CHAPTER 1: Synthesis of Conjugated Linoleic Acid in Ruminants and Humans

There is increasing evidence that nutrition plays an important role in the development of human chronic diseases including cancer, cardiovascular disease, insulin resistance and obesity. Developing foods and diets that promote human health is central to public health initiatives for preventing and lowering the economic and social impact of chronic disease.¹  Direct and indirect costs of cardiovascular disease (CVD) have been estimated at $445 billion in the United States²  and €200 billion within the European Union.³  Global costs of CVD in 2010 totalled US$863 billion.⁴  These costs are projected to increase several-fold by 2030, reaching unsustainable levels due to people living longer and the rapid increase in obesity in developed and developing countries.^2,5 

Following the identification of the anti-mutagenic properties of conjugated linoleic acid (CLA) isomers in cooked beef,^6–8  numerous studies have investigated the biological activity of CLA isomers in cell culture and animal models (http://fri.wisc.edu/cla.php). Much of the research has focused on the effects of cis-9, trans-11 18:2 (trivial name rumenic acid) or trans-10, cis-12 18:2 due to the cost and availability in a range of mammalian and avian species. In addition to the inhibition of mutagenesis, specific isomers of CLA have been demonstrated to modulate energy metabolism, immunity, inflammation, insulin resistance and bone metabolism in several animal models.^9–21  However, evidence that the physiological effects described in vitro or in other mammalian species are also replicated in humans remains inconclusive.^22–28 

The optimal intake of one or more isomers of CLA in humans remains to be established. Direct or exponential extrapolation of data from studies in the rat model of carcinogenesis implicate intakes of cis-9, trans-11 18:2 between 95 and 3500 mg per day being required for significant decreases in cancer risk in human populations.^29,30  Estimates of cis-9, trans-11 18:2 consumption in human populations vary between 15 and 1500 mg per day^29,31–39  depending on the methodology used to estimate dietary intakes, with marked differences between countries, gender and socio-economic groups. Isomers of CLA are present in a wide range of foods including milk, beef and lamb, and to a much lesser extent in pork, poultry, fish and eggs, with trace amounts in some vegetable sources.^32,35–43  Milk and dairy products are the major source of CLA in the human diet contributing to between 66 and 80% of total intake.^31,35–39  Typically concentrations of CLA in pork, chicken and fish are lower than 0.1 g per 100 g lipid,⁴⁴  whereas the consumption of lamb, beef and other ruminant meat products account for 15–32% of average daily CLA intakes in developed countries.^{34–36,38,44} 

The CLA status of humans can be increased using oral supplements or fortification of foods with synthetic sources, which typically contain equal amounts of cis-9, trans-11 18:2 and trans-10, cis-12 18:2, or from a higher consumption of ruminant-derived foods. In contrast to synthetic sources, meat and milk from ruminants contain numerous positional and geometric isomers of CLA with conjugated double bonds at positions 6,8 through to 13,15, with cis-9, trans-11 18:2 being the major isomer, and trans-7, cis-9 18:2 or trans-11, cis-13 18:2 as the second most abundant.^44–47 

Producing ruminant-derived foods containing higher amounts of CLA offers the opportunity to increase the consumption of CLA, principally cis-9, trans-11 18:2, without requiring major changes in the habitual diet or eating habits. For this reason, a considerable amount of research has been dedicated to understanding the nutritional, physiological and genetic factors influencing CLA concentrations in meat and milk. The present chapter provides a comprehensive review of the most recent evidence on the biochemical, microbial, nutritional and physiological factors influencing the biosynthesis of CLA isomers in ruminants and humans.

Ruminant diets vary in composition depending on species, physiological state, and the cost and availability of feed ingredients. Diets often contain forage species (grasses, legumes or forage maize) of variable maturity and nutritional value, and differ in composition from those containing forages as the sole feed to combinations of forage, cereals and protein supplements. By-products of the food industry or lipid supplements may also be included. A general characteristic of ruminant diets is the relatively high fibre content (>300 g cell wall constituents per kg dry matter (DM)) and low amounts of lipid (<50 g per kg DM).⁴⁵  Lipid in cereal grains, plant oils, marine lipids and by-products are predominantly in the form of triacylglycerol (TAG). Most of the lipid in grasses and legume forages is present as phospholipids (PL) and glycolipids (GL) located within thylakoid membranes of chloroplasts.^48–50  In forages, GL are the major lipid class, with galactolipids (mono- and di-galactosyl diacylglyerol) being the most prevalent. These differ from TAG in that one or more carbohydrate molecules are linked to one position of the glycerol backbone. Forages and oilseeds also contain several PL species, (phosphatidylcholine, phosphatidylglycerol and phosphatidylethanolamine) as structural components of cell membranes. Most PL contain a diacylglycerol covalently bonded to a phosphate group, which is often esterified to a simple organic molecule such as choline. In general, both fatty acid moieties bound to glycerol in glycolipids or PL are unsaturated. Non-esterified fatty acids (NEFA) are minor components of most ruminant feeds, but are the major lipid class in ensiled grasses and forage legumes and in certain proprietary fat supplements. Changes arising during fermentation in silo are characterized by a substantial decrease in the relative abundance of polar membrane lipid and an increase in NEFA, TAG, diacylglycerol (DAG) and monoacylglycerol (MAG) fractions attributable to the activity of plant and microbial lipases.^51–54 

The amount and composition of constituent fatty acids differs substantially between ingredients in ruminant diets (Table 1.1). For grasses and legume forages fatty acid content is generally lower than 50 g per kg DM with cis-9, cis-12, cis-15 18:3 (α-linolenic acid; 18:3 n-3) as the major fatty acid (>50 g per 100 g fatty acids).^55–59  However, conservation of grass by drying results in substantial decreases in lipid content,^58,60,61  principally due to the disappearance of cis-9, cis-12 18:2 (linoleic acid; 18:2 n-6) and 18:3 n-3 by oxidation and leaf shatter.⁴⁹  Oxidation arises from the activity of plant lipoxygenases, which catalyse the incorporation of molecular oxygen in non-esterified 18:2 n-6 and 18:3 n-3, forming 9- and 13-hydroperoxy polyunsaturated fatty acids, respectively, that are highly reactive and rapidly metabolized into a series of oxylipins including volatile leaf aldehydes and alcohols, hydroxy- and epoxy-fatty acids and jasmonates.^62,63  In contrast, 18:2 n-6 is the predominant fatty acid in forage maize, whole crop silages and cereal grains.^45,57,64  Ruminant diets may also contain up to around 50 g per kg DM of additional lipid in the form of oils or oilseeds. Oils from rapeseeds, high oleic sunflowerseeds, olives and peanuts are a rich source of cis-9 18:1 (oleic acid),^65,66  cottonseeds, safflowerseeds, soyabeans and sunflowerseeds are abundant in 18:2 n-6,^65,67  whereas linseeds^65,68  and camelina^69,70  contain relatively high proportions of 18:3 n-3 (Table 1.1). In contrast, coconut oil is rich in 12:0 (lauric acid), whereas 16:0 (palmitic acid) is the major fatty acid in palm oil.⁶⁵  Ruminant diets may also be supplemented with fish oil containing cis-5, cis-8, cis-11, cis-14, cis-17 20:5 (eicosapentaenoic acid; 20:5 n-3) and cis-4, cis-7, cis-10, cis-13, cis-16, cis-19 22:6 (docosahexaenoic acid; 22:6 n-3)⁶⁸  or marine algae enriched in 22:6 n-3.⁷¹ 

Following ingestion and mastication, the ester linkages of TAG, PL and GL are rapidly hydrolysed in the rumen.^72–74  Hydrolysis of dietary TAG occurs predominantly as a result of microbial lipases.⁷⁴  Forage plant tissues are also rich in endogenous galacto- and phospholipases, which remain active once ingested into the rumen for several hours, suggesting that senescence of the plant material itself in the rumen may contribute to ruminal lipolysis in grazing animals.⁷⁵  Dawson et al.⁷⁴  challenged this idea and concluded that microbial lipases were more important than plant enzymes. These conclusions were drawn using autoclaved grass as a substrate, which the authors noted was not ideal because of the many effects that autoclaving has in addition to enzyme denaturation. This debate has been revisited by Lee et al.,⁷⁶  who reported increased NEFA and decreased polar lipid abundance after 6 h incubation of fresh ryegrass leaves in buffer, confirming that plant-catalysed lipolysis occurred. Further, Van Ranst et al.⁷⁷  reported lipolysis of up to 60% after 8 h incubation of fresh red clover leaves. Both studies suggested the observed lipolysis to be due to active plant lipases that could contribute to overall ruminal lipolysis. However, until plant lipase activity is compared directly with that of ruminal micro-organisms, there will remain an uncertainty about their relative importance. Nonetheless, it may be useful to breed forage plants low in lipase activity in order to increase the amount of polyunsaturated fatty acids escaping the rumen. Plant lipids may also be compartmentalized in the ingested plant material, effectively protecting them from both endogenous and microbial hydrolysis.⁷⁸ 

Among the various types of ruminal micro-organisms, the bacteria are considered to be most active in lipolysis.⁴⁸  The most active bacterial species isolated selectively using TAG as a substrate was Anaerovibrio lipolytica,⁷⁹  with which most research has been carried out. More recently, Jarvis et al.⁸⁰  isolated two bacteria from red deer that hydrolysed TAG and grew on glycerol. The bacteria, which were highly active against tallow, tripalmitin and olive oil, were most closely related to the genus Propionibacterium and clostridial cluster XIVa. Cirne et al.⁸¹  isolated Clostridium lundense sp. nov. from the bovine rumen. Clostridium lundense exhibited lipolytic activity against olive oil, but neither its activity nor its numbers was reported, so it is difficult to assess its likely importance. Thus, a wider range of bacteria than is usually considered may be involved in the lipolysis of TAG in the rumen.

