Induction of Gut Mucosal Immune Responses - Wiley Online Library

Scand. J. Immunol. 47, 401–407, 1998

FRONTLINES

Induction of Gut Mucosal Immune Responses: Importance of Genetic Background and Th1/Th2 Cross-Regulation M. KJERRULF*, D. GRDIC*, M. KOPF† & N. LYCKE* *Department of Medical Microbiology and Immunology, University of Go¨teborg, Sweden and †Basel Institute for Immunology, Switzerland

(Received 12 December 1997; Accepted 5 January 1998)

Kjerrulf M, Grdic D, Kopf M, Lycke N. Induction of Gut Mucosal Immune Responses: Importance of Genetic Background and Th1/Th2 Cross-Regulation. Scand J Immunol 1998;47:401–407 The reciprocal regulation of T-helper cell (Th) subsets is widely documented in various animal models of infectious diseases. In this study IFN-g/IL-4 double knockout (DKO) mice were used to analyse the role of Th subsets in mucosal immune responses. We found that the DKO mice had normal IgA differentiation but impaired induction of specific gut mucosal antibody responses after oral immunization using cholera toxin adjuvant. Both Th1 and Th2 responses were reduced compared with wild-type mice. Despite the absence of both IFN-g and IL-4 in the DKO mice the overall results were similar to previous observations in IFN-g receptor-knockout (IFN-gR¹/¹) mice and did not suggest a strict cross-regulation of the two Th subsets in the gut mucosa. To further examine the role of IFN-g in mucosal immunity we compared two different mouse strains lacking IFN-g, i.e. IFN-g¹/¹ (C57BL/6) and IFN-gR¹/¹ mice (129/Sv). We found that IFN-gR¹/¹ mice exhibited reduced mucosal antibody responses and decreased Th1 and Th2 activity after oral immunization, while IFN-g¹/¹ mice had intact antibody responses and increased Th2 responses. Thus, genetic differences were found to critically affect the development of a specific gut mucosal immune response. An enhanced Th2 activity in the Peyer’s patches following oral immunization was associated with an ability to mount strong intestinal IgA immunity. Nils Lycke, Department of Medical Microbiology and Immunology, University of Go¨teborg, Guldhedsgatan 10, S-413 46 Go¨teborg, Sweden

INTRODUCTION þ

The discovery of a polarization of CD4 T-helper cells (Th) into Th1 and Th2 subsets has been of crucial importance to our understanding of host responses against infectious microorganisms. Studies of Leishmania major infections in inbred mice have revealed that, whereas most strains, e.g. C57BL/6 or 129/Sv mice (H-2b), mount Th1 responses and resolve the infection, BALB/c mice (H-2d) mount Th2 responses and succumb to the infection [1–4]. However, protection against other infections, such as Borrelia burgdorferi, requires a Th2 response, while a Th1 response may be deleterious [5]. Thus, depending on the infectious agent, either a Th1 or Th2 response can be critical for protection. The Th1 response is characterized by production of IFN-g, lymphotoxin and IL-2, while the Th2 response is dominated by secretion of IL-4, IL-5, IL-6 and IL-10. The Th1 and Th2 differentiation pathways are mutually exclusive and regulated by secretion of IFN-g/IL-12 and IL-4/IL-10, respectively [3]. q 1998 Blackwell Science Ltd

The influence of the Th1 and Th2 subsets on mucosal IgA responses has caught comparatively little attention. We recently demonstrated that IL-4¹/¹ mice have impaired mucosal IgA responses following oral immunization with soluble protein antigens, indicating that Th2 cells are critical for specific gut IgA responses [6, 7]. On the other hand, Th2 cells appear not to be required for the induction of oral tolerance, because IL-4¹/¹ mice were well-tolerized by feeding antigen [8, 9]. These results suggest that IL-4 has an important function in the stimulation of mucosal IgA responses, but not for oral tolerance. The role of IFN-g in the regulation of mucosal immune responses is less clear. IFN-g is abundantly produced in the gut mucosa [10–14] and has a well-documented effect on the expression of major histocompatibility complex (MHC) class I and II molecules [15, 16], as well as on the production of secretory component [17, 18]. In IFN-gR¹/¹ mice we observed impaired IgA responses and reduced Th1 as well as Th2 activity after oral immunizations with keyhole limpet haemocyanin (KLH) plus cholera toxin (CT) adjuvant [19]. Importantly, we

