Our observations with RBC in human platelets suggested
Our observations with RBC8 in human platelets suggested a more wide-ranging role for Rals in platelet function compared to our observations in Ral deficient mouse platelets. Using lumi-aggregometry, 10 μM RBC8 significantly reduced platelet aggregation and ATP secretion responses in both WT and DKO platelets using the threshold concentration of CRP (0.6 μg/mL), while 3–10 μM RBC8 also significantly reduced ATP release (Fig. 2A and B). Further investigations using FACS analysis to assess integrin activation (Fig. 2C) revealed almost identical, dose-dependent reductions in both WT and Ral DKO platelets with RBC8 treatment using either CRP or PAR4-AP as agonist. Here, inhibitory responses with RBC8 were more sensitive to lower (1 μM) concentrations of Quinacrine hydrochloride hydrate compared to the aggregation/dense granule secretion assay. The reduction in CRP-mediated P-selectin exposure in WT platelets with RBC8 appeared to be dose-dependent, but 10 μM RBC8 was required to significantly suppress P-selectin to levels observed in Ral DKO platelets in the absence (or presence) of RBC8 (Fig. 2D). Furthermore, RBC8 could significantly suppress WT platelet-leukocyte aggregation formation, an effect principally mediated by platelet P-selectin interaction with PSGL-1 on leukocytes (Supplementary Fig. 1) . We had previously demonstrated that Ral DKO platelets have a near complete ablation of CRP-mediated platelet-leukocyte interaction, making it challenging to determine off-target effects of RBC8 with this assay . Using PAR4-AP as agonist, RBC8 appeared less potent at reducing P-selectin levels in WT platelets, although 10 μM RBC8 did significantly decrease the response. However, 10 μM RBC8 did also significantly suppress PAR4-AP-mediated P-selectin exposure in Ral DKO platelets (Fig. 2D). Overall, our data show that RBC8 elicits off-target effects in mouse platelets as evidenced by numerous inhibitory effects in Ral DKO platelets (Fig. 2A-D). It is therefore possible that similar off-target effects exist for RBC8 in human platelets, however we cannot definitively say that this is the case since the differences in our data may just reflect fundamental differences in Ral function between human and mouse platelets. For instance, Rals may have a more critical role in regulating human platelet dense secretion, as reported by Kawato et al., whereas our observations in Ral deficient mouse platelets suggest a very weak role for Rals in dense granule release, which did not alter platelet aggregation or integrin activation responses [16,17]. If such a difference between species were true, it would help explain why inhibition of human Rals (with RBC8) have a more profound effect on human platelet activation responses, which are critically reliant on secreted ADP amplification signals. Also, compensatory upregulation of specific signalling pathways have been previously reported in transgenic mice and therefore it cannot be excluded that similar issues are present in Ral DKO transgenic mice that could potentially mask Ral specific functions in platelets . However, even at 1 μM RBC8, which has weak inhibitory effects on Ral activation (Fig. 1Aii, approximately 10% inhibition), we observed significant effects of the compound on CRP- and/or PAR4-AP-induced human platelet integrin activation and P-selectin exposure (Fig. 1D and E). While our experiments suggest that RBC8 is targeting signalling component(s) other than Rals in mouse platelets, it is not clear what those target(s) are likely to be. In the Yan paper which identified RBC8 as a Ral inhibitor, the compound showed no off-target activity towards Ras or RhoA, both of which are activated in response to platelet stimulation [39,40]. The GTPase Rac1 has been shown to be important specifically for GPVI-mediated platelet responses, but Rac1 deficient platelets have defective Ca2+ mobilisation and RBC8 did not alter CRP-mediated Ca2+ signalling responses (Fig. 1H) . Our observations suggest the target(s) is likely to be a Ca2+ sensitive component of platelet signalling pathways that is critical for integrin activation and dense granule secretion, the latter being reinforced by our observations that exogenous ADP could largely recover the platelet aggregation defects with RBC8 treatment (Fig. 1B). Based on this, we suspected the Rap1 isoforms, Rap1a and Rap1b, as likely candidates. Like Rals, they are members of the Ras family of GTPases and are specifically regulated by the calcium (and DAG) sensitive guanine nucleotide exchange factor (GEF), CalDAG GEF1, and are critical regulators of integrin activation and platelet secretory responses [6,42]. However, we did not observe any inhibitory effect of RBC8 (between 1 and 10 μM) on CRP-induced Rap1 activation suggesting the off-target effects are not mediated by Rap1 (Supplementary Fig. 2). We are therefore currently uncertain of the Ral-independent mechanism of RBC8 in platelets.