Cryptosporidium species are apicomplexan parasites that can cause mild to severe diarrheal illnesses in animals and humans (Griffiths, 1998; Leav et al., 2003). In particular, immunosuppressed hosts infected with this protozoan, including patients with acquired immunodeficiency syndrome (AIDS), may undergo chronic intractable diarrhea and life-threatening illnesses (O'Donoghue, 1995; Chai et al., 1999; Guk et al., 2003, 2005). In immunocompetent hosts, the infection is generally subclinical and resolved within 2-4 weeks, and they become resistant thereafter (O'Donoghue, 1995). From studies of human and animal cryptosporidiosis, it is apparent that adoptive immunity, in the forms of CD4+ T cells and IFN-γ, is important for resistance to and clearance of the infection (Abrahamsen, 1998; Theodos, 1998; Leav et al., 2005).

The major effectors responsible for the self-resolution of the host from C. parvum infections have been reported to be lymphocytes in the small intestine, i.e., intraepithelial lymphocytes (IEL), lamina propria lymphocytes (LPL), and mesenteric lymph node lymphocytes (MLNL) (O'Donoghue, 1995; Wyatt et al., 1996; Abrahamsen et al., 1997; Pasquali et al., 1997). Our previous studies have reported that CD8+ IEL were involved in protection against a challenge infection with C. parvum, through enhanced cytotoxicity, whereas LPL played a more important role in protection against a primary infection with C. parvum, through production of IFN-γ and IgA (Chai et al., 1999; Guk et al., 2003). In mucosal immunity, detection of antigens from the gut is essential, and Peyer's patches (PP) are the major site of antigen sampling and presentation by microfold (M) cells, and transportation to T- and B- cell areas (Boher et al., 1994; Kelsall and Strober, 1996; Aleksandersen et al., 2002; Macdonald, 2003; Mariotte et al., 2004). Nevertheless, there have not been many reports on the role of PP lymphocytes (PPL) against Cryptosporidium infection in mice. Thus, the present study aimed to investigate the role of PPL in self-resolution of mice from primary and challenge infections with C. parvum.

Among the PPL, the proportions of CD4+, CD8+, and sIgA+ T cells increased significantly (P < 0.05) in primary-infected mice compared to the controls (Table 1). In challenge-infected mice, the proportions of these cells were only slightly higher than the controls (CD4+ PPL and sIgA+ PPL), or significantly (P < 0.05) lower than the controls (CD8+ PPL). The proportion of sIgG1+ PPL did not change significantly in any of the infected groups.

The proportions of the PPL expressing IFN-γ+ or IL-2+ in primary-infected mice increased greatly (P < 0.05) compared to the controls (Table 2). However, in challenge-infected mice, the proportion of the PPL expressing IFN-γ+ became lower than that in primary-infected mice, though it was maintained at a significantly higher (P < 0.05) level than the controls. Meanwhile, the proportion of the PPL secreting IL-2 decreased significantly (P < 0.05) compared to both primary-infected mice and the controls (Table 2).

The in vitro productions of IgG1 and IgA from the isolated PPL in both primary-infected and challenge-infected mice were significantly greater (P < 0.05) than those of the controls (Table 3). The levels of both antibody isotypes were higher in challenge-infected mice than in primary-infected mice, although the differences were statistically not significant.

Our previous studies showed that mice began to shed C. parvum oocysts on day 4 PI, peaking on day 10 PI, and oocysts were not observed after day 13 PI in primary-infected mice (Guk et al., 2003). In challenge-infected mice, oocyst excretion peaked on day 4 after challenge, i.e., day 10 PI, the number of oocysts being smaller than that in primary-infected mice, and oocysts were not observed after day 10 after challenge, i.e., day 16 PI (Guk et al., 2003). These results suggested that secondary immune responses of the hosts are more rapid than primary immune responses.

In the present study, experimental C. parvum infections in immunocompetent C57BL/6 mice resulted in alterations of lymphocyte subsets, cytokine expressing cell proportions, and antibody secreting patterns among the PPL. In primary-infected mice, increases in CD4+, CD8+, sIgA+ and IFN-γ+ PPL were observed at the beginning of oocyst clearing on day 10 PI, but the proportions of these cells, with the exception of IFN-γ+ PPL, were normalized or decreased to lower levels than those of the controls in challenge-infected mice on day 13 PI, i.e., day 7 post-challenge. This result suggests that the mechanisms implicated in the control of a challenge infection may be different from those of a primary infection with C. parvum.

