental design with 14-d periods used in exp. 1 may be criticized due to the confounding effects of treatment with time. However, time-related effects associat- ed with short experimental periods were considered rather minor defect compared with the importance of the detailed FA composition analyses of FO and omasal digesta com- bined with measurements of nutrient flow at the omasum that provided the first quanti- tative estimates on the ruminal outflow of long-chain n-3 PUFA biohydrogenation inter- mediates in lactating cows. For exp. 2 and 3, experimental periods lasted for 28 and 21 days, respectively. 3.2. Experimental treatments All experimental treatments comprised restrictively fermented grass silage and a cereal- based concentrate (forage:concentrate ratio 60:40, on a dry matter (DM) basis) offered at 95 % of ad libitum intake measured during 14 d before the start of the experiment. Forages were prepared from primary growths of mixed timothy (Phleum pratense) and meadow fescue (Festuca pratensis) in experiments reported in I and III, and from the primary growth of tall fescue (Festuca arundinacea) in II and IV. The ultra-refined herring and mackerel oil (FO; EPAX 3000 TG, Pronova Biocare AS, Norway) was used in all exper- iments (I-IV) as a dietary supply of very long-chain n-3 PUFA. In the experiment docu- mented in III, SO (Raisioagro Ltd., Finland) and LO (Elixi Oil Ltd., Finland) were used as sources of 18:2n-6 and 18:3n-3, respectively. All oil supplements were fed in equal amounts twice a day by mixing thoroughly with concentrate ingredients just before feeding. Experiment reported in I was conducted to characterize the intermediates formed during ruminal biohydrogenation of long-chain PUFA in FO. The structure and composi- tion of 20- to 24-carbon FA in FO and the omasal digesta was determined and ruminal outflow quantified in cows fed a basal diet containing no additional oil (control) or sup- plemented with 250 g/day of FO (treatment FO). Experiment documented in II and IV Natural resources and bioeconomy studies 74/2020 30 involved a physiological study in which the effects of incremental dietary FO supplemen- tation (0, 75, 150, or 300 g FO/d) on the flow of FA at the omasum (II), changes in the key bacterial species known to be capable of biohydrogenation (II), milk production (IV), and milk FA composition (IV) were investigated. Experiment reported in III investigated the effects of diet containing no additional oil (control), FO alone (200 g FO/day; treat- ment FO) or FO in combination with SO (rich in 18:2-6; 200 g FO and 500 g of SO/day; treatment SFO) or LO (rich in 18:3n-3; 200 g of FO and 500 g of LO/day; treatment LFO) on the flow of FA at the omasum and bacterial populations in lactating cows (III). 3.3. Experimental measurements Feed intake was determined daily as the difference between the amount of feeds of- fered and the amount of refused feeds. For rumen fermentation measurements rumen fluid was collected at regular intervals through rumen cannula using a vacuum pump and a flexible tube (II; III). In addition, subsamples of filtered ruminal fluid were collected for visual assessment of protozoal numbers (II; III). Rumen bacterial populations known to be capable of biohydrogenation were determined from omasal digesta by quantita- tive polymerase chain reaction (qPCR) analysis (II; III). In I-III, the omasal sampling tehnique in combination with a triple marker system were used to assess nutrient flow entering the omasal canal as previously described by Ahvenjärvi et al. (2000). Diet di- gestibility was measured by total faecal collection (II; III). Daily milk yields of all experi- mental cows were recorded throughout each experiment. Samples for the analysis of milk composition in IV were collected and composited according to yield over two con- secutive milkings on d 17, 20, 24, and 27 of each experimental period. Ruminal admin- istration of LiCo-EDTA as indigestible marker to estimate flow of FA at the omasum in the experiment reported in II were found to alter milk fat composition (Shingfield et al., 2006b); therefore, only samples of milk collected on d 17, immediately before marker administration, were submitted for detailed FA analysis (IV). A triple marker system based on Cr-EDTA, Yb-acetate, and indigestible neutral detergent fibre (iNDF) as markers for liquid, small, and large particles of omasal digesta, respectively, were used in exper- iment reported in III. For all experiments, cows were housed in individual tie-stalls with- in a dedicated metabolism unit with continuous access to water and milked twice daily. Fatty acid content of grass silage, concentrates, and omasal digesta was determined using internal and external standards (I-IV). Tritridecanoin (T-135; Nu-Chek-Prep Inc., Elysian, MN) was used as an internal standard for feeds (I-IV) and milk (IV) and tride- canoic acid (13:0, N-13A, Nu-Check-Prep Inc.) for omasal digesta (I-III). Trihexadecanoin (T-5888, Sigma-Aldrich, St. Louis, MO) was used to set up an external calibration curve for feed ingredients (I-III) and milk (IV) and hexadecanoic acid (S-4751, Sigma-Aldrich) for omasal digesta (I-III). Fatty acid methyl esters (FAME) of lipids in oil supplements and freeze-dried samples of silage and concentare were prepared in a one-step extraction- methylation procedure (Shingfield et al., 2003). The freeze-dried samples of reconstitut- ed omasal digesta were adjusted to 2.0 by hydrochloric acid and extracted in duplicate Natural resources and bioeconomy studies 74/2020 31 with a mixture (3:2; vol:vol) of hexane and isopropanol. Fatty acid methyl esters were prepared using a two-step base-acid catalysed procedure as described in I. Lipids in milk samples were extracted in duplicate using a mixture of ammonia, ethanol, diethyl ether, and hexane (0.2:2.5:2.5, vol/vol) and converted to FAME using methanolic sodium methoxide as a catalyst (Shingfield et al., 2003). Fatty acid methyl esters were quantified using a gas chromatography (GC) and a high-performance liquid-chromatography (HPLC) (I-IV). Fatty acid methyl esters were quantified using a GC (6890; Agilent Technologies, Wilmington, DE) fitted with a 100-m fused silica capillary column (CP-Sil 88; Chrompack 7489, Chrompack International BV, Middelburg, the Netherlands) using a temperature gradient program and hydrogen as the carrier gas (Shingfield et al., 2003). Individual isomers of 18:1 and 18:2 were further resolved in a separate analysis under isothermal conditions at 170°C (Shingfield et al., 2003). Under these conditions, trans-10,cis-15 18:2 and trans-11,cis-15 18:2 eluted as a single peak (I; II; IV). To resolve these isomers in experiment reported in III, analysis of FAME was repeated using a ionic liquid column (SLB-IL111; 100 m × 0.25 mm i.d., 0.2 μm film thickness, Sigma-Aldrich) and helium as a carrier gas (Alves and Bessa, 2014; Ventto et al., 2017). Fatty acids were initially identified based on retention time comparisons with au- thentic FAME standards: Nu-Chek-Prep GLC #463; 21:0, #N-21-M; 23:0, #N-23-M; 24:0, #N-24-M; trans-11 20:1, #U-64-M; cis-12-21:1, #U-85-M; cis-14 23:1, #U-87-M, Larodan Fine Chemicals 18:4n-3, #10-1840; cis-9 20:1, #20-2001-1-4; 20:4n-3, #20-2024-1; 21:3n- 3, #20- 2103-1-4; 21:5n-3, #20-2105-1-4; trans-13 22:1, #20-2210-9; 22:5n-6, #20-2265- 7; 23:5n-3, #20-2305-1-4; 24:5n-3, #20-2405-4; 25:0, #20-2500-7; 29:0, #20-2900-7; 9-O- 18:0, #14-1800-5-4; 12-O-18:0, #14-1800-6-4; 13-O-18:0, #14-1800-7-4; 16-O-18:0, #14- 1800-8-39 (Malmö, Sweden), and Sigma-Aldrich 26:0, #H-6389; 27:0, #H-6639; 28:0, #O- 4129; 30:0 #T-1902; mixture of 18:2n-6 isomers, #L-8404; mixture of conjugated 18:2 isomers, #O-5632; mixture of 18:3n-3 isomers, #L-6031. Fatty acid methyl esters not available as commercial standards were identified based on the fractionation of FAME by complementary silver-ion (Ag+) thin-layer chromatography (TLC) and GC-mass spec- trometry (MS) analysis of FAME and corresponding 4,4-dimethyloxazoline (DMOX) de- rivatives (I-IV). The distribution of CLA isomers was determined by Ag+-HPLC (I-IV). Anal- ysis was repeated using 2.0 % (vol:vol) acetic acid in heptane to resolve cis-10,trans-12 CLA and trans-10,cis-12 CLA (Ventto et al., 2017). 3.4. Statistical analyses The statistical methods for each experiment have been described in detail in I-IV, only a brief outline is presented herein. Differences in FA flow at the omasum and ruminal FA balance between the control and FO treatments in I were evaluated statistically by a paired t test (SAS Institute Inc., Cary, NC, USA). A 95 % confidence interval was used as the default for the hypothesis test and assumptions of data normality were validated using the univariate procedure of SAS. Natural resources and bioeconomy studies 74/2020 32 For II-IV, data were analysed by ANOVA with a statistical model that included the fixed effects of period and treatment and the random effect of cow using the mixed procedure of SAS. Measurements of rumen pH and fermentation characteristics were analysed by ANOVA for repeated measures with a model that included the fixed effect of period, treatment, time and treatment × time interactions and the random effect of cow assuming an Auto Regressive Order One Covariance Structure. Denominator de- grees of freedom were calculated using the Satterthwaite option and the Kenward- Rogers method. For II and IV, sums of squares were further separated into polynomial contrasts to test for the significance of linear and quadratic responses to dietary FO, and for III into single degree of freedom orthogonal contrasts to test the effects due to oil supplementation (control vs. FO, SFO, and LFO), addition of 18-carbon PUFA-rich plant oils to the FO (FO vs. SFO and LFO), and different sources of 18-carbon PUFA (SFO vs. LFO). Associations of FA flow at the omasum with milk fat yield and the output of FA in milk for individual cows were evaluated by linear regression analysis using REG proce- dure of SAS in IV. In addition, relationships of milk fat content and milk fat yield with concentrations of all identified FA in milk were further analysed by partial least squares regression (PLS) using the PLS procedure of SAS, with milk fat yield and milk fat content as response variables and the proportion of individual FA in milk as an explanatory vari- able to produce correlation loading plots (IV). The number of factors was set to 2 (sum of FA in milk based on carbon chain length and degree of unsaturation, 18-carbon FA, and isomers of 20- to 22-carbon FA) or all 3 factors (for all identified FA) using the leave- one-out cross validation procedure, based on minimizing the predicted residual sum of squares. Natural resources and bioeconomy studies 74/2020 3.5. Summary of experimental designs Table 1 Description of experiments reported in this thesis Publ Exp. Animals Design Dietary ingredients Treatments Measurements and objectives I 1 5 rumen fistu- lated dairy cows 2 subsequent periods, 2 diets Grass silage (GS) Standard concentrate (C) No additional lipid (Control) 250 g/d additional fish oil (FO) GS:C ratio = 60:40 Control FO Ruminal metabolism of long-chain fatty acid (FA) in FO Identification of novel long-chain FA intermediates on the digesta of cows fed FO II, IV 2 4 rumen fistu- lated dairy cows 4 × 4 Latin Square Grass silage (GS) Standard concentrate (C) No additional lipid (Control) 75 g/d additional FO (FO75) 150 g/d additional FO (FO150) 300 g/d additional FO (FO300) GS:C ratio = 60:40 Control FO75 FO150 FO300 Effect of incremental amounts of FO in the GS based diet on o DM intake o Ruminal lipid metabolism o Rumen fermentation characteristics o Rumen microbial communities o Milk production o Milk FA composition o Milk fat depression o Transfer efficiency of FA from feed to milk III 3 4 rumen fistu- lated dairy cows 4 × 4 Latin Square Grass silage (GS) Standard concentrate (C) No additional lipid (Control) 200 g/d additional FO 200 g/d FO plus 500 g/d sunflower oil (SFO) 200 g/d FO plus linseed oil (LFO) GS:C ratio = 60:40 Control FO SFO LFO Effect of FO supplementation alone or in combination with plant oils rich in 18:2n-6 and 18:3n-3 in the GS based diet on o DM intake o Ruminal lipid metabolism o Rumen fermentation characteristics o Rumen microbial communities 33 Natural resources and bioeconomy studies 74/2020 34 4. Results and discussion 4.1. Analysis of long-chain fatty acids by chromatography As a consequence of the complex nature of ruminant derived lipids and FA, the analyti- cal methods used are important when studying lipid metabolism of dairy cows. This chapter describes the analytical methods and challenges of this thesis work. Chromatography is one of the most effective analytical procedures for separating and analysing the properties of lipids. In chromatographic analysis molecules are sepa- rated based on different affinities while passing through a matrix. After being separated, the concentration of each of the molecules is determined as they pass by a suitable de- tector (e.g., UV-visible, fluorescence, or flame ionization) or the molecules are collected in different fractions for further analysis usually by other chromatographic methods. Various forms of chromatography are available to analyse the lipids in feeds and biologi- cal samples, such as tissues and body fluids, e.g. TLC (I), GC (I-IV), and HPLC (I-IV). Thin layer chromatography is used mainly to separate and determine the concen- tration of different types of lipid groups, e.g. free FA, tri-, di-, and monoacylglycerols, cholesteryl esters, and PL. A TLC plate is coated with a suitable absorbing material. A small amount of the lipid sample to be analysed is applied onto the TLC plate. The dif- ferent lipid fractions are separated on the basis of their affinity for the absorbing mate- rial. It is possible to identify the lipid fractions in the sample by using reference com- pounds. Fractions are scraped off from the plate and analysed further using e.g. GC- flame ionization detection (FID) or GC-MS. For details see more information from the LipidWeb, 2019. It is difficult to analyse intact lipid molecules and free medium- and long-chain FA using GC because they are not very volatile. For this reason, the free FA and esterified FA for example in TAG, PL and GL are usually derivatized prior to analysis to increase their volatility. The most commonly used FA derivatives in GC analysis are FAME. As the FAME pass through the GC capillary column, they are separated based on differences in their molecular weights and polarities, and the FAME are quantified using FID or MS as detec- tors. The identification of biohydrogenation intermediates of FA by GC is challenging due to the lack of reference compounds and small concentration of some intermediates. The chromatographic resolution may also be insufficient due to the extensive number of FA compounds present in the samples. Especially FAME that differ in double bond position and double bond configuration may co-elute in GC analysis. Long capillary columns with highly polar cyanopropyl siloxane stationary phase, e.g. the 100m long CP-Sil 88 column (I-IV), can be used in the successful separation of most of the cis and trans 18:1, 18:2 and 18:3 isomers that are found in ruminant-derived samples. However, even with this column type different GC oven temperature programs are sometimes needed to achieve a sufficient separation. However, in recent years the SLB-IL111 column has been report- Natural resources and bioeconomy studies 74/2020 35 ed to provide improvements on the resolution of FAME that are co-eluting with highly polar cyanopropyl siloxane stationary phases (Alves and Bessa, 2014). Gas chromatography in combination with MS (I-IV) has been used most often for structure determination of FA as different derivatives, such as the use of nitrogen con- taining derivatives, e.g. picolinyl (Harvey, 1984; 1992; Christie et al., 1986; Christie, 1998) and DMOX (Zhang et al., 1988; Spitzer, 1997) derivatives. In this research project FA were quantified as FAME using GC-FID and reference FAME compounds were used in identification. Methyl ester and DMOX derivatives of FA were also analysed by GC-MS using electron ionization in order to characterize the FA structures (I-IV). The various ionic species produced from a given FA derivative by electron ionization are separated according to mass/charge (m/z) ratio in which z = 1 in a magnetic field, and a spectrum is obtained that shows the masses of the fragment ions and their abundances relative to the most abundant ion (base ion). Initial identification of FA carbon chain length and number of double bonds in ex- perimental samples were based on GC–MS analysis of FAME. For example molecular ions at m/z 324, 322, 320, 318, 316, 338, 334, 332, 330, 352, 350, 348, 346, 344, 342, 358, 372 and 370 confirmed the appearance of 20:1, 20:2, 20:3, 20:4, 20:5, 21:1, 21:3, 21:4, 21:5, 22:1, 22:2, 22:3, 22:4, 22:5, 22:6, 23:5, 24:5 and 24:6 methyl esters, respec- tively, but the mass spectrum did not contain sufficient characteristic ion fragments to locate the position of double bonds or other specific structural characteristics. There- fore, the structure of long-chain FA contained in FO, omasal digesta and milk fat from cows fed FO were formally identified based on GC–MS analysis of DMOX derivatives (I- IV). As an example, the mass spectrum of the DMOX derivative of trans-9,cis-14,cis-17 20:3 is shown in Figure 3. The McLafferty ion at m/z = 113 is the expected base peak, accompanied by a prominent ion at m/z = 126. Fragmentation of the carbon chain of the FA gives rise to a series of abundant ions that are 14 atomic mass units (amu) apart. The molecular ion (m/z = 359) was used to define the chain length and number of double bonds, and three ions separated by gaps of 12 amu at m/z 196 and 208; 264 and 276; 304 and 316 (Figure 3) were used to locate double bonds at positions ∆9, ∆14, and ∆17, respectively. From the obtained spectra it was possible to confirm that the peaks were FA with a certain molecular weight using the McLafferty ions and molecular ions, but in all cases it was not possible to deduce the detailed structure of the FA, e.g. locate the double bond positions, in which case the peak was considered as an unidentified FA. Characteristic ion fragments in the mass spectrum of DMOX derivatives used to locate the position of double bonds of 20-, 21- and 22-carbon unsaturated FA in omasal digesta are listed in detail in I. The signals from diagnostic ions usually become weaker and the number of interfering ions from other fragments increases with the increasing number of double bonds. Thus, interpretation of spectra from highly unsaturated FA is challeng- ing. Some of the biohydrogenation intermediates of highly unsaturated FA were present in minor amounts in the samples obtained in this research and thus it was often difficult to acquire spectra of sufficient quality. Natural resources and bioeconomy studies 74/2020 36 Figure 3. Gas chromatography-electron ionization mass spectrum of the 4,4-dimethyloxaline (DMOX) of trans-9,cis-14,cis-17-20:3 detected in omasal digesta of cows fed grass silage based diet containing fish oil (modified from I). By cross-referencing peaks in the total ion chromatogram of DMOX and FAME de- rivatives with the relative retention time and elution order of FAME in the GC-FID chro- matogram, minor FA in omasal digesta and milk could be identified. While it is generally accepted that DMOX derivatives exhibit chromatographic properties comparable to those of FAME (Spitzer, 1997; Christie, 1998) under the chromatographic conditions used in this thesis work we observed slight variations in the relative retention and elu- tion order of FAME and DMOX derivatives which introduced major challenges in the attempts to identify the structure of minor components in samples containing a highly complex mixture of FA. An extensive overview of selected regions of the total ion chro- matogram for DMOX derivatives prepared from total lipid in omasal digesta of cows fed the FO treatment is given in I. For example, during the analysis of DMOX derivatives 18:3n-3 eluted after cis-13 20:1 and 22:0 eluted before cis-9,trans-11,cis-15 18:3, cis- 14,cis-17 20:2 and trans-9,trans-14,trans-17 20:3, whereas the reverse was true during the analysis of methyl esters. Relative retention times of trans-11,cis-14,cis-17 20:3, 20:3n-3, 20:5n-3, cis (∆11-15) 22:1, trans-12,trans-17 22:2, ∆10,13,17-22:3, trans- 10,trans-13,cis-16,cis-19 22:4, 22:4n-3, trans-5,cis-10,cis-13,cis-16,cis-19 22:5, 23:0, 24:0, cis-16 25:1 and 26:0 were also found to differ when analysed as a methyl ester or DMOX derivative. The number and positions of double bonds can also be determined by pre- fractionations of the samples by different HPLC or TLC techniques. Development of Ag+- TLC plates with hexane and diethyl ether allowed 18-, 20, and 22-carbon cis-monoenoic Natural resources and bioeconomy studies 74/2020 37 isomers to be resolved from trans monoenoic methyl esters, but did not result in the complete separation of methyl esters of long-chain dienoic and trienoic FA (I). In addi- tion, a fraction containing FA with four to six double bonds remained at or near the origin. However, fractionation of FAME according to the degree of unsaturation and double bond geometry by Ag+-TLC allowed cis and trans monoenoic FA to be completely resolved, confirming the configuration of double bonds for 20:1 and 22:1 isomers. Despite the fractionation of FAME prior to the conversion of DMOX derivatives, it was not possible to elucidate the structure of all 20- and 22- carbon PUFA containing two of more double bonds due to a very low abundance and/or co-elution with other components. A mixture of hexane and diethyl ether was used to develop Ag+-TLC plates based on reports that have shown this solvent system to resolve trans 18:1 and cis 18:1 isomers in bovine milk fat (Precht and Molkentin, 1996; Cruz-Hernandez et al., 2004; 2006), and therefore applied in the analysis of omasal digesta due to the occurrence of cis and trans 20:1 and cis 22:1 isomers. Previous investigations have reported the sepa- ration of methyl esters of cis 20:1, cis 22:1, trans 20:1 and trans 22:1, and geometric isomers of 20:5n-3 and 22:6n-3 by Ag+-TLC using plates developed with a mixture (50:50, v/v) of toluene and hexane (Wilson et al., 2000) or a mixture (85:15, v/v) of toluene and methanol (Fournier et al., 2006a; 2006b) but the use of alternative solvents was not explored further in our studies. Methyl esters of 22-carbon FA containing four or more double bonds were not resolved by Ag+-TLC, which, combined with the inability of GC– MS analysis of DMOX derivatives to distinguish between geometric isomers, meant that inferences on double bond geometry of specific biohydrogenation intermediates in omasal digesta of cows fed FO had to be drawn on the basis of retention times and or- der of elution relative to authentic standards during GC analysis (I). The main drawbacks of the above-mentioned methods (Ag+-TLC, GC-FID, GC-MS) are that pre-fractionation of samples, preparation of derivatives and visual interpretation of mass spectra are labori- ous processes. Techniques applied to the analysis of FA composition allowed carbon chain length and double bond position for most of the 20- and 22-carbon biohydrogenation interme- diates in omasal digesta to be identified, but the structure of several minor FA remained unknown (I). Furthermore, double bond geometry of most polyenoic biohydrogenation intermediates had to be inferred rather than unequivocally determined. Additional in- vestigations based on GC–MS analysis of other FA derivatives including picolinyl esters (Christie et al., 1986; Christie, 1998), or analysis of FAME for example by covalent adduct chemical ionization tandem MS (Michaud et al., 2003; Gómez-Córtes et al., 2009), two- dimensional GC analysis (Vlaeminck et al., 2007), GC-Fourier infrared spectroscopy-MS (Wahl et al., 1994), reversed-phase HPLC (Banni et al., 1996) and Ag+-HPLC (Fournier et al., 2006b) could further help in verifying and characterizing the double bond geometry of polyenoic biohydrogenation intermediates formed during the hydrogenation of long- chain unsaturated FA in the rumen. Natural resources and bioeconomy studies 74/2020 38 4.2. Impact of dietary fatty acids on nutrient intake, digestibility and rumen fermentation characteristics 4.2.1. Fatty acids in feeds Dairy diets are mixtures of fresh or conserved forages and concentrates, all of which contain lipids. These lipids can be characterised as structural or polar membrane lipids, such as GL and PL, free FA, TAG, and sterol esters. In forages, GL and PL predominate, whereas the main components in cereals, oil seeds, animal fats, and by-product feeds are TAG (Harfoot and Hazlewood, 1988). Diets consumed by lactating cows are low in fat content, generally containing only about 40 to 50 g/kg DM of total fat. The predominant PUFA in ruminant diets are 18:2n-6 and 18:3n-3, with 18:2n-6 being a major component of maize silage, plant oilseeds and cereals, whereas 18:3n-3 is a major component of grass forages and linseed (Elgersma, 2015). Moreover, some plant oilseeds provide MUFA (mainly cis-9 18:1), whereas marine products (FO, marine algae) provide long- chain 20- and 22-carbon n-3 PUFA (mainly 20:5n-3 and 22:6n-3; Chilliard et al., 2007). Summary of mean FA composition (g/100 g FA) and total FA content (g/kg of DM) of some common feed ingredients and oil supplements is presented in Table 2. 