Peptide and amino acid transport in Streptococcus bovis

Appl Microbiol Biotechnol (1990) 34:97 102

Applied Microbiology Biotechnology © Springer-Verlag1990

Peptide and amino acid transport in Streptococcus boris Kenneth Westlake 1 and Roderick Ian M a c k i e z

Rumen Biochemistry,Animal and Dairy Science Research Institute, Private Bag X2, Irene 1675, Republic of South Africa Received 13 February 1990/Accepted 26 June 1990

Summary. In amino acid transport studies with S t r e p t o coccus boris using ~4C-labelled amino acids, it has been

shown that between 87% and 95% of cell-associated radioactivity was located in the cytosol. In similar studies with unlabelled peptides, most test peptide associated with S. bovis was truly intracellular. Using sodium dodecyl sulphate-polyacrylamide gel electrophoresis, the proteolytic activity in S. boris was found to be largely cell-associated and of the serine-protease type, but stimulated by dithiothreitol. A wide range of extracellular peptide hydrolysing activities was demonstrated against the pentapeptide Leu-Trp-Met-Arg-Phe, which was completely hydrolysed to eight products after 10min incubation. Some of this pentapeptide was transported intact, indicating the existence of mechanisms for the transport of peptides up to 751 Da. In studies with Arg-Phe-Ala, only Phe (F) and Ala (A), and to a much lesser extent Phe-Ala (FA) were transported after extracellular hydrolysis to FA, Arg (R), F and A. In this case, amino acid transport was much more predominant than peptide transport. The extent and nature of peptide transport was affected by the addition of protease inhibitors.

Introduction

It has been well documented that many bacteria can utilize peptides as a source of amino acids and, for several bacterial systems, peptide transport followed by intracellular cleavage has been directly demonstrated (Payne 1980). Further, the location and specificity of several bacterial peptidases and permeases has been elucidated (Payne 1980). However, despite the contri~ Present address: AFRC Institute of Food Research, Colney Lane, Norwich NR4 7UA, UK 2 Present address: Department of Animal Sciences, University of Illinois, Urbana-Champaign, 1207 W. Gregory Drive, Urbana, IL 61801, USA Offprint requests to: R. I. Mackie

bution that such studies would make to a better understanding of nitrogen flow in the tureen and hence to identifying possible points at which bacterial protein utilization could be controlled, little work has been conducted with ruminal bacteria. Some data is available for these bacteria from radiolabelling studies and studies in which altered growth rates in the presence of various nitrogen sources has been monitored (Pittman and Bryant 1964; Pittman et al. 1967; Russell 1983). Because the latter technique relies on indirect measurement and the former cannot account for amino acid efflux from the cell or extracellular hydrolysis prior to uptake, neither will give an accurate picture of amino acid or peptide transport. Results obtained using hydrophilic and hydrophobic fractions from casein hydrolysate showed that the hydrophilic peptides were catabolized twice as fast as hydrophobic peptides by mixed ruminal bacteria and also by pure cultures of peptide-utilizing rumen bacteria (Chen et al. 1987). Peptide nitrogen was taken up more than twice as fast as amino acid nitrogen by mixed ruminal bacteria in this study. More recently, Broderick et al. (1988) have utilized peptidep-nitroanilides to monitor peptide uptake in diluted ruminal fluid. However, by utilizing a dansylation procedure (Payne and Bell 1979) in which visual analysis after two-dimensional TLC can be performed, the fate and transport characteristics of native peptides rather than peptide derivatives can be obtained. Since Streptococcus boris has been shown to be an important proteolytic bacterium in the rumen (Russell et al. 1981; Wallace and Brammall 1985), the aim of the present paper was to determine the extent of extracellular proteolytic activity and to further monitor peptide and amino acid transport in this bacterium.

Materials and methods Bacterial. Streptococcus boris K11.21.09.6C was isolated from the ruminal contents of Merino sheep fed lucerne hay and was from our own collection.

