1388
Chem. Res. Toxicol. 2007, 20, 1388–1392
Metabonomic Evaluation of Schaedler Altered Microflora Rats Cynthia M. Rohde,*,† Dale F. Wells,† Lora C. Robosky,† Matthew L. Manning,† Charles B. Clifford,‡ Michael D. Reily,† and Donald G. Robertson† Metabonomics EValuation Group, Pfizer Global Research and DeVelopment, 2800 Plymouth Road, Ann Arbor, Michigan 48105, and Charles RiVer Laboratories, 251 BallardVale Street, Wilmington, Massachusetts 01887 ReceiVed May 25, 2007
Previously, we identified two distinct metabonomic phenotypes in Sprague–Dawley rats sourced from two different rooms (colonies) in the Charles River, Raleigh facility [Robosky, L. C., Wells, D. F., Egnash, L. A., Manning, M. L., Reily, M. D., and Robertson, D. G. (2005) Metabonomic identification of two distinct phenotypes in Sprague–Dawley (Crl:CD(SD)) rats. Toxicol. Sci. 87, 277–284]. On the basis of literature reports and cohabitation experiments, we concluded that the differing phenotypes were due to different gut flora populations. One hypothesis explaining this phenomenon was attributed to the practice of initiating new colonies with rats derived from foundation colonies that had limited gut floral populations, the Charles River altered Schaedler flora (CRASF) rats. We hypothesized that the lack of differentiation of CRASF rats to the full complement of microflora was responsible for the altered phenotype characterized by increased urinary chlorogenic acid metabolites and decreased hippurate (CA rats) as opposed to the prevalent phenotype characterized by the inverse ratio of these metabolites (HIP rats). Upon receipt, it was confirmed that the CRASF rats exhibited a metabonomic profile similar to CA rats that remained constant while animals were housed individually in a dedicated animal room. However, exposure of CRASF rats to HIP rats, or their bedding, led to a relatively rapid but variable rate of reversion to the historic HIP type metabolic profile. On the basis of the results, we conclude that CRASF rats have a unique metabolic profile due to their limited gut flora constitution. If rigorous isolation procedures are not employed, the CRASF phenotype will eventually differentiate into the more typical HIP phenotype with a time course that may be quite variable. Given the marked metabolic heterogeneity between the phenotypes, this work highlights the importance of monitoring rat metabolic profiles. Introduction Humans, as well as other mammals, contain large microbial populations on the skin, in the mouth, in the genital area, and in the gastrointestinal (GI)1 tract. The majority of these bacteria, approximately 1013, reside in the GI tract (2). The full complement of this microflora population is acquired shortly after birth but is adjusted throughout life depending on such factors as antibiotic administration, infection, stress, hygiene, and diet ( 3–6). These gut flora have coevolved with their hosts and play a vital role in a number of mammalian processes including the immune response, intestinal development and health, and metabolism of nutrients and drugs/toxins (2, 7–11). Because of their involvement in these various activities, differences in these microflora populations may be a source of interanimal variation in safety assessment studies (11, 12). Recently, two distinct phenotypes of Sprague–Dawley rats, originating from the Charles River Raleigh Facility, were discovered by examination of urinary metabolite profiles (1). One phenotype, designated HIP, was characterized by high hippurate and low chlorogenic acid metabolite levels, while the second phenotype, designated CA, exhibited the inverse ratio of these urinary metabolites. The HIP rats represent the conventional metabonomic phenotype for Sprague–Dawley rats
typically determined in our laboratory. Literature reports (13, 14) and a cohabitation experiment (1) suggested that the difference in the two phenotypes was due to differing gut flora populations between CA and HIP rats. This postulated variation in gut flora in the two colonies could be explained by a difference in the rate of conventionalization of gut microorganisms. When a new production colony of Sprague–Dawley rats is initiated at Charles River Laboratories, a specific process is followed. The new colony is seeded using foundation colony rats, which have a specific complement of gut flora. In the case of the Sprague–Dawley rats, the altered Schaedler flora population of bacteria is used and the foundation colony rats are referred to as Charles River altered Schaedler flora (CRASF) rats (15, 16). Over time, the foundation rats in the new production colony acquire a full complement of gut bacterial species. As the previously described CA rats came from a production colony initiated in December 2002, as opposed to the HIP rats whose colony was formed in March 2000 or earlier, we postulated that the CA rats had not yet fully converted to the conventional gut flora population. Therefore, the following study was performed to determine if the CRASF rats were the source of the CA phenotype.