4.2.1.1. Fatty acids in grass silage and concentrates Consistent with previous reports (Halmemies-Beauchet-Filleau et al., 2013a; Elgersma, 2015), for all forages, 18:3n-3 represented on average 46–62 % of total FA, followed by 16:0 (14–17 %) and 18:2n-6 (12-19 %) (I–IV). Typically, concentrates, in particular cereals and plant oilseeds, contain relatively high concentrations of 18:2n-6, cis-9 18:1 and 16:0 (Elgersma, 2015). In the experiments reported in I and III, concentrate supplements fed were comprised of rolled barley, solvent extracted rapeseed meal and molassed sugar beet pulp and therefore the lipid in the diets was relatively rich in 18:2n-6 (on average, 37 % of total FA), cis-9 18:1 (29 %) and 16:0 (16 %) with few differences in FA composi- tion and content between these experiments. However, in experiment documented in II and IV the content of cis-9 18:1 was 36 % because rapeseed expeller was used instead of solvent extracted rapeseed meal and concentrate supplement contained rolled oats. 4.2.1.2. Fatty acids in oil supplements In experiment reported in III, SO and LO were fed in combination with FO as sources of plant-derived 18:2n-6 and 18:3n-3, respectively. As expected, ultra-refined herring and mackerel oil supplements fed in I-III were relatively rich in 20:5n-3, 22:5n-3, 22:6n-3, and appropriate sources of several other long-chain PUFA not contained in other feed ingre- dients. The FA composition of refined FO is in a good agreement with other published values in the literature showing that marine lipids from fish, mammals, plankton or algae are rich sources of long-chain (20- to 22-carbon) PUFA, of which 20:5n-3 and 22:6n-3 are the most important (Ackman, 1992). However, the FA composition of FO may vary great- Natural resources and bioeconomy studies 74/2020 39 ly according to fish species, season and geographical location, with values of 4–32 % of total FA for 20:5n-3 and 2–27 % for 22:6n-3, and there may be also a great variability for the other major FA, i.e. 14:0, 16:0, 16-, 18-, 20- and 22-carbon MUFA (Belling et al., 1997; Moffat and McGill, 1993; Vlieng and Body, 1988). 4.2.2. Effect on nutrient intake and digestibility Increasing levels of FO (II) and when supplementing the FO diet with plant oils (III) de- creased intakes of grass silage DM, total DM, organic matter, neutral detergent fibre (NDF), potentially digestible NDF and crude protein, which were lowered yet further when FO was fed with plant oils (III), even though cows were fed at a restricted level of intake in both cases (II; III). It is well established that FO and FO plus plant oils cause dose-dependent decreases in intake in lactating cows (Keady et al., 2000; Palmquist and Griinari, 2006; II; III), but the mechanisms involved are not well defined. The adverse effects of higher PUFA intake in cows have been attributed to several mechanisms in- cluding changes in ruminal fermentation pattern and microbial communities involved in fibre digestion as well as a tendency to shift the site of nutrient digestion from the ru- men to the intestines, and elevated plasma gut peptide concentrations (Allen, 2000). The decrease in DM intake to oil supplements were not accompanied by adverse ef- fects on ruminal or total tract nutrient digestion, but seemed to increase whole tract DM digestion (II; III), which may be explained by longer ruminal retention time caused by lower intake (Huhtanen and Kukkonen, 1995). Overall, our data suggest that inclusion of FO alone and in combination with plant oils in the diet has no negative impact on rumen function, consistent with earlier studies indicating that moderate amounts of FO (Lee et al., 2008; Shingfield et al., 2010b), plant oils (Bayat et al., 2018; Shingfield et al., 2008a) or their blend (Toral et al., 2010) do not depress nutrient digestion in ruminants. As expected, dietary oil supplements increased the intakes of several MUFA and PUFA (II; III). Supplementing the diet with FO increased the intake of most long-chain PUFA other than 18:2n-6, including 16:2n-4, 16:3n-4, 16:4n-1, 18:1n-9, 18:4n-3, 20:3n-3, 20:4n-6, 20:4n-3, 20:5n-3, 21:5n-3, 22:3n-3, 22:5n-3, and 22:6n-3 (II; III), whereas the intakes of cis-9 18:1 and 18:2n-6 were substantially higher for SFO than LFO by design (III), and accordingly 18:3n-3 intake was greater for LFO compared with SFO (III). 4.2.3. Effect on rumen fermentation characteristics Dietary FO supplements alone or in mixture with plant oils had no effect on mean rumen pH (6.58–6.68) (II; III). This is consistent with previous findings in cattle fed different lipid sources, including FO (Keady and Mayne, 1999) and SO (Shingfield et al., 2008a), where- as in some cases FO supplementation of 250 g/day (Shingfield et al., 2003) or 420 g/day (Toral et al., 2016a) has been reported to increase ruminal pH in lactating cows. Ammonia N concentrations were not affected by oil supplements (II; III) consistent with earlier findings in lactating cows fed FO (Shingfield et al., 2003; Toral et al., 2016a) or in nonlactating sheep fed a high-concentrate diet supplemented with a combination Natural resources and bioeconomy studies 74/2020 40 of FO (10 g/kg) and SO (20 g/kg; Toral et al., 2009). However, previous investigatios have demonstrated that increasing level of dietary SO tends to decrease ammonia N concen- trations in the rumen of lactating cows (Shingfield et al., 20008b), whereas supplement- ing the diet with FO increases it over sampling time in cattle (Keady and Mayne, 1999). In the experiments of the present thesis, oil supplements decreased (II) or had no effect (III) on ruminal volatile FA (VFA) concentrations, but FO at high amounts (II) or when supplemented with plant oils (III) promoted an increase in molar proportions of propionate (3:0) (II; III) and butyrate (4:0) (II) at the expense of acetate (2:0) (II; III). From these findings, reduced acetate and total VFA concentrations and increases in propio- nate confirmed earlier observations of rumen fermentation characteristics in lactating cows (Doreau and Chilliard, 1997; Shingfield et al., 2003; Toral et al., 2016a). On the contrary with observed increases in the ruminal proportions of butyrate in our research (II), butyrate concentrations have been also reduced in lactating sheep (Frutos et al., 2018) and goats (Toral et al., 2016a) fed FO. Observed decreases in ruminal butyrate to FO supplements have often been attributed to the adverse effects of unsaturated FA on the growth and microbial activity of specific populations of rumen cellulolytic bacteria, including Butyrivibrio-like bacteria, which produce butyrate and are known to be sensi- tive to the bacteriostatic effects of PUFA (Jenkins et al., 2008), and the selective changes in the ruminal bacterial community capable of biohydrogenation may partly explain the variable effects of FO supplements on rumen VFA profiles (Belenguer et al., 2010). Among long-chain FA, unsaturated FA are more detrimental to rumen microbes than saturated ones (Harfoot and Hazlewood, 1988). The main effects on rumen fermentation characteristics were relatively minor in re- sponse to 75 or 150 g FO/d, the main changes occurring in diets supplemented with 300 g FO/d (II). At relatively low amounts FO has usually rather small effects on rumen fer- mentation parameters in lactating cows, but higher levels of FO or FO fed with SO, have been shown to induce larger changes in the relative proportions of gluconeogenic (3:0 as a major hepatic gluconeogenic substrate) and lipogenic (also called as ketogenic compounds, including 2:0 and 4:0) fermentation end products (Palmquist and Griinari, 2006; Shingfield et al., 2003; 2008b). Furthermore, observed changes in molar VFA pro- portions were more pronounced for SFO and LFO compared with FO, with no evidence of differences due to the source of plant oil (III). Natural resources and bioeconomy studies 74/2020 Table 2 Mean fatty acid composition (g/100 g fatty acids) and total fatty acid content (g/kg of dry matter) of some common feed ingredients and oil supplements 14:0 16:0 16:1 cis-9 18:0 18:1 cis-9 18:2n-6 18:3n-3 20:1 cis-11 20:5n-3 22:5n-6 22:5n-3 22:6n-3 ∑Fatty acids Reference Grass silage1 0.4 16 0.3 1.3 2.9 16 55 22 I, IV Maize silage 0.5 17 0.3 2.3 21 49 5.6 0.2 24 Shingfield et al., 2011 Red clover silage2 0.6 21 0.1 5.6 3.9 19 37 31.8 Vanhatalo et al., 2007 Grass hay3 0.5 21 0.3 1.9 3.3 19 40 8.6 Halmemies-Beauchet- Filleau et al., 2013a Alfalfa 2.4 25 5.6 2.3 14 41 Toral et al., 2016b Barley 20 1 12 58 9 26 Jakobsen, 1999 Maize grain 13 2 33 50 2 45 Jakobsen, 1999 Oats 19 1 33 44 3 44 Jakobsen, 1999 Wheat 21 2 14 58 5 22 Jakobsen, 1999 Camelina oil 0.1 5.6 0.1 2.4 12 16 37 15 954 Bayat et al., 2015 Linseed oil 4.2 2.7 17 16 58 0.2 953 IV Maize oil 0.1 13 0.1 1.9 32 49 0.2 0.3 Ai et al., 2014 Olive oil 10 0.7 2.7 77 5.4 0.5 0.2 Ai et al., 2014 Palm kernel oil4 15 17 2 15 1 Jakobsen, 1999 Palm olein5 1.0 40 0.3 4.2 43 11 0.3 Dorni et al., 2018 Rapeseed oil 4.5 1.4 61 23 9.9 Rego et al., 2009 Safflower oil 6.7 2.3 15 76 953 Bell et al., 2006 Soybean oil 12 3.9 25 54 5.2 Dorni et al., 2018 Sunflower oil 0.1 6.1 0.1 3.6 27 60 0.1 0.2 902 IV Fish oil6 7.3 15 7.8 2.8 9.9 1.3 1.0 1.5 16 0.3 1.8 10 942 I, II, IV Marine algae7 9.9 25 1.6 5.8 1.4 15 0.3 37 955 Toral et al., 2012 1Mixture of meadow fescue x timothy (Festuca pratensis x Phleum pretense); 2The mean of fatty acid content of early and late cut of red clover (Trifolium pratense); 3The mean of fatty acid content of grass hays from two different experiments; 4Palm kernel oil is derived from the kernel of the palm fruit of E. Guineesis, and contains also a substantial amount of medium-chain fatty acids (8:0, 10:0, and 12:0); 5Palm oil is is extracted from the flesh of the fruit of E. Guineesis. When the semi-solid palm oil is refined, it separates into liquid palm olein and palm stearine. Although semi-solid palm oil and liquid palm olein are produced from the same plant and share many similar properties, the main difference between them is their chemical state at room temperature. 6Ultra-refined herring and mackerel oil used in the experiments 1-3; 7Fatty acid composition reported as % of free fatty acids. 41 Natural resources and bioeconomy studies 74/2020 42 4.3. Lipid metabolism in the rumen 4.3.1. General changes in the omasal flow of fatty acid biohydrogenation products and saturated fatty acids Changes in omasal flow of NEFA in response to FO supplements were characterized by increases in 14:0, 16:0, total 16:1, 16:2, 18:1, 18:2, MUFA, PUFA, and a wide range of trans FA (II; Figure 4). Accordingly, inclusion of FO alone or plant oils with FO in the diet increased the flow of 16:0, total 16:1, 18:1, 18:2, trans FA, MUFA, and PUFA at the oma- sum (III; Figure 4). In addition, increasing levels of FO decreased omasal flow of 18:0 and total SFA (II), although when compared with the control, oil supplements had no effect on the amount of 18:0 and SFA leaving the rumen (III). The same was true when FO was compared with SFO and LFO, but the flows of 18:0 and SFA differed between SFO and LFO (III). Moreover, plant oils increased the amount of trans 18:1, total 18:2, trans FA, MUFA, and PUFA at the omasum compared with FO alone, with the amounts of total 18:2, trans FA, and PUFA being higher for LFO than SFO (III). These findings are con- sistent with previous investigations in lactating (Shingfield et al., 2003) and growing cat- tle (Kim et al., 2008; Lee et al., 2008, Shingfield et al., 2010b; 2011) indicating that die- tary PUFA are generally having a common influence on ruminal metabolism by inhibiting the last step of biohydrogenation and resulting in the accumulation of several monoe- noic, dienoic and polyenoic FA in the rumen. Incremental levels of FO in the diet increased the amount of certain OBCFA at the omasum, including iso 13:0, anteiso 13:0, iso 15:0, anteiso 15:0, 3,7,11,15-tetra-methyl- 16:0, cis (∆6, 7, 10 and 11) 17:1, trans-10 17:1, and cis (∆10–12) 19:1 (II), that originate from membrane lipid of rumen bacteria, and have been suggested to represent a proxy of the microbial community in the rumen (Vlaeminck et al., 2006; Fievez et al., 2012). However, dietary oil supplements had relatively minor effects on ruminal escape of OB- CFA (III). In cattle, marine lipids in the diet has been reported to decrease (Boeckaert et al., 2007), have no effect (Or-Rashid et al., 2008), or increase (Boeckaert et al., 2008) concentrations of ruminal OBCFA, indicating that dietary lipid supplements in lactating cows may have variable effects on ruminal outflow of specific OBCFA. Consistent with previous reports in sheep (Kitessa et al., 2001; Toral et al., 2010), FO in the diet increased the ruminal outflow of 10-OH-18:0+9-O-18:0, varying from 0.12- 0.85 (control) to 0.43-1.41 g/d (75-300 g/d FO) (I-II), and 10-O-18:0, varying from 1.48- 2.00 (control) to 5.97-19.4 g/d (75-300 g FO/d; I-II), but had no effect on 13-O-18:0 leav- ing the rumen (I-III). Inclusion of plant oils with FO enhanced yet further the ruminal outflow of 10-O-18:0, being 0.77, 7.29, 32.3 and 17.3 g/d for the control, FO, SFO and LFO respectively (III). Incubations of FA substrates with rumen fluid or pure cultures of rumen bacteria have shown that cis-9 18:1 (Hudson et al., 1995; Jenkins et al., 2006; McKain et al., 2010) and trans-10 18:1 (McKain et al., 2010) can be hydrated to yield 10- OH-18:0, which is further oxidized to 10-O-18:0. Rumen bacteria are also known to be Natural resources and bioeconomy studies 74/2020 43 capable of converting 18:2n-6 to cis-9,13-OH 18:1 (Hudson et al., 1998). Identification of 9-0-18:0, 10-OH-18:0, 10-O-18:0 and 13-O-18:0 in omasal digesta of cows fed FO (I-III) provided clear evidence that dietary FO causes an increase in ruminal outflow of oxy- genated 18-carbon FA in cattle, but it remains uncertain if the accumulation of oxygen- ated 18-carbon FA is due to higher cis-9 18:1 intakes and/or ruminal trans-10 18:1 con- centrations (Jenkins et al., 2006; McKain et al., 2010), or whether one or more FA in FO promote the hydration of unsaturated 18-carbon PUFA, or alternatively inhibit further tranformations of 18-carbon oxygenated FA in the rumen. 4.3.2. Biohydrogenation intermediate products of 16-carbon fatty acids Supplementing the diet with FO increased the flow of 16-carbon FA (Figure 4) and re- sulted in the appearance of several trans-16:1 (∆6–14) and trans-16:2 (Δ9,14; 10,14; 11,15) isomers at the omasum (II; III) confirming earlier findings in cattle (Shingfield et al., 2010b; 2011) that one or more FA in FO inhibit also the complete reduction of 16- carbon unsaturated FA to 16:0 in the rumen. The metabolic fate of 16-carbon unsatu- rated FA in the rumen is not well defined, but comparisons of the FA composition of FO and omasal digesta suggest that the formation of trans-16:1 and -16:2 isomers originate from incomplete biohydrogenation of 16:2n-4, 16:2n-7, 16:3n-4, 16:4n-1 and 16:4n-3 present in FO (II; Shingfield et al., 2010b; 2011). Inclusion of plant oils with FO altered the biohydrogenation of 16-carbon unsaturates, promoting the formation of trans-10- containing products, such as trans-10 16:1 (-, 0.54, 0.70 and 1.30 g/d for control, FO, SFO and LFO, respectively) and trans-10,trans-14 16:2 (-, 0.10, 0.15 and 0.26 g/d, respec- tively) in the rumen (III), suggesting that the mechanisms involved are common to both 16- and 18-carbon unsaturated FA (refer to next chapter 4.3.3). 4.3.3. Biohydrogenation intermediate products of 18-carbon fatty acids Incubations with rumen fluid have established that very long-chain PUFA originating from FO, such as 20:5n-3 and 22:6n-3, inhibit the complete hydrogenation of 18-carbon unsaturated FA causing trans 18:1 isomers to accumulate (AbuGhazaleh and Jenkins, 2004; Klein and Jenkins, 2011). The data presented in this research confirms the contri- bution of dietary supply of unprotected FO to the inhibition of ruminal 18-carbon FA metabolism in vivo in lactating cows (I-II; Toral et al., 2016a). Supplementing the diet with additional FO alone resulted in higher flows (g/d) of 18:1 and 18:2 isomers at the omasum (Figure 4 and 5), changes characterized by an increase in cis (Δ11, 13, 15, 16) 18:1, trans (Δ6–8, 10–13, 15) 18:1, cis-9,trans-12 18:2, trans,cis (Δ9,12; 10,15; 11,15; 12,15) 18:2, and trans,trans (Δ9,14; 10,15; 11,14) 18:2 at the omasum (II; III; Table 3 for most trans 18-carbon MUFA and PUFA). The ruminal outflow of non-conjugated 18:2 biohydrogenation intermediates, such as 18:2n-6 (6.48, 4.59, 3.87 and 6.84 g/d for 0, 75, 150 and 300 gFO/d, respectively), unresolved trans-11,cis-15 18:2 and trans-10,cis-15 18:2 (Table 3), and trans,trans (∆11,14; 11,15) 18:2 (Table 3), increased especially when supplementing the diet with incremental levels of FO (II). At the same time the flow of Natural resources and bioeconomy studies 74/2020 44 18:3n-3 (1.85, 1.34, 1.07 and 1.41 g/d, respectively) at the omasum decreased, but had no effect on cis-9,trans-11,cis-15 18:3 leaving the rumen (II). These findings are con- sistent with previous investigations in lactating (Shingfield et al., 2003) and growing cat- tle (Kim et al., 2008; Lee et al., 2008; Shingfield et al., 2010b; 2011), indicating that one or more FA in dietary FO inhibit the complete hydrogenation of 18-carbon unsaturated FA to 18:0 in the rumen, resulting in the accumulation of numerous trans 18:1 and trans 18:2 biohydrogenation intermediates (Table 3). On the other hand, increased ruminal outflow of cis 18:1 isomers, cis-13 18:1 particularly, may also indicate that supplementa- tion of marine lipid sources rich in 22:6n-3 not only limited trans 18:1 (Figure 4; Table 3) but also cis 18:1 saturation (e.g., II; Toral et al., 2012; 2017). Inclusion of plant oils with FO increased the amount of cis (Δ15, 16) 18:1, trans (Δ6– 8, 10) 18:1 (Figure 5; Table 3), and 18:2 intermediates other than cis-11,cis-14 18:2 com- pared with FO (III). Furthermore, the source of plant oil had a major influence on the amounts of specific 18:1, such as cis (Δ11, 15) 18:1, trans-11 18:1 (Figure 5; Table 3) and 18:2 intermediates leaving the rumen (III). Despite of a similar intake of 18-carbon PUFA and similar flow of trans 18:1, the flow of 18:0 (Figure 4) at the omasum was lower and accumulation of trans 18:2 intermediates (Table 3) greater for LFO than SFO (III). Such findings indicate that the inhibitory effects of FO on the reduction of 18-carbon FA to 18:0 are influenced by the relative amounts of 18:2n-6 and 18:3n-3 in the diet, i.e. the number of double bonds in the 18-carbon FA supplements. Oil supplements decreased flow of 18:2n-6 (17.8, 11.9, 9.94 and 6.76 g/d for control, FO, SFO and LFO, respectively) and 18:3n-3 (7.42, 5.86, 2.67 and 4.37 g/d, respectively) at the omasum, with the amount of 18:3n-3 being lower when plant oils were fed compared with FO alone (III). Comparison of these findings with other published research results is challenging be- cause direct measurements of biohydrogenation intermediates formed in the rumen of lactating cows under in vivo conditions are limited, especially when FO has been used as a lipid source in a combination with plant oil supplements rich in 18:2n-6 and 18:3n-3. Fish oil supplemented treatments (II; III) had no substantial influence on the flow of cis-9,trans-11 CLA (Table 3) or total CLA (Figure 5; Table 3) at the omasum, but altered the relative abundance of positional and