Comparison of proteins involved in chondroitin sulfate utilization by three colonic Bacteroides species.

Three species of colonic bacteria can ferment the mucopolysaccharide chondroitin sulfate: Bacteroides ovatus, Bacteroides sp. strain 3452A (an unnamed DNA homology group), and B. thetaiotaomicron. Proteins associated with the utilization of chondroitin sulfate by B. thetaiotaomicron have been characterized previously. In this report we compare chondroitin lyases and chondroitin sulfate-associated outer membrane polypeptides of B. ovatus and Bacteroides sp. strain 3452A with those of B. thetaiotaomicron. All three species produce two soluble cell-associated chondroitin lyases, chondroitin lyase I and II. Purified enzymes from the three species have similar pH optima, Km values, and molecular weights. However, peptide mapping experiments show that the chondroitin lyases from B. ovatus and Bacteroides sp. strain 3452A are not identical to those of B. thetaiotaomicron. A cloned gene that codes for the chondroitin lyase II from B. thetaiotaomicron hybridized on a Southern blot with DNA from B. ovatus or Bacteroides sp. strain 3452A only when low-stringency conditions were used. Antibody to chondroitin lyase II from B. thetaiotaomicron did not cross-react with chondroitin lyase II from B. ovatus or Bacteroides sp. strain 3452A. Chondroitin lyase activity in all three species was inducible by chondroitin sulfate. B. ovatus and Bacteroides sp. strain 3452A, like B. thetaiotaomicron, have outer membrane polypeptides that appear to be regulated by chondroitin sulfate, but the chondroitin sulfate-associated outer membrane polypeptides differ in molecular weight. Despite these differences, the ability of intact bacteria to utilize chondroitin sulfate, as indicated by growth yields in carbohydrate-limited continuous culture and the rate at which the chondroitin lyases were induced, was the same for all three species.     

Use of targeted insertional mutagenesis to determine whether chondroitin lyase II is essential for chondroitin sulfate utilization by Bacteroides thetaiotaomicron.

Bacteroides thetaiotaomicron produces two inducible chondroitin lyases (I and II) when it is grown on chondroitin sulfate. Both enzymes have very similar biochemical properties. To determine whether both enzymes are required for growth on chondroitin sulfate, we constructed a Bacteroides suicide vector, pE3-1, and used it to create an insertional mutation that interrupts the chondroitin lyase II gene of Bacteroides thetaiotaomicron. pE3-1 contains a 4.4-kilobase cryptic B. eggerthii plasmid (pB8-51), the Escherichia coli cloning vector pBR328, and the EcoRI D fragment from the conjugative B. fragilis plasmid pBF4. A 0.8-kilobase fragment from the center of the B. thetaiotaomicron chondroitin lyase II gene was inserted in pE3-1 to create pEG817. Although, pEG817 is stably maintained in E. coli and can be mobilized into B. thetaiotaomicron by the IncP plasmid R751, pEG817 is not maintained as a plasmid in Bacteroides spp. When pEG817 was mobilized into B. thetaiotaomicron, with selection for a drug marker on pEG817, transconjugants were obtained which had pEG817 inserted into the chondroitin lyase II gene. Western blot analysis was used to confirm that intact chondroitin lyase II is not produced in the mutant. The mutant was able to utilize chondroitin sulfate as a sole source of carbon, although no active chondroitin lyase II was produced. Thus chondroitin lyase I alone appears to be sufficient for growth on chondroitin sulfate. The mutant also had some minor changes in its outer membrane protein profile. However, there was no evidence that any of the major chondroitin sulfate-associated polypeptides in the outer membrane were affected by the insertion in the chondroitin lyase II gene.     

Evidence that polygalacturonic acid may not be a major source of carbon and energy for some colonic Bacteroides species

