The role of holotrichs in the metabolism of dietary linoleic acid in the rumen

17 Biochimica et Biophysics Acta, 528 (1978) 17-27 0 Elsevier/North-Holland Biomedical Press BBA 57109 THE ROLE OF HOLOTRICHS IN THE METABOLISM OF DIETARY LINOLEIC ACID IN THE RUMEN V. GIRARD * and J.C. HAWKE Department of Chemistry, Biochemistry and Biophysics, Massey University, Palmerston North (New Zealand) (Received June 13th, 1977) (Revised manuscript received September 23rd, 1977) Summary The uptake and metabolism of linoleic acid by rumen holotrichs (mainly Isotricha prostoma and I. intestinalis) has been examined in in vitro infusion experiments. Maximum absorption and metabolism of [ l-14C]linoleate by 2 * lo6 Isotricha suspended in 100 ml buffer was obtained using an infusion rate of 1.6 mg linoleate/h. After 90 min, 84% of the added substrate was recovered within the cells, mainly as free fatty acid or phospholipid. There was a rapid incorporation of radioactivity into phospholipid, mainly phosphatidyl- choline, at the commencement of linoleate infusion but no further incorpora- tion after about 40 min. The presence of bacteria during incubations, in approximately the same Isotricha/bacteria ratio as found in the rumen, reduced the uptake of linoleate and the accumulation of free fatty acid by holotrichs but the incorporation into phospholipid remained similar to that obtained in the absence of bacteria. Very little biohydrogenation of linoleic acid occurred in incubations with holotrichs alone. Bacterial suspensions converted linoleic acid to mainly trans monoene and a small amount of stearic acid, but in incubations containing both bacteria and holotrichs, both stearic acid and trans monoene were major products. Using the latter mixed culture, about 20% of the added [1-14C]- linoleic acid was present in holotrich phospholipid of which 62% remained as octadecadienoic acid. The Isotricha population was 3 - 103-2 * 104/ml rumen fluid and it con- tributed about 23% of the linoleic acid in the rumen of a cow on a hay diet. * Permanent address: Envers 39, 2400 Le Lo&. Switzerland. 18 Introduction Emphasis in the studies of lipid metabolism in the rumen has centred on the processes of lipolysis [ 1,2] and biohydrogenation [ 361 in order to explain the extensive conversion of dietary linolenic and linoleic acids into more saturated products before absorption by the ruminant. Little is known about the manner in which the essential fatty acid requirements of ruminants are assured in the presence of this highly efficient process of biohydrogenation but a minimum of 5-10 g/day of linoleic acid must escape hydrogenation in the rumen of the lactating bovine simply to account for the daily secretion of this amount in the milk [7]. Bacterial phospholipids containing small amounts of polyenoic fatty acids were isolated following the in vitro incubation of rumen fluid with triglyceride containing 14C-labelled linoleic acid [S] and some of the essential fatty acid requirement may be provided by digestion of bacterial lipids. Furthermore Keeney [9] found low levels of linoleic in the total lipids of rumen digesta much of which is concentrated in the phospholipid fraction [lo]. However rumen protozoa make a greater contribution than rumen bacteria to the total microbial lipids and are known to absorb linoleic acid [ 111 which is abundant in their phospholipids [ 121. This has led to the suggestion that some species of protozoa prevent the complete biohydrogenation of polyunsaturated fatty acids released from dietary lipids by their rapid uptake and incorporation into cellular lipids [ 91. Isolated holotrichs, consisting mainly of Isotricha species, were selected for study of linoleic acid uptake and metabolism since it has been demonstrated that these rumen protozoa do not hydrogenate linoleic acid [ 131, whereas preparations of protozoa containing both oligotrichs and holotrichs have this hydrogenating ability [ 141. Our further aim was to investigate linoleic acid uptake by holotrichs in the presence of rumen bacteria and the influence of holotrichs on linoleic biohydrogenation by bacteria in short term incubations. Experimental Isolation of holotrichs and bacteria A non-lactating cow fitted with a rumen fistula, and fed once daily with 10 kg of good quality meadow hay, was the source of rumen fluid. Uneaten hay was removed after 3 h. Water and salt were available ad libitum. Holotrichs were precipitated [15] from strained rumen fluid, obtained 1 h before feeding. Approx. 