geometric isomers of Δ9,11 and Δ10,12 CLA in omasal digesta (Table 3), confirming that trans-10 18:1 (Figure 5; Table 3) would not arise from the common biohydrogenation pathway of 18:2n-6 (Figure 2). These findings are consistent with previous investigations in ruminants fed diets containing FO alone (Shingfield et al., 2003) or in a combination with plant oils rich in 18:2n-6 or 18:3n-3 (Toral et al., 2010; Shingfield et al., 2011). Irrespective of diet, cis-9,trans-11 was the major isomer of CLA in omasal digesta (58-72 and 26-51 % of total CLA for II and III, re- spectively), confirming previous reports in lactating cows (Shingfield et al., 2003; 2008a; Loor et al., 2005c). Incremental levels of FO in the diet increased the omasal flow of trans-8,cis-10 CLA, but lowered the amount of trans,trans (Δ11,13; 12,14; 11,13) at the omasum (II; Ta- ble 3). Compared with control, the omasal flow of trans-7,cis-9 CLA and trans,trans (Δ8,10; 9,11; 10,12) CLA increased, and trans-11,cis-13 CLA and trans-11,trans-13 CLA Natural resources and bioeconomy studies 74/2020 45 decreased in response to oil supplementation (III; Table 3). Supplementing the diet with incremental levels of FO had no effect on trans-10,cis-12 CLA flow, but quadratically increased trans-9,cis-11 CLA at the omasum, reaching a maximum for additional FO of 150 g/d (II; Table 3). However, relative to FO, plant oils plus FO resulted in higher trans- 10,cis-12 CLA and trans,trans (Δ8,10; 10,12) CLA and lower trans-7,cis-9 CLA at the oma- sum (III; Table 3). Furthermore, the ruminal outflow of trans-10,cis-12 CLA was greater for SFO than LFO (III), whereas the reverse was true for the flows of trans-11,cis-13 CLA and trans,trans (Δ11,13; 12,14; 13,15) CLA (III; Table 3). Typically, FO has minimal influ- ence on ruminal trans-10,cis-12 CLA formation (Shingfield et al., 2003; 2010), whilst on the contrary increases in dietary 18:2n-6 content often promote trans-10,cis-12 CLA synthesis (Sackmann et al., 2003; Shingfield et al., 2008a). Similarly, trans-11,cis-13 CLA and trans-11,trans-13 CLA at the omasum were higher on LFO than SFO (III; Table 3), confirming previous studies in vitro (Jouany et al., 2007; Honkanen et al., 2016). All these findings together indicate that geometric ∆11,13 CLA isomers are formed during incomplete biohydrogenation of 18:3n-3. The major pathway of ruminal 18:3n-3 metabolism involves an initial isomerisation to cis-9,trans-11,cis-15 18:3 and sequential reduction of double bonds to yield trans- 11,cis-15 18:2 and trans-11 18:1 as intermediates (Figure 2). Relative to the control, oil supplements had no effect on cis-9,trans-11,cis-15 18:3 (1.15, 1.20, 0.79 and 3.60 g/d for control, FO, SFO and LFO, respectively) at the omasum, whereas the flow was higher on LFO than SFO (III) due to the higher intake of 18:3n-3 in LFO. Relative to SFO, the amounts of alternative novel trans 18:2 isomers, including trans-10,cis-15 18:2 and trans,trans (Δ9,14; 11,15; 10,15; 9,12) 18:2 were increased on LFO (III; Table 3), provid- ing further evidence that multiple products are formed during the biohydrogenation of 18:3n-3 (Jouany et al., 2007; Alves and Bessa, 2014). Inclusion of LO with FO altered the major pathways described for 18:3n-3, resulting in trans-10,cis-15 18:2 replacing trans- 11,cis-15 18:2 (III; Table 3) as the major 18:2 intermediate escaping the rumen, which provides more support to the existence of alternative trans-10 pathway of 18:3n-3 me- tabolism in the rumen (Alves and Bessa, 2014; Ventto et al., 2017). In II, revisiting the analysis of 18:2 isomers in omasal digesta of experimental cows revealed that trans- 10,cis-15 18:2 was erroneously reported to coelute with trans-9,cis-12 18:2, cis-12,trans- 16 18:2 and cis-8,cis-12 18:2, but the revised identification is taken into account in Ta- ble 3. Supplementing the diet with incremental levels of FO resulted in increases in trans 18:1 at the omasum (II; Figure 4; Table 3), with the flows of most isomers, other than trans-10 18:1, reaching a maximum to additional FO of 150 g/d, whereas 300 g FO/d promoted the largest increase in the flow of trans-10 18:1 (II; Figure 5; Table 3). In grow- ing cattle, dietary FO supplements are known to cause a progressive increase in trans 18:1 at the duodenum, with no clear indication that ruminal accumulation of trans-10 18:1 would be as pronounced as in the research presented in this thesis (Kim et al., 2008; Lee et al., 2008; Shingfield et al., 2010b). However, alterations in ruminal biohy- drogenation pathways leading to an increase in trans-10 18:1 formation are known to Natural resources and bioeconomy studies 74/2020 46 occur typically in cows fed high-starch, low-fibre diets with or without supplements of plant oils or oilseeds (Shingfield and Griinari, 2007). Under these circumstances, both trans-10 18:1 and trans-10,cis-12 CLA accumulate in the rumen (Shingfield and Griinari, 2007) which is thought to be related to low rumen pH (Fuentes et al., 2009). In contrast, diet supplemented with 300 g FO/d caused a shift in ruminal 18-carbon FA biohydro- genation, resulting in extensive accumulation of trans-10 18:1 at the omasum (II; Figure 5; Table 3), accounting for 33.6 % of total trans 18:1, in the absence of changes in trans- 10,cis-12 CLA at the omasum (II; Table 3) or decreases in rumen pH (II). Previous investigations in vitro have demonstrated that trans-10 18:1 is formed from 18:2n-6 via reduction of ∆10,12 CLA isomers (Kepler et al., 1966; McKain et al., 2010) and may also be formed by isomerisation of cis-9 18:1 (Mosley et al., 2002). How- ever, detailed analysis of lipid in omasal digesta of cows fed FO (I-III) highlighted the possibility that trans-10 18:1 may also be formed from the reduction of other precursors in vivo, including trans-8,trans-10 CLA, trans,cis (∆10,14; 10;15) 18:2, and trans-10,trans- 14 18:2 (II; III; Table 3). Despite of the apparent accumulation of trans-10 18:1 at the omasum with dietary FO, trans-11 was the major 18:1 isomer in omasal digesta, ac- counting for 49.1 and 40.3 %, 41.0, 40.0, 54.2 and 39.6 % and 46.7, 51.3, 26.2 and 39.4 % of total trans 18:1 for each experimental treatments in I, II and III, respectively (Figure 5; Table 3). Interestingly, relative to SFO, the amount of trans-11 18:1 increased on LFO, where- as the reverse was true for the flow of trans-10 18:1 (4.95, 5.45, 44.3 and 35.6 % of total trans 18:1 for control, FO, SFO and LFO, respectively) when compared SFO with the LFO (III; Figure 5; Table 3). In previous investigations, supplements of sunflower seeds or linseeds in combination with FO have been shown to elevate the proportion of trans-11 18:1 in ruminal (AbuGhazaleh et al., 2003) and duodenal digesta (Shingfield et al., 2011) of cattle. Some of these apparent discrepancies between our results and earlier investi- gations can be explained by differences in the effects of lipid supplements on the major biohydrogenation pathways in the rumen due to variation in 18-carbon FA intake and physical form of supplemental lipids (oil vs. oil seeds). However, ruminal accumulation of trans-10 18:1 when supplementing FO with sources of 18:2n-6 or 18:3n-3 is con- sistent with earlier reports in lactating cows fed FO (II; Shingfield et al., 2003; Loor et al., 2005c) or high-concentrate diets with plant oils (Loor et al., 2004; Ventto et al., 2017). These findings indicate that one or more FA in FO inhibit the reduction of trans 18:1 (as well as trans 18:2) intermediates by ruminal micro-organisms, promoting the formation of trans-10 18:1 and other trans-10 containing intermediates such as trans-10,cis-12 CLA and trans-10,cis-15 18:2 (Table 3). However, the underlying causes remain unknown. Natural resources and bioeconomy studies 74/2020 47 Figure 4. Effect of experimental treatments on the omasal flow of selected fatty acids and fatty acid groups at the omasum in lactating cows fed grass silage-based diets in A) II and B) III. Natural resources and bioeconomy studies 74/2020 48 Figure 5. Effect of experimental treatments on the omasal flow of selected unsaturated fatty acids at the omasum in lactating cows fed grass silage-based diets in A) II and B) III. Natural resources and bioeconomy studies 74/2020 Table 3 Effect of dietary fish oil supplements on the flow of selected 18-carbon fatty acids containing at least one trans double bond at the oma- sum or duodenum in lactating or growing cattle fed grass-silage based diets Shingfield et al., 2003 II Shingfield et al., 2010b Shingfield et al., 2011 III Loor et al., 2005c C FO C FO 75 FO 150 FO 300 C FO 82 FO 163 FO 245 C FO L LFO C FO SFO LFO FO L S Sampling site Omasum Omasum Duodenum Duodenum Omasum Omasum Animal Dairy cow Dairy cow Steer Steer Dairy cow Dairy cow Oil inclusion rate, g/d 0 250 0 75 150 300 0 82 163 245 0 266 265 FO132 L132 0 200 FO200 S 500 FO200 L500 428 860 965 Forage inclusion rate, g/kg DM 600 600 580 580 580 580 600 600 600 600 600 600 600 600 600 600 600 600 350 350 350 Duration, d 14 14 28 28 28 28 21 21 21 21 21 21 21 21 21 21 21 21 28 28 28 Animals per group 5 5 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 3 3 3 Flow, g/d ∑trans 18:1 38.2 182 53.6 80.7 149 168 40.3 68.8 107 116 24.5 84.4 108 125 46.3 135 255 289 213 293 282 t4 0.45 0.65 0.62 0.63 0.74 0.42 0.41 0.47 0.52 0.42 0.23 0.69 0.15 0.56 0.66 0.71 0.86 0.81 0.80 1.92 2.03 t5 0.37 0.56 0.44 0.50 0.62 0.42 0.30 0.39 0.39 0.34 0.16 0.42 0.11 0.42 0.48 0.60 0.72 0.69 0.67 1.39 1.74 t6,-7,-8 1.67 6.65 3.46 5.14 7.94 6.83 2.6 4.2 5.6 5.7 1.58 4.67 5.08 5.60 2.14 6.57 11.5 10.3 6.80 12.1 14.1 t9 1.07 6.21 2.02 3.54 6.37 5.72 1.8 3.3 4.4 4.7 1.09 2.84 4.45 4.88 1.36 5.38 11.9 6.67 4.60 6.39 7.83 t10 1.71 14.3 4.25 5.59 9.95 56.4 3.3 4.3 5.0 6.0 2.05 3.91 19.2 6.44 2.29 7.36 113 103 84.1 32.8 114 t11 17.0 121 22.0 32.3 80.8 66.5 21.6 38.7 69.5 79.1 13.8 51.2 66.7 82.9 21.6 69.2 66.8 114 84.7 115 87.3 t12 2.21 9.41 4.31 7.59 11.8 9.41 2.8 5.9 7.8 7.9 1.47 4.79 5.85 8.70 2.73 9.80 11.7 11.5 9.01 15.8 15.3 t13,(-14) 6.46 14.0 11.3 17.2 21.0 15.9 2.1 3.3 4.3 4.3 1.01 4.47 2.68 4.77 7.02 19.6 23.2 26.1 13.7 60.3 23.3 t15 3.15 5.70 5.24 8.21 9.20 6.68 2.5 4.7 5.5 5.1 1.40 5.76 3.25 6.25 3.65 8.98 9.68 11.5 6.20 29.0 9.57 t161 3.99 3.05 - - - - 2.8 3.7 3.5 2.7 1.63 5.63 0.98 4.29 4.37 6.39 5.40 5.60 2.53 18.2 6.57 ∑CLA 4.36 3.50 4.44 5.13 4.92 3.82 0.39 0.52 0.47 0.37 0.38 3.18 0.27 0.55 4.67 4.76 4.85 3.89 3.94 6.50 8.31 c9,t11 2.86 2.08 2.99 3.23 2.83 2.75 0.23 0.28 0.28 0.21 0.19 0.63 0.11 0.26 3.24 3.76 3.74 2.56 0.40 1.39 2.13 c10,t12 - - - - - - - - - - - - - - - - - - 0.13 0.21 0.29 c11,t13 0.01 0.01 - - - - - - - - - - - - - - - - 0.33 0.11 0.19 c12,t14 0.05 0.0 0.05 0.03 0.0 0.01 0.0 0.0 0.0 0.0 0.0 0.04 0.01 0.0 - - - - - - - t7,c9 - - - - - - - - - - - - - - 0.0 0.04 0.01 0.01 - - - t8,c10 - - 0.01 0.0 0.08 0.11 0.01 0.01 0.01 0.02 0.0 0.01 0.01 0.01 - - - - 0.27 0.21 0.29 49 Natural resources and bioeconomy studies 74/2020 Shingfield et al., 2003 II Shingfield et al., 2010b Shingfield et al., 2011 III Loor et al., 2005c C FO C FO 75 FO 150 FO 300 C FO 82 FO 163 FO 245 C FO L LFO C FO SFO LFO FO L S t9,c11 - - 0.14 0.79 0.89 0.28 - - - - - - - - - - - - 0.27 0.21 0.29 t10,c12 0.10 0.02 0.08 0.02 0.03 0.04 0.01 0.02 0.01 0.01 0.02 0.02 0.03 0.01 0.08 0.01 0.21 0.08 0.27 0.32 1.84 t11,c131 0.46 0.20 0.22 0.25 0.19 0.05 0.03 0.03 0.03 0.01 0.03 1.19 0.01 0.10 0.64 0.46 0.13 0.39 0.27 0.85 0.29 t13,c15 - - - - - - - - - - 0.0 0.06 0.0 0.00 - - - - - - - t7,t9 0.0 0.05 0.07 0.11 0.27 0.17 - - - - - - - - - - - - - - - t8,t10 0.01 0.10 0.05 0.07 0.13 0.07 0.01 0.01 0.01 0.02 0.01 0.01 0.02 0.01 0.02 0.04 0.07 0.07 - - - t9,t11 0.22 0.55 0.21 0.23 0.33 0.23 0.05 0.08 0.07 0.05 0.05 0.16 0.06 0.08 0.19 0.25 0.41 0.29 - - - t10,t12 0.05 0.06 0.05 0.03 0.03 0.05 0.03 0.03 0.02 0.02 0.03 0.06 0.01 0.01 0.04 0.03 0.14 0.13 - - - t11,t13 0.40 0.09 0.41 0.21 0.09 0.05 0.02 0.03 0.02 0.01 0.03 0.69 0.01 0.03 0.32 0.09 0.06 0.17 - - - t12,t14 0.19 0.08 0.18 0.14 0.07 0.04 0.02 0.02 0.02 0.01 0.01 0.29 0.01 0.03 0.13 0.07 0.06 0.15 - - - t13,t15 - - - - - - - - - - 0.0 0.02 0.0 0.0 0.02 0.01 0.01 0.03 - - - ∑trans 18:2 - - - - - - - - - - 1.46 4.34 4.26 21.3 - - - c9,t122 - - 0.16 0.16 0.25 0.65 - - - - - - - - 0.14 0.37 0.51 1.11 0.53 3.09 1.16 c11,t15 - - 0.11 0.20 0.36 2.80 - - - - - - - - - - - - - - - t9,c123 0.18 1.60 0.27 0.53 0.87 1.83 - - - - - - - - 0.24 0.82 1.60 4.35 1.33 3.30 1.26 t10,c15 - - - - - - - - - - - - - - 0.0 0.81 5.20 35.8 - - - t11,c154 2.81 18.8 3.73 5.38 8.96 31.5 1.34 2.34 3.29 3.62 0.77 6.73 4.45 7.87 3.72 8.56 6.74 21.5 17.9 50.1 8.31 t12,c15 - - 0.24 0.38 0.39 1.14 - - - - - - - - 0.24 0.60 0.68 1.32 - - - t9,t12 0.03 0.93 0.09 0.16 0.26 0.68 0.32 0.57 0.99 1.25 0.20 2.88 1.50 2.91 - - - - 1.47 2.98 0.68 t10,t155 - - - - - - - - - - - - - - 0.15 0.49 1.02 8.54 - - - t11,t156 - - 0.87 1.43 2.73 4.87 - - - - - - - - 0.96 3.16 2.66 10.2 - - - t11,t147 - - 0.27 0.39 0.72 1.69 - - - - - - - - 0.36 0.87 0.84 3.70 - - - 1Co-elutes with cis-9,cis-11 18:2 in Loor et al., 2005c; 2Co-elutes with cis-9,trans-14 18:2 in II; 3Co-elutes with cis-12,cis-16 18:2 and trans-8,cis-12 18:2 in II, and co-elutes with cis-12,cis-16 18:2 and trans-11,cis-16 18:2 in III; 4Trans-11,cis-15 18:2 co-elutes with trans-10,cis-15 18:2 in II; 5Co-elutes with trans- 9,trans-12 18:2; 6Co-elutes with trans-9,trans-14 18:2; 7Co-elutes with trans-10,trans-13 18:2 in II, co-elutes with trans-10,trans-13 18:2, cis-9,trans-13 18:2 and cis-10,trans-14 18:2 in III; C, control diet; CLA, conjugated linoleic acid; FO, fish oil; L, linseed oil; S, sunflower oil; For clarity purposes, abbreviat- ed names of fatty acids are reported in the table (c, cis; t, trans). 50 Natural resources and bioeconomy studies 74/2020 51 4.3.4. Biohydrogenation intermediate products of 20- to 22-carbon fatty acids Incubations with mixed rumen bacteria have demonstrated that 20:5n-3 and 22:6n-3 disappear over time (Dohme et al., 2003; AbuGhazaleh and Jenkins, 2004; Wasowska et al., 2006; Aldai et al., 2012; 2018; Escobar et al., 2016; Toral et al., 2018c), but to date, only few investigations have characterized 20- and 22-carbon intermediates formed during ruminal biohydrogenation of marine lipid sources high in 22:5n-3 and/or 22:6n-3, and evaluated the putative pathways under in vivo conditions (e.g. I-III; Toral et al., 2010; 2012). Detailed analysis of lipid in omasal digesta of cows fed grass silage-based diets sup- plemented with different levels of refined FO (I-III) along with measurements of ruminal FA balance (I) and nutrient flow at the omasum (I-III; Shingfield et al., 2003) provided the first quantitative estimates of the ruminal outflow of long-chain 20- to 22-carbon PUFA intermediates in lactating cows. In total, 27 previously unidentified 20-, 21- and 22- carbon FA containing at least one trans double bond (Table 4) plus several unique all-cis long-chain PUFA not contained in refined FO were detected in variable concentrations in omasal digesta of cows fed FO for the first time (I; II), with the implication that the hy- drogenation of long-chain unsaturated FA in the rumen involves a complicated series of metabolic reactions and the formation of numerous biohydrogenation intermediates. Overall, the data demonstrated extensive ruminal biohydrogenation of 20:5n-3, 22:5n-3 and 22:6n-3 in vivo resulting in the accumulation of numerous 20-, 21- and 22- carbon unsaturated FA at the omasum (I-III). The total omasal flows of 20-, 21- and 22- carbon PUFA varied between 5.66-46.5, 0.00-1.00 and 3.78-25.8 g/d, respectively, for all experimental treatments in this research (I-III; Figure 6). The differences in the ruminal biohydrogenation of 20:5n-3, 22:5n-3 and 22:6n-3 and the different flows of total 20-, 21- and 22-carbon intermediates at the omasum on LFO versus SFO (III; Figure 6), pro- vided a clear evidence that the source of 18-carbon PUFA (18:2n-6 vs. 18:3n-3) influ- ences the kinetics of ruminal 20:5n-3, 22:5n-3, and 22:6n-3 metabolism. It is well estab- lished that during incubations with rumen contents, 18:2n-6 inhibited 20:5n-3 and 22:6n-3 biohydrogenation (Wasowska et al., 2006). In line with this, studies in growing cattle (Shingfield et al., 2011) and nonlactating sheep (Toral et al., 2010) fed diets con- taining FO and plant oils have also demonstrated that the supply of 18-carbon PUFA may influence the extent of ruminal 20- and 22-carbon PUFA biohydrogenation. Natural resources and bioeconomy studies 74/2020 Table 4 Effect of dietary fish oil supplements on the proportion of selected 20- and 22-carbon fatty acids containing at least one trans double bond in the omasal or ruminal fluid in ruminants Sampling site I Toral et al., 20101 II III Toral et al., 2012 C FO C SFO C FO 75 FO 200 FO 300 C FO SFO LFO C S SMA1 SMA2 SMA3 Sampling site Omasum Rumen Omasum Omasum Rumen Species Bovine Ovine Bovine Bovine Ovine Oil inclusion rate, g/d none 250 none2 FO(10) S(20) none 75 150 300 none 200 FO200 S500 FO200 L500 none S (25) S(25)2 MA(8) S(25)2 MA(16) S(25)2 MA(24) Forage inclusion rate, g/kg DM 600 600 350 350 580 580 580 580 600 600 600 600 485 485 485 485 485 Duration, d 14 14 11 11 28 28 28 28 21 21 21 21 28 28 28 28 28 Animals per group 5 5 5 5 4 4 4 4 4 4 4 4 5 5 5 5 5 Fatty acid profile (g/100 g fatty acids) trans 20:1 t6,-7,-8 - - 0.02 0.02 - - - - 0.02 0.01 0.01 0.01 0.03 0.01 0.01 0.02 0.02 t9,-10 0.01 0.04 - 0.04 0.01 0.04 0.06 0.07 0.01 0.06 0.03 0.03 - - - - - t11 0.02 0.07 <0.01 0.07 0.03 0.06 0.12 0.08 0.03 0.15 0.07 0.08 - - - - - t12 0.01 0.08 <0.01 0.07 0.02 0.06 0.12 0.11 0.01 0.15 0.09 0.11 0.01 0.01 0.01 0.01 0.01 t13 0.02 0.11 <0.01 0.08 0.02 0.09 0.19 0.10 0.01 0.22 0.09 0.10 0.01 <0.01 0.01 0.01 0.01 t14 0.02 0.09 0.03 0.11 0.02 0.11 0.20 0.10 0.02 0.25 0.10 0.09 0.02 0.01 0.01 0.02 0.02 t15 0.03 0.25 <0.01 0.14 0.04 0.30 0.55 0.23 0.02 0.66 0.32 0.17 - - - - - trans 20:2 c10,t153 - 0.04 - 0.03 - 0.02 0.05 0.06 <0.01 0.13 0.07 0.07 - - - - - t11,c15 - 0.16 - - - 0.01 0.05 0.18 - 0.07 0.05 0.18 - - - - - t11,c17 - 0.08 <0.01 0.03 - 0.01 0.04 0.08 - - - - - - - - - t13,c17 - 0.36 <0.01 0.03 0.02 0.09 0.21 0.69 0.20 0.45 0.42 0.42 - - - - - t14,c17 - 0.18 <0.01 0.02 - 0.05 0.11 0.15 <0.01 0.19 0.10 0.05 0.01 <0.01 <0.01 0.01 <0.01 t9,t15 - 0.24 - - 0.03 0.10 0.17 0.27 0.06 0.30 0.20 0.28 - - - - - t10,t164 - 0.21 - - - - 0.06 0.09 - 0.07 0.04 0.09 - - - - - t11,t15 - 0.15 - - 0.05 0.08 0.11 0.15 - 0.07 0.05 0.12 - - - - - t13,t17 - 0.07 - - - 0.01 0.04 0.09 <0.01 0.04 0.04 0.08 - - - - - trans 20:3 c11,c14,t17 - 0.07 - - - 0.02 0.04 0.12 - - - - - - - - - t11,c14,c17 - 0.14 - - 0.05 0.21 0.18 0.41 - - - - - - - - - 52 Natural resources and bioeconomy studies 74/2020 Sampling site I Toral et al., 20101 II III Toral et al., 2012 C FO C SFO C FO 75 FO 200 FO 300 C FO SFO LFO C S SMA1 SMA2 SMA3 t11,t14,c17 - - - - - - - - - 0.04 0.03 0.04 - - - - - ∆10,14,175 - 0.05 - - - - 0.02 0.09 - - - - - - <0.01 <0.01 0.01 ∆11,14,175 - 0.03 - - - - 0.01 0.05 - 0.01 0.01 0.02 - - <0.01 <0.01 0.01 ∆11,14,185 - 0.01 - - 0.03 0.06 0.05 0.02 - 0.03 0.01 0.01 0.02 0.01 0.01 0.01 0.02 t10,t14,c7 - 0.13 - - - 0.05 0.06 0.33 - - - - - - 0.02 <0.01 0.01 c9,t14,t17 - - - - - - - - 0.64 0.66 0.83 0.36 - - - - - t9,c14,c17 - - - - - - - - 0.01 0.17 0.11 0.13 - - - - - t9,c14,t17 - - - - - - - - - 0.03 0.22 0.04 - - - - - t9,t14,t17 - 0.03 - - - - - - 0.01 0.01 0.02 0.04 - - <0.01 0.01 0.01 t10,t14,t17 - 0.05 - - - 0.01 0.03 0.07 - - - - - - - - - trans 20:4 t7,c11,c14,c17 - 0.02 - - - 0.04 0.05 0.15 - 0.06 0.04 0.03 - - - - - trans 22:2 t12,t17 - 0.03 - - <0.01 0.03 0.06 0.05 0.09 0.12 0.08 0.09 - - - - - t13,t18 - - - - <0.01 0.03 0.05 0.05 - - - - - - - - - trans 22:3 c10,t14,c19 - 0.08 - - 0.02 0.02 0.04 0.11 0.06 0.07 0.04 0.09 0.04 0.03 0.01 0.03 0.02 t12,c16,c19 - 0.14 - - - 0.04 0.12 0.23 <0.01 0.19 0.14 0.15 - - 0.12 0.05 0.05 trans 22:4 c7,t13,c16,c19 - 0.05 - - - 0.02 0.02 0.04 <0.01 0.03 0.01 0.03 - - 0.01 0.09 0.07 t8,c13,c16,c19 - 0.13 - - - 0.01 0.07 0.30 0.18 0.24 0.15 0.22 - - - - - t10,t13,c16,c19 - 0.03 - - - 0.01 0.04 0.12 - - - - - - - - - trans 22:5 (t)5,c10,c13,c16,c196 - 0.05 - - - - - - - 0.26 0.06 0.11 - - 0.12 0.11 0.13 1Values are mean of experimental days 3 and 10; 2Intake of dry matter or lipid supplements not reported. Concentrations of oil in the diet (g/kg diet dry matter) indicated in parentheses; 3Co-elutes with trans-11,cis-17 20:2 in III; 4Co-elutes with 21:0 in II; 5Retention time comparisons inferred a cis,trans,trans or trans,cis,trans double bond configuration; 6cis or trans double bond configuration undetermined in III, refer to ∆5,10,13,16,19 22:5; C, control diet; CLA, conjugated linoleic acid; FO, fish oil; L, linseed oil; MA1-3, increasing levels of marine algae in the diet, concentrations of marine algae in the diet (g/kg diet dry matter) indicated in parentheses; S, sunflower oil; For clarity purposes, abbreviated names of fatty acids are reported in the table (c, cis; t, trans). 