CD26 Truncation of GM-CSF and IL-3 Alters Their Functional Interactions

Abstract 1240

CD26 (DPPIV) is a dipeptidyl peptidase that functions by enzymatically cleaving the penultimate proline or alanine of proteins, resulting in functional alterations. The expression and activity of CD26 is relevant in many disease states including obesity and cancer. Modulation of CD26 activity has been shown to increase homing and engraftment of both cord blood and bone marrow cells after transplant indicating the impressive therapeutic potential of CD26 activity altering compounds. Due to its importance in disease states and their subsequent treatments, it is relevant to study how the activity of CD26 alters the functions of the molecules it cleaves, and subsequently their interactions with each other.

Mass spectrometry data from our laboratory has shown that CD26 can cleave the penultimate proline of GM-CSF resulting in the truncated form which has blunted signaling and function. Additional data has recently confirmed that CD26 can cleave IL-3 and results in its diminished function as well. Further, and more importantly, since GM-CSF and IL-3 are members of the IL-3 receptor family, and share a common receptor beta chain, we investigated if CD26 truncation of GM-CSF (TGM) or IL-3 (T3) could inhibit the functional activity of the full length (FL) alternate compound (i.e TGM inhibition of FL IL-3 activity or T3 inhibition of FL GM-CSF activity) in the TF-1 cell line and cord blood cells. We determined that both T3 and TGM could inhibit the colony formation induced by either FL GM-CSF or FL IL-3. This inhibition of function correlated with alterations in reactive oxygen species (ROS) levels that mimicked the truncated versions of either GM-CSF or IL-3 even in the presence of the full length molecules. Strikingly, this inhibition of colony formation did not require a 1:1 ratio of the full length to truncated. Rather, approximately 4–10 fold less truncated could be used to efficiently inhibit the colony formation activity of the full length, even across molecules. Interestingly, the ratio of T3 needed to block the full length GM-CSF (1.25ng/ml T3: 10ng/ml FL GM-CSF) was less than the amount of TGM needed to block the full length IL-3 (2.5ng/ml TGM: 10ng/ml FL IL-3) suggesting that T3 is better at blocking FL GM-CSF than TGM is at blocking FL IL-3. However, the ratios of truncated needed to block the function of self FL molecules are identical for both GM-CSF and IL-3 (1.25ng/ml truncated: 10ng/ml FL).

Signaling and receptor binding studies were performed for GM-CSF with TF-1 and CD34+cord blood cells, and showed that the truncated GM-CSF inhibited the Stat-5 and JAK2 signaling of FL GM-CSF at less than a 1:1 (10ng/ml FL: 1.25 ng/ml TGM) ratio. Receptor binding studies found that TGM bound to the GM-CSF receptor more efficiently than the FL form but concentrations required to produce 50% maximum inhibition of binding (IC50) is 8-fold lower for TGM compared to FL-GM-CSF, indicating that T-GM-CSF is a better competitor for binding, than is FL-GM-CSF itself suggesting that this may be how TGM is blocking the effects of FL GM-CSF, and potentially IL-3, in our model. Finally, cells treated with TGM had diminished respiratory and glycolytic rates compared to those treated with full-length cytokine. These data provide the first evidence of relevant interactions, with functional consequences, of the importance of full length and CD26 truncated cytokines across molecules.