Dec 22, 1972 - Mark A. Peppercorn. The numbers in parentheses refer to designations of these strains used previously. (5). c Obtained from Thomas D. Moore.
Vol. 114, No. 2 Printed in U.S.A.
JOURNAL OF BAcrERIOLOGY, May 1973, p. 641-644 Copyright i 1973 American Society for Microbiology
Role of Intestinal Microflora in the Metabolism of Guanidinosuccinic Acid SHELDON MILSTIEN
AND
PETER GOLDMAN'
National Institute of Arthritis, Metabolism, and Digestive Diseases, National Institutes of Health, Bethesda, Maryland 20014 Received for publication 22 December 1972
Among a variety of bacteria isolated from the gastrointestinal tracts of rats and humans, only streptococci of group N are capable of degrading guanidinosuccinic acid added to their culture medium. The urinary excretion of guanidinosuccinic acid by germfree rats is greater than that of conventional rats. The excretion of this compound by gnotobiotic rats correlates with the capacity of their intestinal microflora to degrade guanidinosuccinic acid in culture. Thus, guanidinosuccinic acid excretion is low in rats infected exclusively with Streptococcus faecalis, and the excretion is not altered when germfree rats are infected with an organism unable to degrade guanidinosuccinic acid (Lactobacillus). These findings suggest that the intestinal microflora, particularly Streptococcus, play a role in the metabolism of guanidinosuccinic acid by the host.
Guanidinosuccinic acid, a constituent of normal urine, is elevated in the urine and serum of azotemic patients; the metabolic pathway leading to the formation of this compound has not been elucidated. The proposal that guanidinosuccinic acid may be derived from the enzymic reduction of dietary canavanine (9) now seems unlikely, because normal human volunteers on a totally defined diet, or during prolonged periods of food deprivation, continue to excrete guanidinosuccinic acid (S. Milstien and P. Goldman, manuscript in preparation). Another possibility, that guanidinosuccinic acid is synthesized by the intestinal microflora of the host, is unlikely because germfree rats, like conventional rats, excrete guanidinosuccinic acid in their urine. Because germfree rats excrete greater amounts of guanidinosuccinic acid than do normal rats, it seems possible that guanidinosuccinic acid might be degraded by the gastrointestinal flora. Turnover of other nitrogenous compounds by the intestinal flora has long been recognized (7, 10), and recently a soil pseudomonad has been isolated which has an enzyme capable of degrading guanidinosuccinic acid to L-aspartate and urea (4). The degradation of guanidinosuccinic acid by a number of intestinal bacteria isolated from the gastrointestinal tract of both rats and huI Present address. Beth Israel Hospital, Boston, Mass. 02215.
mans has now been investigated, and guanidinosuccinic acid excretion has been compared in germfree and conventional rats. In addition, guanidinosuccinic acid excretion has been investigated in rats specifically infected with intestinal bacteria whose ability to degrade guanidinosuccinic acid in vitro has been determined. These studies on the role of the intestinal microflora in determining the guanidinosuccinic acid excretion of the host are the subject of this report. MATERIALS AND METHODS The preparation of L-guanidinosuccinic acid from L-aspartate and 2-methyl-2-thiopseudourea and its purification have been described (4). 14CGuanidinosuccinic acid, prepared from "4C-aspartic acid and 2-methyl-2-thiopseudourea, was purified by chromatography on Dowex 1-acetate (8). The purity of the "4C-guanidinosuccinic acid prepared in this manner was confirmed by the complete release of "4C-aspartate after subjecting a sample of the material to the action of N-amidino-L-aspartic acid amidino hydrolase, an enzyme which is specific for guanidinosuccinic acid (4). Guanidinosuccinic acid in rat urine was quantified on the basis of the color developed in the Sakaguchi reaction on appropriate fractions of the eluate from a Dowex 1 column (8). A 24-h sample of rat urine was added to a column of Dowex 1-acetate and eluted with acetic acid in increasing concentrations (8). The guanidinosuccinic acid content in a sample of bacterial growth medium, which had been clarified by centrifugation, was also measured by this method. 641
