Skip to main content
The American Journal of Pathology logoLink to The American Journal of Pathology
. 1998 Dec;153(6):1717–1722. doi: 10.1016/s0002-9440(10)65686-6

Cryptosporidium parvum Initiates Inflammatory Bowel Disease in Germfree T Cell Receptor-α-Deficient Mice

Randy E Sacco 1, Joseph S Haynes 1, James A Harp 1, W Ray Waters 1, Michael J Wannemuehler 1
PMCID: PMC1876657  PMID: 9846962

Abstract

Flora-bearing mice with targeted disruption of T cell receptor (TCR)-α or -β genes spontaneously develop intestinal inflammation with features similar to ulcerative colitis in humans. TCR-α-deficient mice maintained germfree or colonized with a limited number of intestinal bacteria failed to develop inflammatory bowel disease (IBD)-like lesions. Evidently, inflammation in these mice does not develop spontaneously or result from a generalized antigenic stimulation, but rather requires induction by a heretofore unidentified specific stimulus. We describe the development of IBD-like lesions in germfree TCR-α-deficient mice monoassociated with the protozoan Cryptosporidium parvum. Lesions were seen in distal ileum, cecum, and colon and were most severe in the cecum. A prominent leukocytic infiltrate within the lamina propria was a common characteristic of the lesions observed in the C. parvum-infected germfree TCR-α-deficient mice. The leukocytic infiltrate was composed of aggregates of B220+ cells, the majority of which expressed surface IgD (ie, conventional B lymphocytes). It has been proposed that antigenic stimulation by a microorganism(s) is needed to initiate intestinal inflammation in TCR-α-deficient mice. Our results indicate that a single microbial species, C. parvum, is capable of triggering the development of IBD-like lesions in germfree TCR-α-deficient mice.


Inflammatory bowel disease (IBD) of humans is primarily characterized by Crohn’s disease (CD) and ulcerative colitis (UC). As is the case with other chronic diseases of unknown etiology, the etiology of IBD is believed to be multifactorial, with genetic susceptibility, environmental factors, dysregulated acquired and innate immunity, and nonspecific inflammation likely contributing to the complex clinical manifestations. Although the pathogenesis of IBD is not completely understood, an immunological and/or infectious etiology has been postulated to contribute to the onset of disease. In this regard, it is of particular interest that several immunodeficient mouse strains created via targeted deletion of cytokine, major histocompatibility complex (MHC) class II, or T cell receptor (TCR)-α or -β genes spontaneously develop IBD-like lesions as they age. 1 The chronic colitis observed to spontaneously develop in TCR-α- or TCR-β-deficient mice resembles UC, beginning after approximately 3 to 4 months of age in a majority of the knockout mice. These mice gradually develop chronic diarrhea, leading to progressive wasting that is often associated with anorectal prolapse. We have recently reported the earlier and more predictable development of IBD-like lesions in flora-bearing TCR-α-deficient mice infected with Cryptosporidium parvum. 2 In this model, TCR-α-deficient mice infected at 1 week of age with C. parvum develop IBD-like lesions within 3 weeks after infection.

C. parvum is a ubiquitous protozoan parasite that is a common cause of gastroenteritis and diarrhea worldwide. C. parvum is now recognized as the most important waterborne contaminant in the United States. Although the disease is generally self-limiting in immunocompetent individuals, it can be severe and life-threatening in immunocompromised individuals, such as AIDS patients. As a case in point, a cryptosporidiosis outbreak occurred in Milwaukee during the spring of 1993 that was traced to contamination of a municipal water-treatment facility and resulted in devastating consequences in HIV-infected individuals and those with severely impaired cellular immunity. 3 Of particular interest, among the patients treated, were 12 individuals with stable UC or CD who suffered abrupt clinical relapse during the waterborne outbreak. Eventually, these patients recovered to their clinical baseline after standard IBD therapy and administration of intravenous fluid. Thus, it is evident that C. parvum can induce an acute relapse in patients with pre-existing IBD and an earlier induction of intestinal inflammation in knockout mice predisposed to developing IBD-like lesions.