53 Natural resources and bioeconomy studies 74/2020 54 Figure 6. Effect of experimental treatments on the total omasal flow of 20-, 21- and 22-carbon fatty acids at the omasum in lactating cows fed grass silage-based diets in A) I, B) II and C) III. Natural resources and bioeconomy studies 74/2020 55 Ruminal outflows of all-cis 20-, 21-, and 22-carbon PUFA, most of them not supplied from the diet, such as 20:2n-3, 20:2n-6, 20:3n-3, 20:3n-6, 20:4n-3, 21:3n-3, 21:4n-3, 22:2n-6, 22:3n-3, 22:3n-6, 22:4n-3, and 22:4n-6, exceeded intake, which would indicate these long-chain PUFA as intermediates of 20:5n-3, 22:5n-3 or 22:6n-3 metabolism. In- clusion of plant oils plus FO decreased 20:4n-6, 21:5n-3, 22:4n-6 and 22:6n-3 at the omasum and increased ruminal outflow of 22:4n-3 compared with FO (III). Furthermore, LFO resulted in higher flows of 22:4n-6, 22:4n-3 and 22:5n-3, but decreased the amount of 22:3n-3 compared with SFO (III). The formation of 18-carbon FA intermediates containing a conjugated double bond system (Figure 1B) is proposed to represent as an initial step of the main biohydrogena- tion pathways of 18-carbon PUFA (Lee and Jenkins, 2011; Alves and Bessa, 2014; Honkanen et al., 2016) and it is assumed that the ruminal metabolism of 20:5n-3, 22:5n- 3 and 22:6n-3 would follow the same pattern involving isomerisation of cis double bond(s) to form intermediates with 5 or 6 double bonds, containing at least one trans double bond (Jenkins et al., 2008) and/or formation of conjugated 20- and 22-carbon intermediates that are sequentially reduced to their saturated end products in this bio- chemical process. The reduction of the cis double bond closest to the carboxyl group (cis-5 in 20:5n-3, cis-7 in 22:5n-3) is proposed to represent an initial step of the main biohydrogenation pathway of 20:5n-3, 21:5n-3 and 22:5n-3 PUFA in the rumen (I; II; Toral et al., 2010; 2018c; Jeyanathan et al., 2016; Escobar et al., 2016). Extensive investigations have not yielded any substantive evidence that the carbon chain of PUFA is elongated or shortened during biohydrogenation in the rumen (Harfoot and Hazlewood, 1988; Jenkins et al., 2008). Therefore, e.g. the appereance of 21:3n-3 and 21:4n-3 in omasal digesta of cows fed dietary FO supplements must arise from the hydrogenation of 21:5n-3 in FO via specific mechanism that involves the sequential re- duction of the cis-6 and cis-9 double bonds in 21:5n-3 and 21:4n-3, respectively (I-III). However, definitive information on the biohydrogenation of 22:6n-3 is still lacking. Some reports show much lower importance of above-mentioned reduction mechanism for ruminal 22:6n-3 metabolism compared to that of 20:5n-3 and 22:5n-3 (Aldai et al., 2018; Toral et al., 2018c) than the others (Jeyanathan et al., 2016). This discrepancy between investigations, together with the high accumulation of numerous unique 20- to 22- carbon PUFA intermediates in the rumen (I-III), could indicate that the ruminal metabo- lism of very long-chain n-3 PUFA may differ within ruminant species as recently outlined after comprehensive incubations of 20:5n-3, 22:5n-3 and 22:6n-3 with rumen contents of cows and ewes (Toral et al., 2018c). Omasal digesta of cows fed FO was devoid of conjugated 20:5, 22:5 or 22:6 biohy- drogenation intermediates (I-III), in agreement with earlier studies reporting no detec- tion of conjugated 20- and 22-carbon FA isomers in vivo with cows, sheep, and goats fed FO or marine lipids (Toral et al., 2010; 2012; 2016a) or in vitro incubations of 20:5n-3 and 22:6n-3 (Aldai et al., 2012; Escobar et al., 2016; Jeyanathan et al., 2016). These find- ings, suggest a different pathway for the biohydrogenation of long-chain 20- and 22- carbon n-3 PUFA compared to that of 18-carbon PUFA. However, these observations are Natural resources and bioeconomy studies 74/2020 56 challenged by the quite recent identification and characterization of minor conjugated 22:6 isomers (Aldai et al., 2018) as well as the appearance of conjugated Δ11,13,17,19- 22:4 intermediate in ovine and bovine rumen inoculum (Toral et al., 2018c) after in vitro incubations of 20:5n-3/22:6n-3 and 22:5n-3, respectively. The reason for the discrepan- cy of the results between our research and the in vitro investigations of Aldai (2018) and Toral (2018a) may be the complexity of in vivo experiments conducted with FO com- pared with the more straightforward in vitro approach with pure 20:5n-3, 22:5n-3 and 22:6n-3. Therefore, the transient formation of conjugated double bond structures as a result of ruminal long-chain 20- and 22-carbon PUFA metabolism cannot be unequivo- cally excluded (Figure 7 and Figure 8). Dietary FO supplements alone (I-III) or in combination with plant oils (III) resulted in corresponding increases in numerous polyenoic 20- and 22-carbon intermediates con- taining at least one trans double bond escaping the rumen (Table 4). The most abundant trans 20- and trans 22-carbon FA intermediates found in FO digesta were trans-13,cis-17 20:2 (II; III), trans-10,trans-14,cis-17 20:3 (II), trans-7,cis-11,cis-14,cis-17 20:4 (II; III), trans-12,cis-16,cis-19 22:3 (III), cis-7,trans-13,cis-16,cis-19 22:4 (III) and trans-8,cis- 13,cis-16,cis-19 22:4 (III). In addition, inclusion of plant oils plus FO increased multiple trans,cis and trans,trans 20- and 22-carbon FA intermediates at the omasum, including (Δ11,15; 13,17) 20:2 and trans-11,cis-14,trans-17 20:3, compared with FO (III). Further- more, LFO resulted in higher flows of trans-11,cis-15 20:2, trans,trans (Δ10,16; 11,15; 13,17) 20:2, cis-10,trans-14,cis-19 22:3 and trans-8,cis-13,cis-16,cis-19 22:4, but de- creased the amounts of trans-14,cis-17 20:2 and cis-9,trans-14,trans-17 20:3, compared with SFO (III). The total flow at the omasum varied between 8.62-23.5 and 1.02-5.58 g/d for trans 20- and trans-22-carbon PUFA, respectively, in different levels of FO (I–III; Ta- ble 4), demonstrating that ruminal biohydrogenation of ≥ 20-carbon PUFA to saturated end products is incomplete and highly variable, involving the formation of several trans PUFA intermediates via different biochemical pathways. Even though oil supplements increased 20-, 21-, and 22-carbon PUFA intake, the amounts of 20:0, 21:0, and 22:0 reaching the omasum were not substantially higher than the control (I–III), indicating that ruminal biohydrogenation of 20- and 22-carbon unsaturated FA to saturated end products is incomplete. These findings are in agree- ment with earlier reports in cattle (Shingfield et al., 2003; 2010; Lee et al., 2008) and sheep (Toral et al., 2010) fed FO or marine lipids. In accordance to this we observed also increases in ruminal outflow of several trans 20:1 and trans 22:1 isomers irrespective of oil supplemented diets (I-III; Table 4). Changes in the flow of trans 20:1 biohydrogena- tion intermediates were also accompanied by an increase in the amount of cis (∆11,13- 15) 20:1 at the omasum (I-III) that may reflect ruminal escape of these FA contained in FO, or formation of these isomers during the hydrogenation of 20-carbon PUFA in the rumen. Ruminal escape of FA in FO may also account for the increase in cis (∆11,13,15) 22:1 in FO omasal digesta (I-III), but the possibility that one or more of these isomers is formed during the penultimate step of 22-carbon PUFA hydrogenation in the rumen cannot be excluded. Natural resources and bioeconomy studies 74/2020 57 The biological significance of the formation and accumulation of numerous 20- and 22-carbon FA containing at least one trans double bond in the rumen of cows fed FO remains uncertain. It is not clear whether long-chain n-3 PUFA by themselves or the trans double bond containing FA intermediates formed during the hydrogenation of very long-chain n-3 PUFA in the rumen, might affect the ruminal metabolism of other, mainly 18-carbon PUFA. Total ruminal outflow of specific 20- and 22-carbon biohydrogenation intermediates was of the same magnitude as the total intake of 20:5n-3, 22:5n-3 and 22:6n-3 (I–III) and it is probable that most of these intermediates are also incorporated into milk fat and tissue lipids, albeit at low concentrations. Adapted with permission from Toral P.G., Hervás, G., Leskinen, H., Shingfield, K.J. and Frutos, P. 2018c. J. Dairy Sci. 101:6109-6121. Copyright (2018) American Dairy Science Association. Figure 7. Putative pathways describing initial 20:5n-3 biohydrogenation in vitro. Thick arrows highlight the potentially major pathway; grey arrows and text represent a hypothetical pathway involving the formation of a conjugated 20:5 intermediate (not identified in digesta yet). For clarity purposes, abbreviated names of fatty acids are reported in the figure (c, cis; t, trans). Natural resources and bioeconomy studies 74/2020 58 Adapted with permission from Aldai, N., Delmonte, P., Alves, S.P., Bessa R.J.B. and Kramer, J.K.G. 2018. J. Agric. Food Chem. 66:842-855. Copyright (2018) American Chemical Society. Figure 8. Proposed pathways of DHA (22:6n-3) metabolism using sheep rumen fluid. One of the five possible MTMI-DHA and two of the MC-DHA metabolites are shown as representative prod- ucts, as well as the five 22:5 products identified, all of which contain an isolated trans double bond. (*) Members of the pathway of DHA metabolism that resulted in the formation of the ma- jor 22:5 product in the incubation mixture. (**) Product reported by Jeyanathan et al.2016 (BMC Microbiol. 16:104) that could not be confirmed. MTMI-DHA, mono trans methylene interrupted- DHA; MC-DHA, monoconjugated DHA. 4.3.5. The extent of apparent ruminal biohydrogenation of dietary unsatu- rated fatty acids Mean estimates of 18:2n-6 and 18:3n-3 biohydrogenation reported in the literature vary between 70-95 and 85-100 %, respectively (Doreau and Ferlay, 1994; Chilliard et al., 2007). In this thesis the apparent biohydrogenation of PUFA was calculated as [(intake, g/d – flow at the omasum, g/d)/intake, g/d] (I-III). The amounts of most unsaturated 18- carbon FA at the omasum were lower than their respective intake, indicating extensive ruminal biohydrogenation of 18-carbon FA for all experimental treatments in this re- search (I–III). Mean estimates of cis-9 18:1, 18:2n-6 and 18:3n-3 biohydrogenation were 87, 96 and 98 %, respectively, in cows fed FO (I; II) and 90, 97 and 98 %, respectively, in cows fed FO plus plant oils (III), consistent with previously reported estimates of 88–89 % in lactating cows fed grass-silage based diets containing moderate levels of FO (250 g FO/d; Shingfield et al., 2003). These estimates were marginally higher than correspond- ing values measured in the basal diet (on average 82, 94 and 97 %, respectively; I–III) Natural resources and bioeconomy studies 74/2020 59 and slightly greater than earlier estimates of 58–97 % reported in cattle (Lee et al., 2008; Shingfield et al., 2010b; 2011). Oil supplements increased cis-9 18:1, 18:2n-6 and 18:3n-3 biohydrogenation in the rumen, which was further enhanced when FO was compared with SFO and LFO (III) con- sistent with the phenomenon previously reported in growing cattle fed diets containing plant oils and FO (Shingfield et al., 2011). However, there was no difference in the ex- tent of ruminal cis-9 18:1 and 18:2n-6 biohydrogenation between the SFO and LFO treatments, but the extent of ruminal 18:3n-3 biohydrogenation was substantially greater for LFO than SFO (III), confirming earlier reports in growing cattle (Shingfield et al., 2011) and nonlactating sheep (Toral et al., 2010) that dietary supply of 18-carbon PUFA may influence the extent of n-3 PUFA biohydrogenation in the rumen. The long-chain 20- and 22-carbon n-3 PUFA were extensively biohydrogenated in the rumen, confirming earlier reports in lactating cows (e.g. Doreau and Chilliard, 1997; Shingfield et al., 2003). However, the extent of 20:5n-3, 22:5n-3 and 22:6n-3 biohydro- genation (98, 88, and 98 %, respectively) at present work (I–III) was higher than previ- ously reported mean estimates of 92 and 89 % for 20:5n-3 and 22:6n-3, respectively, in lactating cows fed a high-concentrate diet containing FO (Loor et al., 2005c), and mean values of 95, 71 and 96 % for 20:5n-3, 22:5n-3 and 22:6n-3, respectively, reported in cattle fed grass-silage based diets supplemented with moderate levels of FO (ca. 50- 270 g FO/d; Lee et al., 2008; Shingfield et al., 2010b; 2011). Variations in the extent of long-chain PUFA metabolism have been suggested to reflect e.g. differences in dietary fibre content and concentration of FO in the rumen (Gulati et al., 1999). Limited data exists reporting mean values of ruminal biohydrogenation for unique long-chain n-3 FA originating from FO, such as 20:4n-3 and 21:5n-3, varying between 67- 100 % (Shingfield et al., 2003; I; II) and 73-81 % (I–III), respectively. Consistent with these findings, the ruminal metabolism of very long-chain all-cis polyenoic FA, such as 22:4n-6, 23:5n-3, 24:5n-3, and 24:6n-3, appeared to be highly extensive as well, ca. 87-100 % at present work (I). Increasing levels of FO decreased the extent of ruminal biohydrogenation of 20:5n-3 (-, 93.8, 92.5, and 91.7 % for 0, 75, 150 and 300 g FO/d, respectively; II), 22:5n-3 (-, 86.8, 79.8, and 51.0 %, respectively; II), and 22:6n-3 (-, 93.8, 92.0, and 89.6 %, respectively; II), whereas the dietary supply of plant oils with FO resulted in more extensive metabolism of 22:6n-3 compared with FO (-, 89.9, 95.6, and 94.9 %, for control, FO, SFO and LFO, respectively; III). The effect of diets containing plant oils and FO on the ruminal biohy- drogenation of 20:5n-3 (-, 82.6, 84.0, and 88.2 %, respectively; III) seemed to be similar, but the effect was not significantly different between diets. However, the biohydrogena- tion of 22:5n-3 (-, 78.5, 87.3, and 83.5 %, respectively; III) was less extensive in LFO than SFO, but showed no difference between FO and diets containing plant oils, offering no support to the lowered biohydrogenation of 22:5n-3 in response to the dietary inclusion of plant-derived 18-carbon PUFA as hypothesized in the beginning of this work (refer to the 3rd hypothesis in chapter 2). Natural resources and bioeconomy studies 74/2020 60 4.3.6. Rumen microbial ecology Dietary FO treatments (II; III) had no effect on protozoal numbers in the rumen. Of the species known to be capable of biohydrogenation, the group B. proteoclasticus was the most abundant, comprising up to 0.74 and 1.27 % of total bacteria for II and III, respec- tively. The group encompassing B. fibrisolvens and known Pseudobutyrivibrio spp. were the next most numerous (0.55–0.96 and 0.93 to 1.47 % of total bacterial DNA, respec- tively), whereas B. hungatei represented only a small proportion (0.002-0.04 and 0.04– 0.05, respectively) of total Butyrivibrio + Pseudobutyrivibrio group. Reported numbers of the group Butyrivibrio + Pseudobutyrivibrio in the experiments presented in this thesis were higher than earlier estimates across a range of FO or marine diets in ruminants (Boeckaert et al., 2008; Huws et al., 2010; 2011; Toral et al., 2012). This may reflect the differences between the analysis of reconstituted omasal digesta in our studies, contain- ing large-particle, small-particle, and liquid fractions in amounts representative of that truly entering the omasal canal, and analysis of isolated samples of ruminal digesta in the previous studies. However, the effect of dietary FO on the numbers of B. fibrisolvens + Pseudobutyrvibrio (II) is consistent with several observations reporting a decline in B. fibrisolvens in the presence of marine lipids or 22:6n-3 (Liu et al., 2011; Potu et al., 2011; Gudla et al., 2012; Toral et al., 2012). Even though diets containing FO alone or in com- bination with plant oils altered the amounts of 18-carbon biohydrogenation intermedi- ates and 18:0 at the omasum, no other changes in analysed ruminal bacterial popula- tions were observed by treatment except a decrease in the abundance of Oribacterium spp. on LFO compared with SFO (III). The population sizes of Streptococcus bovis and Propionibacterium acnes, ruminal species known to hydrate rather than hydrogenate cis-9 18:1 and 18:2n-6 (Hudson et al., 2000; Kim et al., 2008), were extremely small in omasal digesta and not affected by treatment (II; III). S. bovis was present in 0.01 and 0.005 % (average) of the bacterial community for II and III, respectively, but the numbers of P. acnes were several orders of magnitude lower (approximately ≤ 0.001 % of total bacteria). Megasphaera elsdenii was below the limit of detection (10−7 of total bacteria) in all samples (II; III). The atypical type strain of B. fibrisolvens, (which historically is the type strain, ATCC 19171), was pre- sent in 0.01 % of the population when incremental levels of FO was fed (II), but below the limit of detection in all samples with diets containing no oil, FO alone or FO plus plant oils (III). Because B. proteoclasticus is the only known bacterium capable of reduc- ing 18:1 isomers to 18:0, it might have been anticipated that the decrease in 18:0 at the omasum to FO (II) or other oil treatments (III) would have been accompanied by a fall in B. proteoclasticus numbers. However, there was no evidence of such an association in this (II; III) or other published reports in the literature (Kim et al., 2008; Huws et al., 2010; 2011; Gudla et al., 2012). Nevertheless, FO has been reported to decrease the abundance of different bacteria belonging to the Butyrivibrio group, including B. proteo- clasticus, in continuous cultures (AbuGhazaleh and Ishlak, 2014). Changes in microbial ecology associated with the trans-10 shift in cows fed 300 g FO/d are difficult to explain. Butyrivibrio spp. catalyze the reduction of trans- Natural resources and bioeconomy studies 74/2020 61 10,cis-12 CLA to trans-10 18:1 (McKain et al., 2010), but did not result in the accumula- tion of trans-10,cis-12 CLA or other 10,12 geometric CLA isomers. Although M. elsdenii has been reported to form trans-10,cis-12 CLA from 18:2n-6 (Kim et al., 2002), there is also evidence to show that no strain of M. elsdenii carries out that reaction (Wallace et al., 2006). Nevertheless, the numbers of M. elsdenii in the experiments of this thesis were below the limit of detection (II; III). However, P. acnes catalyses the reduction of 18:2n-6 to trans-10,cis-12 CLA (Liavonchanka et al., 2006) and FO supplements tended to cause a dose-dependent increase in this population. However, it remains unclear if P. acnes could be solely responsible for trans-10,cis-12 CLA and trans-10 18:1 formation, because the specific activities of this bacterium have not been reported. Alternatively, the growth and activity of a microorganism capable of converting trans-11 18:1 to trans- 10 18:1 may be promoted by one or more FA in FO, but the identity of such species is not known. It has often been assumed that the main FA in FO are responsible for the inhibitory effects on biohydrogenation. Based on our observations in vivo (II; III) and earlier incu- bations of FA substrates with rumen fluid (Wasowska et al., 2006) it seems that FO in- hibits ruminal biohydrogenation through a mechanism that is not solely explained by effects on B. fibrisolvens. Other bacteria, such as Ruminococcus albus, Eubacterium spp., and Treponema spp. (Yokoyama and Davis, 1971; Kemp et al., 1975), and strains of Ori- bacterium (S. Muetzel and R. J. Wallace, unpublished data, personal communication) have been described to convert 18:2n-6 to cis-9,trans-11 CLA and trans-11 18:1. Includ- ing FO in the diet of growing cattle has been shown to induce changes in the number of other rumen bacterial strains, such as Anaerovirbrio lipolytica, Fibrobacter succinogenes, and Ruminococcus flavefaciens (Huws et al., 2010), suggesting a broader influence on ruminal lipid metabolism (Kim et al., 2008; Huws et al., 2010). Furthermore, a negative relationship between DNA concentrations from B. proteoclasticus strains and 18:0 at the duodenum was reported in earlier studies (Kim et al., 2008; Huws et al., 2010). No such relationship was observed in our research (II; III). It is possible, therefore, that unknown microbial species, not quantified here, are involved in FA biohydrogenation. Alternative- ly, direct inhibition of the biohydrogenation pathway, which does not lead to changes in the biohydrogenating community but alters the biohydrogenation products, could be the principal mode of action of the plant PUFA (Maia et al., 2007; Lourenço et al., 2010). 4.4. Lipid metabolism in the mammary gland 4.4.1. Effect of dietary fatty acids on milk production and composition Supplements of FO decreased yields of energy-corrected milk, milk fat and protein, and milk fat content (IV) confirming earlier observations of milk production characteristics in lactating cows (Donovan et al., 2000; Keady et al., 2000; Rego et al., 2005). It is well es- tablished that dietary FO supplements often lower milk yield, with the decrease being dependent on the amount (Donovan et al., 2000; Rego et al., 2005) and source of FO Natural resources and bioeconomy studies 74/2020 62 (Keady et al., 2000). Observed decrease in milk yield, together with depressed DMI (re- fer to chapter 4.2.2), in response to lipid supplements have often been attributed to the adverse effects of unsaturated FA on the growth of specific populations of rumen cellu- lolytic bacteria (Jenkins, 1993). Diets supplemented with 300 g FO/d lowered milk protein yield, whereas 150 g FO/d decreased milk protein content consistent with earlier findings in lactating cows (Offer et al., 1999; Donovan et al., 2000; Keady et al., 2000). In high amounts, FO tended to lower milk lactose secretion, whereas earlier studies have reported variable effects on milk lactose output (Offer et al., 1999; Loor et al., 2005a; Donovan et al., 2000; Keady et al., 2000). Increases in FO supplementation progressively decreased milk fat content and yield, which represents a typical response in lactating cows (Offer et al., 1999; Donovan et al., 2000; Keady et al., 2000; Rego et al., 2005) and is associated with the effects of long- chain PUFA in FO on ruminal lipid metabolism (Shingfield et al., 2003; Loor et al., 2005c). Compared with 0 g FO/d, milk fat yield and content were decreased by 40.6 and 30.1 %, respectively, on diets supplemented with 300 g FO/d. Specific responses on milk fat con- tent during FO-induced MFD are discussed more detail in chapter 4.4.3. 4.4.2. Fatty acids in milk Analysis of fractionated FAME and corresponding DMOX derivatives by GC-FID and GC- MS allowed 196 baseline-separated or coeluting FA to be detected in the milk fat GC-FID chromatogram as presented in IV. Of these, 174 were able to be identified, including 37 unique 20- and 22-carbon intermediates not previously reported in milk from cows fed FO. Overall, dietary FO supplements elevated milk PUFA, mono- and polyenoic trans FA concentrations (IV), and at high amounts altered the distribution of individual trans 16- to 22-carbon FA isomers (IV), which can be attributed to the incomplete biohydrogena- tion of unsaturated 16- to 22- carbon PUFA in the rumen (II). Earlier reports had relied on the same elution order in GC analysis and isomer com- position of partially hydrogenated plant oils to identify changes specifically in milk trans 18:2 isomers to FO (Shingfield et al., 2003; Loor et al., 2005a), but in the experiments presented in this thesis a combination of preparative Ag+-TLC of FAME and GC-MS analy- sis of FAME and corresponding DMOX derivatives was used to analyze milk fat composi- tion to avoid these uncertainties (IV). 