30 min after adding sucrose, to give a concentration of 0.3% (w/v), approx. 7 ml of a white sediment of holotrichs was transferred into a 100 ml glass column (diameter 1 cm) filled with anaerobic buffer maintained at 39°C. The lower layer of the sediment was practically free of small holotrichs (Dasytricha) and bacteria and when counted by the method of Clarke and Hungate [ 161 contained approx. 6 * lo5 holotrichs/ml (mainly Isotricha prostoma and I. intestinalis). The composition of the incubation buffer (pH 7.0) used to dilute the sedimented holotrichs was: Indigocarmine, 0.1 mg; 950 ml of a salt solution containing (g/l) KH,P04, 2.0/NazHP04 . 12Hz0, 15.0/ 19 MgS04, 0.5/CaC12, 0.01/NaHC!03, l.O/NaCl 3.8; 50 ml clarified rumen fluid [16]. Before use, it was gassed for 2 h with 02-free 5% CO* in Nz and 0.02% (w/v) cysteine-HCl added to maintain the redox potential low enough to decolourise Indigocarmine. This buffer had a slightly higher osmolarity and buffering capacity than that used in similar isolations [16] and gave less distended cells. A bacterial suspension was obtained by homogenizing rumen fluid for 1 min and centrifuging at 1000 X g for 10 min to precipitate the food particles and protozoa. The turbid supernatant contained approximately 1.0 * 10” cells/ml (range 2 - lo’-23 - lo9 cells/ml) and when diluted with one-tenth its volume of incubation buffer was used as an alternative diluent to prepare holotrich suspensions. The ratio of bacteria to Isotricha in these suspensions was approx. 4.5 * 10’ which approximated that found in the bovine rumen [ 171. Incubation procedure Approximately 2 * lo6 Isotricha were suspended in 100 ml buffer or buffer/ bacterial suspension and incubated at 39°C in a water bath fitted with a platform shaker operating at 1 cycle/s. 02-free 5% CO2 in Nz was passed through the incubation mixture at 40 ml/min. Before adding substrate the number of protozoa was adjusted to 2.0 * 104/ml (+lO%) by adding either sedimented protozoa or buffer. Fatty acid salts ([ l-‘4C]linoleic or palmitic acids (The Radiochemical Centre, Amersham)) plus puriss grade non-labelled fatty acids (Koch-Light Laboratories Ltd., Colnbrook) or [ 14C]phosphatidylcholine (prepared from spinach grown in 14C02) were dispersed in anaerobic 0.2 M phosphate, containing 1.7% (w/v) methionine, at pH 7.0 (1 mg (1 &i) lipid/ml) by sonication under a Nz atmosphere. The freshly dispersed substrate was continuously infused into the holotrich suspension through a peristaltic Technicon Autoanalyser pump. Measurement of fatty acid uptake and incorporation into holotrich lipids 4-ml portions of the incubation mixture were centrifuged at 1000 X g for 5 min and the precipitated holotrichs washed three times with 4 ml of ice-cold incubation buffer containing 0.002% (w/v) non-labelled substrate. The liquid scintillation procedure [ 181 was used, with appropriate quenching corrections, to determine the radioactivity of the cells resuspended in an isoosmotic solu- tion of NaCl and of chloroform-methanol extracts of the freeze-dried super- natant. The nature of the lipids was determined in freeze-dried samples of holotrichs by refluxing for 6 h in 30 ml of chloroform/methanol (2 : 1, v/v) followed by 3 h in 30 ml of chloroform/methanol/l% HCl (60 : 30 : 0.5, v/v). The com- bined chloroform-methanol extracts were reduced to 20 ml in vacua and washed three times with 4% (w/v) NaCl solution. The chloroform extracts were chromatographed on layers of silica gel G using hexane/diethyl ether/acetic acid (70 : 30 : 1, v/v) (solvent 1) or chloro- form/methanol/water (65 : 25 : 4, v/v) (solvent 2) as developing solvents. The lipids were located by exposure to IZ vapour, and radioactivity measured in the identified