98

Culture techniques. Anaerobic conditions were maintained as described by van Gylswyk and Hoffman (1970). All resting cell studies were conducted in an anaerobic cabinet (Forma Scientific Model 1024, Marietta, Ohio, USA; with 65% N2 - 30% CO~ - 5% H2 gas phase) except where centrifugation was required, in which case sealed anaerobic centrifuge tubes were employed. Culture media and growth conditions. S. boris was grown at 37 ° C, without shaking on a liquid medium based on medium 10 of Caldwell and Bryant (1966) but containing 20% rumen fluid clarified by centrifugation (4°C; 13,000g; 20 min) and 0.25% (w/v) B-vitamin solution (Scott and Dehority 1965). Bacterial inoculum was added in 0.5 ml anaerobic diluent (Caldwell and Bryant 1966) after washing the overnight growth (14 h) from an agar maintenance slant of the same medium. The medium was gassed with a mixture containing 65% N2 - 30% COz - 5% H~ and was incubated at 39 ° C. Harvest conditions. Cells were harvested (10,000g; 20 min) under anaerobic conditions after attaining an optical density at 578 nm (OD578) of 1.0 (mid-logarithmic phase of growth). After washing ( × 2) in cold 50 m M phosphate buffer, pH 6.8, the final bacterial pellet was resuspended in incubation buffer (50 mM phosphate buffer containing 20 m M glucose) as required. Protease inhibition in polyacrylamide gels. S. boris was grown as described above but modified by the use of 5-1 fermentor vessels inoculated with 200 ml bacterial suspension. Subsequent treatments were conducted on freshly prepared bacterial suspensions. Supernatant fluid (growth medium) was sterile filtered through a 0.22-~tm filter (Millipore Intertech, Bedford, Mass., USA) and then concentrated using a Minitan ultrafilter and 10,000-Da molecular mass cut-off filter (Millipore Intertech). The concentrated retentate of molecular mass > 10,000 Da was kept frozen for further use. Bacterial cells were resuspended in a minimal volume of 0.02 M phosphate buffer, pH 6.9 (suspension buffer), prepared using boiled, distilled, deionized water which had been purged with nitrogen gas. The cell suspension was incubated on ice for 30 min with 200 ~tg/ml RNase and DNase (Boehringer, Mannheim, FRG) and then ruptured in a French pressure cell (Aminco model 20K, Urbana, Ill., USA) at a pressure of 34.45 MPa. The lysed fraction was collected in an evacuated and subsequently nitrogenpurged sealed vessel. Unbroken cells were removed by centrifugation (5,000g; 10 min; 4 ° C) and the supernatant passed through a 0.8-~tm sterile filter (Millipore Intertech). The filtrate was then centrifuged (110,000g; 4°C; 60 rain) to sediment membrane and capsular material which formed discrete pellets. Supernatant fluid (cytosolic fraction) was dispensed (3 × 0.5 ml) into 1 ml Eppendorf tubes to which was added 0.5 ml sample solvent containing (w/v) 2% sodium dodecyl sulphate (SDS), and 5% 2-mercaptoethanol. The remainder of the supernatant fluid was discarded and the pellet washed three times with suspension buffer and finally resuspended in 1.2 ml of the same buffer (membrane fraction). Samples of this particulate material and previously prepared and thawed sterile filtered medium samples (2 × 0.5 ml of each) were prepared for SDS-polyacrylamide gel electrophoresis (PAGE) as described above. All samples were subjected to SDSPAGE as described by Strydom et al. (1986) using gels containing co-polymerised gelatine. Protease activity was detected as zones of hydrolysis after staining background gelatine with Amido Black. After electrophoresis, replicate samples were cut into strips and incubated in the presence of various protease inhibitors. Phenylmethylsulphonylfluoride (PMSF), ethylenediamine tetraacetic acid (EDTA) and dithiothreitol (DTT) were made up in 100 mM phosphate buffer, pH 6.8, to a final concentration of 1 mM and 5 mM. Parachloromercuribenzoate (PCMB) was made up in dimethylsulphoxide to the same final concentration. Transport of radiolabelled amino acids. A 1-1 volume of S. boris, grown and harvested as described previously, was resuspended in