Materials and Methods * To whom correspondence should be addressed. Tel: 518-846-6415. E-mail:
[email protected]. † Pfizer Global Research and Development. ‡ Charles River Laboratories. 1 Abbreviations: CRASF, Charles River altered Schaedler flora; GI, gastrointestinal; PCA, principal component analysis.
Animals. CRASF rats, approximately 6–8 weeks old and 125–275 g, were graciously supplied by Charles River Laboratories (Wilmington, MA). Routine Animal Husbandry. Animals were housed in an AAALAC-accredited facility, and all in-life studies were reviewed
10.1021/tx700184u CCC: $37.00 2007 American Chemical Society Published on Web 09/28/2007
Communications
Chem. Res. Toxicol., Vol. 20, No. 10, 2007 1389
and approved by IACUC. Animals were housed in temperature (70–78 °C) and humidity (30–70% relative humidity) controlled rooms with a 12 h light cycle throughout the study. Animals were fed Certified Rodent Chow 5002 (Laboratory Diet, Purina Mills, Richmond, IN) and provided with drinking water ad libitum throughout the study. When urine was not being collected, animals were housed in individual cages. Study Timeline. Immediately upon arrival, CRASF rats, six males and six females, were placed in an isolated animal room, where they remained for 10 days. Urine was collected for 24 h periods for the first 3 days after receipt of the animals and once at the end of the 10 days. After approximately 10 days in the isolated animal room, animals were moved to a room containing a stock colony of HIP phenotype Sprague–Dawley rats. The CRASF rats were placed in individual hanging stainless steel cages in a cage rack directly below stock colony rats, and routine animal husbandry practices were employed taking no special precautions to prevent cross-contamination of the two phenotypes. CRASF rats were handled (for daily clinical procedures) in numerical order (1–12) after handling of the HIP phenotype rats. Urine was subsequently collected approximately once a week for the subsequent 39 days after the transfer to the stock colony. Beginning on day 50, rats still exhibiting the original Schaedler urinary metabolic profile were exposed to bedding from standard stock colony animals for 4 days. Urine samples were collected daily during this time. Urine Collection. When urine was collected, animals were placed in individual metabolism cages (Harvard Apparatus, Holliston, MA). Except for the first urine collection, which was begun immediately upon receipt, animals were acclimated to metabolism cages for 24 h prior to urine collection. Urine was collected into chilled (0 °C) tubes containing a total of 1 mL of 1% sodium azide. After the collection was complete, samples were spun down (3000 rpm, 4 °C) for 10 min to remove particulate matter. Samples were then either immediately analyzed or stored at approximately -70 °C until NMR analysis. One-Dimensional 1H NMR Spectroscopy. Urine samples were prepared for NMR analysis as previously described (1). Briefly, urine was diluted 2:1 with normalization buffer [0.2 M sodium phosphate buffer, pH 7.4 (80:20 H2O:D2O), 1 mM TSP, 3 mM sodium azide] in a 96 well format. Plates were mixed and centrifuged before analyzing. 1H NMR spectra were acquired using a Varian Inova NMR spectrophotometer operating at 600.36 MHz for 1H and equipped with a 1H-[15N, 13C] flow cryoprobe and a Varian automated transport accessory. The probe was washed twice with buffer [(2:1) water:0.2 M sodium phosphate buffer, pH 7.4] between each sample. One-dimensional 1H NMR spectra were acquired at 27 °C using a one-dimensional NOESY pulse sequence including water presaturation and a mixing time of 100 ms. A total of 64 scans were collected with 64K data points, an acquisition time of 2.73 s, an interpulse delay of 1 s, and a sweep width of 12 kHz. NMR Data Analysis. Spectra were processed using 1 Hz line broadening and integrated over 0.04 ppm regions (or bins) from 0.2 to 10 ppm using either Bruker’s Amix software or our internal Matlab processing package. The water-containing region from 4.5 to 6 ppm was excluded in further analyses. Principal components analysis (PCA) was completed using Pirouette software, version 3.02 (Infometrix, Inc., Bothell, WA), for calculation and Spotfire DecisionSite 8.1 software (Spotfire, Inc., Somerville, MA) for visualization. The PCA map was compiled using the full NMR data set for the CRASF (isolated) male rats from days 3 and 5 out of the six CRASF (exposed to stock colony) male rats from day 53. Also included in the PCA analysis were NMR spectral data taken from CA (n ) 10) and HIP (n ) 8) male rats in May 2004.
