pubim1

Analysis of events in B and T cell activation and differentiation, with some emphasis on the interface between mucosal and systemic immunity, constitute the main aims of the research projects in this laboratory, which are being addressed by a number of related experimental approaches. One approach, that directly addresses interactions between mucosal and systemic immunity, involves dissecting the consequences of oral exposure to soluble antigens of microbial and non-microbial origin on subsequent systemic immune responses in mice. Included are the ability of such mice to clear a challenge infection, in vitro T cell recall responses, and the role of intestinal flora and various cytokines and adhesion molecules in modulating these responses. Other approaches include analysis of the role of specific ligand-receptor interactions in controlling B and T cell activation, and dissection of signals that may contribute to their selective differentiation into immediate effectors versus long-lived memory cells.

The proportion of B cells in lymphoid organs draining the site of immunisation that is specific for the immunogen rarely exceeds 1-2% even at the peak of the immune response. This makes it almost impossible to follow antigen-specific responses at the clonal level under physiological conditions. Over the past year, we have been attempting to establish a method that would allow the identification of antigen-specific B cells in vivo. We report that immunisation of mice with phycoerythrin (PE) in adjuvant generates a PE-specific antibody response that is characterized by proliferation and activation of PE-specific B cells, a robust germinal centre reaction, isotype switching, affinity maturation and terminal differentiation to plasma cells. Importantly, PE-specific B cells can be identified by flow cytometry and their activation and differentiation can be followed over time in groups of mice. We have used this system to extend our earlier reported observation that ligation of CD27 during B cell priming inhibits their terminal differentiation into Ig-secreting plasma cells, suggesting that the clonal progeny may be recruited into the memory pool. We reasoned that if the decreased plasma cell generation was indeed accompanied by enhanced memory cell generation we should be able to score these as long lasting PE-binding B cells in vivo.

We report that immunisation of mice in the presence of anti-CD27 induces a larger number of PE-binding B cells than in control, saline-treated, mice by 2 weeks post-immunisation and the difference persists at later times (Table-1).

Further analysis revealed that the presence of anti-CD27 does not influence germinal centre formation, as assessed by the proportion of B cells that bind high levels of peanut agglutinin (PNA) and have downregulated surface Ig and CD38 (PNAhigh, sIglow, CD38low). About 35-45% of antigen-specific B cells had germinal centre characteristics 7-10 days after immunisation, and the proportions were similar in anti-CD27-treated and -untreated groups. In the course of an immune response, cells that exit the proliferative foci of germinal centres lose their ability to bind high levels of PNA and may emerge either as pre-plasma cells that bind PE poorly and have downregulated B220 (PElow, B220low, sIglow, CD38low) or as secondary/memory B cells that bind antigen well and have upregulated CD38 (PEbright, B220high, sIghi, CD38hi). In preliminary experiments, we have found that the presence of anti-CD27 during priming leads to the generation and persistence of a larger number of secondary B cells in vivo. Our data suggest that it may be possible to skew immune responses in vivo away from primary effectors and towards memory cells.

Days after immunisation	% PE- specific B cells following immunisation with
	BSA (controls)		PE + saline		PE + anti-CD27
	PLN	Blood	PLN	Blood	PLN	Blood
14	0.40	0.07	3.82	0.17	7.78	0.91
28	0.04	0.05	0.53	0.46	1.11	0.80
42	0.12	0.14	0.85	0.42	3.01	0.97