Previously, it was shown that TCR-α-deficient mice maintained under germfree conditions do not develop IBD-like lesions. 4 Moreover, in that study, no intestinal inflammation was observed in gnotobiotic TCR-α-deficient mice colonized with Lactobacillus plantarum, Streptococcus faecalis, S. faecium, and/or Escherichia coli. Thus, as has been proposed, inflammation is likely initiated by a specific environmental organism or group of organisms that have yet to be identified. Given our previous observation that flora-bearing TCR-α-deficient mice rapidly develop IBD-like lesions after C. parvum infection, the primary objective of the present experiments was to determine whether germfree TCR-α-deficient mice monoassociated with C. parvum would develop intestinal inflammation. Furthermore, as germfree immunocompetent and immunodeficient mice are more susceptible to C. parvum infection than are flora-bearing mice, a secondary objective was to examine whether germfree TCR-α-deficient mice are more susceptible to C. parvum than are flora-bearing TCR-α-deficient mice.

Materials and Methods

Mice

Adult flora-bearing TCR-α-deficient mice 1 were obtained from Jackson Laboratories (Bar Harbor, ME). A breeding colony of these mice was established and maintained at Iowa State University. Germfree TCR-α-deficient mice were derived at the University of Wisconsin Gnotobiotic Laboratory, Madison, WI, and a breeding colony was established and maintained at Iowa State University. Mice maintained under germfree conditions were monitored for contamination by culturing fecal samples on blood agar plates under aerobic or anaerobic conditions. Furthermore, swabs were periodically taken from random locations inside the isolators for both aerobic and anaerobic bacteriological culture.

Cryptosporidium parvum

Purified oocysts were isolated from feces collected from calves experimentally inoculated with C. parvum oocysts using a previously described method. 5 Before use, oocysts were incubated for 30 minutes with 2.5% peracetic acid to kill any contaminant bacteria and then washed three times with PBS. Adult germfree and flora-bearing mice were challenged with 10 6 C. parvum oocysts via gastric intubation. Additional flora-bearing and germfree mice were maintained as noninfected controls. Fecal pellets were collected weekly from individual mice. Fecal impression smears were prepared, stained with carbol fuschin, and examined for the presence of C. parvum oocysts.

Histology

At 6 to 12 weeks after challenge, mice were euthanized, and the entire intestinal tract of each mouse was infused with 10% neutral buffered formalin (NBF). In addition, the pancreas, liver, lungs, and gall bladder from individual mice were fixed in NBF. After fixation, samples of these tissues were processed and stained with hematoxylin and eosin (H&E) for routine histological evaluation. In addition, tissue sections were stained with Warthin-Starry silver stain and examined for the presence of bacterial populations.

Antibodies

The following monoclonal antibodies (MAbs) were used in these experiments: GL3, hamster IgG anti-mouse γδ TCR; H57-597, hamster IgG anti-mouse TCR-β; 217-170, mouse IgG1 anti-mouse IgDb (Igh-5b); RA3-6B2, rat IgG2a anti-mouse B220; and 53-7.313, rat IgG2a anti-mouse CD5 (Ly-1). RA3-6B2 was purified from supernatants of hybridoma culture and biotinylated using standard procedures. The other MAbs were purchased as biotinylated conjugates from PharMingen (San Diego, CA).

Immunohistochemistry

To assess the cellular composition of the mucosal lesions, immunohistochemical staining was performed on 8-μm frozen cecal sections. Briefly, tissue was snap-frozen in OCT medium (Miles, Elkhart, IN) and stored at −70°C. Sections were then cut and placed on poly-l-lysine-coated slides, fixed with 95% ethanol, and stored frozen at −70°C for later use. Frozen sections were washed in 0.05 mol/L Tris, pH 7.6, and blocked with 5% serum. Sections were incubated overnight at 4°C with biotinylated primary antibodies. Sections were incubated with streptavidin-peroxidase and developed with diaminobenzidine substrate/chromogen (Biomeda, Foster City, CA).

