In this study, microarrays were used to perform a genome-wide comparison of the effects of ROCK2 selective inhibitor SLx-2119 with the effects of atorvastatin in primary cultures of human EC, SMC, and fibroblasts. The study showed that, although SLx-2119 and atorvastatin induced altered expression of genes belonging to the same biological processes, individual genes affected by these drugs were mostly different. These results suggest that there may be synergistic effects between statins and ROCK2-inhibitors, and that a possible therapeutic benefit may be achieved by combining the 2 classes of drugs.
When interpreting effects of drugs on cells in culture, it is important to consider to what extent in vitro
concentrations are attainable in vivo
. In our laboratory, the investigation of pleiotropic effects of statins such as TM up-regulation [20
] and CTGF down-regulation [21
] are a major focus. Hence, we have shown previously that atorvastatin at a concentration of 10 µM consistently up-regulates TM gene and protein expression in cell cultures [20
]. Moreover, administration in vivo
of another HMG-CoA reductase inhibitor to mice at a nontoxic dose also caused increased TM gene expression levels in lung tissue (unpublished observation, 2008). In human subjects, peak plasma concentrations after oral administration of clinically relevant doses of atorvastatin, administered for its lipid-lowering effects (10 – 40 mg), are 0.5 – 10 nM [22
]. Peripheral tissue concentrations and intracellular concentrations of atorvastatin have, to our knowledge, not been described. In the current study SLx-2119 down-regulated Tsp-1 and CTGF transcripts but only at a concentration of 40 µM. Hence, the in vitro
microarray studies reported here were performed with concentrations of atorvastatin and SLx-2119 of 10 µM and 40 µM, respectively. These concentrations are likely higher than tissue and/or intracellular concentrations that can be achieved in vivo
. On the other hand, cells in vivo
are generally exposed to a drug for much longer periods of time than cells in culture. Because of these and other considerations, studies in animals and/or humans will ultimately be required to examine whether the synergistic effects of SLx-2119 and atorvastatin suggested by the present study apply to the in vivo
In radiometric assays, The IC50
value of SLx-2119 for inhibition of ROCK1 was 24 µM. Partial inhibition of ROCK1 may thus be expected at the concentration of SLx-2119 used in the microarray studies (40 µM). However, SLx-2119 is an ATP-competitive inhibitor of ROCK. Therefore, in cell based assays where intracellular levels of ATP are 10–100 fold higher than the concentration of ATP used in the radiometric assay, the potency of SLx-2119 for ROCK2 may be expected to be in the 1–10 µM range and for ROCK1
200 µM. Hence, at the concentrations used in this study, SLx-2119 selectively inhibits ROCK2.
An example of how ROCK2-inhibitors and statins may be complimentary to each other can be found in the effects of SLx-2119 and atorvastatin on genes involved in blood coagulation. Statins have known anti-fibrinolytic effects on vascular cells [24
]. Moreover, the current and former studies have shown that statins up-regulate expression and function of TM [12
]. TM is located on the luminal surface of EC in most normal blood vessels, where it forms a complex with thrombin. When complexed to TM, thrombin no longer cleaves fibrinogen to form fibrin and no longer activates cellular thrombin receptors, but instead activates protein C. Activated protein C counteracts the pro-coagulant, inflammatory, and fibroproliferative effects of thrombin. Recent studies have demonstrated that increased levels of TM attenuate inflammatory responses in a variety of settings, including endotoxin-induced tissue damage and atherosclerosis [25
]. Statins may be therapeutically beneficial in various disorders by up-regulating expression of TM. Interestingly, SLx-2119 did not affect expression of TM in any of the investigated cell types. However, this ROCK2-inhibitor could possibly enhance the therapeutic benefit of statins, by affecting the expression of other anti-fibroproliferative or anti-coagulant factors that are not affected by statins, such as urokinase plasminogen activator (PLAU) and urokinase plasminogen activator receptor (PLAUR). SLx-2119 does not alter prothrombin time (J.L. Ellis, unpublished observation).
Statins are also well-known to down-regulate genes coding for Tsp-1 and CTGF [18
]. These factors are important mediators in the transforming growth factor (TGF) pathway, as Tsp-1 is an activator of TGF and CTGF is one of the main downstream mediators of TGF. The TGF pathway plays a central role in many conditions that lead to fibrosis [29
]. Current microarray data show that SLx-2119 alters the expression of a plethora of genes that may be related to fibrosis and tissue remodeling, including genes coding for several collagen types and MMPs, in PASMC and NHDF. These genes were not altered by atorvastatin, again suggesting that the two classes of drugs may have synergistic effects. By altering CTGF and other fibroproliferative mediators, both statins and ROCK-inhibitors have indeed been shown to alter the fibrotic phenotype of SMC and fibroblasts [15
]. In this study, we compared the effects of SLx-2119 and atorvastatin on SMC isolated from radiation enteropathy biopsies. As shown previously [15
], these cells exhibit a profibrogenic phenotype characterized by increased formation of actin stress fibers and an increased expression of CTGF when compared to SMC isolated from normal intestine. Both atorvastatin and SLx-2119 reduced stress fiber formation and CTGF expression specifically in the RE-SMC. Similar results were found in EC in previous studies, in which ROCK-inhibitors and statins reduced stress fiber formation and endothelial permeability in response to irradiation or thrombin [33
Interestingly, N-SMC from intestine and normal PASMC differed in their response with regard to CTGF. Although this difference may be explained by differences in culture conditions and differences in origin between the two SMC types, different responses of different cells may also be caused by the many targets and diverse effects of Rho proteins. This was also found in a previous study, in which microarrays were used to investigate the effects of ROCK-inhibitor Y-27632 on three different human fibroblast phenotypes [35
]. Similar to the current study, the ROCK-inhibitor affected many genes involved in cell cycle regulation and cytokinesis. Furthermore, it was concluded that the ROCK-inhibitor had phenotype-specific effects on these cells. A similar difference was found on effects of Rho/ROCK inhibition on normal and fibrotic phenotypes of intestinal SMC in the current study.
As reported before for statins [36
], the different cell types in the current study differed greatly in their gene expression response to atorvastatin or SLx-2119, including the total number of genes affected. For example, atorvastatin had little effect on gene expression in fibroblasts whereas SLx-2119 had a major effect. This predominant response of SLx-2119 on fibroblasts likely parallels the role of ROCK2 in the fibrotic response [10
] and the potential for ROCK2-inhibitors to be beneficial in fibrotic disease.
In conclusion, the results of the current gene expression profiling study show that atorvastatin and the ROCK2-inhibitor SLx-2119 exhibit little overlap, but instead are mainly complimentary in their effects on gene expression in several primary human cell cultures. These data are consistent with a potential synergistic effect between statins and ROCK-inhibitors. It should also be noted that the ROCK pathway affects the function of several target proteins by posttranslation modification. Hence, more in vitro and in vivo studies are needed to further explore the possible therapeutic benefit of treatments in which ROCK-inhibitors and statins are combined.