Toxicity of unsaturated fatty acids to the biohydrogenating ruminal bacterium, Butyrivibrio fibrisolvens

Unsaturated fatty acids, particularly α-linolenic acid (LNA; cis-9, cis-12, cis-15-18:3) and linoleic acid (LA; cis-9, cis-12-18:2), are abundant in grass and other ruminant feedstuffs, yet are present at low concentrations in meat and milk. Furthermore, tissue lipids of ruminants have been known for a long time to be more saturated than those of non-ruminants [1]. As the consumption of saturated acids in dairy products and ruminant meats is often associated with an increased incidence of coronary heart disease in man [2], the transformation of unsaturated fatty acids to saturated fatty acids, or biohydrogenation, in ruminants presents a major human health issue. The biohydrogenation process has long been known to occur in the rumen as the result of microbial metabolic activity [3, 4]. Thus, if ruminal biohydrogenation of unsaturated fatty acids can be controlled, it may be possible to improve the healthiness of ruminant meats and milk by increasing their unsaturated fatty acids composition in general and the n-3 fatty acids in particular [5]. One of the unsaturated fatty acids that appears most desirable is conjugated linoleic acid (CLA; cis-9, trans-11-18:2) because of its anticarcinogenic and other health-promoting properties [6, 7]. Major advances have been made in achieving the desired changes in fatty acid content of meat and milk experimentally, via dietary manipulation in ruminants, generally by adding oils containing unsaturated fatty acids to the diet [5, 8–10]. The inclusion of fish oil in particular seems to alter biohydrogenating activity in the rumen [11].

Butyrivibrio fibrisolvens was identified many years ago to undertake biohydrogenation of fatty acids [12] and to form CLA as intermediate in the process [13]. Kim et al. [14] noted that LA inhibited growth of B. fibrisolvens A38, an effect that depended both on the concentration of LA and the growth status of the bacteria. Growing bacteria were more tolerant of LA. In a study of CLA production in different strains of B. fibrisolvens, Fukuda et al. [15] found that the most tolerant strain had the highest linoleate isomerase (forming CLA from LA) specific activity. Different members of the Butyrivibrio/Pseudobutyrivibrio phylogenetic grouping, all of which biohydrogenate PUFA, had different sensitivities to growth inhibition by LA, the most sensitive possessing the butyrate kinase rather than the acyl transferase mechanism of butyrate production [16]. For reasons that were unclear, lactate exacerbated the toxicity of LA to Clostridium proteoclasticum[17], now renamed Butyrivibrio proteoclasticus[18]. The aim of this work was to understand the nature of the toxic effects of PUFA on B. fibrisolvens JW11. Strain JW11 is located in the middle of the numerous B. fibrisolvens/Pseudobutyrivibrio cluster, members of which share the ability to form CLA and vaccenic acid (VA; trans-11-18:1) but which also lack the ability to biohydrogenate VA to stearic acid (SA; 18:0) [16]. Understanding these effects could have important indirect implications for human health by enabling ruminal biohydrogenation of dietary PUFA to be manipulated in order to provide healthier ruminant-derived foods.

B. fibrisolvens was originally described as a small, Gram-positive bacterium particularly prevalent in the rumen of grazing animals [19]. Many strains are proteolytic and involved in fibre breakdown [19, 20]. B. fibrisolvens JW11 was originally isolated as a proteolytic strain [21]. It has been many years since the importance of B. fibrisolvens in the process of PUFA reduction, or biohydrogenation, was first documented [12]. Although other bacteria have been implicated [22], biohydrogenating activity is high among all members of what is now known to be an extensive Butyrivibrio phylogenetic tree [16]. Indeed, in our experience, its activity is many times higher than in other species [17]. 'Type B' bacteria, which complete the reduction of 18:1 isomers to SA, was identified as C. proteoclasticum[23], which has recently been renamed Butyrivibrio proteoclasticus[18]. The pattern of metabolism of LA and LNA observed here, and the identity of the intermediates, follows the pathways established first by Kepler et al. [13] and confirmed later by others [24–26]. The observations linking growth and LA metabolism with B. fibrisolvens JW11 are consistent with those obtained with B. fibrisolvens A38 [14] and B. fibrisolvens TH1 [15]. What is novel about the present observations is that they clearly demonstrate that biohydrogenation is a detoxification process, necessary to escape from the bacteriostatic effects of PUFA. Indeed, they explain that the concentration-dependence of LA toxicity and its apparently lower toxicity in growing cultures, observed by Kim et al. [14], have to be considered in terms of time required by different biomass concentrations to hydrogenate, and thereby detoxify, different concentrations of fatty acids.

Henderson [27] examined the effects of fatty acids on ruminal bacteria. A Butyrivibrio sp. was generally most sensitive to fatty acids, but only saturated and monoenoic acids were included in the study. OA was much more toxic than the saturated fatty acids. Marounek et al. [28] found that C-12 and C-14 fatty acids were more toxic to ruminal and rabbit caecal bacteria than other chain lengths, but again the study was of saturated acids and oleic acid. In non-ruminal bacteria, LA and LNA were much more toxic than saturated or monoenoic acids [29]. The present paper describes the effects of the more abundant poly- and monounsaturated fatty acids on B. fibrisolvens. The PUFA were found to be much more toxic than more saturated fatty acids.