Results

C. parvum Oocyst Shedding

The number of oocysts in feces of germfree and flora-bearing TCR-α-deficient mice were examined each week as a measure of susceptibility to C. parvum. Fecal samples collected from germfree TCR-α-deficient mice infected with C. parvum were typically soft and contained mucus. As shown in Figure 1 , greater than 80% of germfree TCR-α-deficient mice were shedding C. parvum oocysts at 1 week postinfection (PI). In contrast, none of the flora-bearing TCR-α-deficient mice were shedding oocysts at 1 week PI. In fact, the number of oocysts in flora-bearing C. parvum-infected TCR-α-deficient mice was relatively low until 6 to 7 weeks PI, when greater than 50% of the mice were found to be shedding oocysts. In contrast, by 4 weeks PI, 100% of germfree mice infected with C. parvum were shedding oocysts, and they continued to do so throughout the 12-week course of the experiment. Moreover, the number of oocysts shed per mouse was greater for germfree TCR-α-deficient mice than flora-bearing mice (data not shown). Thus, it is readily apparent that in comparison with flora-bearing TCR-α-deficient mice, germfree mice were more heavily infected with C. parvum as assessed by oocyst shedding.

Figure 1.

Figure 1.

Percentage of TCR-α-deficient mice shedding C. parvum oocysts after oral inoculation. Fecal pellets were collected weekly on individual mice, smeared onto glass slides, stained with carbol fucshin, and examined for the presence of C. parvum oocysts.

Lesions in Germfree TCR-α-Deficient Mice Infected with C. parvum

Previously, lesions were observed in liver, gall bladder, and bile ducts of immunodeficient NIH III-nu/nu or SCID mice after infection with C. parvum. 6 In addition, involvement of respiratory epithelium, bile ducts, gall bladder, and pancreas due to C. parvum in AIDS patients has been reported. 7-13 Therefore, we examined sections of pancreas, gall bladder, and lungs from germfree TCR-α-deficient mice infected with C. parvum. No parasites or significant lesions were seen in these tissues at 6 or 12 weeks PI. However, aggregates of lymphocytes with or without plasma cells or macrophages were observed in the liver parenchyma of mice at 12 weeks PI.

Given our previous observation that flora-bearing TCR-α-deficient mice develop IBD-like lesions after C. parvum infection, we examined the ability of C. parvum to induce intestinal inflammation in germfree TCR-α mutant mice. There was no inflammation present in any of the intestinal sections of control adult germfree TCR-α-deficient mice. In addition, no bacterial populations were observed in intestinal tissue sections of germfree TCR-α-deficient mice based on Warthin-Starry silver staining. As shown in Figure 2A , the cecal histoarchitecture was normal with few mononuclear cells, which is similar to that previously described for the colonic architecture of germfree TCR-α mutant mice. 4 At 6 weeks PI, no lesions were observed in the duodenum of germfree TCR-α-deficient mice infected with C. parvum. However, in the distal jejunum and ileum, there was mild crypt epithelial cell hyperplasia. Mild to moderate diffuse infiltration of lymphocytes and plasma cells was observed in the jejunal and ileal lamina propria of these mice. As shown in Figure 2B , cecal mucosa of germfree TCR-α-deficient mice infected for 6 weeks with C. parvum were moderately thickened due to proliferative glands and hyperplasia of surface epithelium. There was a diffuse infiltrate of lymphocytes, macrophages, and plasma cells in the cecal lamina propria. Colonic mucosa was thickened due to proliferation of glands and mild infiltration of lymphocytes in the lamina propria.

Figure 2.

Figure 2.

Histopathological features of intestinal inflammation in C. parvum-infected germfree TCR-α mutant mice. A: Cecal tissue from a noninfected adult germfree TCR-α-deficient mouse showing few mononuclear cells in the lamina propria. H&E; magnification, ×64. B: Cecal tissue from an adult germfree TCR-α mutant mouse infected for 6 weeks with C. parvum showing a moderately thickened mucosa with increased numbers of leukocytes in the lamina propria. H&E; magnification, ×64. C: Cecal tissue from an adult germfree TCR-α mutant mouse infected for 12 weeks with C. parvum. There is a marked thickening of cecal mucosa and submucosa, which is characteristic of the lesions observed in these mice. Note the striking number of lymphocytes infiltrating the lamina propria. H&E; magnification, ×64. D: Numerous C. parvum can be seen attached to the apical surface of the cecal epithelial cells in this germfree TCR-α-deficient mouse infected for 12 weeks with the parasite. H&E; magnification, ×160.