402 M. Kjerrulf et al. found that the absence of IFN-g did not favour Th2 development in these mice. According to the view that IFN-g may affect CD4þ T-cell development by suppressing Th2 differentiation, this finding was unexpected [20]. In the present study we used single or double knockout mice to investigate if IFN-g and IL-4 exert cross-regulatory functions in the induction of intestinal IgA responses. Furthermore, we asked to what extent genetic differences may affect mucosal immune responses in mice lacking IFN-g.

MATERIALS AND METHODS Mice and immunizations. Heterozygous breeding pairs of IFN-g knockout (IFN-g¹/¹) mice [21] on the C57BL/6 background were kindly provided by Genentech Inc. (San Francisco, CA, USA). The mice were interbred to obtain breeding pairs homozygous for the disrupted IFN-g gene. Loss of the wild-type allele was confirmed by polymerase chain reaction analysis of tail biopsies. C57BL/6 wild-type control mice were provided by B&K (Stockholm, Sweden). IFN-g receptor knockout (IFN-gR¹/¹) [22] and wild-type 129/Sv mice were kindly provided by M. Aguet (Genetech Inc.). IFN-g receptor/IL-4 double knockout mice (DKO, 129/Sv) [23] were obtained by breeding of IFN-gR¹/¹ with IL-4¹/¹ mice [6]. All mice were maintained in microisolator cages under specific pathogen-free conditions at our department. Groups of six mice were immunized three times perorally, at 10-day intervals, with 2 mg of KLH (Sigma Chemical Co., St. Louis, MO, USA) given together with 10 mg of CT (List Biological Laboratories, Campbell, CA, USA) adjuvant [7] and analysed six days after the final immunization. Preparation of gut lavage. A modification of the method originally described by Elson et al. was used [24]. Briefly, the small intestine was removed, ligated at both ends and injected with 2 ml phosphate buffered saline (PBS) containing soybean trypsin inhibitor (Sigma) at 0.1 mg/ml. After 10 min the solution was transferred to a test tube and centrifuged for 10 min at 650 × g. The supernatant was transferred to a microfuge tube and 10 ml of 100 mM phenyl methyl sulfonyl fluoride (PMSF: Sigma) in PBS was added per ml of lavage. The tube was then centrifuged at 14 000 × g for 20 min and the supernatant was transferred to a new microfuge tube. Ten ml of 100 mM PMSF and 10 ml 1% sodium azide were added per ml of lavage. After a 15 min incubation at r.t., 50 ml fetal calf serum (FCS) was added per ml of lavage and the tube was centrifuged at 14 000 × g for 20 min. The supernatants were stored in aliquots at ¹ 70 8C. Analysis of antigen-specific B- and T-cell responses. Spleen (Sp), Peyer’s patch (PP) and lamina propria lymphocytes (LPL) were prepared as previously described [25]. Serum and gut lavage antibody concentrations were determined by ELISA [26]. KLH- and CT-specific antibodyforming cells were determined by the antigen-specific spot-forming cell (SFC) assay [27]. For in vitro cytokine production triplicate cultures of splenocytes or PP cells were incubated for 48 or 96 h in 96-well microtiter plates (Nunc, Roskilde, Denmark), whereupon cell-free supernatants were collected. Splenocytes were cultured in the absence or presence of KLH at 0.1 mg/ml. PP cells were stimulated with 10% of an anti-CD3 containing supernatant (145–2C11) [28] combined with antiCD28 (Pharmingen, San Diego, CA, USA) at 5 mg/ml and rIL-2 [29] at 100 U/ml. Cytokine concentrations were determined by ELISA [26]. Briefly, for