components [ 181. Alternatively, lipids* were recovered from the layers for the preparation of methyl esters [19,20]. The methyl esters were 20 separated according to their degree of unsaturation on silica gel G impregnated with AgNO, [21] and separated by gas-liquid chromatography on 6 ft X 0.25 inch glass columns packed with 13% DEGS on 100-120 mesh chromosorb W at 170°C with Nz as carrier gas. A stream splitter prior to the flame ionization detector allowed the collection of 5/6ths of each component in glass tubes containing glass wool moistened with scintillant. Samples were then eluted into scintillation vials with 10 ml of scintillation liquid [ 181. Results The in vivo linoleic acid content of holotrichs The contribution of Isotricha to the total linoleic acid content of the rumen in the post-feeding period was determined in representative samples of rumen digesta af weekly intervals over a 6 week period. The 18 : 2 content of rumen digesta remained fairly constant over this period of sampling (mean values 8.1, 8.6 and 7.6 mg 18 : 2 per 100 g rumen digesta at 0.5, 1.0 and 3.0 h post-feeding, respectively). Soon after feeding Isotrichs (3 * 103-2 . 104/ml strained rumen fluid) contained about 40% of the total 18 : 2 in wet digesta, but contributed 20-25% at other sampling times. Uptake of linoleate by holotrichs When washed holotrichs were suspended in buffer containing 4 pg sodium linoleate/ml their movement increased, the cells swelled and, after about 60 min, burst. The bursting stage coincided with the appearance of a stable foam in the incubation mixture through which 5% CO* in Nz was bubbled. Bursting was avoided if the suspending buffer contained 3 pg or less sodium linoleate/ml. A similar behaviour was obtained with glucose, in that more than 1.0% (w/v) glucose led to bursting but this was avoided during short-term incubations if the concentration of glucose was below 0.5%. However the total amount of sodium linoleate added during incubations could be increased by continuously infusing it into the buffer containing suspended holotrichs. Optimal rates of infusion into 100 ml suspensions were investigated between the limits of 0.8 and 3.2 mg linoleate/h. High rates of infusion (2.6 mg and 3.2 mg linoleate/h) gave high initial rates of uptake but led to the early bursting of holotrichs. However infusion at 1.6 mg linoleate/h resulted in an absorption of 2.0 mg linoleate by 2 * lo6 Isotrichs in 90 min at a linear rate without bursting and these conditions were chosen for subsequent experiments. Attempts to obtain cell division were unsuccessful despite the use of a number of alternative buffers. Table I summarizes the uptake of [l-i4C]- linoleic acid infused at the rate of 1.6 mg/h by suspensions of 2 * lo6 Isotricha alone in anaerobic incubation buffer and the distribution of label in the cellular lipids. An average of 83.9% of the added label was recovered from the protozoa1 cells after 90 min. Most of the absorbed linoleate remained as free fatty acid (69.0%), but 25.2% was recovered as a constituent of phosphatidylcholine with a small proportion in lysophosphatidylcholine. Total uptake of linoleate by the Isotrichae was reduced in the presence of bacteria, with lower proportions being incorporated into free fatty acids and higher proportions into phospho- lipids. 21 TABLE I INCORPORATION OF RADIOACTIVITY INTO LIPIDS OF HOLOTRICHS INCUBATED ALONE OR WITH BACTERIA FOLLOWING THE CONTINUOUS INFUSION OF [1-‘4C]LINOLEATE Sodium linoleate (1 pCi/mg) in phosphate buffer (1 mg/ml) was continuously infused at the rate of 1.6 mg/h for 90 min into a suspension of holotrichs (2. 106 Isotricha) in 100 ml of anaerobic incubation buffer or buffered bacterial suspension at 39%. The bacterial suspension was free of food particles and protozoa and contained approx 1 . 10 12 bacteria/100 ml. FFA, free fatty acids; PC, phosphatidylcholine; LPC. lysophosphatidylcholine; PE. phosphatidylethanolaine: UPL, unidentified phospholipid. Microorganisms incubated Uptake of 14C Percentage distribution of 1 4 C in lipids absorbed by by holotrichs holotrichs (% of added I4C) * FFA PE UPL PC LPC ** Holotrichs (2.106) 83.9 _+ 3.1 69.0 f 2.7 trace trace 25.2 ? 