4.4.2.1. Saturated fatty acids and cis-9 18:1 in milk Indirect comparison of dietary FO supplements on milk fat synthesis and milk FA compo- sition of lactating cows reported in the literature is presented in Table 5. In this thesis work, supplements of FO progressively decreased the relative proportions of short- and medium-chain SFA (4- to 16-carbon FA; IV) and 18:0 in milk (IV; Table 5) that were ac- companied by an increase in 14:0, 16:0 and 18:0 at the omasum (II; Figure 4). Previous investigations in lactating cows fed FO have reported similar decreases in the propor- Natural resources and bioeconomy studies 74/2020 63 tions of milk FA synthesized de novo (Donovan et al., 2000; Keady et al., 2000; Rego et al., 2005), which can be explained by the inhibitory effects of long-chain FA on ACACA activity and the de novo synthesis of short- and medium-chain SFA in mammary secreto- ry cells with the effects being more potent when the number of carbon atoms and/or the degree of unsaturation, especially the number of trans double bonds in the carbon chain, increases (Chilliard et al., 2000). It is also well established that FO lowers mamma- ry mRNA abundance for ACACA and FASN in lactating cows (Ahnadi et al., 2002). Fur- thermore, in our study dietary FO decreased cis 18:1 concentration in milk that were, for the most part, due to decreases in cis-9 18:1 (IV; Table 5). Observed decreases in milk 18:0 and cis 18:1 concentrations are consistent with previous findings in cows fed FO (Offer et al., 1999; Donovan et al., 2000; Rego et al., 2005), due to lowered availability of 18:0 for direct incorporation in milk fat or for endogenous synthesis of cis-9 18:1 in the mammary gland. Basal activity of SCD, evaluated using the milk 14:1/14:0 desaturation index as a proxy (Bernard et al., 2008), was not affected by FO inclusion (IV; Table 5). However, observed increases in cis-9 18:1/18:0 concentration ratio (IV; Table 5), without substan- tial changes in cis-9 18:1 at the omasum (II; Figure 5) could be considered as evidence that the alterations in the cis-9 18:1:18:0 ratio may potentially arise from changes in SCD substrate specificity. Reports of FO having a direct effect on SCD transcription are equiv- ocal. Inclusions of FO (15 g/kg DM) and soybean oil (30 g/kg DM) have been reported to decrease (Harvatine and Bauman, 2006) or increase (Invernizzi et al., 2010) mRNA abun- dance for SCD1 in mammary gland tissue. When added to the diet or fed as rumen pro- tected form, FO tended to decrease mammary SCD mRNA abundance (Ahnadi et al., 2002), whereas SCD1 transcription was more recently demonstrated to be unaffected by abomasal infusion of FO (Dallaire et al., 2014). Dietary FO inclusion had rather small or no effect on the milk fat OBCFA concentra- tions (IV) consistent with observed OBCFA flows at omasum (II) and confirmed recent observations in lactating sheep fed marine lipids (Toral et al., 2018b). However, these findings are questioned by some other investigations that have reported also increases (Toral et al., 2015) or decreases (Toral et al., 2018c) in milk OBCFA in lactating cows or sheep, respectively. The inconsistency between these studies may reflect changes in duodenal flow of rumen bacteria, because FA synthesis by rumen bacteria is considered to be the main source of OBCFA in milk fat (Vlaeminck et al., 2006). 4.4.2.2. 16-carbon fatty acids in milk Dietary FO supplements increased milk fat cis 16:1 and trans 16:1 concentrations, with the majority of the increase due to enrichment of cis-9 16:1 and trans (∆9-12) 16:1. In addition, increases in 10-O-16:0 and all identified milk 16:2 isomers were observed. These findings are in line with previous investigations in lactating ruminants fed diets supplemented with FO (Toral et al., 2015; 2018b). Increases in 16:1 (II; Figure 4) and 16:2 isomers at the omasum (II) were attributed to incomplete biohydrogenation of Natural resources and bioeconomy studies 74/2020 64 16:2n-4, 16:3n-4, 16:4n-1, and 16:4n-3 reflecting the distribution of 16-carbon interme- diates in milk (Destaillats et al., 2000). 4.4.2.3. 18-carbon fatty acids in milk Incremental amounts of FO in the diet had no effect on milk 18:2n-6 or 18:3n-3 (Ta- ble 5), but altered the distribution of non-conjugated 18:2 and 18:1 FA intermediates in milk fat, changes characterized by substantial increases in the total concentrations of 18:2 isomers containing one or two trans double bonds (0.85, 1.23, 1.91 and 3.73 g/100g FA for 0, 75, 150 and 300 g FO/d, respectively) and trans 18:1 isomers (Table 5) confirming earlier findings in lactating cows (Chilliard et al., 2001; Shingfield et al., 2003; Loor et al., 2005a; Toral et al., 2015). Most of the increase in total milk trans FA content to FO (52.3, 77.4, 122 and 115 g/d, respectively; IV) was associated with specific enrich- ment of trans 18:2 (Δ9,11; Δ9,12; Δ11,15) and trans 18:1 (Δ8–12) isomers (IV) which can be attributed to the incomplete biohydrogenation of unsaturated 18-carbon FA in the rumen (II; Table 3). Previous studies in vitro have demonstrated that both 20:5n-3 and 22:6n-3 inhibit the reduction of 18-carbon unsaturated FA to 18:0 causing trans 18:1 and trans 18:2 intermediates to accumulate (AbuGhazaleh and Jenkins, 2004; Klein and Jen- kins, 2011). Complementary Ag+-TLC fractionation of FAME and GC-MS analysis of DMOX deriva- tives indicated that FO results in the appearance of 21 minor trans 18:2 isomers in milk not reported previously (IV). Even though separation using the CP-Sil 88 capillary column was not possible, further work in our laboratory based on GC-MS and GC-FID analysis with the 100-m SLB-IL111 capillary column (Alves and Bessa, 2014) has allowed trans- 10,cis-15 18:2 to be isolated in milk fat of cows fed 200 g of FO/d alone or in combina- tion with 500 g of SO or LO/d (Kairenius et al., unpublished). Observed increases in milk trans-10,cis-15 18:2 secretion (0.00, 0.18, 1.28 and 5.99 g/d for control, FO, SFO and LFO, respectively) in experiment 3 (Kairenius et al., unpublished), have provided further evidence that FO alters the biohydrogenation of 18-carbon PUFA, and plant-derived oil supplements can be used to increase the ruminal outflow of spesific 18-carbon PUFA intermediates (III) available for absorption and incorporation into milk and meat fat. Incremental supplementation of FO elevated milk fat cis-9,trans-11 CLA content in a quadratic manner, reaching a maximum on the level of 150 g FO/d (from 0.61 to 2.15 g/100 g of FA for 0 and 150 g/d of FO, respectively) but not on 300 g FO/d supplementa- tion level. This was a consequence of the apparent alterations in ruminal biohydrogena- tion on the diet supplemented with 300 g FO/d, which caused trans-10 18:1 to accumu- late with no change in trans-11 18:1 leaving the rumen (II; Figure 5; Table 3). The sup- plementation level of 300 g FO/d further increased trans-10 18:1 (from 1.06 to 4.20 g/100 g of FA for 150 and 300 g/d FO, respectively) with no change in trans-11 18:1 concentration (from 5.31 to 5.46 g/100 g of FA, respectively) (Table 5). Increases in milk cis-9,trans-11 CLA content were consistent with previous findings in lactating cows (Of- fer et al., 1999; Donovan et al., 2000; Shingfield et al., 2003; Rego et al., 2005) and can Natural resources and bioeconomy studies 74/2020 65 be explained by an increase in the supply of trans-11 18:1 (II; Figure 5; Table 3) for en- dogenous cis-9,trans-11 CLA synthesis in the mammary gland in 150 g/d of FO supple- mentation level (Griinari et al., 2000). Trans-7,cis-9 CLA was not detected in omasal digesta (II; Table 3) but it increased in milk reaching a maximum on the level of 150 g FO/d (from 0.80 to 1.05 g/100 g of FA for 0 and 150 g/d FO, respectively; IV), which can be explained by greater amounts of trans- 7 18:1 at the omasum (II; Table 3) being used for endogenous synthesis of trans-7,cis-9 CLA in the mammary gland (Corl et al., 2002). Secretion of cis-9,trans-12 18:2, cis- 9,trans-13 18:2, and cis-9,trans-14 18:2 in milk (IV) also exceeded the flow at the oma- sum (II; Table 3), providing further evidence that these isomers are formed via the action of SCD on trans (∆12-14) 18:1 in the bovine mammary gland (Griinari et al., 2000; Shing- field et al., 2008b). For all treatments, trans-11 18:1 was quantitatively the most important 18:1 isomer in milk fat, accounting for proportionately 0.35, 0.40, 0.48, and 0.40 of total trans 18:1 for 0, 75, 150 and 300 g/d of dietary FO, respectively, although at high amounts, FO in- creased milk trans-10 18:1 concentration several-fold as noted above (up to 0.31 of total trans 18:1 for 300 g FO/d; IV). Both of these observations are in close agreement with other reports in lactating cows fed FO (e.g. Loor et al., 2005a; Palmquist and Griinari, 2006; Toral et al., 2015). Specific effects of FO on milk trans-10 18:1 content, and the level of other trans-10 containing FA isomers in milk of cows experiencing FO-induced MFD are discussed more detail in chapter 4.4.3. In the experiment presented in IV, cows fed FO produced milk containing higher concentrations of FA containing a hydroxyl or oxo group located on carbons 9 and 10 relative to the carboxyl group, including 10-OH-18:0 and 10-O-18:0, but no other signifi- cant changes in the amounts of 13- and 15-O-18:0 in milk were observed. The enrich- ment of 10-O-18:0 in FO-milk is consistent with previous findings in lactating ruminants fed rations supplemented with marine lipids (Bichi et al., 2013; Toral et al., 2014; 2015; 2018b), but the number of in vivo experiments reporting concentrations of oxygenated 18-carbon FA in bovine milk with diet-induced MFD is limited (IV; Toral et al., 2015; Leskinen et al., 2019). 4.4.2.4. 20- to 22-carbon fatty acids in milk Dietary FO supplements enriched 20:5n-3, 22:5n-3 and 22:6n-3 in milk (IV; Table 5) and resulted in the appearance of multiple 20- to 22-carbon FA in milk (IV). Increases in milk long-chain n-3 PUFA concentrations are consistent with several previous reports in lac- tating cows fed FO under different feeding strategies (Table 5). However, the main find- ings in most of these reports are related to the FO-induced changes in the distribution of 18-carbon FA in milk of cows fed FO (e.g. Shingfield et al., 2003; Loor et al., 2005a), and the number of investigations reporting the abundance of cis and trans polyenoic FA in milk is limited (IV; Toral et al., 2015). Natural resources and bioeconomy studies 74/2020 66 Milk from cows fed incremental levels of FO contained cis-14 20:1, trans (∆9-15) 20:1, positional and geometric isomers of 20:2 (n = 9) and 20:3 (n = 7), and an unusual 20:4 intermediate not contained in FO (IV). Dietary FO supplements increased also milk 20:2n-6, 20:3n-3, 20:4n-6, 20:4n-3, cis-11 22:1 and cis-13 22:1 concentrations and the abundance of 22:2 (n = 2), 22:3 (n = 2), and 22:4 (n = 3) isomers containing at least one trans double bond. In total, the appearance of 56 novel 20-, 21-, or 22-carbon interme- diates containing at least a single trans double bond and several all-cis long-chain un- saturated FA were found to be increased in milk from cows fed FO (IV), consistent with other findings in lactating cows (Toral et al., 2015). No conjugated 20-, 21-, and 22- carbon FA were detected in milk (IV; Toral et al., 2015), consistent with an absence of these intermediates in omasal digesta of experimental cows offered FO-supplemented diets (I-III). Incremental amounts of FO resulted in dose-dependent increases in 20- and 22- carbon trans FA concentrations (range 0.16–1.61 and 0.01–0.28 g/100 g of FA, respec- tively) in milk, with trans 20:1, trans 20:2 and trans 20:3 isomers being the most abun- dant (0.18, 0.27, and 0.17 g/100 g of FA, respectively), confirming other observations in cows fed FO (Toral et al., 2015). A close relationship between flow at the omasum (II) and output in milk (IV) suggests that most, if not all, of the unusual minor 20- to 22- carbon FA in milk from cows fed FO originated from the rumen rather than elongation and desaturation of FA in nonmammary tissues or possible endogenous synthesis in the mammary gland. However, possible benefits of milk 20- to 22-carbon FA to human health need to be considered in the context of the overall changes in milk FA to FO. 4.4.2.5. Transfer efficiency of long-chain n-3 fatty acids from feed to milk The transfer efficiency of long-chain PUFA from the diet into milk was determined by evaluating the slope of regression of the amount of 20:5n-3, 22:5n-3 and 22:6n3 in milk versus intake. Supplements of FO elevated milk 20:5n-3, 22:5n-3 and 22:6n-3 concentra- tions in a dose-dependent manner, increases that were associated with a mean appar- ent transfer efficiency of 1.3, 8.1, and 1.8 %, respectively, being consistent with values in the literature, ranging between 1.4-3.3 and 1.9-4.8 % for 20:5n-3 and 22:6n-3, respec- tively (Lock and Bauman, 2004; Rego et al., 2005; Palmquist and Griinari, 2006; Toral et al., 2015), but much lower than the apparent transfer efficiency of 25.5 % for 22:5n-3 reported for lactating cows fed FO (Toral et al., 2015). Typically transfer efficiencies re- ported in the literature are calculated from concentrations of the individual FA in diet and milk, but also estimates evaluated by regression are reported, being 5.7 and 0.98 % for 20:5n-3 and 22:6n-3, respectively (Palmquist and Griinari, 2006). Even though it is generally accepted that apparent transfer of 20:5n-3, 22:5n-3 and 22:6n-3 is rather low, reported literature estimates are subject to considerable variation (Palmquist and Gri- inari, 2006). Dietary FO supplements increase 20:5n-3 and 22:6n-3 concentrations in ruminant milk fat and tissue lipids (Palmquist, 2009), but enrichment is limited. The rather low Natural resources and bioeconomy studies 74/2020 67 apparent transfer of 20:5n-3 and 22:6n-3 from the diet into milk in this, and other re- ports in lactating cows fed dietary FO supplement, can in the most part be attributed to the extensive biohydrogenation of these highly unsaturated FA in the rumen (Shingfield et al., 2013). However, the difference in the enrichment of these long-chain n-3 PUFA in milk also reflects the preferential incorporation of PUFA into plasma CE and PL instead of TAG, which is the primary source of FA for milk fat synthesis (Shingfield et al., 2013). It is also well established that the use of different feeding strategies to increase milk fat n-3 PUFA content via conversion of dietary 18:3n-3 to very long-chain n-3 PUFA, such as 20:5n-3 and 22:6n-3, has little or no effect on milk fat levels of 20:5n-3 and 22:6n-3 (Palmquist, 2009) because of limited activity of ∆5- and ∆6-desaturase, enzymes respon- sible for this conversion, and elongase in the mammary gland of lactating cows (Hage- meister et al., 1991). A higher efficiency of transfer for 22:5n-3 compared with 20:5n-3 or 22:6n-3, is postulated to be in part a result of the potential elongation of 20:5n-3 to 22:5n-3 and retroconversion of 22:6n-3 to 22:5n-3 in body tissues (Palmquist, 2009). 4.4.3. Milk fat depression Dietary FO decreased milk fat synthesis and milk fat content up to -40.6 and -30.1 %, respectively (IV; Table 5) consistent with earlier findings in lactating cows fed FO (Ta- ble 5) or marine products (Ahnadi et al., 2002) alone or in combination with plant lipids (Invernizzi et al., 2010; Angulo et al., 2012). Incremental levels of FO decreased progressively the proportions of short- and me- dium chain SFA and their secretion in milk (IV; Table 5). However, oil supplements dec- areased (II) or had no effect (III) on rumen VFA concentrations, but at high amounts FO promoted an increase in molar propionate and butyrate proportions at the expense of acetate (II). These alterations in rumen function may have contributed to the observed differences in the proportions of short- and medium-chain SFA in milk fat (IV). Plasma acetate and BHBA were not analysed in our experiments, but earlier investigations have reported changes in the mammary uptake of these metabolites in cows fed FO (Loor et al., 2005a) and high-concentrate diets (Loor et al., 2005b). However, possible changes in rumen fermentation, e.g. shift in the rumen VFA profile toward less acetate and more propionate, plasma concentrations and mammary uptake of acetate and BHBA, and reductions in de novo FA synthesis cannot fully explain reductions in milk fat synthesis (Bauman and Griinari, 2003; Loor et al., 2005b). To explore in more detail the possible mechanisms underlying the FO-induced MFD in lactating cows, relationship between FA flow at the omasum and milk FA output was investigated, but ruminal outflows of trans-9,cis-11 CLA (II; Table 3) and trans-10,cis-12 CLA (II; Table 3) did not explain the lowered milk fat synthesis (IV). Decreases in milk fat output in response to FO were not either associated with increases in milk trans-10,cis- 12 CLA (IV; Table 5), in agreement with earlier findings reporting extremely low concen- trations of milk trans-10,cis-12 CLA in response to FO (Loor et al., 2005a). However, these observations are challenged by a recent meta-analysis that investigated correla- Natural resources and bioeconomy studies 74/2020 68 tion between MFD and milk fat proportions of multiple biohydrogenation intermediates, and reported the highest correlation of MFD with trans-10 18:1 and trans-10,cis-12 CLA (Conte et al., 2018). No cis-10,trans-12 CLA was detected in omasal digesta (I; II; Table 3) or in milk (IV), consistent with other findings in lactating cows experiencing diet-induced MFD (Shingfield et al., 2003; Ventto et al., 2017; Leskinen et al., 2019). Previous investigations in vivo have also demonstrated the potential role of trans- 10,cis-15 18:2 in MFD (Alves and Bessa, 2014; Ventto et al., 2017; Leskinen et al., 2019), although no direct measurements of the potential antilipogenic effects of trans-10,cis- 15 18:2 on milk fat synthesis have been reported. However, further studies using mice adipocyte cell cultures have shown that trans-10,cis-15 18:2 isolated from beef fat does not exert same anti-adipogenic properties as trans-10,cis-12 CLA (Vahmani et al., 2016). Trans-10,cis-15 18:2 co-eluted with trans-11,cis-15 18:2 on the CP-Sil 88 column used for GC-analysis of FA in omasal digesta (I; II; Table 3) and milk fat (IV; Table 5), and no defini- tive conclusions of the incremental levels of dietary FO on trans-10,cis-15 18:2 for- mation can be drawn based on the results of this thesis. However, the appearance of trans-10,cis-15 18:2 in the omasal digesta, as well as increased omasal flow of trans- 10,cis-15 18:2 in response to inclusion of additional level of 200 g FO/d alone or in com- bination with plant oils (0.00, 0.81, 5.20 and 35.8 g/d for control, FO, SFO and LFO, re- spectively; III) demonstrates the potential involment of ruminal trans-10,cis-15 18:2 formation in lactating cows fed diets often associated with MFD (III). In support of this, the milk fat synthesis (-16.7, -41.4, and -48.7 %) and milk fat content (-11.8, -19.3, and - 27.7 %) decreased in experiment 3 for FO, SFO and LFO compared with diet containing no oil, respectively (Kairenius et al., unpublished). The inhibitory effects of dietary FO supplementation on milk fat synthesis were as- sociated with a remarkable increase in trans-10 18:1 at the omasum from control diet to increasing levels of FO (II; Table 3) and elevated trans-10 18:1 concentrations in milk fat, respectively, with the greatest increase to the additional level of 300 g FO/d (IV; Ta- ble 5). These findings are in agreement with previous reports supporting the potential role of trans-10 18:1 in FO-induced MFD (Bauman and Griinari, 2003; Shingfield and Griinari, 2007; Conte et al., 2018). Across all treatments, a close negative association existed between trans-10 18:1 at the omasum and milk fat secretion (IV), which may be considered evidence that increased ruminal formation of trans-10 18:1 contributes to FO-induced MFD in lactating cows. Even though increases in milk trans-10 18:1 content is a consistent feature of diet- induced MFD in lactating cows (Bauman and Griinari, 2003; Shingfield and Griinari, 2007; Conte et al., 2018), previous reports on the physiological effects of trans-10 18:1 in the lactating cow are inconsistent. Abomasal infusion of 42.6 g/d of trans-10 18:1 was shown to have no influence on milk fat secretion (Lock et al., 2007), findings that have been challenged on the basis that the enrichment of trans-10 18:1 in milk during postruminal infusions was too low (1.11 g/100 g FA; Lock et al., 2007) for possible ef- fects on mammary lipogenesis to be detected (Kadegowda et al., 2008). Further investi- gations have demonstrated that postruminal infusion of a mixture of 18:1 methyl esters Natural resources and bioeconomy studies 74/2020 69 supplying 92.1 g/d of trans-10 18:1 over a 5-d period induced an approximate 20 % de- crease in milk fat yield, associated with a higher proportion of trans-10 18:1 in milk (on average 4.37 g/100 g FA; Shingfield et al., 2009). Even though a direct cause and effect could not be established, comparisons with reports in the literature and the relative abundance of constituents in the methyl ester preparation infused implicated trans-10 18:1 as the isomer responsible. In most cases a dramatic deacrease in milk fat yield (even up to -50%; Invernizzi et al., 2010) is associated with a lower expression of most of the lipogenic genes involved in bovine mammary de novo synthesis, uptake of preformed long-chain FA, and TAG synthesis and lower expression of transcription factor SREBP1 that is a regulatory ele- ment of genes involved in major mammary lipogenic pathways (Bernard et al., 2018). Incubations of mammary epithelial cells with trans-10 18:1 have been shown to de- crease lipogenic gene expression of FASN, SCD, and SREBF1 (Kadegowda et al., 2009), suggesting that trans-10 18:1 may be active in the mammary gland. In support of this, a close negative association between milk fat secretion and concentration of trans-10 18:1 in milk was observed in the loading plots for correlations between milk FA composition and milk fat content and yield in our study (IV). Decreases in milk fat secretion with increasing FO supplementation were also asso- ciated with an increase in milk fat cis-11 18:1 concentration (from 0.46 to 1.27 g/100 g of FA) consistent with a negative relationship between these parameters reported pre- viously for cows with FO-induced MFD (Gama et al., 2008; Toral et al., 2015). Cis-11 18:1 is a component of bacterial and dietary lipids and can be synthesized from the elonga- tion of cis-9 16:1, and the enrichment of cis-11 18:1 in milk on FO treatments can be explained by an increase in cis-11 18:1 at the omasum (from 2.53 to 10.5 g/d; II). Earlier experiments established that FO promotes ruminal escape of cis-11 18:1 (I; Shingfield et al., 2003), but the biological activity of cis-11 18:1 as a potent milk fat inhibitor