The present experiments help to resolve the purpose of biohydrogenation in the ruminal bacteria that undertake this reductive metabolism. Our results provide support for the conclusions of Harfoot and Hazlewood[22], Kemp and Lander [30] and Kemp et al. [31] that biohydrogenation is a detoxification mechanism rather than a means of disposing of reducing power, as proposed earlier [32]. The reductase which converts CLA to VA in B. fibrisolvens comprises 0.5% of the total cell protein [33], a very significant expenditure of cellular resources that signifies a vital function. It should be noted that, although more research emphasis is placed on its metabolism of LA because CLA is an intermediate, biohydrogenation is probably more important for B. fibrisolvens to survive high LNA concentrations, as LNA is more toxic than LA and is usually present at higher concentrations than LA in forages (e.g. [3]). Also to be noted is that CLA is almost as toxic as LA, as found before [14].

There are several possible reasons why unsaturated fatty acids are generally more toxic than saturated fatty acids. The double bonds alter the shape of the molecule, such that kinked unsaturated fatty acids disrupt the lipid bilayer structure [34]. The finding that different PUFA isomers, such as LNA and γ-LNA, had different toxicity would be consistent with such an interpretation. However, it is not clear that the toxicity was necessarily a membrane effect. The free carboxyl group was necessary for growth inhibition to take place. Methyl esters, which might be expected to be sufficiently hydrophobic to be incorporated into a membrane just as efficiently as a free fatty acid, were non-toxic. They were metabolized in the same way as the free fatty acids, however, as they were hydrolysed by bacterial esterase activity. The free carboxyl group was also necessary for disruption of cell integrity, as measured by PI ingression. However, PI ingression was affected little by the number of double bonds in the fatty acid molecule, as detected by both fluorescence spectroscopy and flow cytometry, and could not explain why 18:3 fatty acids were more toxic to growth than 18:2 fatty acids, and especially why 18:1 was not toxic to growth. Furthermore, sodium lactate exacerbated growth inhibition by LA, in a similar manner to that observed with B. proteoclasticus[23], but had no similar effect on the influence of LA on cell integrity of B. fibrisolvens. A similar conclusion was reached by Maia et al. [17] when comparing the toxic effects of fatty acids on growth and cell integrity in different species of ruminal bacteria. Thus, although a toxic mechanism involving disruption of the extraordinarily thin cell envelope of B. fibrisolvens[35] seems an attractive and logical possibility, the evidence suggests that the primary effect of PUFA lies elsewhere.

An alternative possibility is that the ready diffusion of the free fatty acid across the membrane causes chemiosmotic difficulties, perhaps uncoupling the proton-motive force [36], dissipating the membrane potential by facilitating ion leakage [37] or decoupling intramembrane pathways [38, 39]. While this remains a possibility, the different effects on acyl CoA and ATP pools on PUFA toxicity suggest a metabolic effect, specifically in acyl CoA metabolism. Measurement of CoA metabolic pools in bacteria is relatively rare. Here, acetyl CoA and butyryl CoA were present at highest concentration and the butyrate pathway intermediates at much lower concentrations, as found also in Clostridium acetobutylicum[40]. All acyl CoAs except acetoacetyl CoA were diminished by >96% when LA was added to the medium. In contrast, the ATP pool was affected later than acyl CoA pools, and remained at about one-third of the control values, presumably due to the contribution of glycolysis.

The toxicity of PUFA in different species of ruminal bacteria was found to be related partly to whether the bacteria produced butyrate; cellulolytic bacteria were the other most sensitive species [17]. Within the Butyrivibrio phylogenetic group, the most sensitive species were those that formed butyrate via the butyrate kinase mechanism rather than acyl CoA transferase [16]. Thus, there seems to be a connection between PUFA toxicity and butyrate formation. A metabonomic analysis [41] might help to identify precisely where the PUFA act. It may also be instructive to determine why trans-11, cis-15-18:2, a product of LNA metabolism, appeared to permit growth while the other dienoic acid investigated here did not.

The influence of sodium lactate in lengthening the lag phase indicates that lactate potentiates the toxic effects of PUFA in B. fibrisolvens, as shown previously with B. proteoclasticus[22]. Such a high concentration of lactate (70 mM) would only occur in animals suffering acidosis [42]. The toxicity may be an osmotic effect, or due to a leakage of ions across the membrane, or may even be a metabolic effect. Lactate is a major product of glucose metabolism in B. fibrisolvens JW11 and many other butyrivibrios [16], and high concentrations may therefore impose metabolic feedback, further stressing bacteria affected by PUFA.

This paper explains the basis of the beneficial effect on meat and milk fatty acid composition of adding oils to the ruminant diet. Ruminal biohydrogenation is modified via differential toxicity to ruminal bacteria of different PUFA, including the fish oil fatty acids, EPA and DHA. If we can understand how selective fatty acid toxicity, or indeed other factors, affects the physiology of biohydrogenating bacteria in the rumen, we may be able to suggest new, rational dietary modifications that will eventually lead to ruminant products that are healthier for human consumption.