Five Bacteroides species that are found in the human colon can utilize polygalacturonic acid (PGA) when they are grown in laboratory media: Bacteroides thetaiotaomicron, Bacteroides vulgatus, Bacteroides ovatus, Bacteroides fragilis subsp. a, and Bacteroides sp. strain 3452A (an unnamed DNA-DNA homology group). PGA-degrading enzymes from B. thetaiotaomicron have been isolated and characterized previously. To determine whether a PGA lyase activity in human feces could be attributed to any of these species, we first determined the properties of PGA lyases from the other four Bacteroides species. PGA lyases from all the Bacteroides species were soluble, cell associated, and inducible by PGA. All had similar pH optima (8.4 to 8.8) and similar molecular weights (50,000). All activities were enhanced by calcium. The PGA lyases from the five species differed with respect to isoelectric point: B. thetaiotaomicron (pI 7.5), B. vulgatus (pI 7.7), B. ovatus (pI 5.8, 7.2), B. fragilis subsp. a (pI 6.1), and Bacteroides sp. strain 3452A (pI 7.7). The PGA lyase activity in human feces resembled those of the Bacteroides PGA lyases in that it had a pH optimum of 8.4 to 8.8 and was enhanced by calcium. However, it differed from the Bacteroides PGA lyases both with respect to isoelectric point (pI 4.2 to 4.4) and molecular weight (100,000). On the basis of these findings, it appears that the PGA lyase activity in human feces is not produced by any of the Bacteroides species surveyed in this survey. Moreover, there was no detectable PGA lyase activity in feces that had the same properties as the Bacteroides enzymes.(ABSTRACT TRUNCATED AT 250 WORDS)     

Evidence that polygalacturonic acid may not be a major source of carbon and energy for some colonic Bacteroides species.

Five Bacteroides species that are found in the human colon can utilize polygalacturonic acid (PGA) when they are grown in laboratory media: Bacteroides thetaiotaomicron, Bacteroides vulgatus, Bacteroides ovatus, Bacteroides fragilis subsp. a, and Bacteroides sp. strain 3452A (an unnamed DNA-DNA homology group). PGA-degrading enzymes from B. thetaiotaomicron have been isolated and characterized previously. To determine whether a PGA lyase activity in human feces could be attributed to any of these species, we first determined the properties of PGA lyases from the other four Bacteroides species. PGA lyases from all the Bacteroides species were soluble, cell associated, and inducible by PGA. All had similar pH optima (8.4 to 8.8) and similar molecular weights (50,000). All activities were enhanced by calcium. The PGA lyases from the five species differed with respect to isoelectric point: B. thetaiotaomicron (pI 7.5), B. vulgatus (pI 7.7), B. ovatus (pI 5.8, 7.2), B. fragilis subsp. a (pI 6.1), and Bacteroides sp. strain 3452A (pI 7.7). The PGA lyase activity in human feces resembled those of the Bacteroides PGA lyases in that it had a pH optimum of 8.4 to 8.8 and was enhanced by calcium. However, it differed from the Bacteroides PGA lyases both with respect to isoelectric point (pI 4.2 to 4.4) and molecular weight (100,000). On the basis of these findings, it appears that the PGA lyase activity in human feces is not produced by any of the Bacteroides species surveyed in this survey. Moreover, there was no detectable PGA lyase activity in feces that had the same properties as the Bacteroides enzymes.(ABSTRACT TRUNCATED AT 250 WORDS)     

Analysis of outer membrane proteins which are associated with growth of Bacteroides thetaiotaomicron on chondroitin sulfate.

By analyzing outer membrane proteins of Bacteroides thetaiotaomicron on two-dimensional polyacrylamide gels, we were able to identify 10 protein spots that were associated with growth on chondroitin sulfate but not with growth on glucuronic acid or other monosaccharides. These proteins were distinct from the outer membrane polypeptides that were associated with growth on two other negatively charged polysaccharides, polygalacturonic acid and heparin. Of the 10 protein spots that were associated with growth on chondroitin sulfate, 4 could be detected on immunoblots with antiserum that had been raised against outer membranes from bacteria grown on chondroitin sulfate and then cross-adsorbed with membranes from bacteria grown on glucose. Synthesis of these four proteins appeared to be regulated coordinately with synthesis of the two enzymes that degrade chondroitin sulfate, chondroitin lyase I and II. Although one of the four proteins (Mr 110,000) was similar in molecular weight to the chondroitin lyases, the cross-adsorbed antiserum which detected this outer membrane protein did not cross-react with either of these two enzymes.     

Identification and characterization of a Bacteroides gene, csuF, which encodes an outer membrane protein that is essential for growth on chondroitin sulfate.