Results Urine NMR spectra (Figure 1) and PCA analysis (Figure 2) from the initial 3 days in the animal facility showed that the CRASF phenotype was similar, but not identical, to the CA phenotype. The CRASF rats had similar hippurate levels to CA
Figure 1. Representative urinary NMR spectra of HIP, CRASF (exposed to stock colony), CRASF (isolated), and CA male rats. The region shown is from 8.2 to 6.6 ppm. Individual CRASF (isolated) and CRASF (exposed to stock colony) urinary spectra were chosen from rats on days 3 and 53, respectively. Individual urinary NMR spectra of HIP and CA male rats are from data obtained in May 2004. Only representative samples from male rats are shown, but female rats exhibited a similar pattern. Resonances for hippurate and phenylacetylglycine are indicated on the spectra. Resonances for 3-(3hydroxyphenyl)propionic acid and 3-(4-hydroxyphenyl)propionic acid are marked on the spectra as 1 and 2, respectively.
Figure 2. PCA map of HIP, CRASF (exposed to stock colony), CRASF (isolated), and CA rats. Each phenotype is indicated as follows: HIP (triangles), CRASF (exposed to stock colony; stars), CRASF (isolated; squares), and CA (circles). For the CRASF (exposed to stock colony) rats, only the five male rats that were fully converted on day 53 were used in the analysis. Only male rats were used in the PCA analysis, but female rats showed a similar pattern.
rats but slightly lower levels of chlorogenic acid metabolites as compared to CA rats (Table 1). There were also slight differences in other metabolites, such as phenylacetylglycine (Table 1). This CRASF phenotype was stable over the course of 10 days, while the rats were maintained in an isolated animal room. After 10 days in isolation, CRASF rats were transferred to a room containing stock colony rats, which exhibit the HIP phenotype by both NMR spectra (Figure 1) and PCA analysis (Figure 2). Seven days after the transfer, all of the CRASF rats still maintained their original urinary metabolite profile. However, 13 days after the transfer to the stock colony, the first rat to be handled after stock colony rats during routine husbandry
1390 Chem. Res. Toxicol., Vol. 20, No. 10, 2007
Communications
Table 1. Relative Levels of Various Urinary Metabolites from HIP, CA, CRASF (Isolated), and CRASF (Exposed to Stock Colony) Ratsa
hippurate N-methylnicotinamide 3-(3-hydroxyphenyl)propionic acid 3-(4-hydroxyphenyl)propionic acid phenylacetylglycine
CRASF CRASF (exposed (isolated) to stock colony)
HIP
CA
1.0 1.0 1.0
0.2 1.5 5.1
0.1 1.8 4.0
1.5 1.7 0.5
1.0
10.7
4.8
0.8
1.0
1.5
2.4
1.6
a
Levels of all metabolites were acquired and normalized as described in the Materials and Methods. The group average for each metabolite for each phenotype was then divided by the group average for each metabolite for the HIP phenotype. Group averages for metabolite levels for HIP and CA rats were obtained from data obtained in May 2004. Group averages for metabolite levels for CRASF (isolated) rats were obtained from data acquired on day 11 and for CRASF (exposed to stock colony rats) on day 53.
Figure 3. Change in hippurate levels in individual CRASF rats over the course of the experiment. Hippurate levels were quantitated from NMR spectra using Metabonomi software, as described in the Materials and Methods, for each urine collection. (A) Individual male rat hippurate levels. (B) Individual female rat hippurate levels.