5 ml incubation buffer (final vol = 5,5 ml) and incubated at 39 ° C for 15 min. To this was added 1 ml of either L-[U-14C] glutamic acid, L-[U-14C] phenylalanine or L-[U-14C] alanine (Radiochemical Centre, Amersham, Bucks., UK) to a final concentration of 0.2846 MBq m l - 1. After a futher 2-min incubation, 4.5 ml of the bacterial suspension was collected and washed twice (2,0009; 4 ° C; 10 rain) in incubation buffer. The supernatant (Sn~) was retained and the pellet (P1) was transferred to a French pressure cell and the cells ruptured at a pressure of 34.45 MPa. The ruptured cells were collected under aerobic conditions on ice and any unbroken cells sedimented after centrifugation (2,0009; 10rain; 4 ° C). This constituted fraction P2. The supernatant (Sn2) was centrifuged (60,0009; 30 min; 4 ° C) after removing 0.2 ml to determine total radioactivity in this fraction. The pellet (P3), which comprised membranous material, was washed twice in 0.02 M phosphate buffer, pH 6.8, and the washings combined (W). The supernatant (Sn3) constituted the cytosolic fraction. After resuspension in deionised water, pelleted material (0.2 ml), was treated with 1 ml Soluene 350 (Packard Instrument Co., Ill., USA) and left at room temperature overnight. To this was added 0.4 ml deionized water and the final volume made up to 5.0 ml with Scintillator 299 (Packard Instrument Co.). Supernatant fractions (0.2 ml) were added to 4.8 ml Scintillator 299 (containing 0.4 ml deionized water) and the radioactivity in each sample was measured in a Packard 2000 CA liquid scintillation counter. The remaining original cell suspension was centrifuged through 0.3 ml silicone oil (3 parts Hysol 556 fluid; 2 parts Hysol 550 fluid obtained from Dexter-Hysol, Olean, NY, USA) in 1.5-ml Eppendorf microtubes. Supernatant (Sn~b) and pellet (Plb) fractions were counted in a liquid scintillation counter.

Transport of unlabelled peptides and the effect of protease inhibitors. A 200-ml volume of S. boris, grown and harvested as described previously, was resuspended in 10 ml incubation buffer. After incubation at 39°C for 15 min, 0.7 ml cell suspension was layered above 0.3 ml silicone oil mixture as described above. To this was added 0.3 ml peptide (Serva, Westbury, NY, USA) in incubation buffer to a final concentration of 10 mM. After 2 rain incubation, the bacteria were centrifuged through the silicone oil according to the method of Hurwitz et al. (1965). Supernatant (200 ~tl) was transferred to a 1/2 dram vial (Supelco, Bellefonte, Pa., USA) prior to dansylation. The microtube was then inverted and allowed to stand for 5 min in order for the silicone oil to drain completely from the bacterial pellet. The tube was frozen in liquid N2 and the tip of the tube, containing the bacterial pellet, removed with a scalpel blade. The frozen pellet was transferred to a 1/2 dram vial, to which was added 200 ml 15% trichloroacetic acid (TCA) and the suspension boiled for 10 min. Suspension (100 ~tl) was then removed from the vial and the remainder, together with the supernatant samples, were dansylated according to the method of Payne and Bell (1979), modified in the case of the cellular fraction by the addition of 10 ~tl 2 M NaOH, to replace bicarbonate. To determine the extent of soluble extracellular proteolytic activity, 10 ml of growth medium was passed through a sterile 0.22 lxm membrane filter. Filtrate samples (0.7 ml) were incubated with peptide (as above) before dansylation. Thin layer chromatography (TLC) of the dansylated extracts was performed on 5 × 5 micropolyamide plates (Schleicher and Schuell, Dassel, FRG). The plates were developed in formic acid: deionized H20 (1.5:50) in the first dimension and benzene: acetic acid (9:1) in the second dimension. Dansylated products were viewed under UV light (366 nm) and photographed using a red filter. To study the effect of protease inhibitors, the above method was modified to allow the addition of 0.1 ml inhibitor to a final concentration of 1 mM (PMSF, PCMB) or 5 mM (EDTA, DTT). This was achieved by increasing the peptide concentration while decreasing the volume added to maintain the same final concentration. Confirmation of unlabelled amino acid and peptide transport, as opposed to cell wall-binding, after cell disruption in a