In the reporting year we have established an experimental model for tracking T cells primed in vivo following oral immunisation in the absence of adjuvant. The model involves transfer of lymphocytes from a transgenic mouse bearing a CD4 T cell receptor specific for an ovalbumin (OA) peptide into normal recipients, and tracking OA-specific T cell numbers and their activation and differentiation in various organs of the chimeric recipients following feeding and subsequent parenteral immunisation. Donor CD4 cells can be identified by flow cytometry with the help of a clonotypic antibody (KJ126) that specifically recognises the transgenic TCR. We report that transfer of 40x106 pooled lymphocytes leads to successful chimerism in all lymphoid organs including Peyer’s patches (PP). Oral administration of OA to these mice leads to substantial expansion of antigen-specific T cells in the PP at 24h (6% in fed mice vs. 2% in unfed controls) and 48h (11% in fed mice), and this is followed by an expansion of specific cells in mesenteric lymph nodes (MLN) at day 4 (9% in fed mice vs. 4% in controls). The clonal expansion is mirrored by activation events including upregulation of CD25 and CD69 and downregulation of CD62L. No perturbations are seen in the spleen or peripheral lymph nodes (PLN). By day 7 after feeding, the response at mucosal sites subsides and there are relatively equivalent proportions of antigen-specific transgenic cells in the two groups. At this point, if the mice are immunized subcutaneously (s.c.) with OA in adjuvant (as in a standard oral tolerance experiment), a huge expansion of OA-specific T cells, which peaks 4 days after immunisation, is observed in the PLN draining the site of s.c. immunisation. However, the expansion is much smaller than occurs in unfed control mice (22% in fed mice vs. 36% in controls) and the differences are seen even 7 days after immunisation (6% in fed mice vs. 10% in controls). Negligible perturbations are seen in non-draining lymphoid organs. Activation events are now restricted to the draining PLNs, with no perturbations at other sites. The data show that activation, proliferation and death of antigen-specific T cells occurs at (and is largely restricted to) mucosal sites following oral administration of soluble antigen. Similar events occur at (and are restricted to) draining lymph nodes following s.c. administration of antigen in adjuvant. Not unexpectedly, antigen emulsified in adjuvant leads to markedly greater proliferative responses than soluble antigen. As expected, cells from the draining PLN of fed mice respond poorly to antigen recall in vitro.

We have also tracked antigen-specific cells following i.p. immunisation of the chimeric mice with soluble OA, and preliminary experiments reveal a pattern that is remarkably similar to what happens after oral immunisation with OA- there is a transient activation and expansion of antigen-specific cells in the spleen, MLN and to some extent in the PLNs, and the response dies down by day 7. No perturbations are seen in PP, confirming that only oral antigen can prime cells at this site. Surprisingly, mice immunised i.p. also showed systemic T cell hyporesponsiveness to a subsequent immunisation with OA in adjuvant, and the degree of hyporesponsiveness was similar to that seen following oral administration of OA. Our results suggest that oral tolerance as a phenomenon may have less to do with the route of antigen uptake and more to do with the absence of adjuvant during initial T cell priming.

Our results allow us to tentatively formulate the following hypothesis to explain oral tolerance induction;- in the absence of adjuvant, T cell priming will be limited, with the generation of a large ‘effector’ component that has limited replicative ability (but can secrete cytokines ) upon subsequent antigen encounter . The ‘effectors’ generated in MLN and spleen will be largely of the Th2 kind, but because PP are chronically activated, it is possible that some Th1 cells may be generated in PP, as has indeed been reported by others. The decreased T cell expansion seen after the second priming in adjuvant may then be due sequestration of antigen by these ‘effectors’ or by their ability to inhibit antigen presentation to naive T cells because of local cytokine secretion (the ‘suppressor’ cells or CD25+ regulatory T cells reported by others). We would like to test this hypothesis over the next year.

1. Mukherjee P, Dani A, Bhatia S, Singh N, Rudensky AY, George A, Bal V, Mayor S and Rath S (2001) Efficient presentation of both cytosolic and endogenous transmembrane protein antigens on MHC class II is dependent on cytoplasmic proteolysis. J Immunol 167:2632-2641.

2. Bhatia S, Mukhopadhyay S, Jarman E, Hall G, George A, Basu SK, Rath S, Lamb JR and Bal V (2002) Scavenger receptor-specific allergen delivery elicits IFN-g-dominated immunity and directs established TH2-dominated responses to a nonallergic phenotype. J Allergy Clin Immunol 109:321-328.