(rat 1) was found to have a phenotype similar to HIP rats (Figures 1 and 2). The urinary hippurate levels had increased, while the chlorogenic acid metabolites had decreased. After 21 days in the stock colony, another male (rat 2), in the cage adjacent to the first converted animal, switched to a phenotype similar to HIP rats. By the 28th day in the stock colony, three rats had spontaneously converted to a phenotype similar to HIP rats, two males (rats 1 and 2) and one female (rat 7). The timeline of this conversion can easily be observed by tracking the hippurate levels of each rat over time (Figure 3). After 35 days in the stock colony, one male rat (rat 3) had partially
changed to the new phenotype. So, rats were being conventionalized, but only gradually. To see if the conversion could be accelerated by exposure to HIP rat fecal material, two male (rats 4 and 5) and three female rats (rats 8–10), which still maintained the CRASF phenotype, were exposed to bedding from HIP rats and monitored daily for phenotype changes beginning on day 50. With this exposure, one male (rat 4) and all three females (rats 8–10) exposed to the HIP bedding quickly converted to the urinary metabolite profile similar to HIP rats (Figure 2). The second male rat (rat 5) exposed to HIP bedding did not convert but did show an increase in hippurate levels by the fourth day of exposure (Figure 3). During this exposure period, the two female rats (rats 11 and 12) not given HIP bedding maintained the CRASF phenotype, while the remaining CRASF male rat (rat 6) spontaneously converted to a new phenotype (Figure 3).
Discussion This communication provides evidence that the differences observed in the urinary metabolite profiles of HIP and CA rats are due to variations in their respective microflora populations caused by differential acquisition of normal gut flora after stock colony initiation from CRASF rats. The first piece of data to support this hypothesis is that the main differences between the two urinary metabolite profiles, CA and HIP, are alterations in the hippurate and chlorogenic acid metabolite levels. Both hippurate and chlorogenic acid derivatives are produced via metabolism of polyphenols, compounds commonly found in fruits, vegetables, and cereals (Figure 4) (17, 18). Metabolism of these compounds is mainly accomplished by gut flora, although certain steps, such as glycine conjugation, are performed by the mammalian host system (Figure 4) (18, 19). It remains unclear which bacteria are responsible for each specific reaction that occurs during the breakdown of polyphenolic compounds in vivo, but it has previously been shown that multiple species of microflora are required to produce the full complement of metabolites found in the urine of conventionalized rats (20, 21). Therefore, changes in the microflora population may alter the types of reactions that can be performed, modify the metabolite profile, and potentially limit or enhance the bioavailability of various ingested compounds. Further support for the hypothesis comes from other studies in the literature where changes were observed in the urinary levels of hippurate and chlorogenic acid metabolites in cases where differences in gut flora populations occurred (13, 14, 22, 23). In these studies, the intestinal bacteria population was altered by using germ-free rats vs conventionalized, adjusting the diet of the animals, or treating the animals with different compounds. Additionally, the urinary metabolite profile of isolated CRASF rats, which are used to seed new colonies of Sprague–Dawley rats at Charles River Laboratories, most closely resembles that of the CA rats as opposed to the HIP rats. Finally, CRASF rats exposed to either a HIP stock colony environment or directly to HIP rat bedding relatively rapidly acquire a urinary metabolic profile consistent with HIP rats. All of these results suggest that the CA rats have yet to acquire the full complement of microflora demonstrated by the conventional HIP rats. It is unclear why the metabolic profile of exposed CRASF rats differs from HIP rats. It is possible that the exposed CRASF rat profile is a transient state due to repopulating gut flora that may revert to the typical HIP phenotype over longer periods of time (>50 days) or perhaps the change reflects the fact that CRASF rats never fully become the HIP phenotype over their individual life spans but eventually conventionalize over
Communications
Chem. Res. Toxicol., Vol. 20, No. 10, 2007 1391
Figure 4. Metabolism of chlorogenic acid. Chlorogenic acid is the most abundant polyphenol in the human diet. After ingestion, chlorogenic acid can be metabolized via intestinal bacteria to either quinic acid or caffeic acid, eventually leading to the production of hippurate or phenylpropionic acid derivatives. Reactions that are performed by gut flora are highlighted by a bold arrow, while reactions that are done by the mammalian host system are indicated by a thin arrow.