At 12 weeks PI, there were no significant lesions in the duodenum or jejunum of germfree TCR-α-deficient mice infected with C. parvum. However, the ileal epithelium was hyperproliferative, and there were a moderate number of apoptotic cells. Ileal crypts of germfree TCR-α-deficient mice infected with C. parvum were hyperplastic. There was a mild to moderate diffuse infiltrate of lymphocytes and plasma cells in the lamina propria. In general, lesions in the small intestine were less severe than observed in the large intestine. In the cecum, mucosa, and submucosa were thickened (Figure 2C) . The cecal mucosal epithelium was focally eroded and contained a moderate number of necrotic cells. The epithelium was thickened and immature. There was a marked infiltrate of lymphocytes, plasma cells, macrophages, and polymorphonuclear leukocytes. Numerous C. parvum were observed along the apical surface of the epithelium (Figure 2D) . Submucosal lymphatics were dilated. There were a number of lymphoid aggregates in the lamina propria. Occasional crypt abscesses were characterized by gland lumina containing neutrophils and necrotic debris. The colonic mucosa was thickened due to proliferation of glands and surface epithelium. A diffuse, moderate infiltration of lymphocytes and plasma cells was seen in the colonic lamina propria.

Immunohistochemical Staining of Leukocytes in Cecal Tissue of Germfree TCR-α-Deficient Mice Infected with C. parvum

We used immunohistochemical staining of cecal tissues isolated at 12 weeks PI from germfree TCR-α-deficient mice to phenotype the leukocytes infiltrating the inflamed lamina propria of these mice. There were few TCR-β+ cells, γ/δ TCR+ cells, or NK1.1+ cells in the cecal lamina propria of germfree TCR-α-deficient mice infected with C. parvum (not shown). However, as shown in Figure 3, A and B , there were a striking number of aggregates of B220+ lymphocytes. The predominant B cell population in the lamina propria of mice is the Ly-1+ B cell (B-1) subset as opposed to conventional B cells (B-2). 14 As mature B-2 B cells are surface (s)IgDhigh, whereas sIgD expression on B-1 B cells ranges from negative to dull positive, 14 we examined the expression of sIgD to distinguish these B lymphocyte subpopulations. Interestingly, the aggregates of B220+ cells were predominantly composed of cells intensely stained by MAb to sIgD (Figure 3C) , indicating that these cells are likely representative of the conventional (ie, B-2) B cell subset.

Figure 3.

Figure 3.

Immunohistochemical staining of lymphocytes infiltrating the cecal lamina propria of germfree TCR-α-deficient mice 12 weeks after oral challenge with C. parvum. A: Numerous aggregates of B220+ cells in the lamina propria stained with monoclonal antibody RA3-6B2. Immunoperoxidase procedure; original magnification, ×25. B: Higher-power magnification of an aggregate of lymphocytes stained with RA3-6B2. Immunoperoxidase procedure; magnification, ×64. C: Serial section stained for sIgD with monoclonal antibody 217-170. Note intense staining of a majority of these cells. Immunoperoxidase procedure; magnification, ×64.

Discussion

The chronic colitis observed to develop spontaneously in TCR-α- or TCR-β-deficient mice, or TCR-β × δ-double-deficient mice resembles UC in humans. Lesions develop in a majority of these TCR-deficient mice after approximately 3 to 4 months of age. These mice gradually develop chronic diarrhea, leading to progressive wasting that is often associated with anorectal prolapse. 1 It has been hypothesized that the disease in these immunodeficient mice is mediated by chronic stimulation of lymphoid cells via an environmental agent that would normally be cleared in immunocompetent animals. In support of this latter hypothesis, it was recently shown that germfree TCR-α-deficient mice do not develop IBD-like lesions. 4 In fact, no intestinal inflammation was observed in TCR-α-deficient mice colonized with one or two microbial species from the normal intestinal flora. We have extended these observations to show that germfree TCR-α-deficient mice monoassociated with C. parvum develop intestinal inflammation. Our results provide a mechanism for the induction of colitis in TCR-α-deficient mice, in that antigenic stimulation by a single microbial species, such as that provided by C. parvum, is sufficient to trigger intestinal inflammation. Furthermore, the chronicity of this antigenic stimulation evidently leads to the progression and perpetuation of the inflammation in these mice.