1.7 3.9 r 0.8 Holotrichs (2. 106) and 51.2 + 2.7 58.8 ? 2.3 trace trace 34.4 * 2.1 4.6 f 0.6 bacteria (1012) * x r S.D., n = 8. ** RF between phosphatidylcholine and phosphatidylethanolamine. However this distribution of label in the cellular lipids after 90 min did not reflect the distribution throughout the period of continuous infusion. Fig. 1 shows that initially, almost all the radioactivity in cellular lipids was associated with phospholipid but there was no further increase in labelled phospholipid after about 30 min. Free fatty acids accumulated at a linear rate 10 min after infusion commenced. The presence of bacteria in the incubation buffer did not affect the incorporation of label into phospholipid but decreased the amount of free fatty acid which accumulated in the holotrichs. This depression of linoleate uptake resulting from the presence of bacteria during the incubation did not occur when palmitate was used as substrate (Fig. 2). The rate of palmitate uptake by holotrichs approximated that of linoleic acid when bacteria were absent from the incubation mixture. In pulse-chase infusion experiments (Fig. 3) [ 1-14C]linoleate was replaced by unlabelled linoleate after 40 min. This was the stage in earlier experiments when the phospholipids had reached virtually a saturation level of radioactivity (Fig. 1) and when the linoleate being absorbed by holotrichs was accumulating as free fatty acid. Fig. 3 shows that the level of accumulated radioactivity in holotrichs associated with free fatty acids remained constant indicating that there was neither exchange of cellular free fatty acids with free fatty acids in the incubation medium nor utilisation of absorbed free fatty acids for phos- pholipid synthesis (Fig. 1). On the other hand, in the presence of bacteria the rate of decline in the cellular free fatty acids of holotrichs when the infusion of unlabelled linoleate was commenced was greater than the conversion into acyl lipids (c.f. Fig. 1) thus indicating an exchange with the free fatty acids in the incubation medium. Such an exchange may provide the reason for the slower rate of accumulation and lower final levels of cellular free fatty acids (Fig. 1). Alternatively the capacity to exchange may be due to biohydrogenation of linoleic acid observed when bacteria were present but not with holotrichs alone. Tome of mfusmn (rmn) Fig. 1. The incorporation of radioactivity into lipids of holotrichs during the continuous infusion of 1.6 mg [l-14C]linoleate/h. See Table I for incubation conditions. Each value was obtained from separate incubations of 100 ml holotrich suspension. - - - - - -, rate of linoleate infusion. Uptake of radioactivity into total lipids (m- 0). phosphatidylcholine (0 ------0). and free fatty acids (i‘---“) by holotrichs incubated alone. Uptake of radioactivity into total lipids (a -g), phosphatidylcholine (0 -0) and free fatty acids (A- A) by holotrichs incubated in the presence of bacteria. Fig. 2. The absorption of [l-14CJlinoleate (1 &i/mg) or [l-l4 Clpalmitate (1 @/mg) by holotrichs when incubated alone or with bacteria. See Table I for conditions of incubation and substrate addition. Linoleate (0 -----D) and palmitate (Ad ) absorbed by a suspension of holotrichs in buffer. Linoleate (m- n ) and palmitate (A ---_A) absorbed by a suspension of holotrichs in buffered rumen fluid containing bacteria. Results are means of two incubations. When Isotricha were incubated in buffer with [ 14C]phosphatidylcholine (1.6 mg/h into 100 ml for 90 min) only 10.7% of the added phosphatidyl- choline was recovered as free fatty acids. Absorbed radioactivity (12.9% of the added substrate) was present as free fatty acids (9.3%) and phospholipid (89%). In the presence of bacteria (1 * lO’O/ml) 32.9% of the added phosphatidyl- choline was converted into free fatty acids. Radioactivity absorbed by Time of lnfuslon (mtn) Fig. 3. Pulse-chase IaDelling of the cellular free fatty acids of holotrlchs in the presence or absence of bacteria. Incubation conditions were as described in Table I except that [1-14C]linoleate was replaced by non-labelled linoleate after 40 min. l -0. radioactivity in free fatty acids of holotrichs which were incubated in buffer; n - l , radioactivity in free fatty acids of holotrichs which were incubated in a buffered rumen fluid containing bacteria. 