The present study demonstrates increased survival, enhanced infarct healing and attenuated LV remodeling after MI in transgenic mice with cardiac-restricted overexpression of CTGF. In accordance with the animal model, study of patients with ST-elevation MI demonstrated attenuated LV remodeling in the cohort that displayed increasing s-CTGF levels after MI. Thus, the study provides novel translational evidence of enhanced infarct healing and anti-remodeling activities of CTGF.
The area at risk following ligation of the left coronary artery was found to be similar in Tg-CTGF mice and NLC mice. This finding is congruent with previously reported data demonstrating that regional myocardial blood flow was similar in Tg-CTGF mice and NLC mice 
. Consistent with these findings permanent ligation of the left coronary artery in Tg-CTGF mice and NLC mice generated myocardial infarctions of similar size after 24 hours of ischemia. Thus, the salutary actions of CTGF in post-MI remodeling appear to rely on mechanisms that operate after this time point. First, based on the cardioprotective actions of CTGF due to its stimulation of the Akt/GSK-3β salvage kinase pathway in cardiac myocytes recently reported from our laboratory 
, CTGF might inhibit infarct expansion in the subacute phase after MI. The reduced occurrence of cellular apoptosis in the remote myocardium of Tg-CTGF mice after MI is consistent with this hypothesis. Secondly, the anti-inflammatory properties reported for CTGF may also attenuate pro-inflammatory responses activated by ischemic tissue necrosis and thus, limit infarct expansion 
. To what extent such anti-inflammatory actions of CTGF is mediated through the Akt/GSK-3 axis is currently unknown. The reduced contents of CD68-positive macrophages, and leucocytes (CD45-positive cells) in general, in the peri-infarct region of Tg-CTGF mice reported in the current study, also support a salutary anti-inflammatory role of CTGF. Consistent with these findings, plasma GDF-15 levels, a TGF-β superfamily cytokine and predictor of developing heart failure 
, as well as plasma CRP levels, were both lower in patients that responded with increasing serum CTGF levels after MI. However, whether the reduced contents of leucocytes 4 weeks after induction of MI, solely reflect anti-inflammatory actions of CTGF or to large extent also enhanced infarct healing and differentiation of scar tissue are yet to be resolved.
Another aspect of CTGF-mediated attenuation of myocardial remodeling is the potential for an effect of CTGF on proliferation of cardiac progenitors or cardiac stem cells, and ultimately, on regeneration of myocardial tissue. Although myocardial tissue contains resident progenitors or stems cells with some regenerative capability, the significance of myocardial regeneration in the natural cause of post-MI healing is limited 
. However, previous reports have indicated a paracrine factor-stimulated Akt/GSK-3β pathway in proliferation and survival of resident cardiac stem cells 
. Thus, increased tissue concentrations of paracrine factors that enhance the activity of the Akt/GSK-3β pathway, like that of CTGF, may increase stem cell activity and regeneration of myocardial tissue. In this study we demonstrate increased number of c-kit+
cells as well as increased number of cells in mitosis (Ki-67+
cells) in the peri-infarct region of Tg-CTGF mice versus that of NLC mice after MI. Of particular interest in this context is a recent report providing evidence of both proliferation of c-kit+
stem cells in the peri-infarct region after MI and contribution of these cells to new myocyte formation 
. A tantalizing interpretation of our data is that the increased contents of c-kit+
cells in the peri-infarct region of Tg-CTGF mice may represent increased cardiac stem cell activity. However, c-kit is a cell surface marker not only present on cardiac stem cells. C-kit is widely expressed on hematopoietic stem cells and inflammatory cells and mast cells of hematopoietic origin. On the other hand, the reduced numbers of CD68-positive macrophages and CD45-positive leucocytes (i.e. cells also expressing the surface marker c-kit) in the peri-infarct region of Tg-CTGF would not be consistent with the increased number c-kit+
cells in peri-infarct region being inflammatory cells. Thus, the increased number of c-kit+
cells in the peri-infarct region of Tg-CTGF mice may represent bona fide cardiac stem cell activity. Supporting a putative role of CTGF-mediated proliferation of cardiac stem cells/cardiac progenitors after MI are recent data demonstrating that CTGF stimulates proliferation of cardiosphere-derived cells in vitro
. However, the implications of the latter findings, i.e. to what extent CTGF may stimulate regeneration of myocardial tissue in vivo
still remain to be settled.