99 French press and subsequent fractionation, was performed after dansylation of prepared fractions.

and, of those tested, was the most effective inhibitor of proteolysis in S. boris.

Results

Distribution of 14C-labelled amino acids Inhibition of proteolytic activity after PAGE The proteolytic activity of various fractions p r e p a r e d from S. bovis is shown in Fig. 1. Most of this activity was present in the cytosolic fraction, whereas lower activity was found in the growth m e d i u m fraction. In the presence of 1 m M D T T proteolytic activity was enhanced in both the cytosolic and m e m b r a n e fraction but not the culture m e d i u m fractions. The addition of P C M B (1 mM) had little effect on proteolytic activity while P M S F (1 mM) caused approximately 80% inhibition by comparison with a control in the absence of inhibitor

The distribution of label between fraction Pab and Sn~b (Table 1) shows that for each amino acid, approximately 35% of the radiolabel had been taken up by the cell after 2 min incubation. The distribution of radiolabel in fractions P2 and Sn2 gives an indication of the extent of cell disruption after passage through the French pressure cell and shows that the percentage cell breakage varied between approximately 91% and 99%. This technique was found to be the most effective at high cell densities and for this reason cells from Plb were resuspended in a m i n i m u m volume of buffer. After ultracentrifugation of the lysed cell fraction (Sn2) , the distribution of label between the m e m b r a nous material (P3) and the cytosolic material (Sn3) confirmed that most (81-95%) of the 14C-amino acid was present in the cytosol. After washing, a small a m o u n t of radioactivity was released from pellet P3 into the wash buffer (W).

Distribution of unlabelled amino acids and peptides

Fig. 1. Protease activity in sodium dodecyl sulphate-polyacrylamide gels containing co-polymerized gelatine of fractions prepared from Streptococcus boris. Lane 1, growth medium (molecular mass > 10,000 DA); lane 2, growth medium (molecular mass < 10,000 Da); lane 3, membrane fraction; lane 4, cytosol fraction. Approx. 25 ~tg protein was applied per lane. All samples were prepared from an equal volume of culture except for growth medium which was prepared from twice the volume of the other samples

In this study, the distribution of small peptides, as well as amino acids, was investigated and confirmed the above data. Results were recorded on a qualitative basis and showed for triglycine, Arg-Phe-Ala (RFA) and Ala (A), that the concentration of each c o m p o u n d in the cell wall fraction was below the detection limits of the assay (10 ~M). The concentration of these compounds in the cytosolic fraction was much higher and could be easily seen as large intense spots when viewed under UV light after two-dimensional chromatography.

Table 1. Distribution of 14C-labelled amino acids in various fractions of Streptococcus botJ/S

Fraction

Glutamate

Alanine

Phenylalanine

Actvitya

% Total

Actvitya

% Total

Actvitya

% Total

Plb Sn~b

21.5 47.2

31 69

23.8 44.4

35 65

24.40 41.40

37 63

P2 Sn2

0.20 21.28

1 99

2.14 21.66

9 91

0.24 24.16

1 99

P3 Sn3

0.21 20.22

1 95

0.22 17.54

1 81

0.24 22.46

1 93

W

0.85

4

0.87

12

1.45

6

~ Radioactivity in dpm × 10 6, from 4.5 ml original cell suspension, after correcting to 100% recovery of label For fraction preparation and identification see Materials and methods

