Inhibition of Proto-Oncogene c-Src Tyrosine Kinase

Abstract

Cellular-Src (c-Src) encodes a plasma membrane-associated tyrosine protein kinase, which plays a vital role in signaling pathways related to cellular development and carcinogenesis (1,2). It was the first proto-oncogene to be described and is the cellular homologue in humans of the viral oncogene of Rous sarcoma virus, the chicken tumor virus discovered by Peyton Rous in 1911 (3). More recently, c-Src has been implicated in connexin43 (Cx43) remodeling in epicardial border zone myocytes following myocardial infarction (MI) (4).

Cx43 is the major ventricular gap junction protein in mammalian hearts, and the intercellular coupling mediated by these gap junctions allow for normal electrical conduction (5–7). Cx43 interacts with a scaffolding protein zonula-occludens-1 (ZO-1) to localize at intercalated disks (8,9). After a MI, phosphorylation of c-Src leads to its activation (p-Src), which then disrupts the ZO-1/Cx43 interaction (4,10). Displacement of the Z0-1/Cx43 interaction can cause Cx43 relocalization from the intercalated disks to myocyte lateral membranes (4,11). Cx43 lateralization in the infarct border zone is associated with altered gap junction coupling and is believed to be an important mechanism underpinning electrical conduction slowing and the development of reentrant arrhythmias (12).

Sovari et al. (11) have previously shown that 1-(1,1-dimethylethyl)-1-(4-methylphenyl)-1H-pyrazolo[3,4-d] pyrimidin-4-amine (PP1), a c-Src inhibitor, can increase Cx43 levels at gap junctions, improve gap junction function, and reduce inducibility of ventricular tachycardia (VT) in angiotensin-converting enzyme overexpression (ACE8/9) mice. Activation of the renin-angiotensin system in these mice led to c-Src up-regulation, loss of Cx43 and gap function coupling, and arrhythmogenesis; these processes were mitigated by inhibition of c-Src (11) (Fig. 1).

Figure 1. Scheme of c-Src Kinase Pathway Modulating Cx43.

Angiotensin II (Ang II) activates and up-regulates c-Src, which in turn causes a dysregulation and degradation of connexin43 (Cx43) with impaired gap junction function. PP1 inhibits c-Src kinase and interrupts Ang II–mediated Cx43 reduction, myocyte uncoupling, and sudden arrhythmic death. ATR-1 = angiotensin II type 1 receptor; RAS = renin-angiotensin system; VF = ventricular fibrillation; VT = ventricular tachycardia; ZO = zonula-occludens. Reproduced with permission from Sovari et al. (11).

In this issue of the Journal, Rutledge et al. (13) expand on these prior findings by applying our current understanding of post-MI Cx43 pathophysiology and hypothesizing that inhibition of c-Src activation can improve Cx43 levels and thereby ameliorate arrhythmia inducibility following MI. Mice undergoing MI after coronary artery ligation were treated with c-Src inhibitors PP1 and AZD0530, which is currently in clinical development and has been used to treat human cancers (14,15). The authors showed that c-Src inhibition moderated Cx43 down-regulation after MI compared with infarcts in mice treated with an inactive analogue or with saline. In addition, the restoration in Cx43 levels in mice treated with c-Src inhibitors was correlated with observed findings of lower p-Src levels, enhanced conduction velocity, and reduced arrhythmia inducibility in these MI mice. Interestingly, the improved Cx43 levels did not appear to be due to redistribution of Cx43, but rather to PP1 effects on increased Cx43 transcription as well as decreased Cx43 degradation. Furthermore, the authors’ results suggest that the underlying post-MI mechanisms leading to reduced Cx43 levels and c-Src activation are similarly triggered throughout the myocardium, rather than being contained only to peri-infarct tissue.

Rutledge et al. (13) should be commended for continuing to elucidate a novel approach toward a potential new antiarrhythmic therapy via inhibition of c-Src. They have added to a growing literature of studies seeking to modulate gap junction remodeling in order to decrease post-infarct arrhythmias. These studies include targeted Cx43 gene transfer to a porcine infarct model (16), engraftment of Cx43-expressing embryonic cardiomyocytes post-infarct (17), use of growth hormone-releasing peptide to modulate Cx43 levels (18), and use of heavy ion radiation to up-regulate Cx43 levels (19,20).

However, further investigation will be necessary. Notably, post-translational phosphorylation is critical to almost all aspects of Cx43 regulation and gap junction function (21,22). The authors did find that c-Src inhibition increased slowly migrating forms of Cx43, which are felt to be related to phosphorylated Cx43 that are associated with enhanced gap junction function (23,24). Nevertheless, further studies are needed to understand the precise mechanisms and sites of Cx43 phosphorylation, and how c-Src inhibition relates to these different forms of Cx43. In addition, as the authors point out, although they found a beneficial outcome by using c-Src inhibitors after 2 weeks post-MI, the effects of c-Src inhibition immediately after an ischemic event has not be examined and, in fact, could be harmful. There is data to suggest that Cx43 down-regulation after an MI may be adaptive because restricting gap junction communication may prevent the spread of deleterious by-products of ischemia and thereby limit infarct size (25). Hence, the timing of c-Src inhibition is important; furthermore, duration of treatment will need to be determined because cardiac Cx43 has a relatively short half-life of about 1 to 2 h (26,27), and the substrate for arrhythmia may persist beyond the 2-week period after an MI. In addition, there are concerns about QT prolongation with this class of tyrosine kinase inhibitors (28), and it is not uncommon for a purportedly antiarrhythmic therapy to have pro-arrhythmic effects instead. Hence, more studies are needed to better understand these potential side effects due to tyrosine kinase inhibitors, including the risks of left ventricular dysfunction and hypertension (both systemic and pulmonary) (29,30). The long-term outcomes and possible cardiovascular side effects of this class of drugs are therefore still unknown for an already potentially vulnerable patient population. Finally, the clinical applicability of a mouse model has its limitations such that the role of c-Src inhibition in a larger animal infarct model is necessary.

In conclusion, c-Src and its importance in tumorigenesis have been widely studied. The findings of Rutledge et al. (13) have provided further insight into the role of c-Src inhibition and arrhythmogenesis. Furthermore, although PP1 is not used clinically, AZD0530 has been studied in several phase 1 and 2 human cancer trials and is therefore closer to clinical application (31–33). The analogy to tumorigenesis and cancer therapeutics may be appropriately fitting. Ventricular arrhythmias are believed to result from multiple factors and stochastic events; thus far, the mainstay treatments have involved relatively blunt tools including internal defibrillators, radiofrequency ablation of infarct border zones, and antiarrhythmic drug therapies nonspecifically targeting various ion channels. Similarly, until the advent of mechanistic-driven therapeutics, oncologic treatments have tended to involve surgical resection, radiation therapy, and relatively nondiscerning chemotherapy that can collaterally affect normal tissue while not fully targeting the underlying cause(s). However, over the last decades, advances in the understanding of the underlying molecular processes involved in tumorigenesis have contributed to more specific and precise therapeutics for a variety of cancers. Analogously, although it is not clear that c-Src inhibition or even modulating Cx43 after an infarct will be the key to reducing arrhythmias, studies such as this from Rutledge et al. (13) seeking mechanism-based therapies for ventricular arrhythmias provide further optimism that we are one step closer to an effective antiarrhythmic treatment.