23 holotrichs comprised 41.6% of the added substrate - 30.3% of this was present as free fatty acids and 65.1% as phospholipids. Biohydrogenation of linoleic acid After infusion of [l-‘4C]linoleic acid for 90 min over 90% of the label remained as linoleic acid (Table II). In the presence of bacteria, however, little [ l-‘4C]linoleic acid remained in the free fatty acids of both the holotrich cells and the supernatant containing bacteria, and most of the radioactivity was present in 18 : 0 and 18 : 1. In contrast, linoleic acid contributed 62% of the radioactivity incorporated into the fatty acids of holotrich phospholipids when bacteria were present during the incubation. The main features of the 14C dis- tribution in fatty acids following infusion of [l-14C]linoleate for 90 min into incubation medium containing bacteria or holotrichs alone or a mixed culture of both bacteria and protozoa (Table III) were: (i) tram-18 : 1 was the main product of the incubation of bacteria with (l-14C)-labelled 18 : 2, with small amounts of 18 : 0 and cis-18 : 1 also formed; (ii) holotrich suspensions exhibited very little hydrogenation of (l-‘4C)-labelled 18 : 2 and the main products were small proportions of tram isomers of 18 : 2 with 80.1% of the (l-‘4C)-labelled 18 : 2 remaining after 90-min incubation; (iii) holotrichs and bacteria incubated together produced the greatest extent of hydrogenation, with the major products being 18 : 0 (41.6%) and tram-18 : 1 (34.6%) with 13% cis-18 : 2 TABLE II DISTRIBUTION OF RADIOACTIVITY IN THE FATTY ACIDS OF THE CELLULAR LIPIDS OF HOLOTRICHS AND IN THE INCUBATION MEDIUM FOLLOWING INCUBATION WITH [1-‘4Cl- LINOLEATE See Table I for conditions of incubation and substrate addition. Microorganisms incubated Lipid fraction and Distribution of radioactivity (percentage of methyl distribution of esters) added radioactiv- ity (% of total) 16:0+16:1 18: 0 18: 1 18: 2 ‘c18 Free fatty acids * 6.2 93.8 - remaining in incu- bation medium (16.1) Holotrichs (2 * 106) Cellular free fatty - 4.6 95.4 acids (57.0) Cellular phospho- 1.4 1.6 5.1 90.2 1.7 lipids (24.9) Holotrichs (2. 106) and bacteria (1 . lOI*) Fatty acids of lipids in bacterial suspension (46.8) Cellular free fatty acids (30.1) CeIIuIar phospho- lipids (20.1) 0.2 53.2 40.0 3.4 3.0 - 43.1 50.8 4.1 3.8 1.5 8.9 24.4 62.4 2.4 * None detected. T A B L E II I T H E D IS T R IB U T IO N O F R A D IO A C T IV IT Y IN T H E F A T T Y A C ID S IS O L A T E D F R O M L IP ID S A F T E R T H E IN C U B A T IO N O F r1 -1 4C 1L IN O L E A T E W IT H B A C T E R IA . H O L O T R IC H S A N D A M IX T U R E O F H O L O T R IC H S A N D B A C T E R IA S ee T a b le I f o r co n d it io n s o f in cu b a ti o n . M ic ro o rg a n is m s p re se n t in t h e in cu b a ti o n D is tr ib u ti o n o f ra d io a ct iv it y in t h e fa tt y a ci d s o f th e to ta l li p id s (% ) S a tu ra te d (1 8 : 0 + tr a ce 16 :0 )* M o n o en e (1 8 : 1 ) tr a m ci s D ie m (1 8 : 2) P o ly en e (1 8: 3) tr a n s- t ra m ci s- ci s a n d c is -t ra m B a ct er ia ( 1 . 10 12 ) 9. 5 i 6. 2 ** B a ct er ia ( 1 * 10 12 ) a n d h o lo tr ic h s (2 . 10 6) 41 .6 ? 7. 3 H o lo tr ic h s (2 . 1 06 ) * G L C an al y si s o f th e sa tu ra te d f ra ct io n i so la te d b y A g + /T L C [ 25 1. ** x f S .D .. n = 6 . * * * N o n e d et ec te d . 70 .1 i 4. 9 6. 3 r 4. 2 tr a ce 7. 1 ?r 1 .2 - ** * 34 .6 k 3 .2 4. 7 f; 2 .1 2. 6 f 0. 2 13 .2 + 2 .4 1. 3 0. 7 t 2. 6 + 0 .3 1. 7 k 0 .2 9. 9 f 1. 1 80 .1 f. 4 .4 5. 7 * 1. 