This study also indicates that reduced cellular apoptosis of the remote myocardium Tg-CTGF mice may contribute to reduced loss of myocardial tissue reflecting in decreased LV dilatation and cardiac myocyte hypertrophy. Indeed, Tg-CTGF mice disclosed reduced cross-sectional area of cardiac myocytes four weeks after MI. However, the latter finding could also be due to direct CTGF-engendered inhibition of myocardial hypertrophy through induction of anti-hypertrophic gene programs previously reported in Tg-CTGF mice 
. Thus, inhibition of myocardial hypertrophy per se
after MI may also contribute to attenuated LV remodeling. The slight increase of microvessel density in the peri-infarct region of Tg-CTGF mice 4 weeks after MI compared with that of NLC mice is consistent with a previous report from our laboratory, demonstrating increased microvessel densities in myocardial tissue of Tg-CTGF mice not subjected to myocardial infarction 
. The significance of such an increase in myocardial microvessel densities of Tg-CTGF mice is uncertain. However, despite the fact that myocardial blood flow of Tg-CTGF mice was not found to be statistically different from that of NLC mice 
, the increase of myocardial microvessel densities of Tg-CTGF mice may provide shorter distances of oxygen diffusion in the tissue, and thus, improved transport of oxygen to the cells. The latter would be particularly relevant under the increased workload of the remaining myocardial tissue after myocardial infarction and would also be consistent with the reduced incidence of cellular apoptosis in remote myocardial tissue of Tg-CTGF versus that of NLC mice reported in this study.
Mice with cardiac-restricted overexpression of CTGF displayed subtle increase of myocardial collagen contents compared with non-transgenic control mice 
. Furthermore, the CTGF-engendered increase of myocardial collagen contents was minor compared with that of non-ischemic myocardial tissue in ischemic heart failure. Indeed, the increase of myocardial collagen contents in ischemic heart failure was less in Tg-CTGF mice that that in non-transgenic control mice, consistent with the attenuated signals for LV remodeling under cardiac exposure to CTGF. An obvious interpretation of the data is that CTGF, even under the settings of ischemic heart failure, is not a major driver of myocardial fibrosis. Such an interpretation may seem at odds with reports demonstrating that myocardial CTGF expression is associated LV remodeling in heart failure 
. However, although myocardial CTGF expression may reflect pathologic LV remodeling and myocardial fibrosis, no studies have yet reported that CTGF is a principal driver of fibrosis. Thus, a function of CTGF as a marker of myocardial damage and remodeling in heart failure is not in contrast to a cardioprotective function of CTGF.
Increase of s-CTGF levels in patients after MI was associated with similar anti-remodeling effects as in Tg-CTGF mice. Despite the fact that relatively few patients were included, this study represents a homogenous cohort of patients with acute ST-elevation MI, due to single vessel thrombus, successfully revascularized by PCI. The use of repeated CMR examinations at discrete time-points allowed a precise description of both infarct healing and LV function. Although the s-CTGF levels were not statistically different at any time point during follow-up among the groups, the statistical analysis revealed significant intra-patient variation during infarct healing with the patient cohort segregating into two sub-groups; one in which s-CTGF levels increased after MI, and the other in which s-CTGF levels remained unaltered or decreased. These striking differences in the intra-patient responses of s-CTGF levels after MI allowed analyses of the putative relations of quantitative parameters of LV remodeling and function to that of the course of s-CTGF levels after MI. Interestingly, only patients who responded with increase in s-CTGF after MI displayed attenuated LV remodeling and improved recovery of LV function. The wide inter-patient variations in circulating CTGF levels are consistent with previous reports and could have multiple explanations 
. On the other hand, the distinct intra-patient responses of s-CTGF levels after MI could be due to single nucleotide polymorphisms in the CTGF gene. Indeed, previous reports have disclosed single nucleotide polymorphisms in the promoter region of CTGF at binding sites for transcription factors 
. Although s-CTGF levels were not studied in relation to these specific polymorphisms, patients homozygous for the G allele at -945 exhibited loss of Sp1 binding to the promoter and lack of Sp1-mediated repression of the promoter 
. Thus, the striking differences in s-CTGF responses to MI and post-MI remodeling in patients may be due to polymorphisms in the promoter of CTGF.
In conclusion, the congruent findings of attenuated LV remodeling after MI in mice with cardiac-restricted overexpression of CTGF as well as in patients that respond with increased s-CTGF levels after MI, support a beneficial role of CTGF in LV remodeling and functional recovery of the heart after MI mediated by attenuation of inflammatory responses and inhibition of apoptosis.