Bacteroides thetaiotaomicron can utilize a variety of polysaccharides, including charged mucopolysaccharides such as chondroitin sulfate (CS) and hyaluronic acid (HA). Since the enzymes (chondroitin lyases I and II) that catalyze the first step in breakdown of CS and HA are located in the periplasm, we had proposed that the first step in utilization of these polysaccharides was binding to one or more outer membrane proteins followed by translocation into the periplasm, but no such outer membrane proteins had been shown to play a role in CS or HA utilization. Previously we have isolated a transposon-generated mutant, CS4, which was unable to grow on CS or HA but retained the ability to grow on disaccharide components of CS. This phenotype suggested that the mutation in CS4 either blocked the transport of the mucopolysaccharides into the periplasmic space or blocked the depolymerization of the mucopolysaccharides into disaccharides. We have mapped the CS4 mutation to a single gene, csuF, which is capable of encoding a protein of 1,065 amino acids and contains a consensus signal sequence. Although CsuF had a predicted molecular weight and pI similar to those of chondroitin lyases, it did not show significant sequence similarity to the Bacteroides chondroitin lyase II, a Proteus chondroitin ABC lyase, or two hyaluronidases from Clostridium perfringens and Streptococcus pyogenes, nor was any CS-degrading enzyme activity associated with csuF expression in Bacteroides species or Escherichia coli. The deduced amino acid sequence of CsuF exhibited features suggestive of an outer membrane protein. We obtained antibodies to CsuF and demonstrated that the protein is located in the outer membrane. This is the first evidence that a nonenzymatic outer membrane protein is essential for utilization of CS and HA.     

Use of randomly cloned DNA fragments for identification of Bacteroides thetaiotaomicron.

Randomly cloned fragments of DNA from Bacteroides thetaiotaomicron were used as hybridization probes for differentiation of B. thetaiotaomicron from closely related Bacteroides species. HindIII digestion fragments of DNA from B. thetaiotaomicron (type strain) were inserted into plasmid pBR322 and labeled with [alpha-32P]dCTP by nick translation. These labeled plasmids were screened for hybridization to HindIII digests of chromosomal DNA from type strains of the following human colonic Bacteroides species: B. thetaiotaomicron, Bacteroides ovatus, reference strain 3452-A (formerly part of B. distasonis), Bacteroides uniformis, Bacteroides fragilis, Bacteroides vulgatus, Bacteroides distasonis, Bacteroides eggerthii, and reference strain B5-21 (formerly B. fragilis subsp. a). Two of the five cloned fragments hybridized only to DNA from B. thetaiotaomicron. Each of these two fragments hybridized to the same DNA restriction fragment in five strains of B. thetaiotaomicron other than the strain from which the DNA was cloned. One of the cloned fragments (pBT2) was further tested for specificity by determining its ability to hybridize to DNA from 65 additional strains of colonic Bacteroides.     

Differences in the architecture of cytoplasmic and intracytoplasmic membranes of three chemotrophically and phototrophically grown species of the Rhodospirillaceae

Chondroitin sulfate lyase (EC 4.2.2.4) was present constitutively at low levels (0.06 to 0.08 U/mg of protein) in cells of Bacteroides thetaiotaomicron which were growing on glucose or other monosaccharides. When these uninduced bacteria were incubated with chondroitin sulfate A (5 mg/ml), chondroitin sulfate lyase specific activity increased more than 10-fold within 90 min. Synthesis of ribonucleic acid and of protein was required for induction, and induction was sensitive to oxygen. The disaccharides which resulted from chondroitinase action did not act as inducers, nor did tetrasaccharides or hexasaccharides obtained by digestion of chondroitin sulfate with bovine testicular hyaluronidase. None of these substances was taken up by uninduced cells; they may not have been able to penetrate the outer membrane. The smallest oligomer capable of acting as an inducer was the outer membrane. The smallest oligomer capable of acting as an inducer was the octassacharide. Oligomers larger than the octassacharide induced chondroitin lyase activity nearly as well as intact chondroitin sulfate.     

Evidence that the Bacteroides thetaiotaomicron chondroitin lyase II gene is adjacent to the chondro-4-sulfatase gene and may be part of the same operon.