repeated generations. It also remains unclear why the CA rats documented in our first report (1) did not develop the conventional complement of gut flora within the 3 year period from colony initiation. In several cases where germ-free or low hippurate-excreting rats were exposed to a standard stock colony environment, rats achieved a conventional urinary metabolite profile after only 21 days (19, 22, 23). This gut flora acclimation occurs even more rapidly when rats are housed with high hippurate-excreting rats ( 1, 22). However, if the CA rats were not exposed to a HIP environment, with the exacting husbandry techniques practiced in today’s animal facilities, it is possible that only slight changes in the microflora environment of the CA rats may have occurred, leading to a urinary profile very similar to the original CRASF foundation rats. These data, taken together, emphasize the importance of microflora in the determination of the overall metabolic phenotype of the host animal. The types of bacteria living in the GI tract in combination with host metabolism ultimately determine the metabolic fate of an ingested compound, whether it is a normal component of the diet or an exogenous material such as a drug or toxin. This means that differences in gut flora populations can influence how an individual animal responds to a particular stimulus, a potential source for interanimal variation. Therefore, screening of populations for phenotypic metabolic differences should be considered as these differences may influence the results of research investigations, especially safety assessment studies. This source of variation also needs to be taken into account when considering the concept of personalized health care, as has been previously noted (11, 24–26). In conclusion, the urinary metabolite profile of CRASF rats maintained in an isolated environment is similar to that of CA rats. When CRASF rats are housed in a room with HIP rats, their metabonomic phenotype gradually changes to one that closely resembles that of the HIP rats. Exposure to HIP bedding greatly accelerates this conversion process. The phenotypic
differences between the CA and the HIP rats appear to be due to alterations in the gut flora populations, most likely due to a slow rate of microflora conventionalization in the CA colony. Although it would be interesting to further explore the differences between the HIP and the CA rats, especially with respect to how their differences might affect drug metabolism, the CA rats are no longer available. The CA rats, as recently reported, have now reverted to the HIP phenotype (27).
References (1) Robosky, L. C., Wells, D. F., Egnash, L. A., Manning, M. L., Reily, M. D., and Robertson, D. G. (2005) Metabonomic identification of two distinct phenotypes in Sprague–Dawley (Crl:CD(SD)) rats. Toxicol. Sci. 87, 277–284. (2) Sears, C. L. (2005) A dynamic partnership: Celebrating our gut flora. Anaerobe 11, 247–251. (3) Alverdy, J., Zaborina, O., and Licheng, W. (2005) The impact of stress and nutrition on bacterial-host interactions at the intestinal epithelial surface. Curr. Opin. Clin. Nutr. Metab. Care 8, 205–209. (4) Collins, M. D., and Gibson, G. R. (1999) Probiotics, prebiotics, and synbiotics: Approaches for modulating the microbial ecology of the gut. Am. J. Clin. Nutr. 69, 1052S–1057S. (5) Hawrelak, J. A., and Myers, S. P. (2004) The causes of intestinal dysbiosis: A review. Altern. Med. ReV. 9, 180–197. (6) Mai, V. (2004) Dietary modification of the intestinal microbiota. Nutr. ReV. 62, 235–242. (7) Hooper, L. V., Midtvedt, T., and Gordan, J. I. (2002) How hostmicrobial interactions shape the nutrient environment of the mammalian intestine. Annu. ReV. Nutr. 22, 283–307. (8) Backhed, F., Ding, H., Wang, T., Hooper, L. V., Koh, G. Y., Nagy, A., Semenkovich, C. F., and Gordan, J. I. (2004) The gut microbiota as an environmental factor that regulates fat storage. Proc. Natl. Acad. Sci. U.S.A. 101, 15748–15723. (9) Kelly, D., Conway, S., and Aminov, R. (2005) Commensal gut bacteria: Mechanisms of immune modulation. Trends Immunol. 26, 326–333. (10) Mikov, M. (1994) The metabolism of drugs by gut flora. Eur. J. Drug Metab. Pharmacokinet. 19, 201–207. (11) Nicholson, J. K., Holmes, E., and Wilson, I. D. (2005) Gut microorganisms, mammalian metabolism and personalized health care. Nat. ReV. Miocrobiol. 3, 431–438.