100 Hydrolysis and transport of peptides In all experiments, a control without added peptide served to distinghish between transported compounds and endogenous amino acids and peptides. A cell-free fraction was also used to ascertain whether chemical degradation during the dansylation procedure could be responsible for apparent peptide hydrolsis and was always negative. Investigations conducted on repeated occasions showed that there was partial extracellular hydrolysis of the peptides triglycine, RFA and Leu-TrpMet-Arg-Phe (LWMRF) after 2 min incubation with S. boris (Table 2). In order to determine whether this activity was cell bound or freely soluble, the peptides were also incubated for 2 min with cell-free medium in which the S. boris had been grown and showed that S. boris possesses a small amount of freely soluble extracellular peptidase activity. Although the identify of the hydrolysis products was the same, the extent of hydrolysis was not as great as that shown by resuspended, washed bacterial cells, as judged by the amount of peptide remaining in the incubation medium and the much lower intensity (concentration) of fluorescent peptide fragments. Triglycine was hydrolysed to glycine alone by medium and resuspended cell preparations with some triglycine remaining after 2 min incubation. Both glycine and triglycine accumulated in the cytosolic fraction (Table 2). The tripeptide RFA was hydrolysed to Phe-Ala (FA), Phe (F), A and Arg (R) by both the medium and resuspended cell fraction (Table 2, Fig. 2). In both fractions, some RFA remained unhydrolysed after 2 min incubation. Of the hydrolysis products, FA, F and A accumulated in the cytosol (Table 2, Fig. 2). The pentapeptide L W M R F underwent extracellular hydrolysis to eight products after 2 min incubation with the resuspended cell fraction (Table 2, Fig. 3A). Of these, two were identified as L W M R and F. The product marked 'U' (unidentified) which appears above 'DNS' (dansYlated product) (Fig. 3B) is thought to be arginine although the fraction constituting the remainder of the petide (LWM) could not be detected. The remainder could not be identified because of lack of suitable standards. The original pentapeptide was not visible in the extracellular fraction after 10 min incubation. When both L W M R F and RFA were added together to the incubation mixture, the intact tetrapeptide

Table 2. Hydrolysis and intracellular accumulation of triglycine

(GGG), Arg-Phe-Ala (LWMRF) in S. boris

(RFA), and

Leu-Trp-Met-Arg-Phe

Peptide added a

Extracellular hydrolysis

Intracellular accumulation

GGG RFA LWMRF

GGG, G RFA,. FA, R, F, A L W M R ,F Six unidentified peptides

GGG, G FA, F, A LWMR, F

a A, Ala; F, Phe; G, Gly; L, Leu; M, Met; R, Arg; W, Trp

Fig. 2 A, B. Extracellular hydrolysis of the tripeptide Arg-Phe-Ala (RFA) and amino acid transport by S. boris. Resuspended cells of S. boris were incubated for 2 min in the presence of RFA. After separation of cells and supernatant by centrifugation, the cells were lysed and dansylated extracts of cellular and supernatant samples were viewed under UV light after two-dimensional chromatography. A Extracellular (supernatant)extract. B Intracellular (cytosol) extract. A, Ala; F, Phe; R, Arg; DNS, dansylated product; U, unidentified compound

(LWMR), as well as FA, F and A, accumulated in the cytosolic fraction. The effects of the protease inhibitors EDTA, DTT, PCMB and PMSF on the transport of L W M R F and RFA are summarized in Table 3. The addition of D T T (5 mM) prevented the intracellular accumulation at detectable levels of L W M R F as well as any peptide fragments derived from this or RFA. In the presence of 5 m M EDTA, L W M R F was transported intact. However, neither RFA, nor fragments of this or L W M R F accumulated with the bacteria after 2 min incubation. RFA did not accumulate within the cell even in the absence of protease inhibitors.