The chondroitin lyase II gene from Bacteroides thetaiotaomicron has previously been cloned in Escherichia coli on a 7.8-kilobase (kb) fragment (pA818). In E. coli, the chondroitin lyase II gene appeared to be expressed from a promoter that was about 0.5 kb from the beginning of the gene. However, when a subcloned 5-kb fragment from pA818 which contained the chondroitin lyase II gene and the promoter from which the gene is expressed in E. coli was introduced into B. thetaiotaomicron on a multicopy plasmid (pEG800), the chondroitin lyase specific activity of B. thetaiotaomicron was not altered. Further evidence that the promoter that is recognized in E. coli may not be the promoter from which the chondroitin lyase II gene is transcribed in B. thetaiotaomicron was obtained by making an insertion in the B. thetaiotaomicron chromosome at a point which is 1 kb upstream from the chondroitin lyase II gene. This insertion stopped synthesis of the chondroitin lyase II gene product, as would be predicted if the gene was part of an operon and was transcribed in B. thetaiotaomicron from a promoter that was at least 1 kb upstream from the chondroitin lyase II gene. A region of pA818 which was adjacent to the chondroitin lyase II gene and which included the region used to make the insertional mutation was found to code for chondro-4-sulfatase, an enzyme that breaks down one of the products of the chondroitin lyase reaction. The upstream insertion mutant of B. thetaiotaomicron which stopped synthesis of chondroitin lyase II had no detectable chondro-4-sulfatase activity. This mutant was still able to grow on chondroitin sulfate, although the rate of growth was slower than that of the wild type.     

Enumeration of polysaccharide-degrading Bacteroides species in human feces by using species-specific DNA probes.

DNA probes that are specific for each of five predominant species of human colonic Bacteroides (B. thetaiotaomicron, B. uniformis, B. distasonis, "Bacteroides group 3452-A", and B. ovatus) were used to detect and enumerate these species in fecal samples from two adult volunteers. These five species are capable of fermenting many of the complex polysaccharides that are thought to be sources of carbon and energy for bacteria in the colon. Estimates for the concentrations of some of these species in feces have not been previously available because of the difficulties in differentiating colonic Bacteroides spp. by conventional biochemical tests. Our results indicate that all the species except B. ovatus were present in high numbers (greater than 10(9)/g [dry weight]) in the feces of both volunteers. However, the concentrations of the more versatile polysaccharide-degrading species within this group of organisms (7.6 X 10(9) to 12.0 X 10(9)/g [dry weight] for B. thetaiotaomicron; 2.9 X 10(9) to 6.3 X 10(9)/g [dry weight] for "Bacteroides group 3452-A") did not differ significantly from the concentrations of less versatile polysaccharide-degrading species (1.2 X 10(10) to 2.0 X 10(10)/g [dry weight] for B. uniformis; 5.8 X 10(9) to 8.4 X 10(9)/g [dry weight] for B. distasonis). B. ovatus was not detectable by our method. Since our lower limit of detection is approximately 1 X 10(9) to 2 X 10(9)/g (dry weight) of feces, this is consistent with earlier estimates that indicated that the concentration of B. ovatus in feces is near or below this value.(ABSTRACT TRUNCATED AT 250 WORDS)     

Enumeration of polysaccharide-degrading Bacteroides species in human feces by using species-specific DNA probes

DNA probes that are specific for each of five predominant species of human colonic Bacteroides (B. thetaiotaomicron, B. uniformis, B. distasonis, "Bacteroides group 3452-A", and B. ovatus) were used to detect and enumerate these species in fecal samples from two adult volunteers. These five species are capable of fermenting many of the complex polysaccharides that are thought to be sources of carbon and energy for bacteria in the colon. Estimates for the concentrations of some of these species in feces have not been previously available because of the difficulties in differentiating colonic Bacteroides spp. by conventional biochemical tests. Our results indicate that all the species except B. ovatus were present in high numbers (greater than 10(9)/g [dry weight]) in the feces of both volunteers. However, the concentrations of the more versatile polysaccharide-degrading species within this group of organisms (7.6 X 10(9) to 12.0 X 10(9)/g [dry weight] for B. thetaiotaomicron; 2.9 X 10(9) to 6.3 X 10(9)/g [dry weight] for "Bacteroides group 3452-A") did not differ significantly from the concentrations of less versatile polysaccharide-degrading species (1.2 X 10(10) to 2.0 X 10(10)/g [dry weight] for B. uniformis; 5.8 X 10(9) to 8.4 X 10(9)/g [dry weight] for B. distasonis). B. ovatus was not detectable by our method. Since our lower limit of detection is approximately 1 X 10(9) to 2 X 10(9)/g (dry weight) of feces, this is consistent with earlier estimates that indicated that the concentration of B. ovatus in feces is near or below this value.(ABSTRACT TRUNCATED AT 250 WORDS)     

Purification and characterization of novel chondroitin ABC and AC lyases from Bacteroides stercoris HJ-15, a human intestinal anaerobic bacterium.