A. lipolyticus lacks the ability to hydrolyse galactolipids and PL and, therefore, other lipolytic species would be expected to predominate in grazing ruminants. The Butyrivibrio spp. appeared to contain all the phospholipase A, phospholipase C, lysophospholipase and phosphodiesterase activities typical of the mixed rumen contents.⁴⁸  Their lipase activity against TAG varies between different Butyrivibrio and Pseudobutyrivibrio strains, but not in a manner that corresponds to their position in the phylogenetic tree.⁸²  Little is known about how other lipases vary across different strains/species, nor whether other recently recognized species may possess such activities in the rumen of grazing ruminants.

There have been few recent studies that investigate protozoal lipolysis. Wright⁸³  suggested Epidinium spp. to be responsible for 30–40% of the lipolytic activity in the rumen. Epidinium ecaudatum was reported to liberate galactose from galactolipids, suggesting galactosidase activity, although lipase activity per se was not demonstrated.⁸⁴  Another protozoal species, Entodinium caudatum, was shown to have phospholipase activity,⁸⁵  but it is possible that this activity was more relevant to the intracellular metabolism of the protozoa than to the digestion of dietary lipids. The earlier studies to determine the contribution of protozoa to the lipolytic activity in the rumen were conducted using fractionated rumen fluid, with the possibility that lipolytic activity in protozoal fractions was more due to the activity of bacteria that the protozoa had ingested than that of the protozoa themselves. Once again, given that protozoa comprise up to half the microbial biomass present in the rumen, their lipolytic properties warrants further investigation.

Much is known about bacterial lipases in general. They comprise eight well-documented families and their modes of action are reasonably well characterized.⁸⁶  The lipase activity of A. lipolyticus was investigated in some detail by methods available at the time, some four decades ago.^87,88  Perhaps surprisingly, no cloning and sequencing studies seem to have been done when the technology became widely available, and the most detailed recent study⁸⁹  was derived from genomic analysis. Three enzymes from A. lipolyticus were identified from genomic analysis of A. lipolyticus. The alipA, alipB and alipC encoded 492-, 438- and 248-amino acid peptides, respectively. Phylogenetic analysis indicated that alipA and alipB clustered with the GDSL/SGNH family II, and alipC clustered with lipolytic enzymes from family V. Subsequent expression and purification of the enzymes showed that they had esterase activities with substrate specificities favouring the hydrolysis of caprylate-, laurate- and myristate-containing substrates. Genomes are available for several species in the Butyrivibrio group, but to date no similar analysis appears to have been carried out for their lipase activities.

Metagenomic methods will be invaluable in order to understand the full complement of lipolytic enzymes that are present in the rumen. Expression libraries may be useful. Liu et al.⁹⁰  screened a metagenomic library from the rumen of grazing Holstein cows for lipase activity, using trioleoylglycerol as substrate. Out of 15 360 bacterial artificial chromosome (BAC) clones, only two were found to have high lipase activity, which seems surprisingly small. The likely origin of the genes was investigated, based on the other open reading frames (ORFs) present in the BAC clones. It was impossible to decipher the host for the first gene, Rlip1, but the second, Rlip2, gave most similar homologues from Thermosinus carboxydivorans, which has not previously been associated with the ruminal ecosystem. Nonetheless, by BLASTing their deposited lipase sequences it was discovered that the genes most likely encoded carboxyl esterases. Thus the prevalence of these lipases in the overall community is far from certain. Clearly many more lipase sequences must be analysed from the metagenome, and assignment to lipases assured, in order to understand the true nature of lipolytic enzymes active in the rumen.

Lipolysis is considered rate limiting for the biohydrogenation of dietary unsaturated fatty acids in the rumen.⁴⁸  During ruminal digestion of grasses, mono- and digalactosyl diacylglycerides are released following the rupture of chloroplasts.⁷³  Early studies involving incubations of ¹⁴C-labelled substrate with strained rumen fluid indicated that lipolysis conforms to first-order kinetics, occurring at a high rate with a short lag time.⁹¹  These observations led Hawke and Silcock⁹¹  to conclude that hydrolysis of dietary acyl lipids in the rumen would be sufficiently rapid not to impede biohydrogenation. However, such conclusions were not supported by observations that the profile of products formed during intra-ruminal infusion of esterified or non-esterified 18:2 n-6 in sheep differed.^92,93 

Much of what is known about the factors influencing ruminal lipolysis is based on incubations of various substrates with rumen fluid. Laboratory-scale experiments have obvious drawbacks with respect to mimicking conditions in vivo. Accepting these limitations and the assumptions therein, a number of factors have been determined to influence the rate and extent of lipolysis in a simulated rumen environment (Figure 1.1). Increases in dietary nitrogen content were found to increase the rate at which triolein was hydrolysed,⁹⁴  whereas the rate of lipolysis was decreased by replacing fibre with starch⁹⁵  or when more mature forages were incubated.⁹⁶  As could be expected, lipolysis increases with incubation time,^97,98  but is often decreased at low rumen pH.^99,100  The extent of lipolysis is also decreased for lipids with a higher melting point containing relatively high proportions of saturated fatty acids.⁴⁵  Increasing the amount of substrate incubated also lowers progressively the rate and extent of lipolysis in vitro.⁹⁷  After 24 h incubations with buffered rumen fluid, between 73.7% and 89.5% of TAG in fish oil, linseed oil or sunflower oil was found to be hydrolysed.⁹⁸  Similarly, lipolysis of lipid in cocksfoot and red clover after 24 h incubations were reported to vary between 65.0% and 82.0%.^101,102 

Measurements of lipolysis in vivo are scarce. Studies in cattle,^54,61  goats¹⁰³  and sheep¹⁰⁴  all affirm that a small fraction of esterified dietary lipid escapes the rumen. In lactating cows fed diets based on grass silage or red clover silage, NEFA, PL, TAG, DAG and MAG fractions were found to account for 80%, 12%, 4.4%, 2.4% and 0.8% of total fatty acids in omasal digesta.⁵⁴  The majority of 18:3 n-3 entered the omasum in the form of polar lipid (42–46%) originating from plant PL and GL, with smaller amounts as NEFA (22–26%), TAG (17–20%), DAG (8–14%) or MAG (2–3%). Between 85% and 93% of esterified lipid was reported to be hydrolysed in the rumen of lactating cows fed diets based on fresh grass or grass silage (forage:concentrate ratio 60:40), the extent of which was decreased to 80–86% and 70% when grass silage was replaced with grass hay⁶¹  or red clover silage,⁵⁴  respectively. While incomplete, ruminal lipolysis is extensive, resulting in NEFA accounting for more than 80% of the total amount of fatty acids leaving the rumen.^{48,54,61,103,104} 

From a historical perspective, the existence and importance of ruminal biohydrogenation have been known for more than 50 years. It has only been in recent years following the discovery of the biological activities of CLA isomers that research on ruminal lipid metabolism has focused more on the biochemical mechanisms involved in the formation of the less abundant fatty acid intermediates. Advances in lipid analysis allowed the marked differences in the fatty acid profile of lipids in the diet (principally cis-9 18:1, 18:2 n-6 and 18:3 n-3) and that of lipids leaving the rumen (mainly 16:0 and 18:0) to be established.,¹⁰⁵  These differences were discovered to be the consequence of the process we now call biohydrogenation, which converts unsaturated fatty acids to saturated fatty acids via firstly a cis-trans isomerization to trans fatty acid intermediates followed by hydrogenation of the double bonds (Figure 1.1).^48,106,107  Little if any oxidation or elongation of the carbon chain is thought to occur.⁴⁸ 

Early evidence of ruminal biohydrogenation was obtained when linseed oil was incubated with sheep ruminal contents.¹⁰⁸  The 18:3 n-3 content of the linseed oil decreased from 30% to 5% with an accompanying increase in the concentration of 18:2. Shorland et al.¹⁰⁵  also noted that the ruminal metabolism of 18:3 n-3 caused the accumulation of 18:2 and 18:1 intermediates, which were then converted to 18:0. The requirement for a free carboxyl group for hydrogenation established that lipolysis must precede biohydrogenation.⁹¹  After these initial studies demonstrated the overall features of biohydrogenation in ruminal contents, various in vivo and in vitro experiments were carried out to examine the pathways involved in ruminal digesta. Among these, the work of Dawson and his co-workers was particularly informative. Wilde and Dawson¹⁰⁹  constructed a metabolic scheme for the metabolism of 18:3 n-3 to 18:0 based on incubations of sheep ruminal contents with U-¹⁴C labelled 18:3 n-3. The initial step was the isomerization of the cis-12 bond to either the Δ11 or Δ13 position. Thereafter, one of the bonds was hydrogenated to leave an 18:2, followed by hydrogenation of another bond producing an 18:1 intermediate. Hydrogenation of the 18:1 intermediate yielded 18:0 as the final product.