Discussion

By following the distribution of ~4C-labelled amino acids after cell lysis and fractionation it was shown that, although the distribution of ~4C label varied ac-

101 Table 3. Intracellular accumulation of amino acids and peptides in S. boris and inhibition by various protease inhibitors Inhibitor (raM)

Intracellular accumulation of peptides or amino acids a

Control (no inhibitor PMSF (1)

FA, F, A, LWMRF, LWMR F, A, LWMRF, LWMR F, A, LWMRF, LWMR LWMRF

PCMB (1) EDTA (5) DTT (5)

No intracellular accumulation

Peptides or amino acids not transported

RFA, FA RFA, FA RFA, FA, F,A, LWMR RFA, FA, F,A, LWMRF, LWMR

a See footnote to Table 2 PMSF, phenylmethylsulphonylfluoride; PCMB, parachloromercuribenzoate; EDTA, ethylenediaminetetraacetate; DTT, dithiothreitol

Fig. 3 A, B. Extracellular hydrolysis of the pentapeptide Leu-TrpMet-Arg-Phe (LWMRF) by S. boris. Resuspended cells of S. boris were incubated in the presence of LWMRF. After separation of cells and supernatant by centrifugation the dansylated supernatant extract was viewed under UV light after two-dimensional TLC. A Endogenous fluorescent metabolites. B Extracellular hydrolysis products from LWMRF cording to the amino acid, most radioactivity was found to be cytosolic and not membrane-bound. In a similar study in which a non-radioactive amino acid and peptides were used and where the distribution of each was followed after dansylation and two-dimensional TLC, the results corroborated the above findings with most of the added amino acid being found in the cytosol. In this case, although the concentration o f added amino acid or peptide was below the detection limit in the membrane fraction (P3), the same compounds were visible in the cytosolic (Sn3) fraction. Together these data confirm that such studies represent a valid measure of amino acid and peptide transport and that the amount of cell-wall binding is negligible. Amino acid transport in S. boris has been demonstrated recently (Russell et al. 1988). Analysis by SDS-PAGE of the location of proteolytic activity in S. boris could detect only a very small amount of extracellular activity and showed that most proteolytic activity was cell-associated of the serine protease type. The presence of small amounts of activity in the extracellular fraction may also be due to release of membrane-bound protease during sample pre-

paration. The presence of a small amount of extracellular activity was corroborated in peptide transport studies where the nature of the cell-free activity was the same as the cell-bound activities and therefore also suggests that the activity was released during sample preparation. Alternatively, cell lysis with concomitant release of intracellular enzymes during batch growth of S. boris for use in these experiments cannot be ruled out. The combined data, using SDS-PAGE and peptide transport techniques, suggest that proteolytic activity in S. boris is loosely bound to the cell wall. In all studies conducted with RFA and in the presence of protease inhibitors, only F and A accumulated intracellularly. However, in studies with L W M R F and in the presence of various protease inhibitors this peptide was transported intact showing that S. boris is capable of transporting amino acids and small peptide fragments up to a molecular mass of 751 Da. In the absence of protease inhibitors L W M R F did not appear to be transported in earlier studies. This was most likely due to hydrolysis to L W M R and F, which accumulated intracellularly. In recent in vitro studies with mixed rumen microorganisms (Broderick et al. 1988), the uptake of trialanine was the most rapid in a series of alanine homopeptides. However, the largest of the series, pentaalanine (373 Da), was also transported. Apparent exclusion, therefore, of small peptides is a consequence of factors other than chain length. The activity and range of cell-bound extracellular peptidase activity in S. boris are indicated by the production of eight breakdown products from the pentapeptide LWMRF. In this case, after 10 min incubation, the original peptide could not be detected above the detection limit of the assay which indicated a high level of peptidase activity in the bacterium. Protease inhibitors were shown to affect peptide transport according to the inhibitor used and it is interesting to note that whereas RFA was never detected within the cell, the transport of the dipeptide FA, but not of the higher oligomers, was inhibited by PMSF and PCMB. It seems likely that al-

