Lymph Node Derived CLL Cells Have a More Activated Phenotype and Better Antigen Presentation Capabilities Compared To Those From The Peripheral Blood

It is now well established that CLL is a highly proliferative disease with replication restricted to lymphoid tissue microenvironments. There is good evidence that interactions involving activated CD4+ T cells influence proliferation of the neoplastic B-cell clone. In this study we investigated the paradox that, whilst chronically activated T cells are found at these sites, in-vitro studies have shown that peripheral blood (PB) CLL cells suppress T cell activation. We hypothesized that the explanation for these contradictory findings is that lymph node (LN) CLL cells have increased antigen-presenting capacity compared to those in the PB. Accordingly, we obtained paired PB and LN fine needle aspirate (FNA) samples from a series of CLL patients and compared the expression of molecules associated with activation, antigen presentation and the ability to stimulate a third party mixed lymphocyte reaction (MLR).

Using multi color flow cytometry, CD5+/CD19+ CLL cells were gated for expression of CD80, CD86, HLA-DR, CD25 and CD69. LN CLL cells had higher expression of markers associated with co-stimulation: CD80 (p=0.002), CD86 (p=0.037, n=6), antigen presentation: HLA-DR (p=0.04) and activation: CD25 (p=0.005) and CD69 (p=0.0018).

We next used an MLR to determine the functional significance of these findings. Irradiated CLL cells from PB and LN were mixed at co-culture ratio’s of 1:1 and 1:10 and the proliferation/activation of CD4+ T cells measured using the expression of Ki67, CD69, and HLA-DR, and thymidine incorporation. T cell expression of the proliferation marker ki67 and activation marker CD69 was higher when co-cultured with LN CLL cells compared to PB derived cells. 1:1 co-culture ratio, mean Ki67 expression induced by LN CLL cells was 16.1% , SD ± 8.0 compared to 11.5%, SD± 7.7 induced by PB CLL cells (p=0.04) and at 1:10 ratio 8.4% , SD ± 3.6 compared to 3.9%, SD± 3.8 (p=0.02). Similarly, T cell expression of CD69 induced by LN CLL cells at 1:1 ratio was 37.0% , SD ± 15.5 compared to 26.6%, SD± 13.8 induced by PB-CLL cells (p=0.025) and at a 1:10 ratio 14.7% , SD ± 7.7 compared to 11.0%, SD± 7.4 (p=0.0027). LN-derived CLL cells also induced higher levels of T cell HLA-DR expression: 1:1 ratio 30.0%, SD ± 14.7 compared to 21.1%, SD± 9.1 (p=0.02) and at a 1:10 ratio 12.3%, SD ± 5.0 compared to 8.4%, SD± 3.7 (p=0.03). In accordance with these findings, T cell proliferation, measured by thymidine incorporation, was significantly higher in LN-CLL co-cultures compared to PB: 1:1 ratio, LN 2252, SD ± 1549 compared to PB 1615, SD± 1302 (p=0.008) and at 1:10 ratio, LN 910, SD ± 746 compared to, PB 416, SD± 366 (p=0.052).

These data support our hypothesis that LN CLL cells have an increased capacity to activate T cells compared to those from the PB. We next investigated whether migration through the endothelium might play a role in promoting the antigen presenting function of CLL cells and hypothesized that this increased antigen presentation capacity of LN-CLL cells would be associated with markers of adhesion and migration. In keeping with this concept, we established that LN-CLL cells had higher expression of CD49d (p=0.008) and CD38 (p=0.018) and reduced CXCR4 (p=0.02) when compared to PB-CLL. To demonstrate this functionally we used a novel in vitro circulation system to compare the phenotype of cells that migrate through the endothelium with those that remain in circulation. CLL cells that migrated into the extravascular compartment in this system had a similar phenotype to LN CLL cells with elevated levels of CD49d and CD38 (P = 0.02 and P = 0.009 respectively), increased levels of CD80 (p=0.0014) and CD86 (p=0.015) and a trend towards increased expression of the activation marker CD69 (p=0.08). CLL cells that remained in the circulating compartment had a phenotype comparable to PB CLL cells.

These results provide an explanation for the fact that CLL lymph nodes contain significant numbers of activated T cells, despite the known inhibitory effects of the tumor. Since activated T cells are in turn thought to promote tumor proliferation, our findings suggest that reciprocal interactions between CLL cells and LN T cells might play an important role in disease progression. In addition, our results show that transendothelial migration appears to activate CLL cells. Whether this, or other interactions such as stimulation of the B cell receptor, are responsible for the enhanced antigen presenting capacity of LN CLL cells remains to be determined.