The other early contribution to delineating biohydrogenation pathways was the seminal work of Tove and his group with the bacterial species, Butyrivibrio fibrisolvens, that had been identified by Polan et al.¹¹⁰  to possess high biohydrogenating activity. When incubated with B. fibrisolvens, 18:2 n-6 was initially isomerized to a conjugated 18:2, thought to be mainly cis-9, trans-11 18:2, but other dienoic isomers were also detected.¹¹¹ The cis-9, trans-11 18:2 intermediate was then hydrogenated to trans-11 18:1 (vaccenic acid) as the final product. However, in mixed ruminal bacteria, the 18:1 intermediate was further hydrogenated to 18:0.¹¹² Kepler and Tove¹¹³  also incubated B. fibrisolvens with 18:2 n-6 and 18:3 n-3. They confirmed that 18:2 n-6 was first isomerized to cis-9, trans-11 18:2 followed by further hydrogenation to a mixture of trans 18:1 isomers. When 18:3 n-3 was used as a substrate, it was first isomerized to cis-9, trans-11, cis-15 18:3 (conjugated linolenic acid). This product was transient and further hydrogenated to a non-conjugated 18:2 intermediate as the final product. Slightly later, bacteria were isolated that converted monoenoic acids to 18:0 and hydrogenated 18:2 n-6 to 18:0, with traces of the intermediates that had been seen with B. fibrisolvens and mixed ruminal microbes.^114,115  The first step of 18:3 n-3 biohydrogenation is analogous to that of 18:2 n-6, namely the formation of a conjugated bond structure, cis-9, trans-11, cis-15 18:3.

Biohydrogenation of cis-6, cis-9, cis-12 18:3 (γ-linolenic acid, 18:3 n-6) during incubations with pure strains of Butyrivibrio and a bacterium identified as Fusocillus babrahamensis has also been investigated.¹¹⁶  The group B bacterium (Fusocillus) hydrogenated 18:3 n-6 to 18:0, whereas incubations with the group A bacterium (Butyrivibrio) yielded cis-6, trans-11 18:2 as an end product. By analogy with the metabolic pathway described for 18:3 n-3, biohydrogenation of 18:3 n-6 is thought to proceed via isomerization of the cis-12 double bond to yield cis-6, cis-9, trans-11 18:3, that is sequentially reduced to cis-6, trans-11 18:2, trans-11 18:1 and 18:0.¹¹⁶  Most feeds fed to ruminants do not contain 18:3 n-6. However, certain oilseeds including evening primrose and borage are relatively abundant in 18:3 n-6, but the effects of these lipids on ruminal lipolysis and biohydrogenation in vivo have not been investigated.

Several recent studies have investigated the metabolism of cis-6, cis-9, cis-12, cis-15 18:4 (stearidonic acid; 18:4 n-3) during incubations with rumen fluid.^117,118  Numerous intermediates were found to be formed. Based on the profile of fatty acids detected, the metabolism of 18:4 n-3 was proposed to proceed via the formation of 5,7,11,15 18:4 or 5,8,10,15 18:4.¹¹⁸  Conjugated 18:4 products were reduced to yield 5,11,15 18:3 or 5,10,15 18:3 and hydrogenated yet further to 11,15 18:2 and 5,10 18:2, respectively.¹¹⁸  Trans-6 to -12 18:1 were found to accumulate with trans-11 18:1 being the major intermediate formed.¹¹⁷ 

Much less is known about the fate of longer chain polyenoic fatty acids in the rumen. Detailed analysis of ruminal^119,120  and omasal digesta^121,122  have indicated that numerous 20- and 22-carbon biohydrogenation intermediates containing one or more trans double bonds are formed during the biohydrogenation of fatty acids in fish oil or marine algae. However, the biochemical pathways responsible are not known. The most recent investigations suggest that the first committed steps of 20:5 n-3 and 22:6 n-3 biohydrogenation in the rumen involve the reduction and/or isomerization of double bonds closest to the carboxyl group.^121,122  No conjugated ≥20-carbon fatty acids have been detected.

Ruminal biohydrogenation of cis-9 18:1, 18:2 n-6 and 18:3 n-3 in vivo varies from 58–87%, 70–95% and 85–100%, respectively.^107,123,124  The extent of ruminal fatty acid biohydrogenation is influenced by the composition and amount of lipid in the diet, retention time in the rumen and the characteristics of the microbial population. Biohydrogenation is extensive, with 18:0 being the major fatty acid leaving the rumen on most diets. However, the reduction of unsaturated 18-carbon fatty acids to 18:0 in the rumen is incomplete and numerous 18:1, 18:2 and 18:3 intermediates accumulate. The final reduction of trans 18:1 is thought to be the rate limiting step in the complete biohydrogenation of 18-carbon unsaturated fatty acids.¹²⁵  Decreases in pH from 6.5 to below 6.0 lowers the extent of cis-9 18:1,^126,127  18:2 n-6^127–131  and 18:3 n-3^127,129,132  biohydrogenation and inhibits the final reduction of trans 18:1 to 18:0^127,132  in vitro. The effects of pH below normal physiological ranges in the rumen may be related to disruption of bacterial cell membranes resulting in the inactivation of membrane bound isomerases and reductases.¹¹³ 

Characterizing the biohydrogenation process in vivo represents a major challenge. Relatively few studies have reported sufficiently detailed measurements of post-ruminal fatty acid outflows to allow for inferences on metabolic pathways. The situation is further complicated due to oils and oilseeds being used to alter substrate supply rather than pure fatty acids. In this regard incubations of fatty acids with rumen fluid or pure strains of ruminal bacteria have proven invaluable in establishing the profile of possible intermediates formed during the biohydrogenation of unsaturated fatty acids (Table 1.2). However, intermediates formed in vitro may not necessarily parallel biohydrogenation pathways in vivo. Discrepancies may arise from the incubation of excessively high amounts of substrate, methods used to introduce fatty acids into batch cultures, source of ruminal fluid and potential loss of microbial activity over an extended incubation period.¹⁰⁷ 

Early studies demonstrated that mixed or pure cultures of ruminal bacteria were capable of hydrogenating cis-9 18:1 to 18:0.⁴⁸  More recent experiments have indicated that cis-9 18:1 is also hydrated to 10-OH 18:0 (10-hydroxystearic acid), an intermediate that is further oxidized to 10-O 18:0 (10-ketostearic acid) during incubations with ruminal contents.^133,134  Hydration of the cis-9 double bond involves the incorporation of one hydrogen atom from water.¹³⁴  A select number of bacteria isolated from the rumen, including species identified as Fusocillus babrahamensis,¹³⁵  Selenomonas ruminantium,¹³⁶  Enterococcus faecalis,¹³⁶  Streptococcus bovis¹³⁷  and Propionibacterium acnes¹³⁴  are capable of hydrating cis-9 18:1. P. acnes is known to catalyse the oxidation of 10-OH 18:0 to 10-O 18:0.¹³⁴  Incubations of 1-¹³C cis-9 18:1 with rumen contents have indicated that cis-to trans isomerization of the Δ9 double bond also takes place, resulting in the formation of trans-6, -7, -9, -10, -11, -12, -13, -14, -15, and -16 18:1 intermediates.¹³⁸  Further investigations of the metabolism of 1-¹³C cis-9 18:1 by mixed rumen microbes grown in continuous culture at pH 5.5 or pH 6.5 and at different dilution rates indicated that decreases in pH and dilution rate prevented the formation of trans 18:1 intermediates with a double bond beyond the Δ10 position.¹²⁶  There is also evidence to suggest that trans to cis isomerization may occur, following reports that incubations of trans-9 18:1 with mixed ruminal bacteria resulted in the production of cis-9 18:1 and cis-11 18:1.¹³⁹  Incubations with pure or mixed cultures of ruminal bacteria have allowed several biochemical pathways of cis-9 18:1 metabolism to be characterized (Figure 1.2).

For many years isomerization of the cis-12 double bond leading to the formation of cis-9, trans-11 CLA has been considered the first committed step of 18:2 n-6 biohydrogenation in the rumen, followed by reduction to trans-11 18:1 and 18:0 (Figure 1.3). More recent studies have shown that geometric isomers of both 9,11 and 10,12 18:2 are formed during incubations of 18:2 n-6 with rumen fluid^{127,132,140,141}  and pure cultures of several ruminal bacteria.¹⁴²  The amounts of 9,11 and 10,12 CLA isomers formed during the isomerization of 18:2 n-6 also appear to be pH-dependent, but reports are conflicting. Some investigations demonstrated that decreases in pH below 6.0 lower the accumulation of cis-9, trans-11 18:2 and trans-10, cis-12 18:2,^127,129  whereas increased formation was reported in others.^130,131  Conjugated 18:2 products formed during the initial isomerization of 18:2 n-6 are transient and reduced to yield both trans (-4, -5, 6-8, -9, -10, -11, -12, -13 and -14) and cis (-9, -10, -11, -12 and -13) 18:1 intermediates^134,140,141 (Table 1.2). Decreases in pH from 6.4 to 5.6 have been shown to promote trans-10 18:1 formation, changes accompanied by a decrease in the accumulation of trans-11 18:1 accumulation.¹²⁷  Increases in the amounts of 18:2 n-6 incubated have also been shown to increase the ratio of trans-10 18:1/trans-11 18:1.^141,143 

Incubations with strains of Enterococcus faecalis isolated from the rumen have demonstrated that 18:2 n-6 may also be hydrated to yield 10-OH, cis-12 18:1 and cis-9, 13-OH 18:1.¹⁴⁴  Recent investigations provided the first evidence that biohydrogenation of 18:2 n-6 may also proceed by a mechanism involving the migration of the cis-9 double bond rather than hydrogen abstraction or isomerization of the cis-12 double bond resulting in the formation of cis-6, cis-12 18:2, cis-7, cis-12 18:2 and cis-8, cis-12 18:2.¹⁴¹ 

Owing to the higher number of double bonds the metabolic pathways responsible for the biohydrogenation of 18:3 n-3 in the rumen are far more complex than for cis-9 18:1 or 18:2 n-6. Most investigations indicate the metabolism of 18:3 n-3 in the rumen involves the formation of a single 18:3 intermediate cis-9, trans-11, cis-15 18:3, that is sequentially reduced to trans-11, cis-15 18:2, followed by trans-11 18:1 with 18:0 as an end product^{48,109,113,145}  (Figure 1.4).