102 though in the absence of protease inhibitor larger peptides of similar hydrophobicity are transported by S. boris, R F A is not. Rapid hydrolysis, after transport, to smaller fragments offers a likely explanation for this observation. Previous work with Escheriehia coli has identified separate dipeptide and oligopeptide permeases (Payne and Bell 1979), and if R F A is rapidly hydrolysed within the cell, then the same situation may be true of S. boris. In the case of DTT where no intracellular peptide accumulation could be detected, it is perhaps possible that increased proteolytic activity in the presence of 5 m M DTT has resulted in the cleavage of added peptides to their constitutive amino acids which, as previously demonstrated (Payne and Bell 1979), can exit from the cell. Thus, by using a contribution of complementary SDS-PAGE and peptide transport techniques, a fuller picture of the extent, nature and location of protease and peptidase enzymes in S. boris has been obtained. By repeating such studies on other major proteolytic ruminal bacteria such as Bacteroides ruminicola and Butyrivibrio fibrisolvens, a clearer picture of potential interspecies substrate transfer and synergistic activities of these bacteria could be developed which would complement growth rate studies. Detailed knowledge regarding protease inhibition and nature of action, as obtained in these studies, will also enable the identification of potential control points for manipulation of rumen fermentation.

References Broderick GA, Wallace RJ, McKain N (1988) Uptake of small neutral peptides by mixed rumen microorganisms in vitro. J Sci Food Agric 42:109-118

Caldwell DR, Bryant MP (1966) Medium without rumen fluid for non-selective enumeration and isolation of tureen bacteria. Appl Microbiol 14:794-801 Chen H, Strobel HJ, Russell JB, Sniffen CJ (1987) Effect of hydrophobicity on utilization of peptides by ruminal bacteria in vitro. Appl Environ Microbiol 53:2021-2025 Gylswyk NO van, Hoffman JPL (1970) Characteristics of cellulolytic cillobacteria from the rumens of sheep fed tef (Eragrostis tef) hay diets. J Gen Microbiol 60:381-386 Hurwitz C, Brown CB, Peabody RA (1965) Washing bacteria by centrifugation through a water-immiscible layer of silicones. J Bacteriol 90:1692-1695 Payne JW (1980) Transport and utilization of peptides by bacteria. In: Payne JW (ed) Microorganisms and nitrogen sources. Wiley, Chichester, pp 211-256 Payne JW, Bell G (1979) Direct determination of the properties of peptide transport systems in Escherichia coli, using a fluorescent labelling procedure. J Bacteriol 137:447-455 Pittman KA, Bryant MP (1964) Peptides and other nitrogen sources for growth of Bacteroides ruminieola. J Bacteriol 88:401-410 Pittman KA, Lakshmanan S, Bryant MP (1967) Oligopeptide uptake by Bacteroides ruminieola. J Bacteriol 93:1499-1508 Russell JB (1983) Fermentation of peptides by Baeteroides ruminicola B14. Appl Environ Microbiol 45:1566-1574 Russell JB, Bottje WG, Cotta MA (1981) Degradation of protein by mixed cultures of rumen bacteria: identification of Streptococcus boris as an actively proteolytic rumen bacterium. J Anim Sci 53:242-252 Russell JB, Strobel HJ, Driessen AJM, Konings WN (1988) Sodium-dependent transport of neutral amino acids by whole cells and membrane vesicles of Streptococcus boris, a ruminal bacterium. J Bacteriol 170:3531-3536 Scott HW, Dehority BA (1965) Vitamin requirements of several cellulolytic tureen bacteria. J Bacteriol 89:1169-1175 Strydom E, Mackie RI, Woods D (1986) Detection and characterization of extracellular proteases in Butyrivibrio fibrisolvens H17c. Appl Microbiol Biotechnol 24:214-217 Wallace RJ, Brammall ML (1985) The role of different species of bacteria in the hydrolysis of protein in the rumen. J Gen Microbiol 131:821-832