detection of IL-4, 5 and 10 we used matched pairs of capturing and biotinylated detecting antibodies (Pharmingen) followed by antibiotin (Vector Laboratories Inc. Burlingame, CA, USA) and extravidin-peroxidase (Sigma). For the IFN-g ELISA we used a capturing rat antimouse IFN-g (Pharmingen) and, for detection, a polyclonal rabbit antimouse IFN-g serum (our own production), followed by an alkaline phosphatase-conjugated antirabbit Ig (Souther Biotechnology, Birmingham, AL). Detection levels for IL-4, 5, 10 and IFN-g were 40 pg/ml, 0.3 ng/ ml, 0.5 ng/ml and 0.7 ng/ml, respectively. Cryosections and immunohistochemical analysis. Sections of the small intestine were removed and placed into Histocon (Histolab Products AB, Go¨teborg, Sweden) at þ 4 8C. The tissues were cut into appropriate pieces, placed in plastic forms (Cryomold, Miles Inc, Elkhart, IN, USA) filled with O.C.T. Compound (Miles Inc.) and subsequently snap frozen in isopenthane in liquid N2 for < 60 s. Frozen cross-sections were prepared on microslides using a cryostat1720 (Leitz, Wetzlar, Germany). For detection of IgAþ cells, sections were labelled with FITC-conjugated anti-IgA (Southern Biotechnology, Birmingham, AL, USA) at 1/50. Slides were mounted on Aquatex (Merck, Darmstadt, Germany) and photographed with a Leica DMLD photomicrography system (Leica Mikroskopie, Wetzlar, Germany). Statistical analysis. We used Student’s t-test for analysis of significance.

RESULTS Impaired gut mucosal immune responses in IFN-gR/IL-4 double knockout mice IFN-gR/IL-4 DKO and wild-type 129/Sv mice were immunized perorally with optimal doses of KLH plus CT adjuvant [7, 19] and intestinal lamina propria antigen-specific antibody forming cell (SFC) activity was assessed. We found that the DKO mice exhibited a 50% reduction in KLH-specific SFC responses, compared with wild-type 129/Sv mice, indicating an impaired induction of gut mucosal antibody responses to soluble protein antigens (Fig. 1). The reduced gut mucosal SFC response was also reflected in significantly lower KLH-specific serum

Fig. 1. IFN-gR/IL-4 double knockout (DKO) mice exhibit impaired specific IgA antibody responses in the intestinal lamina propria following oral immunization with a soluble protein antigen. DKO (white bars) and wild type 129/Sv mice (black bars) were immunized three times orally with KLH (2 mg) plus CT adjuvant (10 mg) and the numbers of lamina propria KLH-specific and CT-specific antibodyproducing cells (SFC) were determined. The results are expressed as the mean SFC/107 cells 6 SEM of five individual experiments. *Denotes P < 0.05 compared with wild-type control mice.

q 1998 Blackwell Science Ltd, Scandinavian Journal of Immunology, 47, 401–407

Induction of Gut Mucosal Responses

403

Fig. 2. Both Th1 and Th2 responses are reduced in DKO mice after oral immunization. DKO (white) and wild-type 129/Sv mice (black) were given three oral immunizations with KLH (2 mg) plus CT adjuvant (10 mg). A: Spleen cells were cultured for 96 h in the presence or absence of recall antigen (KLH at 100 mg/ml) and supernatants were analysed for cytokine production by ELISA. Cytokine concentrations are expressed as mean values 6 SD of three pairs of mice per group. Cells cultured in the absence of KLH did not produce detectable levels of cytokines. B: Serum anti-KLH Ig isotypes are expressed as mean log10 titres 6 SD of six mice per group. One experiment of three giving similar results is shown. *Denotes P < 0.05 compared with wild-type control mice.

antibody log10 titres in the DKO mice, i.e. 2.8 6 0.3, as compared to wild-type controls, 3.6 6 0.3, (P < 0.05). Interestingly, the response to the adjuvant itself was unaltered as anti-CT SFC responses of similar magnitudes were observed in DKO and wild-type mice (Fig. 1).

responses following oral immunization was not supported by these results. Rather, the data indicated that the lack of Th2 activity, as previously observed by us in IFN-gR¹/¹ and IL-4¹/¹ mice [7, 19], was responsible for the impaired gut mucosal IgA response.

