Following our initial observation of an unexpected increase

Following our initial observation of an unexpected increase in the maximal rates of cGMP formation by NO-stimulated sGC in the presence of cytosolic PCA preparations [15], we attempted to purify, identify and characterize the responsible protein(s). The purification protocol described in the present study did not result in isolation of a single protein but yielded a protein mixture with major components corresponding to 40, 70, and 100 kDa on SDS-PAGE. LC-MS/MS analysis revealed that the final preparation of sGC-AF contained albumin, actin, annexin A6 and gelsolin. Except for albumin, these proteins are all related to or part of the cytoskeleton and contractile elements of smooth muscle [24], [25], [26], [27], [28], [29], [30]. None of the single proteins mimicked the effect of the partially purified preparations (not shown), indicating that sGC-AF is a protein complex rather than a single protein. In view of attempts to isolate sGC from blood platelets by one of us (B.M.) in the mid 1980s, the apparent modulation of sGC activity by cytoskeleton-related proteins is an intriguing finding. Platelets and smooth muscle cells share common mechanisms with respect to cGMP signaling and Ca2+ handling, including high expression levels of the α1β1 sGC heterodimer [31], [32]. In the second half of the 1980s we attempted to purify sGC from bovine platelets using a wide variety of methods, including the protocol applied for the successful purification of the enzyme from bovine lung [33]. However, we failed to obtain a pure enzyme, and the pooled fractions from final purification steps typically contained Fesoterodine Fumarate and gelsolin as major contaminating proteins (Mayer, B., Guthmann, F. and Böhme, E. unpublished). This observation was not further pursued and never published, but together with the present findings it might support the view that sGC is modulated by an interaction with contractile elements in smooth muscle cells and platelets. The various isoforms of actin are not only the basic framework of the cytoskeleton in all cells, but also constitute the main part of the thin filaments and, together with myosin (and other proteins), form the contractile apparatus of smooth muscle [24], [25]. Actin is also essentially involved in the shape change of activated platelets [34]. Gelsolin is an actin-binding and -severing protein, which regulates actin filament elongation or depolymerization [26], [27], [28]. Interestingly, the vasodilator-stimulated phosphoprotein (VASP), an established target of cGMP-dependent protein kinase that is associated with the cytoskeleton [35], was reported to stabilize actin filaments to the severing effect of gelsolin without interfering with gelsolin binding [36]. Annexins are a family of Ca2+- and phospholipid-binding proteins [29]. Among these, isoform A6 regulates membrane trafficking and signal transduction. Besides other functions, it appears to serve as an anchoring protein, linking actin filaments to the cell membrane [30] and to interact with small Ca2+-binding proteins of the S100 family [37], [38]. These proteins act as sensors for free cytosolic [Ca2+] and are involved in the regulation of a wide variety of cellular processes [39], [40]. In both smooth muscle cells and platelets, the sGC/cGMP pathway counteracts the effects of intracellular free Ca2+, resulting in relaxation and inhibition of aggregation, respectively, processes that essentially involve the cytoskeleton. Thus, it is likely that the interaction of sGC with the activating factor described in the present study is modulated by Ca2+ ions. It is certainly warranted to clarify this issue in future work. Our data indicate that sGC-AF increases maximal rates of cGMP formation by NO-saturated sGC without affecting basal sGC activity. The effect was independent of the NO donor (DEA/NO or GSNO) applied and was similar in magnitude when the enzyme was exposed to 0.3 and 10 µM DEA/NO. Thus, effects on the rate of NO release or NO consumption can be excluded, indicating that sGC-AF converts the NO-bound enzyme into a more active conformation. The increase in activity caused by sGC-AF was similar to that observed with saturating concentrations of the sGC stimulator BAY 41-2272, but their effects are clearly distinguishable. The pronounced, about 10-fold increase in basal activity caused by BAY 41-2272 [41] was not observed with sGC-AF. Moreover, the effects of the two agents were additive even though BAY 41-2272 was present at a saturating concentration of 10 µM [42], indicating that their interaction with NO-bound sGC involves distinct mechanisms. BAY 41-2272 supports CO activation of sGC by inducing the formation of a 5-coordinate heme-CO complex [43], but the physiological mechanism underlying stimulation of sGC by endogenous CO is unknown. Work is in progress in our laboratory to test the intriguing possibility that sGC-AF functions as endogenous sGC stimulator mediating CO activation of the enzyme in smooth muscle.