1392 Chem. Res. Toxicol., Vol. 20, No. 10, 2007 (12) Li, H., Ni, Y., Su, M., Qiu, Y., Zhou, M., Qiu, M., Zhao, A., Zhao, L., and Jia, W. (2007) Pharmacometabonomic phenotyping reveals different responses to xenobiotic intervention in rats. J. Proteome Res. 6, 1364–1370. (13) Gavaghan, C. L., Nicholson, J. K., Connor, S. C., Wilson, I. D., Wright, B., and Holmes, E. (2001) Directly coupled high-performance liquid chromatography and nuclear magnetic resonance spectroscopic with chemometric studies on metabolic variation in Sprague–Dawley rats. Anal. Biochem. 291, 245–252. (14) Phipps, A. N., Stewart, J., Wright, B., and Wilson, I. D. (1998) Effect of diet on the urinary excretion of hippuric acid and other dietaryderived aromatics in rat. A complex interaction between diet, gut microflora and substrate specificity. Xenobiotica 28, 527–537. (15) Dewhirst, F. E., Chien, C. C., Paster, B. J., Ericson, R. L., Orcutt, R. P., Schauer, D. B., and Fox, J. G. (1999) Phylogeny of the defined murine microbiota: altered Schaedler flora. Appl. EnViron. Microbiol. 65, 3287–3292. (16) Sarma-Rupavtarm, R. B., Ge, Z., Schauer, D. B., Fox, J. G., and Polz, M. F. (2004) Spatial distribution and stability of the eight microbial species of the altered Schaedler flora in the mouse gastrointestinal tract. Appl. EnViron. Microbiol. 70, 2791–2800. (17) Gonthier, M. P., Verny, M. A., Besson, C., Remesy, C., and Scalbert, A. (2003) Chlorogenic acid bioavailability largely depends on its metabolism by the gut microflora in rats. J. Nutr. 133, 1853–1859. (18) Manach, C., Scalbert, A., Morand, C., Remesy, C., and Jimenez, L. (2004) Polyphenols: Food sources and bioavailability. Am. J. Clin. Nutr. 79, 727–747. (19) Goodwin, B. L., Ruthven, C. R., and Sandler, M. (1994) Gut flora and the origin of some urinary aromatic phenolic compounds. Biochem. Pharmacol. 47, 2294–2297.
Communications (20) Peppercorn, M. A., and Goldman, P. (1971) Caffeic acid metabolism by bacteria of the human gastrointestinal tract. J. Bacteriol. 108, 996– 1000. (21) Peppercorn, M. A., and Goldman, P. (1972) Caffeic acid metabolism by gnotobiotic rats and their intestinal bacteria. Proc. Natl. Acad. Sci. U.S.A. 69, 1413–1415. (22) Williams, R. E., Eyton-Jones, H. W., Farnworth, M. J., Gallagher, R., and Provan, W. M. (2002) Effect of intestinal microflora on the urinary metabolic profile of rats: A 1H-nuclear magnetic resonance spectroscopy study. Xenobiotica 32, 783–794. (23) Nicholls, A. W., Mortshire-Smith, R. J., and Nicholson, J. K. (2003) NMR spectroscopic-based metabonomic studies of urinary metabolite variation in acclimatizing germ-free rats. Chem. Res. Toxicol. 16, 1395–1404. (24) Clayton, T. A., Lindon, J. C., Cloarec, O., Antti, H., Charuel, C., Hanton, G., Provost, J. P., Le Net, J. L., Baker, D., Walley, R. J., Everett, J. R., and Nicholson, J. K. (2006) Pharmaco-metabonomic phenotyping and personalized drug treatment. Nature 440, 1073–1077. (25) Goodacre, R. (2007) Metabolomics of a superorganism. J. Nutr. 137, 259S–266S. (26) Nebert, D. W., and Vesell, E. S. (2006) Can personalized drug therapy be achieved? A closer look at pharmaco-metabonomics. Trends Pharmacol. Sci. 27, 580–586. (27) Robosky, L. C., Wells, D. F., Egnash, L. A., Manning, M. L., Reily, M. D., and Robertson, D. G. (2006) Communication regarding metabonomic identification of two distinct phenotypes in Sprague– Dawley (Crl:CD(SD)) rats. Toxicol. Sci. 91, 309.
TX700184U