3 25 remaining unaltered. Over a 50 min incubation period an approximately equal amount of cis- and truns-18 : 1 was obtained with a bacterial culture with appreciable 18 : 2 remaining unhydrogenated. With holotrichs, and a mixed culture, the distribution was similar to that obtained after 90 min with less overall hydrogenation. Discussion It was intended that the range of rates of infusion of 18 : 2 into Isotricha suspension should approximate the release of polyunsaturated fatty acids in the rumen following a pasture diet. However the maximal rate of infusion without causing an early disruption of protozoa was 1.6 mg/h into 100 ml, which is considerably less than the calculated in vivo rate of release of polyunsaturated fatty acids [ 221. The binding of free fatty acids by particulate material [ 231 possibly explains why protozoa in strained rumen fluid were able to tolerate 30-60 mg linseed oil fatty acids/100 ml rumen fluid [24]. In contrast we found that 0.4 mg 18 : 2/100 ml was toxic towards isolated holotrichs. The first indication of 18 : 2 toxicity was the swelling of Isotricha which corre- spond to an accumulation of free fatty acids within the cell. The similarity in the behaviour of holotrichs towards high glucose [25] and 18 : 2 uptake sug- gests that disorganisation was due to the accumulation of substrate, i.e. free fatty acids, within the cells. Visual examination showed that swelling of holotrichs was less pronounced in the presence of bacteria and it is likely that the infusion rate of 18 : 2 could have been increased beyond 1.6 mg/h under these conditions. The pulse-chase studies indicated that the free fatty acids absorbed by Isotricha when incubated in mixed culture with bacteria were less well retained than 18 : 2 absorbed by Isotricha incubated alone. Competition by bacteria for available substrate, and exchange with suspending medium may provide mechanisms for the protection of holotrichs against the unphysiological accumulation of fatty acids. Although the ratio of free fatty acids : phospholipid in Isotricha cells obtained after 90 min incubations in holotrichs - bacterial cultures was less than when holotrichs were incubated alone, this ratio was still higher than that observed by Katz and Keeney [ 121 in protozoa obtained from the rumen 4 h after feeding. However in the early stages of infusion, the ratio was consider- ably less since most of the available 18 : 2 was incorporated in phospholipid (Fig. 1). The absence of cell division in the in vitro experiments probably lessened the requirement for structural lipids and, as a consequence, the bio- synthesis of phospholipid. Much of the dietary linoleic acid is present as constituents of phospholipids. However little hydrolysis of phosphatidylcholine occurred when Isotricha were incubated with this substrate alone. In contrast, incubation mixtures containing both Isotricha and bacteria hydrolysed about one-third of the substrate in 20 min. This would indicate that Isotricha rely on the activity of bacterial [ 261 or plant lipases [2] for the release of acyl groups from dietary phospholipids prior to their absorption. In recent years there has been considerable interest in the different capacities 26 of the rumen components and the individual organisms to hydrogenate dietary unsaturated fatty acids [4,5,27]. The role of the heterogeneous particulate fraction, which contains partly digested food particles, adhering bacteria and protozoa, has been emphasised by Harfoot et al. [23]. Even after removal of this fraction by centrifugation at 160 X g, the remaining supernatant still con- verts polyunsaturated fatty acids mainly to 18 : 1 rather than to 18 : 0 which occurs when total rumen fluid is used [ 51. Although protozoa are included in the particulate fraction, the Holotrich protozoa themselves produced very few products of 18 : 2 biohydrogenation and none of these were fully saturated. In view of likely bacterial contamination this small activity in the biohydrogena- tion of 18 : 2 could not be attributed to the protozoa. In view of the inability of Isotricha to hydrogenate 18 : 2 the greater hydrogenation of 18 : 1 to 18 : 0 by bacteriallsotricha suspensions than by bacteria alone was unexpected. Moreover Chalupa and Kutches [13] were un- able to obtain the biohydrogenation of 18 : 1 by Isotricha reported earlier by Gutierrez et al. [ll]. The rapid uptake of free fatty acids by Isotricha and the