IFN-gR/IL-4 double knockout mice exhibit reduced Th2 responses

The genetic background influences the ability to mount mucosal immune responses in IFN-g-deficient mice

We subsequently asked whether the observed impairment in the induction of intestinal anti-KLH responses in the DKO mice resulted from an imbalance in the reciprocal regulation of mucosal Th1 and Th2 cells. Following oral immunizations, spleen T cells were isolated and their cytokine production to recall antigen was analysed. We found that, compared with wild-type mice, the DKO mice had decreased production of IFN-g, IL-5 and IL-10, and no production of IL-4 (Fig. 2A). Also, KLH-specific IgG1 and IgG2a serum titres were 8–12 fold lower in the DKO mice, reflecting the absence of both IL-4 and IFN-g activity (Fig. 2B). Serum IgG2b, IgG3 and IgA titres were not significantly different in DKO and wild-type mice. These results clearly demonstrated impaired Th1 as well as Th2 activity in the DKO mice. Thus, the assumption that IFN-g and IL-4 may exert reciprocal regulation of gut mucosal immune

Given that antigen-specific mucosal immune responses were impaired in IL-4¹/¹ [7] as well as in DKO mice – both demonstrating poor Th2 induction – we concluded that IL-4 may be a key element for the induction of mucosal immunity. However, the role of IFN-g in the regulation of mucosal immune responses was still not clear. To further investigate the role of IFN-g in mucosal immunity, we compared responses to oral immunization in IFN-g-deficient mice of different genetic background, i.e. IFN-gR¹/¹ mice (129/Sv) [22] and IFN-g¹/¹ mice (C57BL/6) [21]. We found that IFN-g¹/¹ mice demonstrated unaltered gut anti-KLH responses compared with wild-type C57BL/6 mice (Fig. 3), while the IFN-gR¹/¹ mice, as previously observed [19], exhibited impaired mucosal responses to oral immunization with KLH plus CT adjuvant (Fig. 3). Strikingly, we observed an increase in Th2 cytokines in the IFN-g¹/¹ mice,


404 M. Kjerrulf et al. Table 2. Ability to develop Th2 responses in the Peyer’s patches is associated with induction of mucosal immunity in IFN-g deficient mice

Mouse strain

Fig. 3. The genetic background appears to influence the ability to develop mucosal IgA responses in different IFN-g-deficient mouse strains. IFN-gR¹/¹ (gR¹/¹), 129/Sv, IFN-g¹/¹ and C57BL/6 (B6) mice were given three oral immunizations with KLH (2 mg) plus CT adjuvant (10 mg). KLH-specific lamina propria SFC responses are expressed as means 6 SEM of three pairs of mice per group. One representative experiment of four giving similar results is shown.

whereas in the IFN-gR¹/¹ mice both Th1 and Th2 cytokine levels were reduced compared with wild-type controls (Table 1). Thus, an increased Th2 activity was associated with the ability to mount a strong mucosal response in the IFN-g¹/¹ mice. Th2 activity in the Peyer’s patches is required in order to mount intestinal IgA responses to soluble protein antigens As the Peyer’s patches (PP) are regarded as important sites for induction of gut immune responses, we asked whether IFN-gR¹/ ¹ and IFN-g¹/¹ mice differed in their ability to mount a Th2 response in the PP. Mice were given a single oral dose of KLH plus CT adjuvant. Ten days later, PP T cells were isolated and stimulated in vitro with anti-CD3 combined with anti-CD28 and IL-2 to induce cytokine production. As shown in Table 2, the PP cells of the IFN-g¹/¹ mice exhibited significant increases in Th2 cytokines compared with wild-type C57BL/6 mice. In contrast, Table 1. The genetic background influences Th2 differentiation in IFN-g-deficient mice

Mouse strain

IFN-g (ng/ml)

IL-4 (pg/ml)

IL-5 (ng/ml)

IL-10 (ng/ml)