Further studies have demonstrated that a diverse range of intermediates are formed during incubations of 18:3 n-3 with ruminal bacteria. In addition to cis-9, trans-11, cis-15 18:3, trans-9, trans-11, cis-15 18:3 is also produced during the initial isomerization of 18:3 n-3.¹⁴⁶  Biohydrogenation of 18:3 n-3 has been reported to result in the accumulation of numerous conjugated 18:2 (cis-9, cis-11 18:2, cis-11, trans-13 18:2 and trans-11, trans-13 18:2), non-methylene-interrupted 18:2 (cis-9, cis-15 18:2, cis-9, trans-13 18:2, trans-9, cis-12 18:2, trans-11, cis-15 18:2, trans-9, trans-12 18:2), trans 18:1 (6-8,-9, -11, -12, -13-14, -15 and -16) and cis 18:1 (-13 and -15) intermediates.¹⁴⁰  Two strains of B. fibrisolvens have also been identified as capable of converting trans-11,cis-15 18:2 to trans-11, cis-13 18:2.¹⁴⁷  The complexity of the reactions responsible for the biohydrogenation of 18:3 n-3 was reinforced following reports that 14 positional and geometric isomers of 18:3 isomers were formed during incubations of U-¹³C 18:3 n-3 with ruminal contents.¹⁴⁸  After 48 h of incubation, ¹³C enrichment was detected in trans-8, trans-10 18:2, trans-11, trans-13 18:2 and geometric isomers of 9, 11 and 10, 12 18:2. This is the first report that cis-9, trans-11 18:2 may also be formed from 18:3 n-3. In addition to studies in vitro, feeding diets enriched in 18:3 n-3 have been shown to increase cis-9, trans-13, cis-15 18:3, cis-9, trans-11, trans-13 18:3, trans-9, trans-11 18:2, trans-13, trans-15 18:2, 11,13 18:2 and 12,14 18:2 concentrations in bovine milk.^149–151  One or more of these fatty acids may originate from the rumen. While the major pathway describing the biohydrogenation of 18:3 n-3 in the rumen is well established, those responsible for the formation of minor intermediates remain incomplete and require further investigation.

We are indebted to the early work of Tove and his colleagues for establishing many of the properties of the first step in the metabolism of 18:2 n-6 and 18:3 n-3.^113,152,153  Linoleate isomerase (EC 5.2.1.5) is described as an enzyme that catalyses the conversion of 18:2 n-6 to other positional and geometric 18:2 isomers, including conjugated intermediates. The properties of this enzyme activity vary according to the host bacterial species and the product that is formed. Kepler and Tove¹¹³  carried out a partial purification of linoleate isomerase activity from B. fibrisolvens and developed a convenient assay method based on the UV absorbance of the conjugated double bond system at 233 nm. The particulate cell-free preparation formed mainly cis-9, trans-11-18:2 from 18:2 n-6; the equilibrium was strongly in favour of cis-9,trans-11 18:2 formation. Perhaps significantly when it comes to understanding the mechanism, the authors failed to isolate a homogeneous preparation that was active. Furthermore, the reaction was short-lived in reaching an endpoint, and the reaction could be restored by the addition of more enzyme, suggesting that a cofactor was required. Isomerization of 18:3 n-3 occurred in a similar manner, but at a faster rate, to cis-9, trans-11, cis-15 18:3. Further refinement of the substrate specificity and mechanism of action followed,^152,153  but little further work was done for the following three decades.

In contrast, cloning, crystallization and structural analysis of the linoleate isomerase catalysing the formation of trans-10, cis-12 18:by non-ruminal P. acnes revealed at the atomic level, with a single protein, a mode of action that involves hydride abstraction by enzyme-bound FAD.^154,155  Isotope studies with a ruminal P. acnes were consistent with this mode of action.¹⁴² 

In non-ruminal lactic acid bacteria, which produce both isomers of CLA, it has been proposed that cis-9, trans-11 18:2 formation from 18:2 n-6 by lactic acid bacteria involves a hydration-dehydration mechanism via a 10-hydroxy,cis-12 18:1 intermediate.¹⁵⁶  Such a mechanism has been eliminated as a possible route of cis-9, trans-11 18:2 synthesis from 18:2 n-6 by B. fibrisolvens, since 10-hydroxy, cis-12 18:1 was not converted to cis-9, trans-11 18:2.¹⁴²  The Lactobacillus linoleate isomerase seems, like the B. fibrisolvens enzyme, to be a multi-component enzyme,¹⁵⁷  one component derived from the membrane fraction and another from the soluble cell content.

Two publications have claimed to have identified a gene encoding a linoleate isomerase that forms cis-9, trans-11 18:2. Park et al.¹⁵⁸  published a sequence purporting to be a linoleate isomerase from B. fibrisolvens A-38. The gene sequence of linoleate isomerase in non-ruminal lactic acid bacteria has been published within a patent.¹⁵⁹  Our own in silico analysis suggests that both claims may be mistaken. Furthermore, BLASTing the sequences against the recently available genomes of human intestinal species that produce cis-9, trans-11 18:2, including B. fibrisolvens, Roseburia inulinivorans and R. hominis (the last two are cis-9, trans-11 18:2 producers from the human intestine), failed to identify credible isomerase genes. In other preliminary studies, we have attempted to use published data from studies of the CLA reductase of B. fibrisolvens to identify neighbouring genes as possible candidates for linoleate isomerase, based on the assumption that CLA reductase and linoleate isomerase might be coordinately expressed. Fukuda et al.¹⁶⁰  isolated a mutant of B. fibrisolvens that formed exceptionally high concentrations of cis-9, trans-11 18:2 from 18:2 n-6. This resulted from the loss of the CLA reductase activity.¹⁶¹  The mutant gene was identified, sequenced and expressed in E. coli, reportedly producing active linoleate isomerase. However, when we and others (G.T. Attwood, personal communication) BLASTed the deposited sequence against protein and nucleotide databases, a high similarity to oxaloacetate decarboxylase was revealed. Indeed, when we investigated the genomes of the previously described human bacteria, greater than 80% identity was obtained with a gene annotated as an oxaloacetate decarboxylase alpha subunit OadA. Clustering with this gene, rather than finding an ORF of unknown function, we found genes encoding transport and other genes relating to oxaloacetate metabolism. Oxaloacetate decarboxylase is very well characterized. Once again, therefore, the published sequence seems to be in error.

There is little doubt that bacteria are the main organisms responsible for biohydrogenation in the rumen, but equally there is no debate that ruminal fungi and particularly protozoa have roles in determining the amounts of CLA and trans 18:1 isomers leaving the rumen.¹⁰⁶ 

Polan et al.¹⁶²  identified B. fibrisolvens as a bacterium that rapidly hydrogenates 18:2 n-6, forming cis-9, trans-11 18:2 and trans-11 18:1 as intermediates. Biohydrogenation of B. fibrisolvens did not result in the formation of 18:0. Subsequent studies^115,163  identified other bacteria that were capable of biohydrogenation, but the results did not provide much information about relative activities. The method used radio-labelled fatty acids as substrates and was therefore highly sensitive, but the concentrations of labelled acids used as substrates were very low (2 μg ml^-1), making comparisons of specific activity and relative contribution of different species difficult. Even minor activity could have led to a positive result. Among the bacteria capable of forming 18:0, two were identified as ‘Fusocillus’ spp.¹¹⁵  Fusocillus is not a genus that has endured in bacterial taxonomy, nor did any of the cultures survive into the DNA-sequencing era. Van de Vossenberg and Joblin¹⁶⁴  isolated a bacterium from a cow at pasture that could also form 18:0 from 18:2 n-6. It was phenotypically similar to ‘Fusocillus' and 16S rRNA analysis indicated that it was phylogenetically close to Butyrivibrio hungatei. Subsequently, a species named Clostridium proteoclasticum was identified as a 18:0 producer with morphological and metabolic properties that were indistinguishable from those reported for Fusocillus.¹⁶⁵  Moon et al.¹⁶⁶  reclassified C. proteoclasticum as Butyrivibrio proteoclasticus from its 16S rRNA gene sequence. The bacteria involved in the different steps of the biohydrogenation process were classified for many years as group A and B.⁴⁸  Group A bacteria hydrogenated 18:2 n-6 and 18:3 n-3 to trans-11 18:1, whereas group B bacteria converted both fatty acids to 18:0. It is now more appropriate to describe the bacteria based on their correct taxonomy (Figure 1.5).