Resistance to C. parvum infection has been shown to be dependent on T lymphocyte function, especially that of CD4+ T cells. 15-17 TCR-α-deficient mice are known to possess functional B cells, γ/δ T cells, and TCR-αβ+ T cells. 18-19 In fact, TCR-αβ+, and in some cases γ/δ T cells, in TCR-α mutant mice are known to express CD4. 20 In contrast to immunocompetent mice, germfree TCR-α-deficient mice, as well as flora-bearing TCR-α mutant mice, develop chronic C. parvum infection and intestinal inflammation. Our results expand on previous research by refining the critical role for CD4+ T cells in resistance to C. parvum infection to further suggest a requirement for CD4+ T cells possessing heterodimeric α and β chains of the TCR.

A striking feature of the intestinal lesions in germfree TCR-α-deficient mice monoassociated with C. parvum were the numerous aggregates of B220+, sIgD+ B cells in the intestine. Intense expression of sIgD would suggest that these B cells are conventional, or B-2, B cells, as opposed to B-1 (Ly-1) B cells. As TCR-α-deficient, TCR-β-deficient, TCR-β × δ-double-deficient mice, or MHC-class-II-deficient mice, but not RAG-1-deficient mice, develop colitis spontaneously, it has been speculated that the disease is caused by an inappropriate regulation of B cells due to the lack of class-II-MHC-restricted CD4+ αβ T cells. Moreover, it has been implied that IBD in TCR-deficient mice could result from an autoimmune attack against the mucosal epithelium, possibly due to cross-reactive antibodies and thus may be regarded as an autoimmune disease. Interestingly, B-2 B cells are known to produce pathogenic autoantibodies. 14 Thus, it remains to be determined whether the B-2 B lymphocytes in the lamina propria of germfree TCR-α-deficient mice monoassociated with C. parvum produce autoantibodies and whether these autoantibodies cross-react with host intestinal epithelium.

The previously described mucosal lesions in C. parvum-infected flora-bearing TCR-α-deficient mice 2 were more severe than were observed in C. parvum-infected germfree TCR-α-deficient mice in the present study. Moreover, the leukocytic infiltrate in flora-bearing mice contained a significant number of γ/δ T cells as well as B cells. 2 Thus, one could speculate that once the intestinal mucosa has been breached in flora-bearing TCR-α-deficient mice, other microbial organisms, including members of the normal intestinal flora, could potentiate the inflammation. The enhanced inflammatory response observed in flora-bearing TCR-α-deficient mice may result in the local proliferation or recruitment of γ/δ T cells or may be a consequence of their local proliferation or trafficking to the inflammatory site.

The results presented herein support the previous hypothesis that the inability of an immunocompromised host to regulate the immune response to chronic antigenic stimulation in the intestinal mucosa results in inflammation. Previous research had failed to identify a single microbial species capable of inducing intestinal inflammation in TCR-α-deficient mice. We have demonstrated that the ubiquitous protozoan parasite C. parvum is capable of inducing IBD in TCR-α-deficient mice that are predisposed to intestinal inflammation. In at least one case report, clinical relapse after exposure to C. parvum has been documented to occur in patients with previously stable IBD. 3 Thus, patients with certain types of transient or more permanent immunodeficiency or with a history of intestinal disease may be at risk for developing IBD-like lesions after exposure to C. parvum or other enteric pathogens.

Acknowledgments

We thank Andrea Dorn and Diane McDonald for their excellent technical assistance with the germfree animals.

Footnotes

Address reprint requests to Dr. Randy E. Sacco, USDA/ARS/National Animal Disease Center, P.O. Box 70, 2300 Dayton Avenue, Ames, IA 50010. E-mail: rsacco@nadc.ars.usda.gov.

Supported by grant DK-52552 from the National Institute for Diabetes and Digestive and Kidney Diseases (R.E. Sacco).