IFN-g¹/– C57BL/6 IFN-gR¹/– 129/Sv

* < 0.7 11 6 3 28 6 6 72 6 17

105 6 30* < 40 54 6 12* 101 6 4

4.5 6 2.8 2.5 6 1.1 1.6 6 0.8 2.3 6 0.4

4.7 6 1.1* 2.2 6 0.7 9.1 6 3.1 16 6 2.7

Mice were given three oral immunizations with KLH plus CT adjuvant. Spleen cells were cultured for 96 h in the presence or absence of recall antigen (KLH at 100 mg/ml) and supernatants were analysed for the presence of cytokines with ELISA. Cytokine concentrations are expressed as mean values 6 SEM of three pairs of mice per group. Cultures without recall antigen had undetectable levels of cytokines. This is one representative experiment of three giving similar results. *Denotes P < 0.05 compared with wild-type control mice.

IFN-g¹/– C57BL/6 IFN-gR¹/– 129/Sv

IFN-g (ng/ml)

IL-4 (pg/ml)

IL-5 (ng/ml)

IL-10 (ng/ml)

< 0.7* 4.4 6 1.0 4.8 6 0.9 3.7 6 1.3

92 6 12* < 40 < 40 62 6 7

0.7 6 0.1 < 0.3 n.d. n.d.

78 6 19* 14.5 6 3.4 21 6 8.2 23 6 6.6

Mice were given a single oral immunization with KLH plus CT adjuvant. PP cells were isolated and cultured in vitro for 96 h in the presence or absence of anti-CD3, anti-CD28 and IL-2 as described in Material and Methods. Cell-free supernatants were analysed for cytokine production with ELISA. Concentrations are expressed as mean values 6 SEM of five mice per group. Cytokine production by unstimulated cells was below detection level. *Denotes P < 0.05 compared with wild-type control mice.

the IFN-gR¹/¹ mice failed to develop Th2 responses, as assessed by the impaired IL-4 production in the PP. Thus, the observed reduction in specific mucosal IgA responses in the IFN-gR¹/¹ mice was associated with a lack of induction of Th2 cells in the PP. While the specific IgA responses to oral immunization were significantly lower in the IFN-gR¹/¹ mice than in the IFN-g¹/¹ mice, we found that the levels of total IgA in intestinal secretions were similar in non-immunized IFN-gR¹/¹ (129/Sv) and IFN-g¹/¹ mice (C57BL/6), as well as in DKO mice (Fig. 4). Moreover, immunohistochemical staining of frozen gut tissues revealed that DKO, IFN-g¹/¹ and IFN-gR¹/¹ mice exhibited distributions of IgA-positive cells in the PP and in the lamina propria that were similar to those observed in C57BL/6 and 129/ Sv wild-type mice (Fig. 4). Thus, IgA B-cell differentiation seemed to be unaffected by the absence of either IL-4 [7], IFN-g or both. Rather, it was an inability to induce specific responses that was responsible for the impaired mucosal antiKLH IgA SFC activity, observed in IFN-gR¹/¹ and DKO mice. DISCUSSION The present study has shown that DKO mice, lacking both IL-4 and IFN-g, had normal mucosal total IgA production but exhibited impaired responses to a soluble protein antigen, KLH, given orally together with CT adjuvant. In this regard the DKO mice demonstrated a similar reduced responsiveness to oral immunization, which was previously found in IFN-gR¹/¹ mice [19]. Both DKO and IFN-gR¹/¹ mice had decreased Th1 and Th2 activity. Thus, the absence of IL-4 in DKO mice did not restore the impaired induction of mucosal immune responses exhibited by IFN-gR¹/¹ mice. There is ample evidence that cross-regulation of Th1/Th2 responses, mediated by IFN-g and IL-4, exists in systemic immune responses. For example, in studies of Leishmania


Induction of Gut Mucosal Responses

405

Fig. 4. Normal IgA B cell differentiation in IFN-gR¹/– , IFN-g¹/– and DKO mice. A: Gut lavages of unimmunized mice were analysed for concentrations of IgA with ELISA. Results are expressed as mean mg/ml 6 SD of six mice per group. B–D: Cryosections of the small intestine from IFN-g¹/– , IFN-gR¹/– and DKO mice were stained with anti-IgA FITC. Micrographs were taken at 100 × magnification. These are the results of at least four separate experiments.