According to our previous study which targeted on murine LPL and IEL, CD4+ LPL played an important role in protection against a primary infection with C. parvum, through induction of IFN-γ+ and IgA (Guk et al., 2003). Whereas, against a challenge infection, CD8+ IEL and CD8+/TCR γδ+ IEL involved more, increasing cytotoxicity, and aided by the sustained activity of the specific IgA (Guk et al., 2003). In the present study, which focused on PPL, CD4+ T helper cells were also important in protective mechanisms of primary-infected mice. Similar reports on the role of PPL in murine cryptosporidiosis have been available. For example, in neonatal BALB/c mice, Cryptosporidium infection induced proliferations of both CD4+ and CD8+ lymphocytes on days 4, 10, and 16 PI in ileal PP, and CD4+ cells increased on day 4 PI in jejunal PP (Boher et al., 1994). In suckling Naval Medical Research Institute (NMRI) mice infected with C. parvum, all CD4+, CD8+, TCR α/β+, and CD45R/ B220+ PPL were shown to be more than 10-fold increases than the uninfected controls on day 17 PI (Mariotte et al., 2004).

In challenge infections, 2 major effecter mechanisms seem to be involved. One is CD8+ IEL and CD8+/ TCR γδ+ IEL, as suggested by our previous study (Guk et al., 2003). However, this is contradictory to the results of the present study. The proportions of CD8+ T cells among the PPL in this study decreased remarkably in challenge-infected mice, down to lower levels than those of the uninfected controls. IEL are known to be derived from dividing precursor cells, originating from PP, circulating in the mesenteric lymph nodes and the thoracic duct lymph, and then selectively returning to the gut wall (Guy-Grand et al., 1978). This discrepancy is difficult to explain. A possible speculation is that CD8+ PPL, which is destined finally to migrate into the gut epithelium, might be depleted after a primary infection, and time is needed for them to restore before a challenge infection or late phages of infection. In this respect, it is worth referring that, in suckling NMRI mice infected with C. parvum, more than 10-fold increased numbers of T and B cells among the PPL decreased rapidly and remained at lower levels than the controls on days 19-23 PI, i.e., the termination of the infection (Mariotte et al., 2004).

The other effecter mechanism in challenge infections is the role of B cells, among the LPL and PPL, which produce IgA. In the present study, the number of IgA expressing cells among the PPL and the level of IgA secretion by the PPL in vitro elevated both in primary- and challenge-infected mice. It is well known that, in the mucosal immune system, most of secretory IgA responses are generated in PP, and the B cell progeny migrates to the lamina propria (Macdonald, 2003). In previous studies on C. parvum, IgA+ and IgG1+ B cells increased among the LPL on day 6 and day 10 PI, and antigen-specific IgA secretion from intestinal mucosa increased markedly in a challenge infection (Huang et al., 1996; Guk et al., 2003).

The IgA secretion is known to be generated by 2 ways. One is T helper cell-dependent and the other is T cell-independent way. With regard to the T-dependent way, it has been suggested that a challenge infection with C. parvum may drive the immune response towards a Th2 type, enhancing IL-4 production by LPL, and may stimulate sustained production of the specific IgA (Guk et al., 2003). Most B cells in PP are known to express IgA; however, other immunoglobulin classes are also present. To enhance IgA production, regulatory T cells in PP promote the immunoglobulin class switch to IgA, and support differentiation of IgA-committed B cells (Reynolds, 1987). As to the T-independent way, it is of note that about 25% of the mouse secretory IgA response is T-independent and largely of the B1 lineage (Macpherson et al., 2000).

In the present study, PPL expressing IFN-γ+ were shown to increase after primary and challenge infections with C. parvum. The extent of IFN-γ+ PPL increase was greater in the primary infection compared with the challenge infection. Reports have indicated importance of IFN-γ in systemic and mucosal immune responses against cryptosporidiosis (Leav et al., 2005), the mechanisms of innate immune responses as well as the source of IFN-γ remain poorly defined. The lack of innate mucosal IFN-γ expression was noted in humans and animals, and this may be an explanation for differences between species in susceptibility to Cryptosporidium (Leav et al., 2005). Further studies of IFN-γ mediated innate immune responses in cryptosporidiosis will be useful to develop strategies for improving resistance to the infection.

In conclusion, we suggest that PPL in the small intestines of mice play an important role in self-clearing of a primary C. parvum infection through proliferation of CD4+, CD8+, IL-2+, and IFN-γ+ T cells, and IgG1 and IgA-secreting B cells. In a challenge infection, the role of T cells seems to be reduced, whereas the role of B cells secreting IgA may be continuously important.