in rumi- nants is equivocal. Incubations with cis-11 18:1 were shown to decrease lipogenesis in bovine adipocytes (Burns et al., 2012), whereas postruminal infusion of a mixture of 18:1 isomers supplying up 12.4 g/d of cis-11 18:1 in cows increased concentrations of this isomer in milk from 0.59 to 1.50 g/100 g of FA but had no effect on milk fat synthesis (Shingfield et al., 2007). During FO-induced MFD, decreases in milk fat synthesis are accompanied by low- ered proportions of 18:0 and cis-9 18:1 in milk fat and higher trans 18:1 concentrations (Offer et al., 1999; Donovan et al., 2000; Table 5). Several reports have suggested that decreases in the availability of 18:0 combined with an increase in trans 18:1 isomers, as well as the regulation of TAG synthesis to maintain milk fat fluidity, may explain, or at least contribute to, the decrease in milk fat synthesis (Loor et al., 2005a; Gama et al., 2008; Toral et al., 2018a,b). It is well established that trans FA have higher melting points than their cis counterparts, being thus more similar to SFA from the physical and functional point of view. In addition, the melting point of FA depends on the chain length and the degree of unsaturation (Ratnayake and Galli, 2009). The larger the num- Natural resources and bioeconomy studies 74/2020 70 ber of carbon atoms, the higher the melting point, and the larger the number of double bonds, the lower the melting point. However, in high amounts FO had no effect on calculated mean milk fat melting point (IV; Table 5), despite of the substantial changes in milk FA composition (IV). This is in a good agreement with previous findings in lactating ruminants fed FO (Toral et al., 2013; 2018a), but in contrast with some other reports in lactating cows demonstrating that dietary FO supplements may also increase (Gama et al., 2008) or decrease (Toral et al., 2015) calculated mean milk fat melting point. However, the inconsistency between our results and previous findings in the literature is difficult to explain. As previous me- ta-analysis concluded, the variations in milk fat melting point in response to a wide range of dietary lipid supplements are much smaller than the variation in milk FA com- position, indicating that the maintenance of milk fat melting point within a normal phys- iological range has a role in the regulation of milk fat synthesis during diet-induced MFD (Toral et al., 2013). Activity of SCD in the mammary gland of ruminants is thought to serve as a mecha- nism to maintain and regulate the fluidity properties of milk fat for efficient secretion from the mammary gland by reducing the melting point of FA, and SCD activity is often estimated by calculating desaturation indexes that are based on the product/substrate FA ratios for the relevant desaturase FA pairs, cis-9 14:1/14:0, cis-9 16:1/16:0 and cis-9 18:1/18:0, in particular. In this research, dietary FO supplements progressively increased milk fat cis-9 18:1/18:0 concentration ratios (by 5.17, 16.7, and 24.7 % compared to control; IV; Table 5), without substantial changes in cis-9 18:1 at the omasum (II; Fig- ure 5), but not cis-9 14:1/14:0 ratios (IV; Table 5), a proxy used for mammary SCD activi- ty (Bernard et al., 2008). This, together with a shortage of available 18:0 (II; Figure 4), caused by the inhibition of the last step of ruminal biohydrogenation, indicates that substrate specifity of SCD may alter FO-induced MFD. In addition, more recently, studies conducted with lactating sheep, have demon- strated that dietary supplementation of 18:0 with FO, regardless of the 18:0 doses, were not able to reverse the negative effect of FO on MFD (Toral et al., 2018a,b). Although the reasons underlying the lack of response to dietary 18:0 are uncertain, these findings suggest that changes in SCD-activity, supply of 18:0 for ∆9-desaturation or milk fat melt- ing point may not, at least alone, explain FO-induced MFD. Cows experiencing FO-induced MFD produced milk containing higher concentra- tions of 10-OH-18:0, 10-O-16:0 and 10-O-18:0 (refer to chapter 4.4.2.3). Partial least square regression analysis indicated that a positive association existed between FO- induced MFD and the milk fat proportions of oxygenated 18-carbon FA, particularly of 10-oxo-18:0 (IV; Table 5), consistent with other findings in lactating ruminants (Bichi et al., 2013; Toral et al., 2015; Frutos et al., 2018). Although marine lipids seem to enrich the concentration of oxygenated FA in milk of ruminants there is no conclusive evidence to establish a direct cause and effect of these FA metabolites on mammary lipogenesis. However, all these findings together suggest the potential involvement of oxygenated 18-carbon FA, with 10-oxo-18:0 in particular, in FO-induced MFD, but the association Natural resources and bioeconomy studies 74/2020 71 between FO-induced MFD and 13- and 15-oxo-18:0 seems to be less relevant (IV; Toral et al., 2015; Frutos et al., 2018). Further exploration of the data, mainly associations between the flow of selected 16- to 22-carbon FA at the omasum to FO treatments (II) and milk fat yield (IV) provided further insight into the possible causes for FO-induced MFD and varying effects of FO on milk FA composition in lactating cows. In the loading plots for correlations between milk FA composition and milk fat yield and content (IV), several close negative associations were identified between milk fat secretion and concentrations of multiple long-chain 16-, 18-, 20-, and 22-carbon PUFA containing one or more trans double bonds. This indi- cates that trans-10,cis-12 CLA alone cannot explain FO-induced MFD. In addition, cows fed FO and experiencing MFD produced milk containing higher concentrations of other FA containing a trans double bond or a hydroxy or oxo group located on carbons 9 and 10 relative to the carboxyl group that included trans-10 16:1, trans-10,trans-14 16:2, unresolved trans-10,cis-15 and trans-11,cis-15 18:2, trans- 10,trans-12 CLA, trans-10,cis-12 CLA, trans-10,trans-14,trans-17 20:3 and 10-OH-18:0 (IV). Furthermore, trans-8,trans-10 CLA, trans (∆9,10) 20:1, and trans-10,trans-16 20:2 were located close to trans-10 18:1, but vertically opposite in the correlation loading plot (IV). Overall, the PLS analysis suggested that FO-induced MFD may arise from changes in the concentrations of multiple FA in milk, including a decrease in 18:0 supply in combination with increased mammary uptake of cis-11 18:1, trans-10 18:1 and mono and polyenoic trans 20- and trans 22-carbon FA (IV). Natural resources and bioeconomy studies 74/2020 Table 5 Indirect comparison of dietary fish oil supplements on milk fat synthesis and milk fatty acid composition of lactating cows Reference Shingfield et al., 20031 Rego et al., 20052 Loor et al., 2005a3 IV 4 Gama et al., 20085 Toral et al., 20156 C FO250 C FO160 FO320 C FO270 C FO75 FO150 FO300 C FO200-HF FO200- LF C FO420 Fish oil source none herring + mackerel none sardine sardine none menhaden none herring + mackerel herring + mackerel herring + mackerel none salmon salmon none anchovy Fish oil inclusion rate, g/d 0 250 0 160 320 0 270 0 75 150 300 0 200 200 0 420 Forage inclusion rate, g/kg DM 600 600 ad lib ad lib ad lib 660 660 580 580 580 580 LF HF LF 400 400 Duration, d 14 14 28 28 28 28 28 28 28 28 28 21 21 21 26 26 Animals per group 5 5 4 4 4 6 6 4 4 4 4 4 4 4 4 4 Milk fat content, g/kg 46.0 42.8 34.8 29.9 23.4 35.4 25.1 41.2 38.9 33.0 28.8 38.6 28.5 26.2 33.4 23.4 Response7 -6.96 -16.4 -32.8 -29.1 -5.58 -19.9 -30.1 -26.2 -32.1 -29.9 Milk fat yield, g/d 788 602 920 780 580 783 567 997 945 846 592 730 450 440 992 706 Response7 -23.6 -15.2 -37.0 -27.6 -5.22 -15.1 -40.6 -38.4 -39.7 -28.8 Milk fat melting point, °C - - - - - - - 38.7 38.4 38.7 38.3 36.1 38.4 37.2 37.3 34.2 Fatty acid profile (g/100 g fatty acids) 4:0 4.58 2.42 3.90 3.13 3.14 1.91 1.83 3.30 3.47 3.17 2.86 - - - 3.39 3.86 6:0 2.23 1.66 1.06 0.95 0.79 1.92 1.75 2.05 2.02 1.77 1.59 - - - 2.38 1.95 8:0 1.11 1.08 0.86 0.70 0.59 1.28 1.23 1.25 1.20 1.05 0.94 - - - 1.30 1.00 10:0 2.22 2.81 1.92 1.54 1.36 3.43 3.54 2.47 2.27 2.06 1.88 - - - 2.80 2.15 ∑4:0,6:0,8:0, 10:0 10.1 7.97 7.74 6.32 5.88 8.54 8.35 9.07 8.96 8.05 7.27 9.60 9.15 8.60 9.87 8.96 12:0 2.40 3.39 2.48 2.02 1.77 4.27 4.49 3.39 3.07 2.88 2.72 - - - 3.03 2.42 14:0 10.2 13.3 9.47 9.36 8.84 13.9 14.8 12.0 11.6 11.4 10.9 - - - 11.2 10.1 16:0 24.7 33.3 23.2 23.1 22.6 34.2 31.8 30.1 29.7 28.2 27.1 - - - 30.1 25.4 ∑12:0,14:0,16:0 37.3 50.0 35.2 34.5 33.2 52.4 51.1 45.5 44.4 42.5 40.7 48.8 56.0 48.8 44.3 37.9 18:0 19.5 4.43 12.0 10.4 7.68 8.69 2.71 10.1 8.90 6.06 4.25 7.88 2.03 2.49 10.2 3.17 10-O-18:0 - - - - - - - 0.11 0.15 0.67 0.73 - - - 0.03 0.75 13-O-18:0 - - - - - - - 0.02 0.02 0.04 0.03 - - - 0.02 0.03 c9-18:1 18.1 4.84 23.6 20.1 15.7 15.8 6.05 17.5 16.2 12.2 8.76 22.3 8.87 10.8 16.8 7.56 72 Natural resources and bioeconomy studies 74/2020 Reference Shingfield et al., 20031 Rego et al., 20052 Loor et al., 2005a3 IV 4 Gama et al., 20085 Toral et al., 20156 C FO250 C FO160 FO320 C FO270 C FO75 FO150 FO300 C FO200-HF FO200- LF C FO420 ∑trans 18:1 4.5 14.4 5.92 8.53 12.0 3.01 13.8 3.91 6.13 11.1 13.5 - - - 3.34 14.5 t10 0.21 1.01 - - - 0.27 1.76 0.34 0.46 1.06 4.20 0.65 2.57 3.76 0.42 4.10 t11 1.80 9.39 - - - 1.08 9.17 1.37 2.43 5.31 5.46 1.92 8.60 11.4 1.25 7.17 ∑CLA 0.56 1.85 2.25 3.23 3.63 0.63 3.36 0.76 1.20 2.38 2.26 - - - 0.72 2.83 c9,t11 0.39 1.66 - - - 0.56 3.20 0.61 1.03 2.15 2.07 0.67 2.74 3.21 0.59 2.56 t9,c11 - - - - - 0.01 0.04 - - - - - - - 0.02 0.10 t10,c12 <0.01 <0.01 - - - - - <0.01 <0.01 <0.01 <0.01 <0.01 0.01 0.02 <0.01 0.01 t9,t11 0.010 0.014 - - - 0.03a 0.08a 0.01 0.02 0.02 0.02 - - - 0.02 0.01 t10,t12 <0.01 0.01 - - - - - <0.01 <0.01 <0.01 0.01 - - - <0.01 0.010 t11,c15 + t10,c15 18:28 0.19 1.56 - - - 0.07 1.01 0.24 0.34 0.64 2.04 - - - 0.11 1.32 18:2n-6 0.90 1.25 2.51 1.99 0.65 1.88 1.36 1.22 1.18 1.21 1.17 3.08 1.62 2.24 2.28 1.96 18:3n-3 0.42 0.45 0.99 1.06 1.03 0.28 0.31 - - - - 0.19 0.17 0.20 0.65 0.56 20:5n-3 0.05 0.11 0.07 0.18 0.33 0.08 0.36 0.06 0.06 0.07 0.17 0.05 0.30 0.27 0.10 0.47 22:5n-3 0.0 0.22 0.07b 0.17b 0.36b 0.14 0.40 0.09 0.08 0.10 0.18 - - - 0.16 0.44 22:6n-3 0.0 0.10 0.06 0.17 0.43 0.04 0.17 0.03 0.03 0.05 0.10 0.11 0.56 0.54 0.05 0.27 Fatty acid ratios c9-14:1/ 14:0 0.06 0.04 - - - 0.08 0.07 0.10 0.10 0.10 0.11 0.06 0.06 0.08 0.08 0.08 c9-18:1/ 18:0 0.94 1.18 - - - 1.82 2.23 1.74 1.83 2.03 2.17 2.88 4.54 4.47 1.73 2.61 c9,t11 CLA/t11- 18:1 0.23 0.18 - - - 0.52 0.23 0.45 0.44 0.41 0.39 0.39 0.31 0.31 0.41 0.37 t10-18:1/t11-18:1 0.12 0.11 - - - 0.25 0.19 0.25 0.19 0.20 0.77 0.34 0.30 0.33 0.34 0.57 1Dry matter intake (DMI) 11.0 and 9.33 kg/d for C and FO250 treatments, respectively; 2DMI not reported; 3DMI 19.8 and 16.2 kg/d for C and FO270 treatments, respectively;4DMI 19.2, 18.9, 18.3 and 16.0 kg/d for treatments FO0, FO75, FO150 and FO300, respectively; 5DMI 19.8, 13.3 and 11.4 kg/d, for treatments C, FO200-HF and FO200-LF, respectively, 6DMI 10.2 and 8.52 kg/d for C and FO420 treatments, respectively; 7Response calculated as [(treat- ment – control)/control]*100; 8Trans-11,cis-15 18:2 co-elutes with trans-10,cis-15 18:2 in IV and Toral et al., 2015, but the possible co-elution not report- ed in Shingfield et al., 2003 or Loor et al., 2005a; ad lib, cows at pasture; C, control diet; CLA, conjugated linoleic acid; FO, fish oil; HF, high fibre diet con- taining 40 % NDF; LF, low fibre diet containing 26 % NDF; aIncludes also trans-8,trans-10 CLA and trans-10,trans-12 CLA isomers; bOnly 22:5 reported; For clarity purposes, abbreviated names of fatty acids are reported in the table (c, cis; t, trans). 73 Natural resources and bioeconomy studies 74/2020 74 5. Conclusions Based on the results of this research the conclusions and appropriate implications are following: i) Dietary FO with or without plant oil supplements modified ruminal biohydro- genation of 16- and 18-carbon unsaturated FA, causing dose-dependent in- creases in trans 16:1, trans 18:1 and trans 18:2 flow and a concomitant decrease in 18:0 at the omasum, and at high amounts promoted trans-10 18:1 accumula- tion at the expense of trans-11 18:1. Supplements of FO increased also the flow of 20:5n-3, 22:5-3 and 22:6n-3 at the omasum and resulted in corresponding in- creases in numerous 20- and 22-carbon unsaturated FA containing one or more trans double bonds at the omasum, providing clear evidence of extensive me- tabolism of 20:5n-3 and 22:6n-3 in the rumen of lactating cows. ii) No conjugated 20-carbon FA were detected in experimental digesta samples, which suggests that biohydrogenation of long-chain PUFA does not involve for- mation of intermediates containing a conjugated double bond system. Never- theless, the hydrogenation of 20:5n-3, 21:5n-3 and 22:6n-3 in the rumen pro- ceeds via two principal mechanisms that involve sequential reduction or isomer- isation of cis double bonds closest to carboxyl group. iii) The plant derived 18-carbon PUFA influenced the ruminal metabolism of long- chain n-3 PUFA in the rumen of lactating cows. The biohydrogenation of cis-9 18:1, 18:2n-6 and 18:3n-3 in cows fed FO diets with or without of plant oils was higher than biohydrogenation of these FA originating from ingredients of a con- trol diet, being even greater when plant oils were fed with FO. The ruminal me- tabolism of 22:6n-3 was more extensive on diets containing higher amounts of 18-carbon PUFA, whereas the biohydrogenation of 22:5n-3 showed no differ- ence between FO and diets containing plant oils. Ruminal outflow of 20:5n-3 was not altered when plant oils were added to FO. Supplements of FO plus plant oils shifted the ruminal biohydrogenation towards a higher production of trans- 10 18:1 at the expense of trans-11 18:1. iv) The inhibitory effects of FO on the reduction of 18-carbon FA to 18:0 were influ- enced by the relative amounts of 18:2n-6 and 18:3n-3 in the diet. Despite of a similar intake of 18-carbon PUFA and similar flow of trans 18:1, the flow of 18:0 at the omasum was lower and accumulation of trans 18:2 and 20- to 22-carbon FA intermediates greater for LFO than SFO. Supplementing FO with sources of 18:2n-6 or 18:3n-3 caused trans-10 and trans-11 18:1 to accumulate, and, on the SFO treatment, trans-10 18:1 was the most abundant biohydrogenation in- termediate escaping the rumen. Natural resources and bioeconomy studies 74/2020 75 v) Alterations in the amount of FA intermediates at the omasum or ruminal biohy- drogenation pathways were not associated with substantial changes in rumen protozoal counts or analysed bacterial populations known to be capable of bio- hydrogenation, but lowered Butyrivibrio spp. numbers in response to incremen- tal levels of FO. vi) Detailed analysis of lipid in omasal digesta and milk fat of cows fed FO enabled the structure identification of 27 and 37, respectively, previously unidentified 20- to 22-carbon FA intermediates containing at least one trans double bond, and the detection of cis-14 20:1, 20:2n-3, 21:4n-3 and 22:3n-6 not contained in FO. Dietary FO supplements can be used to enrich 20:5n-3, 22:5n-3 and 22:6n-3 in milk, with associated decreases in 4- to 18-carbon SFA, several-fold increases in CLA, mono- and polyenoic trans FA, and PUFA concentrations. Changes in the abundance and distribution of 16-, 18, 20-, and 22-carbon FA containing a single or several trans double bonds in milk were analogous with alterations in the ru- minal supply of n-3 PUFA, with enrichment of trans 18:1 and trans 18:2 being quantitatively the most important. vii) Increasing levels of FO decreased milk fat yield (up to -40.6 %) and milk fat con- tent (up to -30.1%). Fish oil-induced MFD was associated with changes in the concentrations of multiple FA in milk, in particular increases in milk fat trans-10 18:1 and cis-11 18:1 concentrations. Decreases in milk fat yield in response to FO were not related to changes in milk trans-10,cis-12 CLA concentration, esti- mated milk fat melting point or the amounts of trans-9,cis-11 CLA and trans- 10,cis-12 CLA at the omasum. In addition, no cis-10,trans-12 CLA was detected in omasal digesta or milk fat. The negative relationship between ruminal outflow of trans-10 18:1 and milk fat secretion confirmed that a shift in ruminal 18- carbon FA biohydrogenation toward trans-10 biohydrogenation pathway has a role in the regulation of milk fat synthesis during FO-induced MFD. A decrease in 18:0 supply in combination with increased mammary uptake of cis-11 18:1, trans-10 18:1, and trans 20- and 22-carbon FA intermediates originating from the rumen may contribute directly or indirectly to the reduction of milk fat ob- served during FO-induced MFD. Natural resources and bioeconomy studies 74/2020 76 6. Future research This thesis work provided a comprehensive analysis of FA metabolites formed in the rumen under different oil-supplemented dietary strategies, especially during FO-induced MFD. This, together with measurements of changes in rumen bacterial population, al- lowed the identification of potential candidates for inhibitors of milk fat synthesis. How- ever, the current theories of diet-induced MFD, including the biohydrogenation theory, remain incomplete. Several experiments support the potential role of trans-10 18:1 in MFD, but no studies have been performed to investigate the potential direct role of trans-10,cis-15 18:2 or oxo-FA, particularly 10-oxo-18:0, in milk fat synthesis. To confirm the role of these specific compounds in MFD, abomasal infusion studies using a relative- ly pure forms of trans-10,cis-15 18:2 and oxo-FA, as well as trans-10 18:1, together with lipogenic gene expression studies are needed. Further research using functional metagenomics and/or metabolomics approaches may provide a greater insight into the diet-induced changes in microbiota and bacterial biohydrogenation within the rumen. The trans-11 to trans-10 shift in microbial biohy- drogenation is not well understood. Butyrivibrio spp. are still the best-known trans-11 producing ruminal bacteria, however, the ruminal bacteria responsible for trans-10 for- mation are unclear. The results of this thesis and inconsistent results in literature sug- gest that other, yet uncultured, bacteria might be involved in this process or that specific rumen conditions are needed to produce trans-10 containing compounds. Plasma lipids and metabolites were not analysed in the experiments of this thesis. Quantitative analysis of FA from plasma lipids could provide a deeper insight into the factors contributing in mammary lipogenesis during FO-induced MFD and reveal the possible changes in the composition and concentration of circulating lipids, mammary uptake (jugular-mammary venous differences) and de novo synthesis of FA, as well as alterations in desaturation of FA in the mammary gland. These analyses could provide results to understand in more detail the mechanisms underlying the diet-induced MFD in cows. Measurements of BHBA, NEFA, plasma glucose and hormones influencing ener- gy metabolism such as insulin and leptin, could also deepen the understanding of the biological processes related to metabolic disorders. The impact of diets and nutritional factors inducing graded levels of MFD on genes and gene networks regulating mammary and tissue lipogenic and desaturation pathways could be used to explore the mechanisms driving important processes of lactating cows (e.g. lactation, energy metabolism). This information could be used in establishing solu- tions for better management of energy balance in dairy cows during late pregnancy and early lactation to improve health, well-being and reproduction performance, as well as in providing new insights for potential strategies to reduce the risk of MFD in dairy cows. The lactating cow represents a unique model to investigate both acute and long- term regulation of MFD, since temporal changes in milk fat secretion and lipogenic gene expression can be readily monitored in the same animal through the collection of se- quential milk samples and a series of tissue biopsies e.g. from subcutaneous adipose, Natural resources and bioeconomy studies 74/2020 77 liver and mammary tissues. Studies in cows fed diets causing MFD are of particular value for investigations on the nutritional regulation of genes that encode for enzymes in- volved in de novo FA synthesis, FA uptake, lipid transport, esterification and desatura- tion. Examining the impact of diet on changes in mammary gene expression and the interrelation between gene expression in adipose, liver and the mammary gland could further the development of long-term breeding strategies for more optimal production of milk constituents and enhanced milk FA composition. Milk production from ruminants is coming under increasing criticism due to the need to simultaneously reduce ruminal methane emissions and the overall impact on the environment, improve the health and welfare of ruminant livestock, and produce increasing amounts of dairy products of high nutritional quality that promote long-term human health. Fundamental research regarding the role of ruminant-derived food in health maintenance and disease prevention and opportunities to develop health pro- moting ruminant-derived foods containing lower proportions of medium-chain SFA and higher concentrations of cis MUFA, PUFA and bioactive lipids, including trans-11 18:1, CLA, 20:5n-3, 22:5n-3 and 22:6n-3, without increasing levels of trans FA in milk, are needed. In addition, development of commercially suitable applications, efficient and safe methods to protect PUFA from ruminal biohydrogenation without impairing their bioavailability for absorption should also be investigated in more detail. There are indications that dairy products may lower the risk of mortality and CVD (Dehghan et al., 2018) and ruminant trans FA may be beneficial for human health (Da Silva et al., 2015). However, little is still known about the nutritional health impact of ruminal and industrial trans 16- to 22-carbon FA and this deserves more research effort. A higher consumption of industrial trans FA in the human diet is known to be associated with increased cardiovascular risk, with some evidence from clinical trials implicating 18- carbon PUFA containing more than one double bond as being particularly harmful (Le- maitre et al., 2006; Muhlenbeck et al., 2017). It is possible that increases in milk poly- enoic trans FA may offset some of the expected benefits to human health from the en- richment of 20:5n-3 and 22:6n-3 in ruminant derived foods, but this remains a question. Further evidence from human clinical and cohort studies would be needed to assess the impact of trans FA originating from different sources (ruminal vs. industrial trans FA) for public health. Natural resources and bioeconomy studies 74/2020 78 References AbuGhazaleh, A.A. & Ishlak, A. 2014. Effects of incremental amounts of fish oil on trans fatty acids and Butyrivibrio bacteria in continuous culture fermenters. J. Anim. Physiol. Anim. Nutr. 98:271-278. AbuGhazaleh, A.A. & Jenkins, T.C. 2004. Disappearance of docosahexaenoic and eicosapentaenoic acids from cultures of mixed ruminal microorganisms. J. Dairy Sci. 87:645- 651. AbuGhazaleh, A.A., Schingoethe, D.J., Hippen, A.R. & Kalscheur, K.F. 2003. Conjugated linoleic acid and vaccenic acid in rumen, plasma, and milk of cows fed fish oil and fats differing in saturation of 18 carbon fatty acids. J. Dairy Sci. 86:3648-3660. Ackman R.G. 1992. Fatty acids in fish ans shellfish. In: Chow, C.K. (eds.) Fatty acids in foods and their health implications. pp. 169-184, New York, NY. Ahnadi, C.E., Beswick N., Delbecchi L., Kennelly J.J. & Lacasse, P. 2002. Addition of fish oil to diets for dairy cows. II. Effects on milk fat and gene expression of mammary lipogenic enzymes. J. Dairy Res. 69:4:521-31. Ahvenjärvi, S., Vanhatalo, A., Huhtanen, P. & Varvikko, T. 2000. Determination of reticulorumen and whole-stomach digestion in lactating cows by omasal canal or duodenal sampling. Br. J. Nutr. 83:67-77. Ai, F.