Membrane-Associated Proteinase 3 On Granulocytes and Myeloid Leukemia Mediates Reversible Inhibition of T Cell Proliferation Associated with Altered ZAP70 and ERK Signaling

Abstract 1031

Proteinase 3 (P3), a serine protease constitutively expressed in primary granules and on the membrane of some resting granulocytes, is the target of T cell-mediated autoimmunity in Wegener's granulomatosis (WG) and of anti-leukemia immunity mediated by PR1 (VLQELNVTV)-specific cytotoxic T lymphocytes (PR1-CTL). We have previously shown anti-CD3/CD28 induced proliferation of healthy donor T-cells to be significantly inhibited by peripheral blood polymorphonuclear neutrophils (PMNs) expressing membrane P3 (mP3) at a ratio of 3 PMNs to 1 PBMC. Our results indicate that mP3⁺ PMNs begin to exert inhibitory effects on T cell proliferation at a ratio of 2.2 mP3⁺ PMNs to 1 PBMC, a ratio greater than that seen in normal homeostatic conditions in peripheral blood. The inhibition was predominantly enzyme-independent and dose-dependent. Notably soluble P3 exerted similar effects on T cells as was seen with mP3. Additionally soluble P3 induced a G₀ cell cycle arrest. Of significance, soluble P3 in acute myeloid leukemia (AML) patient serum can be up to 5-fold higher than that seen in healthy control serum.

To confirm mP3 specificity, we FAC-sorted PMNs based on the mP3 co-expressed CD177 molecule to obtain highly purified (>98%) mP3⁺ and mP3⁻ PMNs. Compared to activated PBMC alone, activated PBMC co-cultured at a ratio of 1:3 with mP3⁺ PMNs showed 58% and 57% inhibition of CD8⁺ and CD4⁺ T cells, respectively (CD8⁺ and CD4⁺: p< 0.003). PBMC co-cultured at that same ratio with mP3⁻ PMNs showed less inhibition - only 29% and 26% inhibition of CD8⁺ and CD4⁺ T cells, respectively (CD8⁺: p<0.05; CD4⁺: p=ns). Inhibition of T cell proliferation by both mP3⁺PMNs and soluble P3 was blocked by anti-P3 mAbs but not by isotype-matched mAb. Furthermore, P3-mediated inhibition of T-cell proliferation is reversible since removal of PMNs or soluble P3 restored the proliferative capacity of the T cells.

Because P3 is over-expressed in AML and chronic myeloid leukemia (CML), we hypothesized that mP3⁺ leukemia may suppress T-cell proliferation. Healthy donor T cell proliferation was studied with CFSE after stimulation with anti-CD3/CD28 mAbs in the presence or absence of mP3⁺ AML for five days. AML mediated a dose-dependent inhibition of T-cell proliferation contingent upon the level of mP3 expression (p<0.0001). Co-incubation of PBMC with AML displaying 91% mP3 positivity, reduced proliferation of CD8⁺ T cells to 29.6%, compared to 94.7% in the PBMC culture alone. This inhibition could be completely abrogated by addition of anti-P3 mAb, restoring the proliferation of CD8⁺ T cells to a level comparable to that seen in control. In contrast, no inhibition of CD8⁺ T cell proliferation was observed in co-cultures of T cells with AML in which only 6% of the AML cells expressed mP3. Thus, there is an inverse correlation between percent proliferation of T cells and the amount of mP3 on AML (R=0.4539, p<0.0001). In addition, AML expressing 91% mP3⁺ induced apoptosis of > 70% of the T cells at a ratio of 10 AML: 1 PBMC as assessed by uptake of aqua dye. The association between the amount of mP3 on AML and percent of apoptosis was significant (R=0.7852, p<0.0001), and apoptosis induced by mP3⁺AML appeared to be specific since T cells did not undergo apoptosis when anti-P3 mAb was added. The same correlation was not seen after PMNs from healthy donors (% mP3⁺: 62%±21.7, n=14) were co-incubated with PBMC. The percentage of cells undergoing apoptosis was less than 10%, regardless of the extent of PMN mP3 positivity. Of note, mP3 expression is significantly higher in bone marrow myeloid derived suppressor cells (MDSC) from leukemia patients compared to MDSC from healthy donors (79.4±5.23% (n=7), 22.4±11.55% (n=3), respectively; p= 0.0007). Because mP3 inhibited proliferation of T cells stimulated via the T cell receptor (TCR), i.e. anti-CD3/CD28, we compared effects of mP3 on ZAP70 and ERK signaling by phosphoflow cytometry. ZAP70 phosphorylation in CD8⁺ and CD4⁺ T cells was reduced by 78% and 80%, respectively, within 5 minutes of co-incubation with mP3⁺ PMNs compared with T cells stimulated with anti-CD3/CD28 mAbs alone, while ERK1/2 phosphorylation was completely blocked within 10 minutes, suggesting that P3-mediated inhibition of T cell proliferation involves downstream TCR signaling pathways. Taken together, these data support an important new function of membrane-bound P3 on PMN and leukemia in controlling adaptive T cell immunity.