Bystander-activated CD4⁺memory lymphocytes: a role in the pathology of rheumatoid arthritis?

Background

Previous studies in this laboratory have highlighted the importance of the proinflammatory cytokine tumour necrosis factor alpha (TNF-α) in the pathogenesis of rheumatoid arthritis (RA) and suggested a role for bystander-activated lymphocytes in its chronic production in the RA joint. Normal peripheral blood (PB) lymphocytes cultured in the presence of a 'cocktail' of inflammatory cytokines (IL-2/IL-6/TNF-α) generated effector cells (T_CK) capable of inducing both immunoglobulin production by B cells [1] and TNF-α production by monocytes in a contact-dependent manner. This was in contrast to lymphocytes activated antigen dependently, via the T-cell receptor (T_TCR), which induced production of both TNF-α and the anti-inflammatory cytokine IL-10 [2, 3]. Lymphocytes isolated from the RA synovial membrane exhibited contact-dependent cytokine induction properties similar to T_CK, not T_TCR, cells [4].

Objective

We have examined the population dynamics and phenotypic changes induced in PB lymphocytes during culture with IL-2/IL-6/TNF-α to determine how such cells resemble those found in the RA joint. As the predominant infiltrating lymphocyte population found to accumulate in the RA synovial membrane is the CD4⁺ T cell we have also explored the possibility of generating monocyte TNF-α-inducing effector cells from purified CD3⁺CD4⁺ populations.

Results

Lymphocytes cultured with IL-2/IL-6/TNF-α over 8 days demonstrated a threefold expansion of the natural killer cell population, while the proportion of T cells remained static. However, both these natural killer and T-lymphocyte populations were independently able to induce contact-dependent TNF-α production in monocytes. Interestingly, effector cells generated by culture of the purified CD3⁺CD4⁺ fraction with IL-2/IL-6/TNF-α for 8 days were able to induce at least threefold more TNF-α production compared with such effectors generated from the corresponding unfractionated lymphocyte population. We subsequently characterised the different naïve and memory cell populations found within the CD4⁺ fraction of our bystander-activated lymphocyte cultures (as defined by Lanzavecchia and colleagues [5]). Despite an overall net loss in the percentage and number of total CD4⁺ T cells, both naïve (CD45RA⁺CCR7⁺) and memory populations (effector:CD45RO⁺CRR7^-; central CD45RO⁺CCR7⁺) were still present after 8 days of culture. However, a preferential retention of effector memory over central memory lymphocytes has been observed in most donors.

Further phenotypic studies have shown that bystander-activated lymphocytes also express high levels of components of the adhesion molecules VLA-1, VLA-4, VLA-5 and LFA-1 (β1, β2 and α4), but lower levels of L-selectin (CD62L). Such properties are associated with the ability of effector lymphocytes to migrate from lymph nodes and into tissues.

Conclusions

Our results suggest that exposure of PB lymphocytes to a cocktail of proinflammatory cytokines induces the outgrowth of an activated effector memory lymphocyte population with the capacity to migrate to inflammatory sites. Studies currently underway will examine the effector function of different CD4⁺ subsets in terms of their ability to induce monocyte TNF-α production.

This work will more closely define pivotal cell types responsible for cognate-dependent cytokine production in diseased joints as well as advancing our understanding of the effects and consequences of prolonged cytokine exposure in chronic inflammation.