Phenotypically, the B. hungatei and B. proteoclasticus groups are much more sensitive to the toxic effects of unsaturated fatty acids than the rest of the Butyrivibrio/Pseudobutyrivibrio cluster, such that their isolation from media containing unsaturated fatty acids is made much more difficult.^82,165  They can also be distinguished phenotypically and genetically based on the mechanism by which they form 4:0 (butyrate), their second most abundant fermentation product after 2:0 (acetate). B. hungatei and B. proteoclasticus had a butyrate kinase activity >600 U per mg protein, while the others had much lower activity⁸²  (Figure 1.5). The butyrate kinase gene was present in B. hungatei and B. proteoclasticus but not in the other group. It has been suggested that the different sensitivities to the toxic effects of CLA isomers and trans-11 18:1 may be linked to the enzyme mechanism by which butyrate is produced. Intracellular acyl-CoA concentrations, principally the precursors of 4:0, in B. fibrisolvens are depleted when the bacterium is exposed to 18:2 n-6.¹⁶⁷ 

Metabolism of 18:2 n-6 by Butyrivibrio results in the formation of cis-9, trans-11 18:2, with smaller amounts of trans-9, trans-11 18:2, and trans-11 18:1, but no trans-10,cis-12 18:2 or trans-10 18:1 is formed.¹³⁴  The bacteria responsible for trans-10 18:1 must therefore be different to the Butyrivibrio spp. The formation of trans-10, cis-12 18:2 also occurs by a different enzymic mechanism to that of cis-9, trans-11 18:2.¹⁴²  Enrichment cultures with starch were reported to isomerize 18:2 n-6 to trans-10, cis-12 18:2,¹⁶⁸  but these also contained abundant large cocci identified as Megasphaera elsdenii. Authors concluded that M. elsdenii as capable of trans-10, cis-12 CLA production in the rumen. Subsequent studies indicated that P. acnes could be responsible for the formation of trans-10, cis-12 18:2.¹⁶⁵  Wallace et al.¹⁶⁵  found no lineate isomerase activity in any pure culture of M. elsdenii. Furthermore, when digesta samples from cows producing high amounts of trans-10, cis-12 18:2 were analysed for M. elsdenii by qPCR of 16S rRNA genes, numbers were <10³ g^-1, while much larger numbers of P. acnes were detectable (R.J. Wallace and S. Muetzel, unpublished observation). P. acnes does not, however, convert trans-10, cis-12 18:2 to trans-10 18:1.¹³⁴  Butyrivibrio spp. are capable of reducing trans-10, cis-12 18:2 and trans-10, trans-12 18:2.¹³⁴  However, it remains difficult to understand how Butyrivibrio would catalyse the reduction of trans-10 18:1 but not the isomerization of 18:2 n-6 to trans-10, cis-12-18:2. Shifts in biohydrogenation pathways favouring the production of trans-10 18:1 are known to occur in ruminants fed high concentrate diets containing unsaturated fatty acids^169–171  or during incubations of 18:2 n-6 with mixed ruminal bacteria containing starch^172,173  or maintained at a low pH.¹²⁷  Recently, switching from high forage to high concentrate diets was shown to be accompanied by major changes in the bacterial community in lactating cows,¹⁷⁴  conditions under which a shift towards the accumulation of trans-10 18:1 at the expense of trans-11 18:1 could be expected.^175,176  Non-ruminal Lactobacillus plantarum converts 18:2 n-6 to cis-9, trans-11 18:2 and trans-9, trans-11 18:2.¹⁷⁷  Of particular relevance to the trans-10 shift, L. plantarum reduces geometric 9,11 18:2 isomers to trans-10 18:1 rather than trans-11 18:1. Lactobacillus spp. have commonly been isolated from the rumen.¹⁷⁸  Their numbers increase with dietary starch content.¹⁷⁹  However, they have been associated with hydration rather than hydrogenation of unsaturated fatty acids.¹⁸⁰  A future challenge will be to associate with certainty the role of individual microbial species with alterations in rumen function and biohydrogenation pathways, such as those indicated from ribosomal intergenic spacer analysis.¹⁷⁴ 

There are often significant differences between bacterial communities attached to solids and those free-swimming, ‘planktonic’ communities that inhabit the liquid phase of the fermentation mixture. The same is true for ruminal lipid metabolism, with the planktonic community converting 18:2 n-6 as far as 18:1 intermediates, whereas the solids-associated community reduced 18:2 n-6 to 18:0¹⁸¹ . B. proteoclasticus was present only in the solids-associated community, at 12% of Butyrivibrio-related clones.¹⁸¹ 

Li et al.¹⁸²  carried out an interesting enrichment study by sequentially subculturing bovine ruminal digesta in the presence of 50 µg per ml trans-11 18:1 in order to investigate species capable of the final reduction of trans 18:1 to 18:0. Amounts of 18:0 were produced at progressively faster rates as the subculturing progressed. 16S rRNA amplicon analysis indicated that bacteria closely related to B. proteoclasticus were enriched, along with a number of previously uncultured bacteria. It was not possible to establish if any of these bacteria hydrogenated trans-11 18:1. Furthermore, there is no obvious reason why trans-11 18:1 would enrich for biohydrogenating bacteria as there would be no benefit conferred to those bacteria able to do so. More likely was that the soluble sugars present in the medium selected in favour of these bacteria.

Several studies have indicated that bacteria other than the Butyrivibrio spp. might be involved in biohydrogenation, particularly in the final reduction of cis 18:1 and trans 18:1 to 18:0. The long-chain unsaturated fatty acids in marine algae inhibit the biohydrogenation of 18-carbon fatty acids in the rumen and alter the ruminal microbial community.¹⁸³  These changes have been taken, by association, to indicate bacterial species involved in biohydrogenation. Inhibition of biohydrogenation by marine algae was found to induce changes in uncultivated bacteria clustering with Butyrivibrio and Pseudobutyrivibrio. Boeckaert et al.¹⁸⁴  found that this changed when biohydrogenation was inhibited. In contrast, no significant decrease in numbers of the known Butyrivibrio-related 18:0 producers were detected in the rumen of sheep fed marine algae.¹²⁰  However, only five animals per treatment were investigated and numerical decreases were detected as the amount of algae supplement increased. Restriction fragment length polymorphism (RFLP) of 16S rRNA amplicons indicated that Lactospiracaea spp. or Quinella-like bacteria increased, which could have occurred for other reasons. In experiments investigating the effects of fish oil supplements on the rumen bacterial community, only weak associations have been observed between digesta 18:0 concentration or ruminal 18:0 outflow and B. proteoclasticus group 16S rRNA concentration.^122,185  Further investigation based on multivariate analysis of data obtained from terminal RFLP (tRFLP) and denaturing gradient gel electrophoresis (DGGE) revealed many terminal restriction fragments (T-RFs) and DGGE bands that were correlated with ruminal cis-9, trans-11 18:2, trans-11 18:1 and 18:0 concentrations.¹⁸⁵  Predictive identification revealed that these linked tRFs were likely to originate from as yet uncultured bacteria, including Prevotella, Lachnospiraceae incertae sedis, and unclassified Bacteroidales, Clostridiales and Ruminococcaceae. DGGE bands led to similar conclusions. Consistent with other recent studies, the bacteria have not been isolated and possible isomerase or reductase activity remains to be confirmed.

Since up to half of the rumen microbial biomass may be protozoal in origin¹⁸⁶  and ca. 75% of the microbial fatty acids present in the rumen may be in protozoa,¹²⁵  it follows that protozoa could represent a very important source of several biohydrogenation intermediates, including isomers of CLA and trans-11 18:1. Early studies concluded that both protozoa and bacteria were involved in biohydrogenation.^187,188  However, the extensive ingestion of bacteria by protozoa caused Dawson and Kemp¹⁸⁹  to doubt this conclusion. Biohydrogenation in ruminal digesta was only slightly decreased following defaunation (i.e. removal of protozoa from the rumen) and the presence of protozoa was not necessary for biohydrogenation to occur.¹⁸⁹  Others also suggested that the minor contribution of protozoa to the biohydrogenation process was due to the activity of ingested or associated bacteria.^190,191  Yet it has been known for a long time that protozoal lipids contain proportionally more unsaturated fatty acids than the bacterial fraction.^48,192  Recently, it was established that these unsaturated fatty acids include cis-9, trans-11 18:2 and trans-11 18:1,¹⁹³  further highlighting the possible significance of protozoa as a means of facilitating the escape of biohydrogenation intermediates from the rumen. Protozoal species differ in fatty acid composition, with larger species like Ophryoscolex caudatus containing more than ten times higher concentrations of cis-9, trans-11 18:2 and trans-11 18:1 than small species such as Entodinium nannelum.¹⁹³  However, the lipid of Isotricha prostoma, a large species and the only holotrich examined, had low concentrations of cis-9, trans-11 18:2 and trans-11 18:1. Holotrichs do not ingest the large particles seen with entodiniomorphs. In incubations with fractionated ruminal digesta, 18:2 n-6 metabolism was similar between strained ruminal fluid and the bacterial enriched fraction, while the fraction containing protozoa had a much lower activity. There was no evidence of ¹⁴C enrichment in trans-11 18:1 or cis-9, trans-11 18:2 during incubations of protozoa with ¹⁴C-18:0. No genes with sequence similarity to fatty acid desaturases from other organisms were found in cDNA libraries from ruminal protozoa (E. Devillard, personal communication). Thus, the protozoa are relatively abundant in trans-11 18:1 or cis-9, trans-11 18:2, but do not synthesize these from 18:2 n-6 or 18:0, confirming earlier considerations.¹⁸⁹  It could be argued that the high unsaturated fatty acid content of protozoa arises from the ingestion of plant particles, especially chloroplasts.^187,194  A recent study showed that the engulfment of chloroplasts is a major contributor to the high 18:3 n-3 concentration of protozoa.⁷⁸  However, these observations do not explain the relatively high concentrations of trans-11 18:1 or cis-9, trans-11 18:2 that originate from biohydrogenation. Lourenço et al.¹⁹⁵  suggested that these findings reflect protozoa preferentially accumulating trans-11 18:1 or cis-9, trans-11 18:2 formed by bacteria associated with lipids in ingested plant cell organelles (Figure 1.6).