642
MILSTIEN AND GOLDMAN
J. BACTERIOL.
In some samples of rat urine, unidentified Sakaguchi-reactive material was found in fractions of the column eluate which could have been confused with guanidinosuccinic acid. For this reason it was sometimes necessary to include "4C-guanidinosuccinic acid as a marker (Fig. 1). When, as in Fig. 1, clear 1000 8 concordance between Sakaguchi reactivity and the radioactive guanidinosuccinic acid marker was lack- E ing, fractions containing the "4C-guanidinosuccinic acid were pooled, lyophilized, and subjected to enzymatic hydrolysis. The amount of Sakaguchi-reacm tive material which disappeared in response to the z action of the specific enzyme was then used to to -4i quantify the amount of authentic guanidinosuccinic 500 acid in the sample. to -f All germfree and gnotobiotic rats were maintained 4 in the National Institutes of Health germfree facility under the direction of H. Bohner. To check for germfree status, fecal material and drinking water were cultured at least every 2 weeks for 14 days both at 37 C and at room temperature in Trypticase soy agar, brain-heart infusion broth, and thiol broth. 0 4 8 12 16 20Bacteria were isolated from the gnotobiotic rats by FRACTION NUMBER (5ml) Thomas D. Moore and shown to be identical to the FIG. 1. Chromatography on Dowex 1 (acetate) of strains used for the infections (5). Urine was collected rat urine. A trace amount of 14C-guanidinosuccinic from germfree, gnotobiotic or conventional rats of the acid was added. The 5-ml fractions of the eluate were Sprague-Dawley strain which weighed between 150 analyzed for 14C and color development in the Sakaand 200 g. The same diet was fed to all rats ad libitum guchi reaction. (rat diet L-356, General Biochemicals, Chagrin Falls, Ohio). No significant difference in average food consumption among the various groups of animals was 400 noted. Cultivation of the feces of the gnotobiotic rats infected with Streptococcus faecalis indicated that S. feciw, (9) E this was the only bacterium present. 0
z --
-v m
z
c
:
300
/
RESULTS
Degradation of guanidinosuccinic acid by intestinal bacteria. A variety of bacteria characteristic of those which can be isolated from the gastrointestinal tracts of humans and rodents were tested for their ability to degrade guanidinosuccinic acid when cultivated on artificial media. The time course of guanidinosuccinic acid degradation in artificial media by a number of strains isolated from the gastrointestinal tract of humans is shown in Fig. 2. A mixed culture of feces, which would be expected to contain group N streptococci, also caused the disappearance of guanidinosuccinic acid. Similarly, guanidinosuccinic acid disappeared when it was added to cultures of Streptococcus faecalis isolated from rodent feces or to a mixed bacterial culture derived from rodent feces (Fig. 3). A number of other organisms isolated from the gastrointestinal tract of humans were examined for their capacity to degrade guanidinosuccinic acid, and these were found to behave as the uninoculated controls of Fig. 2 and 3. All of the results are summarized in Table 1. Guanidinosuccinic acid excretion by germfree, gnotobiotic and conventional rats.
Impure culture
200 n
/iquefecinss(l
200
z
Un loo
//
6S. fecalis(lo)
o7 0 3 6
Control
INCUBATION TIME (Days) FIG. 2. Time course of the degradation of guanidinosuccinic acid by bacteria isolated from the human gastrointestinal tract. Bacteria were cultivated in thiol broth under conditions previously described (5), except that guanidinosuccinic acid was added to the medium at a concentration of 0.6 mg/ml. The guanidinosuccinic acid remaining at the times indicated was assayed in samples of the culture media, and the amount of guanidinosuccinic acid consumed was calculated. In all examples shown, maximal bacterial growth, as measured by light scattering, was reached within the first 12 to 24 h. The control consisted of uninoculated medium.
The demonstration of guanidinosuccinic acid degradation by pure cultures of intestinal bacteria, as well as by impure cultures derived from
VOL. 114, 1973
/ ~ Contrl
GUANIDINOSUCCINIC ACID METABOLISM
643
free rats were specifically infected with a strain of S. fecalis which can degrade guanidinosuccinic acid in vitro. For comparison, germfree E Impure culture rats were also infected with a strain of Lactobacillus which is incapable of degrading guanidinosuccinic acid in vitro. The infection of the germfree rats with S. fecalis caused a significant drop in the excretion I00 of guanidinosuccinic acid, whereas no effect on guanidinosuccinic acid excretion was found when infection was with the Lactobacillus sp. (Table 2). It is curious that after several weeks of exposure to the conventional animal room, the gnotobiotic (infected with S. fecalis) rats did not exhibit the low level of guanidinosuccinic acid excretion characteristic of the conventional or germfree rats who had entered the conventional animal room without prior S. The experimental procedure was as in Fig. 2, except fecalis infection. Perhaps this is a manifestation that guanidinosuccinic acid was added to the medium of the long-term effects of the initial establishment of the bacterial flora that has been noted at a concentration of 0.45 mg/ml. by Dubos et al. (1). Nevertheless, these experiTABLE 1. Ability of various intestinal bacteria to ments indicate that a bacterial flora capable of degrade guanidinosuccinic acida degrading guanidinosuccinic acid in vitro is also capable of lowering guanidinosuccinic acid exIdentity of bacteria degrade GSA cretion of the host. 200
w
-S.
Isolated from humansb Bacteroides fragilis (1) Clostridium perfringens (2, 8) Corynebacterium acnes (3) Streptococcus faecium (9) S. faecalis (10) Escherichia coli (11) S. faecalis var. liquefaciens (12) Mixed culture Isolated from rodents Lactobacillus sp. 1 Lactobacillus sp. 2 Bacteroides sp. S. faecalis Mixed culture
ecolis
+ + + +
TABLE 2. Excretion of guanidinosuccinic acid (GSA) by germfree, gnotobiotic and conventional rats
+ +
a Bacteria were cultivated in the presence of guanidinosuccinic acid (GSA) as described in Fig. 2 and 3. Assays of GSA were done every 2 to 3 days. Ability to degrade GSA was considered to be lacking if no degradation was detected in 1 week compared with a similar medium which had not been inoculated. b Strains were obtained from Charles Zierdt and Mark A. Peppercorn. The numbers in parentheses refer to designations of these strains used previously
(5). c
DISCUSSION Rats whose intestinal flora degrade guanidinosuccinic acid when cultivated in vitro have a lower rate of excretion of guanidinosuccinic acid than do germfree rats. Although other differences in the biology of germfree and conventional rats may explain these observations, it seems likely that guanidinosuccinic acid is at least partially degraded by the intestinal flora
Obtained from Thomas D. Moore.