exchange of these free fatty acids with the incubation medium in the presence of bacteria allows the possibility that hydrogenation of 18 : 1 in mixed culture requires the uptake of substrates by Isotrichu or alternatively its adsorption at the cell surface. The presence of large proportions of 18 : 0 in the free fatty acids of both the bacterial suspension and of Isotrichu (Table II) support an exchange of free fatty acids between the two groups of organisms but does not apportion the ability to hydrogenate. Although the mechanism is unclear, the present work indicates that the biohydrogenation of unsaturated fatty acids in the rumen may involve a symbiotic relationship between bacteria and protozoa. These observations, involving bacteria and Isotrichu, are somewhat analogous to the adsorption characteristics of food particules towards free fatty acids [27] and the considerable capacity of associated microorganisms for biohydrogena- tion [23]. Although the free fatty acids of Isotricha which had been incubated with bacteria were extensively biohydrogenated, appreciable proportions of (l-14C)-labelled 18 : 2 were incorporated into the protozoa1 phospholipids. Under many dietary conditions it is doubtful if Isotrichu play a vital role in providing 18 : 2 for cattle and other ruminants, since, although they contain more 18 : 2 than the oligotrichs [ 281, the latter are more numerous. The role of Isotrichu may be of some importance when cattle or other ruminants are grazing on poor quality pasture when the holotrich/oligotrich ratio is higher [ 171 than on high quality pasture [ 291. Acknowledgement V. Girard was supported by a Grant for Scientific Research awarded by the Federal Institute of Technology, Zurich, Switzerland. References 1 Garton, G.A., Hobson. P.N. and Lough. A.K. (1958) Nature 182, 1511-1512 2 Faruque, A.J.M.O., Jan&. B.D.W. and Hawke, J.C. (1974) J. Sci. Food Agric. 25.1313-1328 3 Reiser, R. (1951) Fed. Proc. 10, 236 27 4 Dawson, R.M.C. and Kemp. P. (1970) in Physiology of Digestion and Metabolism in the Ruminant (Phillipson. A.T., ed.), pp. 504-518, Oriel Press, Newcastle-upon-Tyne 5 Hawke. J.C. and Silcock, W.R. (1970) Biochim. Biophys. Acta 218, 201-212 6 Hunter, W.J., Baker, F.C., Rosenfeld. 13.. Keyser, J.B. and Tove, S.B. (1976) J. Biol. Chem. 251, 2241-2247 7 Moore, J.H. (1974) in Industrial Aspects of Biochemistry (Spender, B.. ed.), PP. 835-863, North- Holland Publishing Co., Amsterdam 8 Hawke, J.C. (1971) Biochim. Biophys. Acta 248. 167-170 9 Keeney. M. (1970) in Physiology of Digestion and Metabolism in the Ruminant (Phillipson, A.T., ed.), PP. 489-503, Oriel Press, Newcastle-upon-Tyne 10 Bath, I.H. and Hill, K.J. (1967) J. Agr. Sci. 68, 139-148 11 Gutierrez. J.. Williams, P.P., Davis. R.E. and Warwick, E.J. (1962) Appl. Microbial. 10, 548-551 12 Katz, I. and Keeney, M. (1967) Biochim. Biophys. Acta 144,102-112 13 Chalupa, W. and Kutches, A.J. (1968) J. Animal Sci. 27, 1502-1508 14 Wright. D.E. (1959) Nature 184.875-876 15 Gutierrez, J. (1955) Biochem. J. 60, 516-522 16 Clarke, R.T.J. and Hungate, R.E. (1966) Appl. Microbial. 14, 340-345 17 Hungate. R.E. (1957) Can. J. Microbial. 3, 289-311 18 Clarke. D.G. and Hawke. J.C. (1970) J. Sci. Food Agric. 21,446-452 19 Metcalfe, L.D. and Schmitz, A.A. (1961) Anal. Chem. 33, 363-364 20 Shehata, A.A.Y.. de Man, J.M. and Alexander, J.C. (1970) Can. Inst. Food Technol. 3, 85-98 21 Wood, R. and Snyder, F. (1966) J. Am. Oil Chem. Sot. 43, 53-57 22 Dawson. R.M.C., Hemington, N., ‘rime, D., Lander, D. and Kemp, P. (1974) Biochem. J. 144, 169- 171 23 Harfoot, C.G., Noble, R.C. and Moore. J.H. (1973) Biochem. J. 132, 829-832 24 Czerkawski, J.W. (1967) Br. J. Nutr. 21. 865-878 25 Sugden, B. and Oxford, A.E. (1952) J. Gem Microbial. 7. 145-153 26 Lough, A.K. (1970) in Physiology of Digestion and Metabolism in the Ruminant (Phillipson, A.T.. ed.). PP. 519-528, Oriel Press, Newcastle-upon-Tyne 27 Harfoot, C.G., Noble, R.C. and Moore, J.H. (1973) J. Sci. Food A&c. 24,961-970 28 Williams, P.P. and Dinusson, W.E. (1973) J. Animal Sci. 36, 151-155 29 Clarke. R.T.J. (1974) Ph.D. Thesis, Massey University, N.Z.. 1964