major infection, susceptible BALB/c mice, treated with anti-IL-4 antibody, resisted the infection with a concomitant shift from a Th2 to a Th1 predominance [30]. Further, in L. major-resistant C57BL/6 mice, disruption of the IFN-g gene induced susceptibility to infection and a shift from Th1 to Th2 predominance [31]. On the other hand, it was recently reported that DKO and IFN-gR¹/¹ mice were equally susceptible to a L. major infection [23]. This latter result corroborates our present finding and demonstrates that the absence of IL-4 in DKO mice cannot restore responsiveness in IFN-gR¹/¹ mice. Given that IFN-g and IL-4 would exert strict cross-regulatory control over mucosal IgA immunity, our hypothesis was that whatever negative effects the IFN-g deficiency would exert, a simultaneous lack of IL-4 might restore the ability to respond. However, our results in DKO mice do not support this notion. Therefore, we would like to propose that the lack of Th2 responses in DKO and IFN-gR¹/¹ mice, could be explained on the basis of default pathways for Th differentiation, favouring a Th1 development [32]. In support of such a possibility, it was reported that the IFN-gR¹/¹ mice were susceptible to L. major infection, although they were still dominated by a Th1 response,

normally known to be protective [33]. Sophisticated studies by Gu¨ler et al. have demonstrated that such strain-dependent default pathways for CD4þ T-cell differentiation may result from differences in early maintenance of IL-12 responsiveness [34]. Whether the IFN-gR¹/¹ or DKO mice are particularly good at sustaining IL-12 responsiveness at mucosal surfaces is currently unknown. The concept of genetically determined default pathways for Th cell differentiation may also be used to explain our present discrepant results in the two IFN-g-deficient mouse strains. Despite their shared MHC locus, IFN-g¹/¹ mice (C57BL/6) and IFN-gR¹/¹ (129/Sv) responded differently to oral immunization. Whereas the IFN-g¹/¹ mice (C57BL/6) demonstrated strong local antigen-specific IgA production and increased Th2 activity, the IFN-gR¹/¹ mice (129/Sv) had reduced mucosal IgA responses and decreased Th2 activity. Similar results have been obtained by other groups [35, 36]. Based on our own studies and observations made by others, we think that the impairment of mucosal IgA responses in the IFN-gR¹/¹ mice (129/Sv) can be explained by default Th pathways, resulting in weak Th2 activity in the 129/Sv mice. Furthermore, the difference in ability to


406 M. Kjerrulf et al. develop Th1 or Th2 responses in the PP in IFN-g¹/¹ (C57BL/6) and IFN-gR¹/¹ (129/Sv) mice may be due to differences in early IL-12 production or maintenance of IL-12R signalling. Studies are underway to address this issue. The nature of the antigen used for mucosal immunizations, i.e. soluble or particulate antigen, may dramatically influence the response. A recent study describes the preferential induction of Th2 cells in the PP, in response to oral vaccination with soluble tetanus toxoid (TT), while TT incorporated in a live Salmonella vector generated strong Th1 responses [36]. Importantly, both these delivery systems stimulated mucosal IgA responses. Moreover, it was also reported that, irrespective of whether IL-4- or IFN-g-deficient mice were used, strong mucosal anti-TT responses could be stimulated by oral Salmonella inoculation [35]. Taken together, these findings illustrate that the nature of the antigen delivery system, soluble or particulate, protein or whole bacteria, will determine whether mucosal immune responses can be elicited. Therefore, it appears that limitations at mucosal inductive sites, dictated by targeted disruptions of the IFN-gR or IL-4 genes, affect the responsiveness to soluble protein but not to whole bacteria. Interestingly, despite impaired specific mucosal IgA responses in DKO and IFN-gR¹/¹, as well as L-4¹/¹ mice [7], we found normal levels of gut lavage total IgA. In fact, DKO mice, as well as IFN-g¹/¹ and IFN-gR¹/¹ mice, had total IgA levels comparable to those observed in wild-type 129/Sv and C57BL/6 mice. Moreover, similar distributions of IgA-positive cells in the lamina propria and in the PP were found. Thus, IgA B-cell differentiation seemed to be unaffected by the absence of IFN-g, IL-4 [7], or both. It may be that, although the mice exhibited impaired specific IgA responses to oral immunization with KLH plus CT adjuvant, they were perfectly able to respond to other antigens. This is particularly well illustrated with the Salmonella antigen delivery system mentioned above, and suggests that IgA B-cell differentiation, if appropriately triggered, can proceed normally in the IL-4¹/¹, IFN-gR¹/¹ or DKO mice.