The relatively low conversion of these intermediates to 18:0 may arise because the bacteria responsible for the reduction of 18:1 intermediates are more vulnerable to protozoal digestive activities. Increased accumulation of 9,11 and 10,12 18:2 isomers in washed protozoa to antibiotic treatment¹⁹⁶  reflects a similar phenomenon, in as much that the enrichment of biohydrogenation intermediates is not due to endogenous protozoal activity. Until the genes encoding the enzymes involved in both bacteria and protozoa are identified, it will be difficult to resolve the issue unambiguously. Recently, I. prostoma was shown incapable of hydrogenating 18:2 n-6,¹⁸¹  but considering its low trans-11 18:1 or cis-9, trans-11 18:2 CLA concentrations, this observation may not be relevant to other species, particularly entodiniomorphs.

Determinations of protozoal fatty acid composition imply that the availability of unsaturated fatty acids, including trans-11 18:1 or cis-9, trans-11 18:2, for absorption by the host animal could depend more on the flow of protozoa rather than bacteria from the rumen. Some ciliate protozoa are retained selectively within the rumen by a migration/sequestration mechanism that depends on chemotaxis.^197,198  As a consequence, protozoal biomass at the duodenum is proportionally less than would be the case if protozoa escaped the rumen attached to digesta particles.^199,200  The flow of microbial nitrogen at the duodenum of steers was recently shown to be 12–15% protozoal in origin. However, protozoal lipid accounted for between 30% and 43% of cis-9, trans-11 18:2 and 40% of trans-11 18:1 reaching the duodenum.²⁰¹  The contribution of protozoa to the flows of 16:0 and 18:0 to the duodenum was less than 20% and 10%, respectively. While there is no doubt that chloroplasts accumulate inside protozoal cells, particularly entodiomorphs, translating that capability to increase unsaturated fatty acids, including trans-11 18:1 or cis-9, trans-11 18:2, available for absorption may prove to be elusive.²⁰² 

Anaerobic fungi were first discovered in the rumen before being found in other environments.²⁰³  It is estimated that they comprise around 7% of rumen microbial biomass, but this value is difficult to ascertain and undoubtedly highly variable.²⁰⁴  Mixed ruminal fungi are capable of the isomerization of 18:2 n-6 to cis-9, trans-11 18:2,^205,206  but activity is very low in comparison with B. fibrisolvens.^205,207  Orpinomyces is the most active genus in biohydrogenation.²⁰⁶ 

Methanogenic archaea are also very significant members of the rumen microbial community, comprising up to perhaps 3–4% of the biomass.²⁰⁸  Until recently, the mechanisms of methane formation have formed the focus for research on ruminal archaea.²⁰⁹  This is changing, however, as genomic analysis reveals new, unsuspected activities.²¹⁰  Lipase genes were not noted in the Methanobrevibacterium ruminantium genome, but many ORFs of unknown function were present.²¹⁰  Whether these may also include enzymes involved in biohydrogenation is not clear.

Quantitative assessment of ruminal CLA biosynthesis requires the measurement of DM flow at the omasum, abomasum or duodenum and a detailed analysis of lipid in digesta entering the sampling site. Measurement of nutrient flow in the ruminant gastro-intestinal tract is reliant on the use of indigestible marker systems to account for sampling errors that occur due to the tendency for digesta to separate during collection.²¹¹  Obtaining representative samples of digesta for the analysis of CLA content is of particular importance owing to the heterogeneous distribution of fatty acids in digesta. It is possible to account for errors due to unrepresentative sampling using a combination of indigestible markers that specifically associate with liquid or particulate fractions of digesta that exhibit different flow characteristics.²¹¹ 

Unless a single marker is uniformly distributed across all digesta phases, unrepresentative sampling will inevitably introduce errors in estimated flows.²¹²  In a number of experiments, Cr₂O₃ has been used as a single marker to determine CLA at the duodenum in lactating cows,^175,213  growing cattle^214–217  and sheep.^104,218,219  In other studies multiple markers have been used for the measurement of CLA at the omasum^{54,61,67,122,220}  or duodenum.^68,221–229 

Analysis of CLA in sampled digesta is reliant on reliable methods for lipid extraction and the preparation of fatty acid derivatives. Conjugated fatty acids are susceptible to isomerization and may disappear during prolonged exposure to acid catalysts.²³⁰  Under these conditions there is also a risk that conjugated products are formed from endogenous sources during methylation. Base-catalysed transesterification has been shown to be the most accurate method for determining CLA composition in range of biological samples.²³¹  However, NEFA are not methylated under these conditions. Virtually all of the CLA leaving the rumen is non-esterified,^54,61  and therefore this catalyst is ineffective for the analysis of digesta fatty acid composition. Isomerization and the production of artifacts with acid-based catalysts can be minimized using lower temperatures during methylation,²³²  but under these conditions the methylation of PL is incomplete.²³¹  Preparation of FAME using methanolic sulphuric acid,^220,233  diazomethane followed by sodium methoxide,²³²  or methanolic sulphuric acid followed by sodium methoxide¹²¹  can be used while avoiding isomerization or the synthesis of allylic methoxy derivatives.

Determination of the relative distribution and abundance of specific CLA isomers also requires the use of long (≥100 m) highly polar capillary columns during gas chromatography (GC) analysis in combination with silver-ion high performance liquid chromatography (HPLC).²³⁴  During GC analysis, cis-9, trans-11 18:2 elutes with the same retention time as trans-8, cis-10 18:2 and trans-7, cis-9 18:2. The trans-11, cis-13 18:2 peak may contain minor amounts of cis-9, cis-11 18:2, while trans, trans isomers with double bonds from 7,9 to 10,12 typically elute as a single peak. The occurrence of 21:0 also complicates the determination of CLA by GC. Depending on the GC column used and the temperature programme applied 21:0 elutes anywhere between cis-11, trans-13 18:2 and cis-10, cis-12 18:2, and may therefore, be erroneously identified as an isomer of CLA.^234–236  An additional complication is that the retention of 21:0 relative to several minor CLA isomers may also differ between GC columns of the same type and changes over the lifetime of the GC column.²³⁴ 

The amount of CLA formed in the rumen has been determined for lactating cows, non–lactating cows, growing cattle and non-lactating sheep (Table 1.3). There are no estimates for lactating sheep, goats or other ruminant species. Synthesis of CLA in the rumen of sheep, growing cattle, non-lactating cows or lactating cows varies between 0.02 and 7.14, 0.14 and 6.59, 0.68 and 3.00 and 1.02 and 15.3 g per day, respectively. It should be noted that the estimates based on GC analysis alone ignore the contribution of minor isomers and are subject to possible errors due to the interference of other fatty acids. Nevertheless, it is clear that diet composition rather than the level of intake, physiological state or species is the major determinant of CLA synthesis in the rumen.

Thus far, 16 isomers of CLA have been detected in ruminal,^119,120  omasal^{54,61,67,122,220}  or duodenal^{68,175,227,229}  digesta with double bonds located at 7,9 through to 13,15. Most reports indicate that ruminal digesta contains only trace amounts of trans-7, cis-9 18:2. Not all 16 isomers have been isolated in a single sample of digesta, indicating that the formation of specific CLA isomers is dependent on the composition of the basal diet and the intake of fatty acid substrates. In lactating cows offered diets based on grass, grass hay, or red clover forages supplemented with cereals and solvent extracted plant protein supplements (Forage: Concentrate (F:C) ratio on a DM basis; 60:40), cis-9, trans-11 18:2 is the major isomer accounting for 48–84% of total CLA at the omasum.^{54,61,67,122,220}  Cis-9, trans-11 18:2 is the main CLA isomer formed in the rumen of cattle fed maize silage based diets^68,229  or in sheep fed diets based on lucerne hay.^119,120  However, the relative abundance of cis-9, trans-11 18:2 was found to be lower (27–31% of total CLA) in lactating cows fed diets (F:C ratio, 60:40) containing lucerne hay and maize silage.¹⁷⁵  Under these circumstances the amounts of trans, trans (7,9 through to 12,14) isomers accounted for 57–63% of total CLA at the duodenum. Reports on ruminal synthesis of specific CLA isomers in non-lactating sheep fed high concentrate diets (F:C 18:82) containing no additional lipid are conflicting. In one investigation, cis-9, trans-11 18:2 was reported to be the major CLA isomer at the duodenum,²¹⁹  whereas in a follow-up experiment trans, trans isomers were quantitatively more important.¹⁰⁴ 

In ruminants fed high proportions of grass, grass silage or red clover silage, where 18:3 n-3 is the predominant dietary fatty acid, trans-11, trans-13 18:2 or trans-11, cis-13 18:2 are the principal CLA isomers formed in the rumen.^227,228,237  Trans-12, trans-14 18:2 is also relatively abundant, whereas cis-9, trans-11 18:2 accounts for between 6.7% and 32% of total CLA synthesis.^227,228 

A number of investigations have explored the potential to increase ruminal CLA synthesis through changes in diet composition. Emphasis has been placed on alterations in the relative proportions of forages and concentrates or marine lipid supplements to manipulate rumen environment or supplementing the diet with plant oils to increase the supply of substrates for ruminal CLA synthesis. For diets containing no additional lipid supplements, increases in concentrate supplementation have variable and rather marginal effects on the amount of CLA at the duodenum (Table 1.3). Decreases in F:C ratio from 60:40 to 25:75 were shown to promote ruminal synthesis of geometric isomers of 9,11 and 10,12 18:2.¹⁷⁵  In contrast, decreases from 65:35 to 35:65 had no effect other than decreasing trans- 9, cis-11 18:2 and increasing formation of several trans, trans isomers.¹⁷⁶ 