Two novel chondroitinases, chondroitin ABC lyase (EC 4.2.2.4) and chondroitin AC lyase (EC 4.2.2.5), have been purified from Bacteroides stercoris HJ-15, which was isolated from human intestinal bacteria with glycosaminoglycan degrading enzymes. Chondroitin ABC lyase was purified to apparent homogeneity by a combination of QAE-cellulose, CM-Sephadex C-50, hydroxyapatite and Sephacryl S-300 column chromatography with a final specific activity of 45.7 micromol.min-1.mg-1. Chondroitin AC lyase was purified to apparent homogeneity by a combination of QAE-cellulose, CM-Sephadex C-50, hydroxyapatite and phosphocellulose column chromatography with a final specific activity of 57.03 micromol.min-1.mg-1. Chondroitin ABC lyase is a single subunit of 116 kDa by SDS/PAGE and gel filtration. Chondroitin AC lyase is composed of two identical subunits of 84 kDa by SDS/PAGE and gel filtration. Chondroitin ABC and AC lyases showed optimal activity at pH 7.0 and 40 degrees C, and 5.7-6.0 and 45-50 degrees C, respectively. Both chondroitin lyases were potently inhibited by Cu2+, Zn2+, and p-chloromercuriphenyl sulfonic acid. The purified Bacteroidal chondroitin ABC lyase acted to the greatest extent on chondroitin sulfate A (chondroitin 4-sulfate), to a lesser extent on chondroitin sulfate B (dermatan sulfate) and C (chondroitin 6-sulfate). The purified chondroitin AC lyase acted to the greatest extent on chondroitin sulfate A, and to a lesser extent on chondroitin C and hyaluronic acid. They did not act on heparin and heparan sulfate. These findings suggest that the biochemical properties of these purified chondroitin lyases are different from those of the previously purified chondroitin lyases.     

Significance of ribosomal ribonucleic acid synthesis for control of the G1 period in the cell cycle of the heterobasidiomycetous yeast Rhodosporidium toruloides.

Bacteroides thetaiotaomicron, a gram-negative anaerobe found in human colons, could utilize chondroitin sulfate, a tissue mucopolysaccharide, as its sole source of carbohydrate. The enzymes responsible for the breakdown of chondroitin sulfate by B. thetaiotaomicron were similar to those produced by Proteus vulgaris and Flavobacterium heparinum and included a lyase (EC 4.2.2.4), which degraded chondroitin sulfate into sulfated disaccharides, sulfatases (EC 3.1.6.4), which removed the sulfate residues, and a glucuronidase, which broke the unsulfated disaccharides into monosaccharide components. Chondroitin sulfate lyase, the first enzyme in the breakdown sequence, was not extracellular. It appeared to be located in the periplasmic space since lyase activity was released by treatment with ethylenediaminetetraacetate and lysozyme. Moreover, sodium polyanethole sulfonate, a high-molecular-weight inhibitor of chondroitin lyase, did not inhibit breakdown of chondroitin sulfate by intact bacteria. The sulfatase and glucuronidase appeared to be intracellular. None of these enzymes was strongly bound to membranes, and none of the steps in the breakdown of chondroitin sulfate was sensitive to oxygen.     

A deletion in the chromosome of Bacteroides thetaiotaomicron that abolishes production of chondroitinase II does not affect survival of the organism in gastrointestinal tracts of exgermfree mice