feces, suggests that the elevated levels of guanidinosuccinic acid excretion in germfree rats might be due to the lack of bacterial degradation of guanidinosuccinic acid in these animals. This hypothesis was tested by noting the effect on guanidinosuccinic acid excretion when germ-
rata Type of of Type
Germfree Gnotobiotic (Lactobacillus sp. 1) Gnotobiotic (Streptococcus faecalis) Gnotobiotic (S. faecalis), conventionalized Germfree, conventionalized Conventional
GSA excreted' (~g/24 h)
130 (120 to 140)
130C 78 (71 to 83) 74 (59 to 83)d 34 29 (22 to 35)
aThe bacteria infecting the gnotobiotic rats are shown in parentheses. ' Unless otherwise indicated, values are the average of at least seven determinations on at least four rats. The range is shown in parentheses. c Average obtained from pooled samples of two rats. d Values obtained on two gnotobiotic (S. faecalis) rats after they had been in the conventional animal room for 2 weeks.
644
MILSTIEN AND GOLDMAN
J. BACTERIOL.
of the host. In this respect, guanidinosuccinic strated in bacterial cultures, but not necessarily acid would be similar to other end products of in gnotobiotic rats infested exclusively with mammalian nitrogen metabolism, such as urea, these bacteria (6). where degradation by the intestinal flora has been demonstrated (3). It has been suggested that guanidinosuccinic LITERATURE CITED acid degradation does not occur in rats because 1. Dubos, R. J., D. C. Savage, and R. W. Schaedlar. 1967. 94% of a 100-mg load of guanidinosuccinic acid The indigenous flora of the gastrointestinal tract. Dis. Colon Rectum 10:23-34. can be recovered in the urine within 5 days (9). Guanidinosuccinic acid degradation could have 2. Ducluzeau, R., P. Raiband, A. B. Dickinson, E. Sacquet, and G. Mocquot. 1966. Hydrolyse de l'uree in vitro et in been overlooked in this protocol, because our vivo, dans le caecum de rats gnotobiotiques, par results suggest that rats are capable of degraddifferentes souches bacteriennes isolees du tube digestif ing less than 0.1 mg of guanidinosuccinic acid de rats conventionnels. C. R. Acad. Sci. 262:944-947. 3. Levenson, S. M., L. V. Crowley, R. E. Horwitz, and 0. J. per day. Malm. 1959. The metabolism of carbon-labeled urea in The diminished excretion of guanidinosucthe germfree rat. J. Biol. Chem. 234:2061-2062. cinic acid observed when germfree rats are 4. Milstien, S., and P. Goldman. 1972. Metabolism of infected with a flora capable of degrading guaguanidinosuccinic acid. I. Characterization of a specific amidinohydrolase from Pseudomonas chlororaphis. J. nidinosuccinic acid is comparable to experiBiol. Chem. 247:6280-6283. ments demonstrating the disappearance of urea 5. Peppercorn, M. A., and P. Goldman. 1971. Caffeic acid from the colons of germfree rats selectively metabolism by bacteria of the human gastrointestinal infected with urease-positive strains of intestitract. J. Bact. 108:996-1000. nal bacteria (2). These experiments, and those 6. Peppercorn, M. A., and P. Goldman. 1972. Caffeic acid metabolism by gnotobiotic rats and their intestinal on guanidinosuccinic acid reported in this pabacteria. Proc. Nat. Acad. Sci. U.S.A. 69:1413-1415. per, suggest that some reactions occurring in the 7. Sorenson, .L. B. 1965. Role of the intestinal tract in elimination of uric acid. Arthritis Rheum. 8:695-701. host can be attributed to the intestinal microflora if these reactions can be demonstrated by 8. Stein, I. M., B. 0). Cohen, and R. S. Kornhauser. 1969. Guanidinosuccinic acid in renal failure, experimental the flora in vitro. However, a simple relationazotemia and inborn errors of the urea cycle. New Engl. ship does not always exist between bacterial J. Med. 280:926-930. reactions in culture and those catalyzed by 9. Takahara, K., S. Nakanishi, and S. Natelson. 1969. Cleavage of canavaninosuccinic acid by human liver to bacteria within the host. For example, certain form guanidinosuccinic acid, a substance found in the reactions of caffeic acid metabolism are carried urine of uremic patients. Clin. Chem. 15:397-418. out by the intestinal microflora and not by the 10. Walser, M., and L. Bodenlos. 1959. Urea metabolism in germfree host. These reactions can be demonman. J. Clin. Invest. 38:1617-1626.