-F., Bin, J., Zhang, Z.-M., Huang, J.-H., Wang, J.-B., Liang, Y.-Z., Yu, L. & Yang, Z.-Y. 2014. Application of random forests to select premium quality vegetable oils by their fatty acid composition. Food Chem. 143:472-478. Aldai, N., Delmonte, P., Alves, S.P., Bessa R.J.B. & Kramer, J.K.G. 2018. Evidence for the initial steps of DHA biohydrogenation by mixed ruminal microorganisms from sheep involves formation of conjugated fatty acids. J. Agric. Food Chem. 66:842-855. Aldai, N., Hervás, G., Belenguer, A., Frutos, P., Mantecón, A.R. & Kramer, J.K.G. 2012. Evaluating the in vitro metabolism of docosahexaenoic acid in sheep rumen fluid. Lipids 47:821-825. Allen, M.S. 2000. Effects of diet on short-term regulation of feed intake by lactating dairy cattle. J. Dairy Sci. 83:1598-1624. Alves, S.P. & Bessa, R.J.B. 2014. The trans-10,cis-15 18:2: a missing intermediate of trans-10 shifted rumen biohydrogenation pathway? Lipids 49:527-541. Angulo J., Mahecha, L., Nuernberg, K., Nuernberg, G., Dannenberger, D., Olivera, M., Boutinaud, M., Leroux, C., Albrecht, E. & Bernard, L. 2012. Effects of polyunsaturated fatty acids from plant oils and algae on milk fat yield and composition are associated with mammary lipogenic and SREBF1 gene expression. Animal 6:1961–1972. AOAC Lipid Library. 2019. Cited on 27 December 2019. https://lipidlibrary.aocs.org/lipid- analysis/selected-topics-in-the-analysis-of-lipids/a-practical-guide-to-the-analysis-of- conjugated-linoleic-acid-(cla) Banni, S., Carta G., Contini M.S., Angioni E., Deiana M., Dessi M.A., Melis M.P. & Corongiu, F.P. 1996. Characterization of conjugated diene fatty acids in milk, dairy products, and lamb tissues. J. Nutr. Biochem. 7:150-155. Bauchart, D., Legay-Cramer, F. & Doreau, M. 1990. Ruminal hydrolysis of dietary triglycerides in dairy cows fed lipid-supplemented diets. Reprod. Nutr. Dev.p. 187. Natural resources and bioeconomy studies 74/2020 79 Bauman, E., Chouinard, P.Y, Lebeuf, Y., Rico, D.E. & Gervais, R. 2016. Effect of lipid supplementation on milk odd- and branched-chain fatty acids in dairy ewes. J. Dairy Sci. 99:6311-6323. Bauman, D.E. & Griinari, J.M. 2001. Regulation and nutritional manipulation of milk fat: low-fat milk syndrome. Livest. Prod. Sci. 70:15-29. Bauman, D.E. & Griinari, J.M. 2003. Nutritional regulation of milk fat synthesis. Annu. Rev. Nutr. 23:203-227. Baumgard, L.H., Corl, B.A., Dwyer, D.A., Sæbø, A. & Bauman, D.E. 2000. Identification of the conjugated linoleic acid isomer that inhibits milk fat synthesis. Am. J. Physiol. 278:R179- R184. Baumgard, L.H., Matitashvili, E., Corl, B.A., Dwyer D.A. & Bauman, D.E. 2002. Trans-10,cis-12 conjugated linoleic acid decreases lipogenic rates and expression of genes involved in milk lipid synthesis in dairy cows. J. Dairy Sci. 85:9:2155-2163. Bayat, A.R., Kairenius, P., Stefański, T., Leskinen, H., Comtet-Marre, S., Forano, E., Chaucheyras- Durand, F. & Shingfield, K.J. 2015. Effect of camelina oil or live yeasts (Saccharomyces cerevisiae) on ruminal methane production, rumen fermentation, and milk fatty acid composition in lactating cows fed grass silage diets. J. Dairy Sci. 98:3166-3181. Bayat, A.R., Tapio, I., Vilkki, J., Shingfield, K.J. & Leskinen, H. 2018. Plant oil supplements reduce methane emissions and improve milk fatty acid composition in dairy cows fed grass silage- based diets without affecting milk yield. J. Dairy Sci. 101:1136-1151. Belenguer, A., Toral, P.G., Frutos, P. & Hervás, G. 2010. Changes in the rumen bacterial community in response to sunflower oil and fish oil supplements in the diet of dairy sheep. J. Dairy Sci. 93:3275-3286. Bell, J.A., Griinari, J.M. & Kennelly, J.J. 2006. Effect of safflower oil, flaxseed oil, monensin, and vitamin E on concentration of conjugated linoleic acid in bovine milk fat. J. Dairy Sci. 89:733- 748. Belling, G.B., Abbeya, M., Campbell, J.H. & Campbell, G.R. 1997. Lipid content and fatty acid composition of 11 species of queensland (Australia) fish. Lipids 32:621-625. Bernard, L., Bonnet, M., Delavaud, C., Delosière, M., Ferlay, A., Fougère, H. & Graulet, B. 2018. Milk fat globule in ruminant: major and minor compounds, nutritional regulation and differences among species. Eur. J. Lipid Sci. Technol. 120:1700039. Bernard, L., Leroux, C. & Chilliard, Y. 2008. Expression and nutritional regulation of lipogenic genes in the ruminant lactating mammary gland. In: Bösze, Z. (eds.) Bioactive components of milk, advances in experimental medicine and biology. Vol. 606, pp. 67-108, New York, US. Springer. Bichi, E., Hervás, G., Toral, P.G., Loor, J.J. & Frutos, P. 2013. Milk fat depression induced by dietary marine algae in dairy ewes: Persistency of milk fatty acid composition and animal performance responses. J. Dairy Sci. 96:524-532. Boeckaert, C., Fievez, V., Van Hecke, D., Verstraete, W. & Boon, N. 2007. Changes in rumen biohydrogenation intermediates and ciliate protozoa diversity after algae supplementation to dairy cattle. Eur. J. Lipid Sci. Technol. 109:767-777. Boeckaert, C., Vlaeminck, B., Fievez, V., Maignien, L., Dijkstra, J. & Boon, N. 2008. Accumulation of trans C18:1 fatty acids in the rumen after dietary algal supplementation is associated with changes in the Butyrivibrio population. Appl. Environ. Microbiol. 74:6923-30. Burdge, G.C. & Calder, P.C. 2005. Conversion of α-linolenic acid to longer-chain polyunsaturated fatty acids in human adults. Reprod. Nutr. Dev. 45:581-97. Natural resources and bioeconomy studies 74/2020 80 Burns, T.A., Kadegowda, A.K.G., Duckett, S.K., Pratt, S.L., & Jenkins, T.C. 2012. Palmitoleic (16:1 cis-9) and cis-vaccenic (18:1 cis-11) acid alter lipogenesis in bovine adipocyte cultures. Lipids 47:1143-1153. Chilliard, Y., Ferlay, A. & Doreau, M. 2001. Effect of different types of forages, animal fat and marine oil in cow's diet on milk fat secretion and composition, especially conjugated linoleic acid (CLA) and polyunsaturated fatty acids. Livest. Prod. Sci. 70:31-48. Chilliard, Y., Ferlay, A., Mansbridge, R.M. & Doreau, M. 2000. Ruminant milk fat plasticity: nutritional control of saturated, polysaturated, trans and conjugated acids. Ann. Zootech. 49:181-205. Chilliard, Y., Glasser, F., Ferlay, A., Bernard, L., Rouel, J. & Doreau, M. 2007. Diet, rumen biohydrogenation and nutritional quality of cow and goat milk fat. Eur. J. Lipid Sci. Technol. 109:828-855. Christie, W.W. 1998. Gas chromatography-mass spectrometry methods for structural analysis of fatty acids. Lipids 33:343-353. Christie, W.W., Brechany, E.Y., Johnson, S.B. & Holman, R. 1986. A comparison of pyrollidide and picolinyl ester derivatives for the identification of fatty acids in natural samples by gas chromatography-mass spectrometry. Lipids 21:657-661. Conte, G., Dimauro, C., Serra, A., Macciotta, N.P.P. & Mele, M. 2018. A canonical discriminant analysis to study the association between milk fatty acids of ruminal origin and milk fat depression in dairy cows. J. Dairy Sci. 101:6497-6510. Corl B.A., Baumgard, L.H., Griinari, J.M., Delmonte, P., Morehouse, K.M., Yurawecz, M.P. & Bauman, D.E. 2002. Trans-7,cis-9 CLA is synthesized endogenously by ∆9-desaturase in dairy cows. Lipids 37:681-8. Corl, B.A., Baumgard, L.H., Dwyer, D.A., Griinari, J.M., Phillips, B.S. & Bauman, D.E. 2001. The role of delta(9)-desaturase in the production of cis-9,trans-11 CLA. J. Nutr. Biochem. 12:622-630. Cruz-Hernandez, C., Deng, Z., Zhou, J., Hill, A,R., Yurawecz, M.P., Delmonte, P., Mossoba, M.M., Dugan, M.E.R. & Kramer, J.K.G. 2004. Methods for analysis of conjugated linoleic acids and trans-18:1 isomers in dairy fats by using a combination of gas chromatography, silver-ion thin-layer chromatography/gas chromatography, and silver-ion liquid chromatography. JAOAC Int. 87:545-562. Cruz-Hernandez, C., Kramer, J.K.G., Kraft, J., Santercole, V., Or-Rashid, M., Deng, Z., Dugan, M.E.R., Delmonte, P. & Yurawecz, M. 2006. Systematic analysis of trans and conjugated linoleic acids in the milk and meat of ruminants. In: Yurawecz, M.P., Kramer, J.K.G., Gudmundsen, O., Pariza, M.W. & Banni, S. (eds) Adv. Conjugated Linoleic Acid Res. pp 45-93, 3rd edn. AOAC, Urbana. Da Silva, M.S., Julien, P., Pérusse, L., Vohl, M.-C. & Rudkowska, I. 2015. Natural rumen-derived trans fatty acids are associated with metabolic markers of cardiac health. Lipids 50:873–882. Dallaire, M.P., Taga, H., Ma, L., Corl, B.A., Gervais, R., Lebeuf, Y., Richard, F.J. & Chouinard, P.Y. 2014. Effects of abomasal infusion of conjugated linoleic acids, Sterculia foetida oil, and fish oil on production performance and the extent of fatty acid ∆9-desaturation in dairy cows. J. Dairy Sci. 97:10:6411-25. Deckelbaum, R.J. & Torrejon, C. 2012. The omega-3 fatty acid nutritional landscape: Health benefits and sources. J. Nutr. 142:3:587S-591S. Dehghan, M., Mente, A., Rangarajan, S., Sheridan, P., Mohan, V., Iqbal, R., Gupta, R., Lear, S., Wentzel-Viljoen, E., Avezum, A., Lopez-Jaramillo, P., Mony, P., Varma, R.P., Kumar, R., Chifamba, J., Alhabib, K.F., Mohammadifard, N., Oguz, A., Lanas, F., Rozanska, D., Bengtsson Natural resources and bioeconomy studies 74/2020 81 Bostrom, K., Yusoff, K., Tsolkile, L.P., Dans, A., Yusufali, AH., Orlandini, A., Poirier, P., Khatib, R., Hu, B., Wei, L., Yin, L., Deeraili, A., Yeates, K., Yusuf, R., Ismail, N., Mozaffarian, D., Teo, K., Anand, S.S. & Yusuf S. 2018. Association of dairy intake with cardiovascular disease and mortality in 21 countries from five continents (PURE): a prospective cohort study. http://dx.doi.org/10.1016/S0140-6736(18)31812-9 Destaillats, F., Wolff, R.L., Precht, D. & Molkentin, J. 2000. Study of individual trans- and cis-16:1 isomers in cow, goat, and ewe cheese fats by gas-liquid chromatography with emphasis on the trans-∆3 isomer. Lipids 35:1027-1032. Dhaka, V., Gulia, N., Ahlawat, K.S. & Khatkar, B.S. 2011. Trans fats - sources, health risks and alternative approach - A review. J. Food Sci. Technol. 48:5:534-541. Dohme, F., Fievez, K., Raes, K. & Demeyer, D.I. 2003. Increasing levels of two different fish oils lower ruminal biohydrogenation of eicosapentaenoic and docosahexaenoic acid in vitro. Anim. Res. 52:309-320. Donovan, D.C., Schingoethe, D.J., Baer, R.J., Ryali, J., Hippen, A.R. & Franklin, S.T. 2000. Influence of dietary fish oil on conjugated linoleic acid and other fatty acids in milk fat from lactating dairy cows. J. Dairy Sci. 83:2620-2628. Doreau, M. & Chilliard, Y. 1997. Effects of ruminal or postruminal fish oil supplementation on intake and digestion in dairy cows. Reprod. Nutr. Dev. 37:113-124. Doreau, M. & Ferlay, A. 1994. Digestion and utilisation of fatty acids by ruminants. Anim. Feed Sci. Tech. 45:379-396. Dorni, C., Paras Sharma, Gunendra Saikia, T. Longvah. 2018. Fatty acid profile of edible oils and fats consumed in India. Food Chem. 238:9-15. Eilander, A., Harika, R.K. & Zock, P.L. 2015. Intake and sources of dietary fatty acids in Europe: Are current population intakes of fats aligned with dietary recommendations? Eur. J. Lipid Sci. Technol. 117:1370-1377. Elgersma, A. 2015. Grazing increases the unsaturated fatty acid concentration of milk from grass- fed cows: A review of the contributing factors, challenges and future perspectives. Eur. J. Lipid Sci. Technol. 117:1345-1369. Enjalbert, F., Combes, S., Zened, A. & Meynadier, A. 2017. Rumen microbiota and dietary fat: a mutual shaping. J. Appl. Microbiol. 123:782-797. Escobar, M., Vlaeminck, B., Jeyanathan, J., Thanh, L.P., Shingfield, K.J., Wallace, R.J. & Fievez, V. 2016. Effect of adsorbants on in vitro biohydrogenation of 22:6n-3 by mixed cultures of rumen microorganisms. Animal 10:1439-1447. Fievez, V., Colman, E., Castro-Montoya, J.M., Stefanov, I. & Vlaeminck, B. 2012. Milk odd- and branched-chain fatty acids as biomarkers of rumen function - An update. Anim. Feed Sci. Tech. 172:1-2:51-65. Fournier, V., Destaillats, F., Juanéda, P., Dionisi, F., Lambelet, P., Sébédio, J.-L. & Berdeaux, O. 2006a. Thermal degradation of long chain polyunsaturated fatty acids during deodorization of fish oil. Eur. J. Lipid Sci. Technol. 108:33-42. Fournier, V., Juanéda, P., Destaillats, F., Dionisi, F., Lambelet, P., Sébédio, J.-L. & Berdeaux, O. 2006b. Analysis of eicosapentaenoic and docosahexaenoic acid geometrical isomers formed during fish oil deodorization. J. Chromatogr. A 1129:21-28. Frutos, P., Toral, P.G., Belenguer, A. & Hervás, G. 2018. Milk fat depression in dairy ewes fed fish oil: Might differences in rumen biohydrogenation, fermentation, or bacterial community explain the individual variation? J. Dairy Sci. 101:6122-6132. Natural resources and bioeconomy studies 74/2020 82 Fuentes, M.C., Calsamiglia, S., Cardozo, P.W. & Vlaeminck, B. 2009. Effect of pH and level of concentrate in the diet on the production of biohydrogenation intermediates in a dual-flow continuous culture. J. Dairy Sci. 92:4456-4466. Gama, M.A.S., Garnsworthy, P.C., Griinari, J.M., Leme, P.R., Rodrigues, P.H.M., Souza, L.W.O. & Lanna, D.P.D. 2008. Diet induced milk fat depression: Association with changes in milk fatty acid composition and fluidity of milk fat. Livest. Sci. 115:319-331. Gebauer, S.K., Chardigny, J.M., Jakobsen, M.U., Lamarche, B., Lock, A.L., Proctor, S.D. & Baer, D.J. 2011. Effects of ruminant trans fatty acids on cardiovascular disease and cancer: A comprehensive review of epidemiological, clinical, and mechanistic studies. Adv Nutr. 2:332-354. Givens, D.I. & Shingfield, K.J. 2006. Optimising dairy milk fatty acid composition. In: Williams, C.M. & Buttriss, J. (eds.) Improving the fat content of foods, pp. 252-280. Cambridge, UK, Woodhead Publishing Ltd. Givens, D.I. 2008. Impact on CVD risk of modifying milk fat to decrease intake of saturated fatty acids and increase intake of cis-monounsaturates. Proc. Nutr. Soc. 67:419-427. Glasser, F., Ferlay, A., Doreau, M., Loor, J.J. & Chilliard, Y. 2010. t10,c12-18:2-Induced milk fat depression is less pronounced in cows fed high-concentrate diets. Lipids 45: 877-887. Glasser, F., Schmidely, P., Sauvant, D. & Doreau, M. 2008. Digestion of fatty acids in ruminants: a meta analysis of flows and varation factors: 2. C18 fatty acids. Animal 2:691-704. Gomez-Córtes, P., Tyburczy, C., Brenna, J.T., Juárez, M. & Angel de la Fuente, M. 2009. Characterization of cis-9,trans-11,trans-15 C18:3 in milk fat by GC and covalent adduct chemical ionization tandem MS. J. Lipid Res. 50:2412-2420. Griinari, J.M. & Bauman, D.E. 1999. Biosynthesis of conjugated linoleic acid and its incorporation into meat and milk in ruminants. In: Yurawecz, M.P., Mossoba, M.M., Kramer, J.K.G, Pariza, M.W. & Nelson, G.J. (eds.) Advances in Conjugated Linoleic Acid Research, pp. 180-200, Vol 1, AOCS Press, Champaign, IL. Griinari, J.M., Corl, B.A., Lacy, S.H., Chouinard, P.Y., Nurmela, K.V. & Bauman, D.E. 2000. Conjugated linoleic acid is synthesized endogenously in lactating dairy cows by ∆9- desaturase. J. Nutr. 130:2285-2291. Griinari, J.M., Dwyer, D.A., McGuire, M.A., Bauman, D.E., Palmquist, D.L. & Nurmela, K.V.V. 1998. Trans-octadecenoic acids and milk fat depression in lactating dairy cows. J. Dairy Sci. 81:1251-1261. Gudla, P., AbuGhazaleh, A.A., Ishlak, A. & Jones, K. 2012. The effect of level of forage and oil supplement on biohydrogenation intermediates and bacteria in continuous cultures. Anim. Feed Sci. Technol. 171:108-16. Gulati, S.K., Ashes, J.R. & Scott, T.W. 1999. Hydrogenation of eicosapentaenoic and docosahexaenoic acids and their incorporation into milk fat. Anim. Feed Sci. Tech. 79:57-64. Hagemeister, H., Franzen, M., Barth, C.A. & Precht, D. 1991. α-Linolenic acid transfer into milk fat and its elongation by cows. Fett/Lipid 93:387-391. Halmemies-Beauchet-Filleau, A., Kairenius, P., Ahvenjärvi, S., Crosley, L.K., Muetzel, S., Huhtanen, P. & Shingfield, K.J. 2013a. Effect of forage conservation method on ruminal lipid metabolism and microbial ecology in lactating cows fed diets containing a 60:40 forage-to- concentrate ratio. J. Dairy Sci. 96:4:2428-2447. Halmemies-Beauchet-Filleau, A., Kairenius, P., Ahvenjärvi, S., Toivonen, V., Huhtanen, P., Vanhatalo A., Givens, D.I. & Shingfield, K.J. 2013b. Effect of forage conservation method on Natural resources and bioeconomy studies 74/2020 83 plasma lipids, mammary lipogenesis, and milk fatty acid composition in lactating cows fed diets containing a 60:40 forage-to-concentrate ratio. J. Dairy Sci. 96:5267-5289. Harfoot, C.G. & Hazlewood, G.P. 1988. Lipid metabolism in the rumen. In: The Rumen Microbial Ecosystem. pp. 285-322 Elsevier Science Publishing, NY. Harris, W.S., Miller, M., Tighe, A.P., Davidson, M.H. & Schaefer, E.J. 2008. Review Omega-3 fatty acids and coronary heart disease risk: Clinical and mechanistic perspectives. Atherosclerosis 197:12-24. Harris, W.S., Mozaffarian, D., Lefevre, M., Toner, C.D., Colombo, J., Cunnane, S.C., Holden, J.M., Klurfeld, D.M., Morris, M.C. & Whelan, J. 2009. Towards establishing dietary reference intakes for eicosapentaenoic and docosahexaenoic acids. J. Nutr. 139:804S-819S Harvatine, K. J. & Bauman, D.E. 2011. Characterization of the acute lactational response to trans- 10,cis-12 conjugated linoleic acid. J. Dairy Sci. 94:6047-6056. Harvatine, K.J. & Bauman, D.E. 2006. SREBP1 and thyroid hormone responsive spot 14 (S14) are involved in the regulation of bovine mammary lipid synthesis during diet-induced milk fat depression and treatment with CLA. J. Nutr. 136:10:2468-2474. Harvatine, K.J., Boisclair, Y.R. & Bauman, D.E. 2009. Recent advances in the regulation of milk fat synthesis. Animal 3:40-54. Harvey, D.J. 1984. Picolinyl derivatives for the structural determination of fatty acids by mass spectrometry: Applications to polyenoic acids, hydroxy acids, diacids and related compounds. Biomed. Mass Spectrom. 11:340-347. Harvey, D.J. 1992. Mass spectrometry of picolinyl and other nitrogen-containing derivatives of fatty acids. In: Christie, W.W. (eds.) Advances in Lipid Methodology - one, pp. 19-80, Ayr, Scotland, The Oily Press. Honkanen, A.M., Leskinen, H., Toivonen, V., McKain, N., Wallace, R.J. & Shingfield, K.J. 2016. Metabolism of α-linolenic acid during incubations with strained bovine rumen contents: products and mechanisms. Br. J. Nutr. 115:2093-2105. Hu, F.B., Manson, J.E. & Walter, C.W. 2001. Types of dietary fat and risk of coronary heart disease: A critical review. J. Am. Coll. Nutr. 20:1:5-19. Hudson, J.A., Cai, Y., Corner, R.J., Morvan, B. & Joblin, K.N. 2000. Identification and enumeration of oleic acid and linoleic acid hydrating bacteria in the rumen of sheep and cows. J. Appl. Microbiol. 88:286-292. Hudson, J.A., MacKenzie, C.A. & Joblin, K.N. 1995. Conversion of oleic acid to 10-hydroxystearic acid by two species of ruminal bacteria. Appl. Microbiol. Biotechnol. 44:1-6. Hudson, J.A., Morvan, B. & Joblin, K.N. 1998. Hydration of linoleic acid by bacteria isolated from ruminants. FEMS Microbiol Lett 169:277-282. Huhtanen, P. & Kukkonen, U. 1995. Comparison of methods, markers, sampling sites and models for estimating digesta passage kinetics in cattle fed at 2 levels of intake. Anim. Feed Sci. Technol. 52:141-158. Huth, P.J., DiRienzo, D.B. & Miller, G.D. 2006. Major scientific advances with dairy foods in nutrition and health. J. Dairy Sci. 89:1207-1221. Huws, S.A., Kim, E.J., Cameron, S.J., Girdwood, S.E., Davies, L., Tweed, J., Vallin, H., & Scollan, N.D. 2015. Characterization of the rumen lipidome and microbiome of steers fed a diet supplemented with flax and echium oil. Microb Biotechnol. 8(2):331–341. Huws, S.A., Kim, E.J., Lee, M.R.F., Scott, M.B., Tweed, J.K., Pinloche, E., Wallace, R.J. & Scollan, N.D. 2011. As yet uncultured bacteria phylogenetically classified as Prevotella, Natural resources and bioeconomy studies 74/2020 84 Lachnospiraceae incertae sedis and unclassified Bacteroidales, Clostridiales and Ruminococcaceae may play a predominant role in ruminal biohydrogenation. Environ. Microbiol. 13:1500-12. Huws, S.A., Lee, M.R.F., Muetzel, S.M., Scott, M.B., Wallace, R.J. & Scollan, N.D. 2010. Forage type and fish oil cause shifts in rumen bacterial diversity. FEMS Microbiol Ecol. 73:396-407. Invernizzi, G., Thering, B.J., McGuire, M.A., Savoini, G. & Loor, J.J. 2010. Sustained upregulation of stearoyl-CoA desaturase in bovine mammary tissue with contrasting changes in milk fat synthesis and lipogenic gene networks caused by lipid supplements. Funct. Integr. Genomics 10:561-575. Jakobsen, K. 1999. Dietary modifications of animal fats: status and future perspectives. Fett/Lipid 101:475-483. Jenkins, T.C. 1993. Lipid metabolism in the rumen. J. Dairy Sci. 76:3851-3863. Jenkins, T.C., AbuGhazaleh, A.A., Freeman, S. & Thies, E.J. 2006. The production of 10- hydroxystearic acid and 10-ketostearic acids is an alternate route of oleic acid transformation by the ruminal microbiota in cattle. J. Nutr. 136:926-31. Jenkins, T.C., Wallace, R.J., Moate, P.J. & Mosley, E.E. 2008. Recent advances in biohydrogenation of unsaturated fatty acids within the rumen microbial ecosystem. J. Anim. Sci. 86:397-412. Jensen, R.G. 2002. Invited review: The composition of bovine milk lipids: January 1995 to December 2000. J. Dairy Sci. 85:295-350. Jeyanathan J., Escobar, M., Wallace, R.J., Fievez, V. & Vlaeminck, B. 2016. Biohydrogenation of 22:6n-3 by Butyrivibrio proteoclasticus P18. BMC Microbiol. 