6

7

8

9

10

11

12

13

14

15

16

17

18

ACKNOWLEDGEMENTS This study was supported by the Swedish Medical Research Council, the Swedish Society for Medical Research, the WHO Transdisease Vaccinology Program, the Nanna Swartz and Magn. Bergwall Foundations, NIH grant #R01 AI40701 and the Swedish Cancer Foundation.

19

20

21

REFERENCES 1 Zlotnik A, Moore TA. Cytokine production and requirements during T-cell development. Curr Opin Immunol 1995;7:206. 2 Rocha B, Guy-Grand D, Vassalli P. Extrathymic T cell differentiation. Curr Opin Immunol 1995;7:235. 3 Reiner SL, Seder RA. T helper cell differentiation in immune response. Curr Opin Immunol 1995;7:360. 4 Carter L, Dutton R. Type 1 and type 2: a fundamental dichotomy for all T cell subsets. Curr Opin Immunol 1996;8:336. 5 Keane-Myers A, Nickell S. Role of IL-4 and IFN-g in modulation of

22 23

24

immunity to Borrelia burgdorferi in mice. J Immunol 1995;155:2020. Kopf M, Le Gros G, Bachmann M, Lamers MC, Bluethmann H, Ko¨hler G. Disruption of the murine IL-4 gene blocks Th2 cytokine responses. Nature 1993;362:245. Vajdy M, Kosco-Vilbois M, Kopf M, Ko¨hler G, Lycke N. Impaired mucosal immune responses in interleukin 4-targeted mice. J Exp Med 1995;181:41. Lycke N, Bromander A, Ekman L et al. The use of knock-out mice in studies of induction and regulation of gut mucosal immunity. Mucosal Immunol Update 1995;3:1. Garside P, Steel M, Worthey EA et al. T helper 2 cells are subject to high dose oral tolerance and are not essential for its induction. J Immunol 1995;154:5649. Bao S, Goldstone S, Husband AJ. Localization of IFN-g and IL-6 mRNA in murine intestine by in situ hybridization. Immunol 1993;80:666. Brett F, Mowat A, Farquharson M et al. The distribution of immunoreactive interferon-g-containing cells in normal human tissues. Immunol 1992;77:515. Breese E, Braegger C, Corrigan C, Walker-Smith J, MacDonald T. Interleukin-2- and interferon-g-secreting T cells in normal and diseased human intestinal mucosa. Immunol 1993;78:127. Quiding M, Nordstro¨m I, Kilander A et al. Intestinal immune responses in humans. Oral cholera vaccination induces strong intestinal antibody responses and interferon-g production and evokes local immunological memory. J Clin Invest 1991;88:143. Ramarathinam L, Shaban R, Neisel D, Kimpel G. IFN-g production by gut-associated lymphoid tissue and spleen following oral immunization. Microb Pathog 1991;11:347. Goes N, Sims T, Urmson J, Vincent D, Ramassar V, Halloran P. Disturbed MHC regulation in the IFN-g knockout mouse. J Immunol 1995;155:4559. Colgan S, Parkos C, Matthews J et al. Interferon-g induces a cell surface phenotype switch on T84 intestinal epithelial cells. Am J Physiol 1994;267:C402. Sollid L, Kvale D, Brandtzaeg P, Markussen G, Thorsby E. Interferong enhances expression of secretory component, the epithelial cell receptor for polymeric immunoglobulins. J Immunol 1987;138:4303. Youngman K, Fiocchi C, Kaetzel C. Inhibition of IFN-g activity in supernatants from stimulated human intestinal mononuclear cells prevents up-regulation of the polymeric Ig receptor in an intestinal epithelial cell line. J Immunol 1994;153:675. Kjerrulf M, Grdic D, Ekman L, Scho¨n K, Vajdy M, Lycke N. Interferon-g deficient mice exhibit impaired gut mucosal immune responses but intact oral tolerance. Immunol 1997;92:60. Scott P. IFN-g modulates the early development of Th1 and Th2 responses in a murine model of cutaneous leishmaniasis. J Immunol 1991;147:3149. Dalton D, Pitts-Meek S, Keshav S, Figari I, Bradley A, Stewart T. Multiple defects of immune cell function in mice with disrupted interferon-g genes. Science 1993;259:1739. Huang S, Hendriks W, Althage A et al. Immune responses in mice that lack the interferon-g receptor. Science 1993;259:1742. Kopf M, Brombacher F, Ko¨hler G et al. IL-4-deficient Balb/c mice resist infection with Leishmania major. J Exp Med 1996;184: 1127. Elson CO, Ealding W, Lefkowitz J. A lavage technique allowing repeated measurement of IgA antibody in mouse intestinal secretions. J Immunol Meth 1984;67:101.