Studies of 18:2 n-6 metabolism in vitro indicate that the amount of cis-9, trans-11 18:2 leaving the rumen would be enhanced in diets that promote high passage rates, maintain rumen pH and contain high amounts of 18:2 n-6.^{127,129,130,132}  Increasing the intake of 18:2 n-6 on ruminal CLA synthesis has been investigated in cattle and sheep. Dietary supplements of sunflower oil, soyabean oil or safflower as a source of 18:2 n-6 has been shown to increase the amount of 8,10, 9,11 and 10,12 18:2 isomers and total CLA at the omasum or duodenum (Table 1.3). However, increases in the the formation of specific isomers to 18:2 n-6 supply is dependent on the composition of the basal diet. In lactating cows fed diets based on grass silage (F:C ratio 60:40), supplements of sunflower oil from 0 to 750 g per day progressively increased cis-9, trans-11 18:2, trans-9, trans-11 18:2, trans-10, cis-12 18:2 and trans-10, trans-12 18:2 at the omasum⁶⁷  (Table 1.4). In contrast, adding higher amounts of sunflower oil from 20 to 40 g per kg DM in high concentrate diets (average F:C ratio 24:76) stimulated ruminal trans-10, cis-12 18:2 formation (0.21–0.37 g per day) but had no effect on other measured CLA isomers at the duodenum in growing cattle.²¹⁵  Ruminal synthesis of trans-10,cis-12 CLA has been reported to be as high as 1.83 g per day in lactating cows fed high concentrate diets (F:C ratio 35:65) containing 50 g per kg DM of sunflower oil.²²³  Consistent with these findings, ruminal infusion of safflower oil in non-lactating sheep fed high concentrate diets (F:C ratio 18:82) at a rate equivalent to 0 to 90 g per kg DM progressively increased the amount (g per day) of cis-9, trans-11 18:2 (<0.01–0.38), trans-9, trans-11 18:2 (0.01–2.08), trans-10, cis-12 18:2 (<0.01–4.49) and trans-10, trans-12 18:2 (0.03–0.42) recovered at the duodenum.¹⁰⁴  At the highest level of infusion, trans-10,cis-12 CLA was identified as the principal isomer formed in the rumen. However, an earlier experiment in non-lactating sheep fed a similar diet reported that incremental soyabean oil supplements (0-94 g per kg DM) had no influence on cis-9, trans-11 18:2 and resulted in limited increases in trans-10, cis-12 18:2 from 0 to 0.22 g per day at the duodenum.²¹⁹ 

Several investigations have examined the effects of dietary linseed oil supplements as a means to increase 18:2 n-6 and 18:3 n-3 intake on ruminal CLA synthesis. In non-lactating cows fed high forage diets (F:C ratio 70:30) supplements of linseed oil (49 g per kg DM) increased duodenal flows of unresolved trans-8, cis-10 18:2 and cis-9, trans-11 18:2, trans-11, cis-13 18:2 and trans-11, trans-13 18:2 from 0.22, 0.12 and 0.13 to 0.64, 0.46 and 0.64 g per day, respectively.²²⁸  In growing cattle fed maize silage based diets, linseed oil (30 g per kg DM) stimulated ruminal cis-9, trans-11 18:2, trans-8, cis-10 18:2, trans-11, cis-13 18:2, trans-13, cis-15 18:2 and trans, trans (9,11 to 13,15) 18:2 synthesis⁶⁸  (Table 1.4). Consistent with reports on the influence of 18:2 n-6 supply, the composition of the basal diet is also a major determinant of the changes in ruminal CLA synthesis to linseed oil. In lactating cows fed high or low forage diets (F:C 65:35 vs. 35:65), linseed oil resulted in similar increases in ruminal cis-9, trans-11 18:2, trans-11, cis-13 18:2 and trans-11, trans-13 18:2 synthesis, irrespective of diet composition.¹⁷⁶  However, trans-8, cis-10 18:2, trans-9, cis-11 18:2 and cis-11, trans-13 18:2 formation was increased on the low forage diet, but remained unchanged on the high forage diet.

It is well established that the concentrations of cis-9, trans-11 18:2 and total CLA are higher in milk in grazing ruminants compared with that on diets based on conserved forages.^45,47,49  Relatively few studies have compared the synthesis of CLA isomers in the rumen of grazing ruminants compared with diets containing conserved forages.^61,237  Even though the number of observations that have been made is limited, there is little evidence that the effects on milk fat CLA concentrations are related to differences in the synthesis of CLA isomers in the rumen. Similarly, replacing grass silage with grass hay⁶¹  or red clover silage^54,227  has limited influence on total CLA synthesis, but may result in differences in the amounts of minor isomers formed in the rumen (Table 1.4).

The effect of dietary marine lipid supplements on ruminal CLA synthesis has also been investigated. Even though the highly unsaturated fatty acids in fish oil^119,122,220  or marine algae^120,184,238  inhibit the complete biohydrogenation of unsaturated 18-, 20-, 21- and 22-carbon unsaturated fatty acids in the rumen, the influence on the accumulation of total CLA is rather small. In some experiments, supplementing grass silage, red clover silage or grass silage based diets with fish oil has been demonstrated to increase trans-7, trans-9 18:2, trans-8, trans-10 18:2 and trans-9, trans-11 18:2 formation and decrease cis-12, trans-14 18:2 synthesis,^220,227  but not in all cases¹²²  (Table 1.4). Studies in growing cattle fed maize silage-based diets indicated that fish oil supplementation up to 16 g per kg diet DM resulted in marginal increases in cis-9, trans-11 18:2 and trans-9, trans-11 18:2 at the duodenum, changes that were not evident when the amount of fish oil in the diet was increased above 24 g per kg diet DM.^68,229 

Secretion of trans-7, cis-11 18:2 and cis-9, trans-11 18:2 in milk is known to be several-fold higher compared with the amounts of these isomers at the duodenum in lactating cows.^45,239,240  These differences are explained by endogenous synthesis in ruminant tissues. It now known that the majority of cis-9, trans-11 18:2 in milk is formed from the desaturation of trans-11 18:1 in the mammary glands. Most, if not all, of the trans-7, cis-11 18:2 found in ruminant milk also originates from the desaturation using trans-7 18:1 as a substrate. The potential to increase the synthesis of CLA in the rumen is relatively limited compared with projections of the enrichment in ruminant foods required to confer benefits to human health. For this reason, considerable emphasis has been placed on understanding the role of diet on the synthesis of CLA precursors, trans-11 18:1 in particular, in the rumen. The reduction of trans 18:1 intermediates to 18:0 is considered rate limiting for the complete biohydrogenation of 18-carbon unsaturated fatty acids in the rumen.¹²⁵  Studies in vitro have demonstrated that increases in the amount of 18:2 n-6 or 18:3 n-3 incubated with mixed ruminal bacteria results in the accumulation of trans 18:1 (Table 1.2). In vivo changes in diet composition tend to have a much greater influence on the amount and relative proportions of trans 18:1 leaving the rumen (Table 1.5) compared with ruminal CLA synthesis (Table 1.3). Therefore, formulating diets that increase ruminal formation of trans-11 18:1 rather than specifically targeting cis-9, trans-11 18:2 synthesis is more effective for increasing the CLA content of ruminant foods.

In ruminants fed diets containing forages and concentrate ingredients, trans-11 18:1 is typically the major biohydrogenation intermediate leaving the rumen. Increases in concentrate supplementation have been shown to cause the accumulation of trans 18:1 intermediates in the rumen.^176,218,241  Switching from grass hay based diets containing 35% to 65% of concentrates increased the flow of trans 18:1 (Δ4 to 16) at the duodenum in lactating cows.¹⁷⁶  Changes in trans-11 18:1 in response to concentrate supplementation were marginal, with most of the increase being associated with the trans-10 isomer (Table 1.5). More extreme increases in the amount of concentrates offered to lactating cows, from 60% to 75%, caused all trans 18:1 isomers to accumulate, other than trans-11 and trans-16.¹⁷⁶  Under these conditions, the most marked influence was on trans-10 18:1 at the duodenum, being increased four-fold compared with the high forage diets (Table 1.5). Switching from a diet based on dehydrated grass to concentrate ingredients was found to result in the almost complete disappearance of trans-11 18:1 and a substantial increase in trans-10 18:1 concentration of abomasal digesta in growing lambs.²⁴² 

Examination of the effects of pH (6.4 vs. 5.6) or dietary F:C ratio (70:30 vs. 30:70) on the production of biohydrogenation intermediates in dual-flow continuous culture suggested that decreases in pH rather that increases in concentrate per se was the principal factor responsible for the shift from trans-11 18:1 to trans-10 18:1 accumulation.¹²⁷  However, this interpretation has been challenged based on recent reports of 18:2 n-6 metabolism by ruminal bacteria²⁴³  and measurements of ruminal fatty acid composition in cows fed diets containing low or high amounts of starch.¹⁷³  The findings from both investigations led to the conclusion that the influence of starch on trans-10 18:1 formation was independent of the decreases in pH that occur during starch fermentation.