16:104. Jouany, J.P., Lassalas, B., Doreau, M. & Glasser, F. 2007. Dynamic features of the rumen metabolism of linoleic acid, linolenic acid and linseed oil measured in vitro. Lipids 42:351- 360. Kadegowda, A.K.G., Bionaz, M., Piperova, L.S., Erdman, R.A. & Loor, J.J. 2009. Peroxisome proliferator-activated receptor-gamma activation and long-chain fatty acids alter lipogenic gene networks in bovine mammary epithelial cells to various extents. J. Dairy Sci. 92:4276- 4289. Kadegowda, A.K.G., Piperova, L.S. & Erdman, R.A. 2008. Principal component and multivariate analysis of milk long-chain fatty acid composition during diet-induced milk fat depression. J. Dairy Sci. 91:749-759. Keady, T.W.J. & Mayne, C.S. 1999. The effects of level of fish oil inclusion in the diet on rumen digestion and fermentation parameters in cattle offered grass silage based diets. Anim. Feed Sci. Technol. 81:57-68. Keady, T.W.J., Mayne, C.S. & Fitzpatrick, D.A. 2000. Effects of supplementation of dairy cattle with fish oil on silage intake, milk yield and milk composition. J. Dairy Res. 67:137-153. Kemp, P., White, R.W. & Lander, D.J. 1975. The hydrogenation of unsaturated fatty acids by five bacterial isolates from the sheep rumen, including a new species. J. Gen. Microbiol. 90:100- 114. Kepler, C.R., Hirons, K.P., McNeill, J.J. & Tove, S.B. 1966. Intermediates and products of the biohydrogenation of linoleic acid by Butyrivibrio fibrisolvens. J. Biol. Chem. 241:1350-1354. Kim, E.J., Huws, S.A., Lee, M.R.F., Wood, J.D., Muetzel, S.M., Wallace, R.J. & Scollan, N.D. 2008. Fish oil increases the duodenal flow of long chain polyunsaturated fatty acids and trans-11 18:1 and decreases 18:0 in steers via changes in the rumen bacterial community. J. Nutr. 138:889-96. Natural resources and bioeconomy studies 74/2020 85 Kim, Y.J., Liu, R.H., Rychlik, J.L. & Russell, J.B. 2002. The enrichment of a ruminal bacterium (Megasphaera elsdenii YJ-4) that produces the trans-10,cis-12 isomer of conjugated linoleic acid. J. Appl. Microbiol. 92:976-82. Kitessa, S.M., Gulati, S.K., Ashes, J.R., Fleck, E., Scott, T.W. & Nichols, P.D. 2001. Utilisation of fish oil in ruminants. I. Fish oil metabolism in sheep. Anim. Feed Sci. Technol. 89:189-99. Kleber, M.E., Delgado, G.E., Lorkowski, S., März, W. & von Schacky, C. 2016. Trans-fatty acids and mortality in patients referred for coronary angiography: the Ludwigshafen risk and cardiovascular health study. Eur. Heart J. 37:1072-1078. Kliem, K.E. & Shingfield K.J. 2016. Manipulation of milk fatty acid composition in lactating cows: Opportunities and challenges. Eur. J. Lipid Sci. Technol. 118: 1661–1683. Klein, C.M. & Jenkins, T.C. 2011. Docosahexaenoic acid elevates trans-18:1 isomers but is not directly converted into trans-18:1 isomers in ruminal batch cultures. J. Dairy Sci. 94:4676- 4683. Lawson, R.E., Moss, A.R. & Givens, D.I. 2001. The role of dairy products in supplying conjugated linoleic acid to man's diet: A review. Nutr. Res. Rev. 14:153-172. Lee, M.R.F., Shingfield, K.J., Tweed, J.K.S., Toivonen, V., Huws, S.A. & Scollan, N.D. 2008. Effect of fish oil on ruminal lipid metabolism in steers fed grass or red clover silages. Animal 2:1859- 69. Lee, Y.-J. & Jenkins, T.C. 2011. Biohydrogenation of linolenic acid to stearic acid by the rumen microbial population yields multiple intermediate conjugated diene isomers. J. Nutr. 141:8:1445-1450. Lemaitre, R.N., King, I.B., Mozaffarian, D., Sotoodehnia, N., Rea, T.D., Kuller, L.H., Tracy, R.P. & Siscovick, D.S. 2006. Plasma phospholipid trans fatty acids, fatal ischemic heart disease, and sudden cardiac death in older adults. Circulation 114:209-215. Leskinen, H., Ventto, L., Kairenius, P., Shingfield, K.J. & Vilkki, J. 2019. Temporal changes in milk fatty acid composition during diet-induced milk fat depression in lactating cows. J. Dairy Sci. 102:5148-5160. The LipidWeb. 2019. Cited on 17 December 2019. https://www.lipidhome.co.uk/ Liavonchanka, A., Hornung, E., Feussner, I. & Rudolph, M.G. 2006. Structure and mechanism of the Propionibacterium acnes polyunsaturated fatty acid isomerase. Proc. Natl. Acad. Sci. USA 103:2576-81. s.61 Lindmark Månsson, H. 2008. Fatty acids in bovine milk fat. Food & Nutrition Research 52, 10.3402/fnr.v52i0.1821. Liu, S.J., Bu, D.P., Wang, J.Q., Sun, P., Wei, H.Y., Zhou, L.Y. & Yu, Z.T. 2011. Effect of ruminal pulse dose of polyunsaturated fatty acids on ruminal microbial populations and duodenal flow and milk profiles of fatty acids. J. Dairy Sci. 94:2977-85. Lock, A.L. & Bauman, D.E. 2004. Modifying milk fat composition of dairy cows to enhance fatty acids beneficial to human health. Lipids 39:1197-1206. Lock, A.L., Parodi, P.W. & Bauman, D.E. 2005. The biology of trans fatty acids: Implications for human health and the dairy industry. Aust. J. Dairy Technol. 60:134-142. Lock, A.L., Tyburczy, C., Dwyer, D.A., Harvatine, K.J., Destaillats, F., Mouloungui, Z., Candy, L. & Bauman, D.E. 2007. Trans-10 octadecenoic acid does not reduce milk fat synthesis in dairy cows. J. Nutr. 137:71-76. Loor, J.J., Doreau, M., Chardigny, J.M., Ollier, A., Sébédio, J.L. & Chilliard, Y. 2005a. Effects of ruminal or duodenal supply of fish oil on milk fat secretion and profiles of trans-fatty acids Natural resources and bioeconomy studies 74/2020 86 and conjugated linoleic acid isomers in dairy cows fed maize silage. Anim. Feed Sci. Technol. 119:227-246. Loor, J.J., Ferlay, A., Ollier, A., Doreau, M., & Chilliard, Y. 2005b. Relationship among trans and conjugated fatty acids and bovine milk fat yield due to dietary concentrate and linseed oil. J. Dairy Sci. 88:726-740. Loor, J.J., Ueda, K., Ferlay, A., Chilliard, Y. & Doreau, M. 2004. Biohydrogenation, duodenal flow, and intestinal digestibility of trans fatty acids and conjugated linoleic acids in response to dietary forage:concentrate ratio and linseed oil in dairy cows. J. Dairy Sci. 87:2472-2485. Loor, J.J., Ueda, K., Ferlay, A., Chilliard, Y. & Doreau, M. 2005c. Intestinal flow and digestibility of trans fatty acids and conjugated linoleic acids (CLA) in dairy cows fed a high-concentrate diet supplemented with fish oil, linseed oil, or sunflower oil. Anim. Feed Sci. Technol. 119:203-225. Lourenço, M., Ramos-Morales, E. & Wallace, R.J. 2010. The role of microbes in rumen lipolysis and biohydrogenation and their manipulation. Animal 4:1008-1023. Maia, M.R., Chaudhary, L.C., Bestwick, C.S., Richardson, A.J., McKain, N., Larson, T. R., Graham, I.A., & Wallace, R.J. 2010. Toxicity of unsaturated fatty acids to the biohydrogenating ruminal bacterium, Butyrivibrio fibrisolvens. BMC microbiol. 10:52. Maia, M., Chaudhary, L.C., Figueres, L. & Wallace, RJ. 2007. Metabolism of polyunsaturated fatty acids and their toxicity to the microflora of the rumen. Antonie van Leeuwenhoek 91:303- 314. McKain N., Shingfield K.J. & Wallace, R.J. 2010. Metabolism of conjugated linoleic acids and 18:1 fatty acids by ruminal bacteria: products and mechanisms. Microbiology 156:579-588. Meier, T., Gräfe, K., Senn, F., Sur, P., Stangl, G.I., Dawczynski, C., März, W., Kleber, M.E. & Lorkowski, S. 2019. Cardiovascular mortality attributable to dietary risk factors in 51 countries in the WHO European Region from 1990 to 2016: a systematic analysis of the Global Burden of Disease Study. Eur. J. Epidemiol. 34:37–55. Michaud, A.L., Yurawecz, M.P., Delmonte, P., Corl, B.A., Bauman, D.E. & Brenna, T. 2003. Identification and characterization of conjugated fatty acid methyl esters of mixed double bond geometry by acetonitrile chemical ionization tandem mass spectrometry. Anal. Chem. 75:4925-4930. Moffat, C.F. & McGill, A.S., 1993. Variability of the composition of fish oils: significance for the diet. Proc. Nut. Soc. 52:441-456. Mosley, E.E., Bahman, S., Moate, P.J. & McGuire, M.A. 2006. Cis-9,trans-11 conjugated linoleic acid is synthesized directly from vaccenic acid in lactating dairy cattle. J. Nutr. 136:570-575. Mosley, E.E., Powell, G.L., Riley, M.B. & Jenkins, T.C. 2002. Microbial biohydrogenation of oleic acid to trans isomers in vitro. J. Lipid Res. 43:290-296. Muhlenbeck, J.A., Butz, D.E., Olson, J.M., Uribe-Cano, D. & Cook, M.E. 2017. Dietary conjugated linoleic acid - c9t11 prevents collagen induced arthritis, whereas conjugated linoleic acid - t10c12 increases. Lipids 52:303. Offer, N.W., Marsden, M., Dixon, J., Speakel, B.K. & Thacker, F.E. 1999. Effect of dietary fat supplements on levels of n-3 poly-unsaturatedfatty acids, trans acids and conjugated linoleic acid inbovine milk. Anim. Sci. 69:613-625. Or-Rashid, M.M., Kramer, J.K.G., Wood, M.A. & McBride, B.W. 2008. Supplemental algal meal alters the ruminal trans-18:1 fatty acid and conjugated linoleic acid composition in cattle. J. Anim. Sci. 86:187-196. Natural resources and bioeconomy studies 74/2020 87 Palmquist, D. L., A. L. Lock, K. J. Shingfield, and D. E. Bauman. 2005. Biosynthesis of conjugated linoleic acid in ruminants and humans. Pp. 179-217 in Advances in Food and Nutrition Research. Taylor S. L., ed. Elsevier Academic Press, San Diego, CA. Palmquist, D.L. & Griinari, J.M. 2006. Milk fatty acid composition in response to reciprocal combinations of sunflower and fish oils in the diet. Anim. Feed Sci. Tech. 131:358-369. Palmquist, D.L. 2009. Omega-3 fatty acids in metabolism, health, and nutrition and for modified animal product foods. Prof. Anim. Sci. 25:207-249. Piperova, L.S., Teter, B.B., Bruckental, I., Sampugna, J., Mills, S.E., Yurawecz, M.P., Fritsche, J., Ku, K. & Erdman, R.A. 2000. Mammary lipogenic enzyme activity, trans fatty acids and conjugated linoleic acids are altered in lactating dairy cows fed a milk fatdepressing diet. J. Nutr. 130:2568–2574. Piperova, L.S., Teter, B.B., Bruckental, I., Sampugna, J., Mills, S.E., Yurawecz, M.P., Fritsche, J., Ku, K. & Erdman, R.A. 2002. Mammary lipogenic enzyme activity, trans fatty acids and conjugated linoleic acids are altered in lactating dairy cows fed a milk fat-depressing diet. J. Nutr. 130:2568-2574. Potu, R.B., AbuGhazaleh, A.A., Hastings, D., Jones, K. & Ibrahim, S.A. 2011. The effect of lipid supplements on ruminal bacteria in continuous culture fermenters varies with the fatty acid composition. J. Microbiol. 49:216-23. Precht, D. & Molkentin, J. 1996. Rapid analysis of the isomers of trans-octadecenoic acid in milk fat. Int. Dairy J. 6:791-809. Precht, D., & Molkentin, J. 1999. C18:1, C18:2 and C18:3 trans and cis fatty acid isomers including conjugated cis-9,trans-11 linoleic acid (CLA) as well as total fat composition of German human milk lipids. Nahrung 43:233-244. Qin, N., Bayat, A.R., Trevisi, E., Minuti, A., Kairenius, P., Viitala, S., Mutikainen, M., Leskinen, H., Elo, K., Kokkonen, T. & Vilkki, J. 2018. Dietary supplement of conjugated linoleic acids or polyunsaturated fatty acids suppressed the mobilization of body fat reserves in dairy cows at early lactation through different pathways. J. Dairy Sci. 101(9):7954-7970. Ratnayake, W.M. & Galli, C. 2009. Fat and fatty acid terminology, methods of analysis and fat digestion and metabolism: a background review paper. Ann. Nutr. Metab. 55:8-43. Rego, O.A., Alves, S.P., Antunes, L.M.S., Rosa, H.J.D., Alfaia, C.F.M., Prates, J.A.M., Cabrita, A.R.J., Fonseca, J.M. & Bessa, R.J. 2009. Rumen biohydrogenation-derived fatty acids in milk fat from grazing dairy cows supplemented with rapeseed, sunflower, or linseed oils. J. Dairy Sci. 92:4530-4540. Rego, O.A., Rosa, H.J.D., Portugal, P., Cordeiro, R., Borba, A.E.S., Vouzela, C.M & Bessa, R.J. 2005. Influence of dietary fish oil on conjugated linoleic acid, omega-3 and other fatty acids in milk fat from grazing dairy cows. Livest. Prod. Sci. 95:27-33. Rodriguez-Estrada, M. & Cilla, A., Cardenia, V. & Guadalupe, G.-L. 2014. Fats and Sterols. Reference Module in Biomedical Sciences. ISBN 9780128012383. Sackmann, J.R., Duckett, S.K., Gillis, M.H., Realini, C.E., Parks, A.H. & Eggelston, R.B. 2003. Effects of forage and sunflower oil levels on ruminal biohydrogenation of fatty acids and conjugated linoleic acid formation in beef steers fed finishing diets. J. Anim. Sci. 81:3174- 3181. Saebø, A., Sæbø, P., Griinari, J.M. & Shingfield, K.J. 2005. Effect of abomasal infusion of geometric isomers of 10,12 conjugated linoleic acid on milk fat synthesis in dairy cows. Lipids 40:823- 832. Natural resources and bioeconomy studies 74/2020 88 Scollan, N., Hocquette, J.-F., Nuernberg, K., Dannenberger, D., Richardson, I. & Moloney, A. 2006. Review. Innovations in beef production systems that enhance the nutritional and health value of beef lipids and their relationship with meat quality. Meat Sci. 74:17-33. Shingfield K.J., Ahvenjärvi, S., Toivonen, V., Vanhatalo, A. & Huhtanen, P. 2007. Transfer of absorbed cis-9,trans-11 conjugated linoleic acid into milk is biologically more efficient than endogenous synthesis from absorbed vaccenic acid in lactating cows. J. Nutr. 137:1154- 1160. Shingfield, K.J., Ahvenjärvi, S., Toivonen, V., Vanhatalo, A., Huhtanen, P. & Griinari, J.M. 2008a. Effect of incremental levels of sunflower-seed oil in the diet on ruminal lipid metabolism in lactating cows. Br. J. Nutr. 99:971-983. Shingfield, K.J., Ahvenjärvi, S., Toivonen, V., Ärölä, A., Nurmela, K.V.V., Huhtanen, P. & Griinari, J.M. 2003. Effect of dietary fish oil on biohydrogenation of fatty acids and milk fatty acid content in cows. Anim. Sci. 77:165-179. Shingfield K.J., Bernard, L., Leroux, C. & Chilliard, Y. 2010a. Role of trans fatty acids in the nutritional regulation of mammary lipogenesis in ruminants. Animal 4:1140-66. Shingfield, K.J., Bonnet, M., & Scollan, N.D. 2013. Recent developments in altering the fatty acid composition of ruminant-derived foods. Animal 77:132-162. Shingfield, K.J., Chilliard, Y., Toivonen, V., Kairenius, P. & Givens, D.I. 2008b. Trans fatty acids and bioactive lipids in ruminant milk. In: Bösze, Z. (eds) Bioactive Components of Milk, Advances in Experimental Medicine and Biology, pp. 3-65, Springer, New York, NY. Shingfield, K. J., & J. M. Griinari. 2007. Role of biohydrogenation intermediates in milk fat depression. Eur. J. Lipid Sci. Technol. 109:799-816. Shingfield, K.J., Lee, M.R.F., Humphries, D. J., Scollan, N.D., Toivonen, V., Beever, D.E. & Reynolds, C.K. 2011. Effect of linseed oil and fish oil alone or as an equal mixture on ruminal fatty acid metabolism in growing steers fed maize silage-based diets. J. Anim. Sci. 89:3728-3741. Shingfield K.J., Lee, M.R.F., Humphries, D.J., Scollan, N.D., Toivonen, V., Reynolds, C.K. & Beever, D.E. 2010b. Effect of incremental amounts of fish oil in the diet on ruminal lipid metabolism in growing steers. Br. J. Nutr. 104:56-66. Shingfield, K.J., Reynolds, C.K., Hervás, G., Griinari, J.M., Grandison, A.S. & Beever, D.E. 2006a. Examination of the persistency of milk fatty acid composition responses to fish oil and sunflower oil in the diet of dairy cows. J. Dairy Sci. 89:714-732. Shingfield, K.J., Saebo, A., Saebo, P.C., Toivonen, V. & Griinari, J.M. 2009. Effect of abomasal infusions of a mixture of octadecenoic acids on milk fat synthesis in lactating cows. J. Dairy Sci. 92:4317-4329. Shingfield, K.J., Toivonen, V., Vanhatalo, A., Huhtanen, p. & Griinari, J.M. 2006b. Short communication: Indigestible markers reduce the mammary 9-desaturase index and alter the milk fatty acid composition in cows. J. Dairy Sci. 89:3006-3010. Spitzer, V. 1997. Structure analysis of fatty acids by gas chromatography - low resolution electron impact mass spectrometry of their 4,4-dimethyloxazoline derivatives. Prog. Lipid Res. 35:387-408. Toral P.G., Belenguer, A., Frutos, P. & Hervás, G. 2009. Effect of the supplementation of a high- concentrate diet with sunflower and fish oils on ruminal fermentation in sheep. Small Rumin. Res. 81:119-125. Toral, P.G., Belenguer, A., Shingfield, K.J., Hervás, G., Toivonen, V. & Frutos, P. 2012. Fatty acid composition and bacterial community changes in the rumen fluid of lactating sheep fed sunflower oil plus incremental levels of marine algae. J. Dairy Sci. 95:794-701 806. Natural resources and bioeconomy studies 74/2020 89 Toral, P.G., Bernard, L., Belenguer, A., Rouel, J., Hervás, G., Chilliard, Y. & Frutos, P. 2016a. Comparison of ruminal lipid metabolism in dairy cows and goats fed diets supplemented with starch, plant oil, or fish oil. J. Dairy Sci. 99:301-316. Toral, P.G., Bernard, L., Chilliard, Y. & Glasser, F. 2013. Short communication: Diet-induced variations in milk fatty acid composition have minor effects on the estimated melting point of milk fat in cows, goats, and ewes: Insights from a meta-analysis. J. Dairy Sci. 96:1232- 1236. Toral, P.G., Chilliard, Y., Rouel, J., Leskinen, H., Shingfield, K.J. & Bernard, L. 2015. Comparison of the nutritional regulation of milk fat secretion and composition in cows and goats. J. Dairy Sci. 98:7277-7297. Toral, P.G., Hervás, G., Carreño, D. & Frutos, P. 2018a. Does supplemental 18:0 alleviate fish oil- induced milk fat depression in dairy ewes? J. Dairy Sci. 99:1133–1144. Toral P.G., Hervás, G., Carreño, D., Leskinen, H., Belenguer, A., Shingfield, K.J. & Frutos, P. 2017. In vitro response to EPA, DPA, and DHA: Comparison of effects on ruminal fermentation and biohydrogenation of 18-carbon fatty acids in cows and ewes. J. Dairy Sci. 100:6187-6198. Toral, P.G., Hervás, G. & Frutos, P. 2018b.Use of high doses of 18:0 to try to mitigate the syndrome of milk fat depression in dairy ewes fed marine lipids. Anim. Feed Sci. Tech. 236:68–75. Toral, P.G., Hervás, G., Missaoui, H., Andrés, S., Giráldez, F.J., Jellali, S. & Frutos, P. 2016b. Effects of a tannin-rich legume (Onobrychis viciifolia) on in vitro ruminal biohydrogenation and fermentation. Spanish Journal of Agricultural Research v. 14, n. 1, p. e0602, mar. 2016. ISSN 2171-9292. Toral P.G., Hervás, G., Leskinen, H., Shingfield, K.J. & Frutos, P. 2018c. In vitro ruminal biohydrogenation of eicosapentaenoic (EPA), docosapentaenoic (DPA), and docosahexaenoic acid (DHA) in cows and ewes: intermediate metabolites and pathways. J. Dairy Sci. 101:6109-6121. Toral, P.G., Rouel, J., Bernard, L. & Chilliard, Y. 2014. Interaction between fish oil and plant oils or starchy concentrates in the diet: Effects on dairy performance and milk fatty acid composition in goats. Anim. Feed Sci. Technol. 198:67–82. Toral, P.G., Shingfield, K.J., Hervás, G., Toivonen, V. & Frutos, P. 2010. Effect of fish oil and sunflower oil on rumen fermentation characteristics and fatty acid composition of digesta in ewes fed a high concentrate diet. J. Dairy Sci. 93:4804-4817. Vahmani, P., Meadus, W.J., Rolland, D.C., Duff, P. & Dugan M.E.R. 2016. Trans10,cis15 18:2 Isolated from Beef Fat Does Not Have the Same Anti-Adipogenic Properties as Trans10,cis12-18:2 in 3T3-L1 Adipocytes. Lipids 51:1231-1239. Vanhatalo, A., Kuoppala, K., Toivonen, V. & Shingfield, K. 2007. Effects of forage species and stage of maturity on bovine milk fatty acid composition. Eur. J. Lipid Sci. Technol. 109:856-867. Ventto, L., Leskinen, H., Kairenius, P., Stefański, T., Bayat, A.R., Vilkki, J. & Shingfield, K.J. 2017. Diet-induced milk fat depression is associated with alterations in ruminal biohydrogenation pathways and formation of novel fatty acid intermediates in lactating cows. Br. J. Nutr. 117:364-376. Vlaeminck, B., Fievez, V., Cabrita, A.J.R., Fonseca, A.J.M. & Dewhurst, R.J. 2006. Factors affecting odd- and branched-chain fatty acids in milk: A review. Anim. Feed Sci. Technol. 131:389- 417. Natural resources and bioeconomy studies 74/2020 90 Vlaeminck, B., Harynuk, J., Fievez, V. & Marriott, P. 2007. Comprehensive two-dimensional gas chromatography for the separation of fatty acids in milk. Eur. J. Lipid Sci. Technol. 109:757- 766. Wahl, H.G., Habel, S.-Y., Schmieder, N. & Liebich, H.M. 1994. Identification of cis/trans isomers of methyl ester and oxazoline derivatives of unsaturated fatty acids using GC-FTIR-MS. J. High Resolut. Chromatogr. 17:543-548. Wallace, R.J., Chaudhary, L.C., McKain, N., McEwan, N.R., Richardson, A.J., Vercoe, P.E., Walker, N.D. & Paillard, D. 2006. Clostridium proteoclasticum: a ruminal bacterium that forms stearic acid from linoleic acid. FEMS Microbiol. Lett. 265:195-201. Wallace, R.J., McKain, N., Shingfield, K.J. et al. 2007. Isomers of conjugated linoleic acids are synthesized via different mechanisms in ruminal digesta and bacteria. J. Lipid Res. 48:2247- 2254. Wang, C., Harris, W.S., Chung, M., Lichtenstein, A.H., Balk, E.M., Kupelnick, B., Jordan, H.S. & Lau, J. 2006. n-3 Fatty acids from fish or fish-oil supplements, but not a-linolenic acid, benefit cardiovascular disease outcomes in primary- and secondary-prevention studies: a systematic review. Am. J. Clin. Nutr. 84:5-17. Wasowska, I., Maia, M.R.G., Niewiedźka, K.M., Czauderna, M., Ribeiro, J., Devillard, E., Shingfield, K.J. & Wallace, R.J. 2006. Influence of fish oil on ruminal biohydrogenation of C18 unsaturated fatty acids. Br. J. Nutr. 95:1199-1211. Whitlock, L.A., Schingoethe, D.J., Hippen, A.R., Kalscheur, K.F., Baer, R.J., Ramaswamy, N. & Kasperson, K.M. 2002. Fish oil and extruded soybeans fed in combination increase conjugated linoleic acids in milk of dairy cows more than when fed separately. J. Dairy Sci. 85:234-243. Wilson, R., Lyall, K., Payne, J.A. & Riemersma, R.A. 2000. Quantitative analysis of long-chain trans- monoenes originating from hydrogenated marine oil. Lipids 35:681-687. Yashodhara, B.M., Umakanth, S., Pappachan, J.M., et al. 2009. Omega-3 fatty acids: a comprehensive review of their role in health and disease. Postgrad. Med. J. 85:84-90. Yokoyama, M.T. & Davis, C.L. 1971. Hydrogenation of unsaturated fatty acids by Treponema (Borrelia) Strain B(2)5, a rumen spirochete. J. Bacteriol. 107:519-527. Zened, A., Enjalbert, F., Nicot, M.C. & Troegeler-Meynadier, A. 2013. Starch plus sunflower oil addition to the diet of dry dairy cows results in a trans-11 to trans-10 shift of biohydrogenation. J. Dairy Sci. 96:451-459. Zhang, J.Y., Yu, Q.T., Liu, B.N. & Huang, Z.H. 1988. Chemical modification in mass spectrometry IV- 2-alkenyl-4,4-dimethyloxazolines as derivatives for the double bond location of long-chain olefinic acids. Biomed. Environ. Mass Spectrom. 15:33-44. Natural resources and bioeconomy studies 74/2020 91 Natural Resources Institute Finland Latokartanonkaari 9 FI-00790 Helsinki, Finland tel. +358 29 532 6000