Bacteroides thetaiotaomicron, an obligate anaerobe normally found in high concentrations in the human colon, is one of the few colon bacteria that can ferment host mucopolysaccharides such as chondroitin sulfate. Previously, we found that a directed insertional mutation in the gene that codes for the chondroitinase II gene of B. thetaiotaomicron did not affect growth on chondroitin sulfate despite the fact that chondroitinase II accounts for 70% of the total cellular chondroitinase activity. Thus, the chondroitinase II gene did not seem to contribute significantly to growth on chondroitin sulfate when the bacteria were grown in laboratory medium. To determine whether this enzyme is important for bacteria growing in the intestinal tract, we tested the ability of a strain that does not produce chondroitinase II to colonize the intestinal tracts of germfree mice and to compete with wild-type B. thetaiotaomicron. The mutant used in these experiments carried a 0.5-kilobase deletion in the chondroitinase II gene and was constructed so that, unlike the original insertion mutant, it contained no exogenous DNA. The deletion mutant colonized the intestinal tracts of germfree mice at the same levels as the wild type. When a mixture of the deletion mutant and wild type was used to colonize germfree mice, the percent wild type, measured by colony hybridization with the deleted 0.5-kilobase fragment as the hybridization probe, did not rise to 100% even after periods as long as 9 weeks. In most experiments, the percent wild type did not rise significantly above the percent in the original mixture.(ABSTRACT TRUNCATED AT 250 WORDS)     

Importance of mucopolysaccharides as substrates for Bacteroides thetaiotaomicron growing in intestinal tracts of exgermfree mice.

We used two approaches to determine whether the mucopolysaccharide chondroitin sulfate is an important source of carbon and energy for Bacteroides thetaiotaomicron in the intestinal tracts of germfree mice. First, we tested the ability of three mutants that grew poorly or not at all on chondroitin sulfate to colonize the intestinal tract of a germfree mouse and to compete with wild-type B. thetaiotaomicron in this model system. One mutant (CG10) was rapidly outcompeted by the wild type. However, since this mutant was unable to grow on chondroitin sulfate because it could not grow on N-acetyl-galactosamine, one of its monosaccharide components, this mutant might also be unable to utilize glycoprotein mucins. Two mutants (46-1 and 46-4) were isolated that grew poorly on chondroitin sulfate but normally on both component sugars. One of them was outcompeted by the wild type, but the percent wild type increased more slowly than with CG10. In one experiment, the percent wild type never reached 100%. The other (46-4) was not outcompeted by the wild type. These results indicate that, although chondroitin sulfate may be a carbon source in the animal, it is not of major importance. Our second approach was to determine by immunoblot analysis whether a 28-kilodalton outer membrane protein that is produced by B. thetaiotaomicron only when it is grown on chondroitin sulfate or hyaluronic acid was being produced at induced level by B. thetaiotaomicron growing in the ceca of exgermfree mice. There was no evidence for induction of this protein in vivo. Thus, the immunoblot results are consistent with results of the mutant competition experiments.     

A deletion in the chromosome of Bacteroides thetaiotaomicron that abolishes production of chondroitinase II does not affect survival of the organism in gastrointestinal tracts of exgermfree mice.

Bacteroides thetaiotaomicron, an obligate anaerobe normally found in high concentrations in the human colon, is one of the few colon bacteria that can ferment host mucopolysaccharides such as chondroitin sulfate. Previously, we found that a directed insertional mutation in the gene that codes for the chondroitinase II gene of B. thetaiotaomicron did not affect growth on chondroitin sulfate despite the fact that chondroitinase II accounts for 70% of the total cellular chondroitinase activity. Thus, the chondroitinase II gene did not seem to contribute significantly to growth on chondroitin sulfate when the bacteria were grown in laboratory medium. To determine whether this enzyme is important for bacteria growing in the intestinal tract, we tested the ability of a strain that does not produce chondroitinase II to colonize the intestinal tracts of germfree mice and to compete with wild-type B. thetaiotaomicron. The mutant used in these experiments carried a 0.5-kilobase deletion in the chondroitinase II gene and was constructed so that, unlike the original insertion mutant, it contained no exogenous DNA. The deletion mutant colonized the intestinal tracts of germfree mice at the same levels as the wild type. When a mixture of the deletion mutant and wild type was used to colonize germfree mice, the percent wild type, measured by colony hybridization with the deleted 0.5-kilobase fragment as the hybridization probe, did not rise to 100% even after periods as long as 9 weeks. In most experiments, the percent wild type did not rise significantly above the percent in the original mixture.(ABSTRACT TRUNCATED AT 250 WORDS)     