Under most situations, metabolism of 18:2 n-6 and 18:3 n-3 in the rumen leads to the formation of trans-11 18:1 during the complete hydrogenation to 18:0 (Figures 1.3 and 1.4). It is perhaps unsurprising that dietary supplements of oils enriched in 18:2 n-6 or 18:3 n-3 have used to increase the amounts of trans-11 18:1 formed in the rumen. Including plant oils in the diet have been shown to result in dose dependent increases in the amount of trans 18:1 intermediates leaving the rumen in sheep^104,218,219 , growing cattle,^68,214–217  lactating^67,176  and non-lactating cows.²²⁸  However, the profile of trans 18:1 intermediates that accumulate is dependent on a complex interaction between the relative proportions of forages and concentrates in the diet and the amount and source of oil supplement (Table 1.5). On high forage diets, supplements of sunflower oil⁶⁷  or linseed oil¹⁷⁶  increase the amount of trans-11 18:1 escaping the rumen. In marked contrast, inclusion of the same oils in high concentrate diets has minimal influence on trans-11 18:1, but results in trans-10 18:1 accumulating several-fold.^176,215  Increases in the concentration of trans-10 18:1 are also accompanied by decreases in the synthesis of fatty acids in the mammary glands of cows and sheep, but not in goats.¹⁰⁷  Changes in milk fatty acid composition also serve as a proxy of the relative amounts of trans 18:1 escaping the rumen, indicating that the major pathways of biohydrogenation are altered under these conditions causing trans-10 18:1 to displace trans-11 18:1 as the principal intermediate formed in the rumen. The most recent evidence has indicated that a low rumen pH and high dietary concentrations of starch and oil are prerequisite in promoting the ‘trans-10 shift’.^172,173 

Palmquist et al.⁴⁵  proposed that ruminal trans-11 18:1 formation is dependent on three inter-dependent processes; 1) substrate supply, 2) inhibition of the reduction of trans 18:1 to 18:0 and 3) prevention of the trans-10 shift. It was argued that substrate supply has a typically permissive role in determining the extent of trans 18:1 fatty acid accumulation in the rumen in response to the other two processes, such that the balance between the inhibition of trans 18:1 reduction and induction of the shift towards trans-10 18:1 at the expense of trans-11 18:1 regulates the magnitude of the overall response to changes in diet composition.

Fish oil and marine algae contain relatively high amounts of long chain (≥ 20-carbon) polyunsaturated fatty acids. When added to the diet of ruminants, both lipid sources lead to an inhibition of the complete biohydrogenation of 18-carbon unsaturated fatty acids causing trans 18:1 and trans 18:2 intermediates to accumulate.^68,120,122  Fish oil is an effective means to increase the amount of trans-11 18:1 leaving the rumen (Table 1.5), but in high amounts can induce a shift towards trans-10 18:1 formation.¹²²  Incubations with mixed ruminal bacteria have shown that both 20:5 n-3 and 22:6 n-3 cause trans 18:1 to accumulate.^244–246  It is possible that other 20-, 22- or 24-carbon unsaturated fatty acids contained in fish oil and marine algae may also inhibit the reduction of trans 18:1 to 18:0 in the rumen.

Relatively few studies have examined the effects of forage conservation or forage species on ruminal biohydrogenation. The most recent investigations indicate that the relative abundance of trans-11 18:1 in the rumen of lactating cows is higher on pasture compared with zero-grazing or grass silage.²³⁷  Flows of trans-11 18:1 at the omasum have also been shown to be higher in lactating cows during zero-grazing compared with diets based on hay or silage prepared at the same time from the same grass swards.⁶¹  Conservation of grass by drying or ensiling had no effect on the amount of trans-11 18:1 leaving the rumen.⁶¹  Replacing grass silage with red clover silage^54,227  also has little influence on trans-11 18:1 or total trans-18:1 formation in the rumen (Table 1.5).

Studies involving the incubation of fatty acids with mixed ruminal bacteria have demonstrated that the biohydrogenation of cis-9 18:1, 18:2 n-6 and 18:3 n-3 results in the formation of trans-7 18:1 as an intermediate (Table 1.2). Because trans-6, -7 and -8 elute as a single peak during GC analysis there are no definitive reports on the effect of diet on trans-7 18:1 formation in the rumen. Direct comparisons of milk fat trans-7, cis-9 18:2 concentrations in ruminants offered plant oil or oilseed supplements have shown a consistently higher enrichment when rapeseed lipids are fed.^46,47  By implication, biohydrogenation of cis-9 18:1 appears to be the main source of trans-7 18:1 in the rumen.

Recent studies have also provided evidence that trans-9 16:1 (palmitelaidic acid) may also serve as a substrate for endogenous cis-9, trans-11 18:2 synthesis in ruminant tissues.²⁴⁷  The metabolic origins of trans 16:1 formed in the rumen have not been extensively investigated. A range of trans 16:1 with ethylenic bonds between positions Δ3 to 14 are known to be formed in the rumen and incorporated into milk fat.^54,248  Fish oil is known to cause trans-9 16:1 to accumulate in the rumen.^{68,122,227,229}  Under these situations, it seems plausible that trans-9 16:1 originates from the incomplete biohydrogenation of cis-9, cis-12 16:2 or cis-6, cis-9, cis-12, cis-15 16:4. On conventional diets, trans 16:1 isomers are thought to originate from the incomplete biohydrogenation of cis-6, cis-9, cis-12 16:3 present in trace amounts in the ruminant diet.²⁴⁸ 

Early reports indicated that the concentrations of cis-9, trans-11 18:2 in milk were increased on pasture^249,250  or on diets supplemented with linseed oil²⁵¹  or fish oil.²⁵²  Such findings were perplexing. The convention at the time was that ruminal biohydrogenation of 18:2 n-6 was responsible for cis-9, trans-11 18:2 formation in ruminants, and yet pasture and linseed oil contain high proportions of 18:3 n-3 and fish oil contains few 18-carbon unsaturated fatty acids. Much earlier studies had demonstrated that several trans 18:1 isomers could be converted to cis-9 containing 18:2 products during incubations with rat liver microsomal preparations.^253,254 

These seminal investigations confirmed that trans 18:1 fatty acids with double bonds at Δ4–7 and Δ11–13 were substrates for the enzyme stearoyl-CoA desaturase (SCD; E.C. 1.14.99.5). Based on the observed changes in milk fat composition to diet and the substrate specificity of the SCD enzyme, Griinari and Bauman²⁵⁵  proposed that the rumen was not the sole source of cis-9, trans-11 18:2 and that endogenous synthesis due to the desaturation of trans-11 18:1 also occurred in ruminant tissues. An overview of the metabolic pathways of endogenous CLA synthesis in ruminants is described in Figure 1.7.

The SCD gene encodes a protein of 359 amino acid residues located in the endoplasmic reticulum (ER) that catalyzes the desaturation of 10- to 19-carbon fatty acyl-CoA substrates, resulting in the formation of a cis double bond located between carbons 9 and 10 of the fatty acid moiety.^256,257  Homologues of the SCD gene have been characterized in several mammalian species, including the mouse, rat, hamster, pig and human. Some mammalian genomes have been shown to contain multiple SCD isoforms. In mice, four isoforms (SCD1, SCD2, SCD3 and SCD4) have been documented.^258–261  Until recently only one isoform of the SCD gene, now referred to as SCD1, had been characterized in sheep,²⁶²  cows²⁶³  and goats.²⁶⁴  A second SCD isoform (SCD5) was recently identified in cattle²⁶⁵  and subsequently found to be expressed in the goat.²⁶⁵ 

Expression of the SCD1 gene is several-fold higher in adipose compared with brain, heart, liver, lung, muscle, spleen or intestinal mucosa,^265,267  with adipose tissue being the major site of desaturase activity in growing ruminants.^268,269  Abundance of SCD1 mRNA in subcutaneous adipose tissue increases after weaning²⁷⁰  with activity of the desaturase enzyme increasing during growth even after SCD gene expression starts to decline.²⁷¹  Catalytic activity of SCD also varies between tissues. Activity of SCD was found to be higher in intestinal mucosa, interfasicular and subcutaneous adipose compared with liver (18.1, 10.6, 5.24 and 1.81 nmol per mg protein per min, respectively).²⁷²  Immediately after parturition, there is a substantial upregulation of SCD1 expression in the mammary glands of sheep²⁶²  and cattle.²⁷³  In lactating goats, relative expression of SCD1 is higher in mammary tissue compared with the liver, subcutaneous, omental and perirenal adipose^274–277  consistent with the mammary glands being the major site of fatty acid desaturation in lactating ruminants.^278,279 

Transcript abundance of SCD5 is several-fold higher in the brain compared with heart, liver, lung, muscle and spleen in growing cattle.²⁶⁵  SCD5 is also expressed in the bovine mammary gland.²⁸⁰  In lactating goats, abundance of SCD5 mRNA is higher in mammary tissue compared with the liver, omental and perirenal adipose.^277,281  Even though SCD5 is expressed in several tissues, the contribution of this SCD isoform to the desaturation of fatty acids in ruminants remains unclear.

The SCD enzyme is the rate-limiting step in the synthesis of monoenoic fatty acids that are utilized as substrates for TAG, PL and cholesterol ester (CE) formation.²⁸²  Stearoyl and palmitoyl- CoA are the preferred substrates but a range of 10- to 19-carbon fatty acyl-CoA are desaturated in various ruminant tissues.^46,256  Cis-9 18:1 is the major product of the SCD enzyme in mammals, the incorporation of which into PL contributes to the maintenance of membrane fluidity.²⁸²  Activity of SCD in the ruminant mammary gland is thought to occur as a mechanism to maintain and regulate the fluidity of milk.²⁸³ 

Fatty acid desaturases are nonheme iron-containing enzymes that introduce a double bond between defined carbons of fatty acyl chains. In common with other desaturases, SCD catalyses a highly regio- and stereo-selective reaction on long-chain fatty acids that consist of essentially equivalent methylene chains in the absence of a distinguishing feature close to the site of desaturation. The Δ9 desaturase homologous to SCD isolated from Ricinus communis (castor) is a homodimeric protein with each monomer folded into a compact single domain comprised of nine helices.²⁸⁴  A diiron active site is located within a core four-helix bundle that is positioned alongside a deep, bent, narrow hydrophobic channel in which the substrate is bound during catalysis.²⁸⁵  It is thought that SCD has 4 transmembrane domains in the ER membrane and that both the -NH₂ and -COOH terminal groups are oriented toward the cytosol.²⁸⁶