References

  • 1.Mombaerts P, Mizoguchi E, Grusby MJ, Glimcher LH, Bhan AK, Tonegawa S: Spontaneous development of inflammatory bowel disease in T cell receptor mutant mice. Cell 1983, 75:275-282 [DOI] [PubMed] [Google Scholar]
  • 2.Waters WR, Palmer MV, Ackermann MR, Harp JA: Accelerated inflammatory bowel disease of TCR-α-deficient mice persistently infected with Cryptosporidium parvum. J Parasitol 1997, 83:460-464 [PubMed] [Google Scholar]
  • 3.Mathey MW, Ross AB, Soergel KH: Cryptosporidiosis and inflammatory bowel disease. Digest Dis Sci 1997, 42:1580-1586 [DOI] [PubMed] [Google Scholar]
  • 4.Dianda L, Hanby AM, Wright NA, Sebesteny A, Hayday AC, Owen MJ: T cell receptor-αβ-deficient mice fail to develop colitis in the absence of a microbial environment. Am J Pathol 1997, 150:91-97 [PMC free article] [PubMed] [Google Scholar]
  • 5.Harp JA, Chen W, Harmsen AG: Resistance of severe combined immunodeficient mice to infection with Cryptosporidium parvum: the importance of intestinal microflora. Infect Immun 1992, 60:3509-3512 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Mead J, Arrowood MJ, Sidwell RW, Healey MC: Chronic Cryptosporidium parvum infections in congenitally immunodeficient SCID and nude mice. J Infect Dis 1991, 163:1297-1304 [DOI] [PubMed] [Google Scholar]
  • 7.Cello JP: Acquired immunodeficiency syndrome cholangiopathy: spectrum of disease. Am J Med 1989, 86:539-546 [DOI] [PubMed] [Google Scholar]
  • 8.Gross TL, Wheat J, Barlett M, O’Connor WO: AIDS and multiple system involvement with Cryptosporidium. Am J Gastroenterol 1986, 81:456-458 [PubMed] [Google Scholar]
  • 9.Hawkins SP, Thomas RP, Teasdale C: Acute pancreatitis: a new finding in Cryptosporidium enteritis. Br Med J 1987, 294:483-484 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Peterson C: Cryptosporidiosis in patients infected with the human immunodeficiency virus. Clin Infect Dis 1992, 15:903-909 [DOI] [PubMed] [Google Scholar]
  • 11.Teixdor HS, Godwin TA, Ramirez EA: Cryptosporidiosis of the biliary tract in AIDS. Radiology 1991, 180:51-56 [DOI] [PubMed] [Google Scholar]
  • 12.Holjlyng N, Jensen BN: Respiratory cryptosporidiosis in HIV-positive patients. Lancet 1988, 1:590-591 [DOI] [PubMed] [Google Scholar]
  • 13.Brady EM, Margolis ML, Korzeniowski OM: Pulmonary cryptosporidiosis in acquired immunodeficiency syndrome. J Am Med Assoc 1984, 252:89-90 [PubMed] [Google Scholar]
  • 14.Murakami M, Honjo T: Involvement of B-1 cells in mucosal immunity and autoimmunity. Immunol Today 1995, 16:534-539 [DOI] [PubMed] [Google Scholar]
  • 15.Ungar BLP, Kao TC, Burris CA, Finkelman FD: Cryptosporidium parvum infection in an adult mouse mode. Independent roles for IFN-γ and CD4+ T lymphocytes in protective immunity. J Immunol 1991, 147:1014-1022 [PubMed] [Google Scholar]
  • 16.Chen W, Harp JA, Harmsen AG: Requirements for CD4+ cells and gamma interferon in resolution of established Cryptosporidium parvum infection in mice. Infect Immun 1993, 61:3928-3932 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Aguirre SA, Mason PH, Perryman LE: Susceptibility of major histocompatibility complex (MHC) class I- and MHC class II-deficient mice to Cryptosporidium parvum infection. Infect Immun 1994, 62:697-699 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Mombaerts P, Clarke AR, Rudnicki MA, Iacomini J, Itohara S, Lafaille JJ, Wang L, Ichikawa Y, Jaenisch R, Hooper ML, Tonegawa S: Mutations in T cell antigen receptor genes α and β block thymocyte development at different stages. Nature 1992, 360:225-231 [DOI] [PubMed] [Google Scholar]
  • 19.Philpott KL, Viney JL, Kay G, Rastan S, Gardiner EM, Chae S, Hayday AC, Owen MJ: Lymphoid development in mice congenitally lacking T cell receptor αβ-expressing cells. Science 1992, 256:1448-1452 [DOI] [PubMed] [Google Scholar]
  • 20.Dianda L, Gulbranson-Judge A, Pao W, Hayday AC, MacLennan IAC, Owen MJ: Germinal center formation in mice lacking αβ T cells. Eur J Immunol 1996, 26:1603-1607 [DOI] [PubMed] [Google Scholar]

Articles from The American Journal of Pathology are provided here courtesy of American Society for Investigative Pathology

RESOURCES