Digestion of proteoglycan by Bacteroides thetaiotaomicron

It has been shown previously that Bacteroides thetaiotaomicron, a human colonic anaerobe, can utilize the tissue mucopolysaccharide chondroitin sulfate as a source of carbon and energy and that the enzymes involved in this utilization are all cell associated (A. A. Salyers and M. B. O'Brien, J. Bacteriol. 143:772-780, 1980). Since chondroitin sulfate does not generally occur in isolated form in tissue, but rather is bound covalently in proteoglycan, we investigated the extent to which chondroitin sulfate which is bound in such a sterically hindered complex can be utilized by intact bacteria. Intact cells of B. thetaiotaomicron were able to digest chondroitin sulfate in proteoglycan, although at a slightly slower rate than free chondroitin sulfate. Prior digestion of proteoglycan with trypsin to produce small fragments of protein with several chondroitin sulfate chains attached did not increase the rate at which the bound chondroitin sulfate was digested. Accordingly, the slower rate of digestion was probably due to attachment of chondroitin sulfate chains to the protein backbone rather than to steric hindrance by other components of the proteoglycan. When proteoglycan which had been incubated with intact bacteria was treated with sodium borohydride to release the undigested fragments of chondroitin sulfate from the protein backbone, the size and composition of the fragments indicated that intact bacteria were able to digest all but three monosaccharides of the chondroitin sulfate chains. Thus, despite steric hindrance due to attachment of the chondroitin sulfate chains to the protein backbone, digestion of bound chondroitin sulfate by intact bacteria was nearly complete.     

Importance of mucopolysaccharides as substrates for Bacteroides thetaiotaomicron growing in intestinal tracts of exgermfree mice

We used two approaches to determine whether the mucopolysaccharide chondroitin sulfate is an important source of carbon and energy for Bacteroides thetaiotaomicron in the intestinal tracts of germfree mice. First, we tested the ability of three mutants that grew poorly or not at all on chondroitin sulfate to colonize the intestinal tract of a germfree mouse and to compete with wild-type B. thetaiotaomicron in this model system. One mutant (CG10) was rapidly outcompeted by the wild type. However, since this mutant was unable to grow on chondroitin sulfate because it could not grow on N-acetyl-galactosamine, one of its monosaccharide components, this mutant might also be unable to utilize glycoprotein mucins. Two mutants (46-1 and 46-4) were isolated that grew poorly on chondroitin sulfate but normally on both component sugars. One of them was outcompeted by the wild type, but the percent wild type increased more slowly than with CG10. In one experiment, the percent wild type never reached 100%. The other (46-4) was not outcompeted by the wild type. These results indicate that, although chondroitin sulfate may be a carbon source in the animal, it is not of major importance. Our second approach was to determine by immunoblot analysis whether a 28-kilodalton outer membrane protein that is produced by B. thetaiotaomicron only when it is grown on chondroitin sulfate or hyaluronic acid was being produced at induced level by B. thetaiotaomicron growing in the ceca of exgermfree mice. There was no evidence for induction of this protein in vivo. Thus, the immunoblot results are consistent with results of the mutant competition experiments.     

Persistence and functional impact of a microbial inoculant on native microbial community structure,nutrient digestion and fermentation characteristics in a rumen model

Small sub-unit (SSU) rRNA-targeted oligonucleotide probes were used to monitor the persistence of a genetically engineered bacterium inoculated in model rumens. Eight dual flow continuous culture fermenters were operated with either standard artificial saliva buffer or buffer with chondroitin sulfate (0.5 g/l) added. After 168 h of operation, fermenters were inoculated with Bacteroides thetaiotaomicron BTX (BTX), at approximately 1% of total bacteria. B. thetaiotaomicron was quantified using a species-specific probe and shown to persist in fermenters 144 h after inoculation (relative abundance 0.48% and 1.42% of total SSU rRNA with standard and chondroitin sulfate buffers, respectively). No B. thetaiotaomicron SSU rRNA was detected in fermenter samples prior to inoculation with strain BTX. Relative abundances of Bacteria, Eucarya and Archaea were not affected by either inoculation or buffer type. Fiber digestion, in particular the hemicellulose fraction, increased after strain BTX addition. Chondroitin sulfate addition to the buffer increased bacterial nitrogen flow in fermenters, but did not alter fiber digestion. Neither inoculum nor buffer type altered total short chain fatty acid (VFA) concentrations but proportions of individual VFA differed. In model rumens, B. thetaiotaomicron BTX increased fiber digestion when added to mixed ruminal microbes, independent of chondroitin sulfate addition; but further study is needed to determine effects on other fiber-digesting bacteria.