Induction of Gut Mucosal Responses 25 Lycke N. Isolation of mouse small intestinal intraepithelial lymphocytes, Peyer’s patch, and lamina propria cells. In: Coligan J., ed., Current Protocols in Immunology. New York: Current Protocols, 1995: 3.19.6. 26 Ho¨rnqvist E, Goldschmidt T, Holmdahl R, Lycke N. Host defence against cholera toxin is strongly CD4þ T cell dependent. Infect Immun 1991;59:3630. 27 Lycke N. Measurement of polyclonal immunoglobulin synthesis using the ELISPOT assay. In: Coligan J., ed., Current Protocols in Immunology, New York: Current Protocols, 1991: 7.14.1. 28 Leo O, Foo M, Sachs D, Samuelson L, Bluestone J. Identification of a monoclonal antibody specific for a murine T3 polypeptide. Proc Natl Acad Sci USA 1987;84:1374. 29 Karasuyama H, Melchers F. Establishment of mouse cell lines which constitutively secrete large quantities of interleukin 2, 3, 4 or 5, using modified cDNA expression vectors. Eur J Immunol 1988;18:97. 30 Sadick M, Heinzel F, Holaday B, Pu R, Dawkins R, Locksley R. Cure of murine leishmaniasis with anti-IL-4 monoclonal antibody. Evidence for a T cell-dependent IFN-g-independent mechanism. J Exp Med 1990;171:115. 31 Wang Z, Reiner SL, Zheng S, Dalton D, Locksley RM. CD4þ

32

33

34

35 36


407

effector cells default to the Th2 pathway in interferon-g deficient mice infected with Leishmania major. J Exp Med 1994; 179:1367. Hsieh C-S, Macatonia SE, O’Garra A, Murphy KM. T cell genetic background determines default T helper phenotype development in vitro. J Exp Med 1995;181:713. Swihart K, Fruth U, Messmer N et al. Mice from a genetically resistant background lacking interferon g receptor are susceptible to infection with Leishmania major but mount a polarized T helper cell 1-type CD4þ T cell response. J Exp Med 1995;181:961. Gu¨ler M, Jacobson N, Gubler U, Murphy K. T cell genetic background determined maintenance of IL-12 signaling effects on BALB/c and B10.D2 T helper cell type 1 phenotype development. J Immunol 1997; 159: 1767. Yamamoto M, VanCott J, Okahashi Z et al. The role of Th1 and Th2 cells for mucosal IgA responses. Ann N Y Acad Sci 1996;778:64. VanCott J, Staats H, Pascual D et al. Regulation of mucosal and systemic antibody responses by T helper subsets, macrophages and derived cytokines following oral immunization with live recombinant Salmonella. J Immunol 1996;156:1504.