The T4 dose used for this study was chosen based on preliminary experiments testing several different doses of T3 and T4. A consistent beneficial effect on LV function and myocyte remodeling with a slight (borderline significant) increase in heart rate was observed with this particular dose. This dose was well tolerated by the MI rats where only 1 animal in the treatment group died after 2 days of treatment during the entire experiment (consistent with a surgical-related death). In this study, eight weeks of T4 treatment after MI significantly increased serum T4 levels and tended to increase serum T3 levels, with a reversal of the α- and β-MHC gene expression patterns caused by MI. T4 treatment increased heart weight, especially LV weight, heart weight-body weight ratio, and LV ±dp/dt, while normalizing TAU. Myocytes isolated from the non-infarcted myocardium showed significant lengthening with MI, and T4 treatment enhanced myocyte cross-sectional area without further increased myocyte length. There was a significant increase (34% increase) in total LV tissue area, especially in the non-infarcted tissue area (41% increase), as well as an increase in the thickness of both infarcted and non-infarcted areas (30% and 36% increase, respectively) in MI rats treated with T4. The total length of smaller arterioles ranging from 5 to 15 μm tended to increase, and collagen deposition tended to decrease with T4 treatment in the LV non-infarcted area.
Many studies have shown that either short term or long term treatment with THs starting early or late after MI can improve LV function and gross measures of LV remodeling
[9, 10, 12, 23, 24]. The present study confirmed that 8 weeks of T4 treatment initiated immediately following MI can improve LV cardiac contractility and relaxation. Although no changes in LV dimension and wall thickness were appreciated with echocardiography in the present study, an increase in LV non-infarcted area thickness was found with T4 treatment in the MI rats by direct measurement of LV tissue cross-sections. In addition, myocytes isolated from LV non-infarcted myocardium were found to have an increase in cell volume after MI due exclusively to cell lengthening, which is consistent with our previous reports from rats and humans
[4, 5, 25]. An increase in cross-sectional area without further cell lengthening was found in myocytes isolated from the same area with T4 treatment, which further increased the cell volume. This selective growth in myocyte transverse area from TH treatment is similar to results from our previous study in Spontaneous Hypertensive Heart Failure rats
 but has never been reported in the MI model. Increased myocyte cross-sectional area contributed to increased wall thickness in the LV non-infarcted area with T4 treatment, a change that should reduce wall stress and improve LV function. It should be noted that T4 treatment led to only a 13% increase in cell volume (exclusively from increased cross-sectional area), which cannot account for the 34% increase in LV weight and the 41% increase in the non-infarct tissue area. Since the Coulter Channelyzer method offers precise measurements of isolated myocyte volume when high quality cells are assessed and whole tissue changes can also be precisely collected, these combined data indicate that more myocytes were present in the non-infarct area with T4 treatment as compared to untreated MI rats. This finding, that isolated myocyte volume accounted for only ~1/3 of the increase in muscle mass from TH treatment, was consistently observed in other preliminary experiments. We previously reported that TH treatment immediately following MI reduced the expression of markers of apoptosis in myocytes in the border area
. New data provided here suggest that the previously observed TH-mediated inhibition of apoptosis led to increased preservation of myocyte number in the border zone of the non-infarcted tissue area.
Small arterioles are very important for myocardium perfusion and oxygen supply. The pro-angiogenic effect of THs has been demonstrated in different models in which the growth of capillaries and arterioles were observed
[14, 15, 26]. A stepwise increase in the total length of smaller arterioles (5 to 15 μm) was found following MI and T4 treatment, indicating a parallel growth of small arterioles along with myocyte hypertrophy, which would help maintain blood supply to the hypertrophic heart.
Increased collagen deposition observed in the non-infarcted area here was also found by others
[27, 28]. Eight weeks T4 treatment following MI tended to decrease collagen deposition in the non-infarct area, which might contribute to the improvement of LV relaxation. The anti-fibrotic effect of TH has been documented in culture conditions as well as in TH-induced hypertrophic hearts which was mediated by either decreased collagen production or increased collagen degradation
[16, 29, 30]. The collagen content in the infarcted area was not measured in this study. However, we observed no increase in cardiac deaths in the T4 treated group during the early stage of post-MI remodeling, indicating that T4 treatment did not significantly interfere with the scar formation process.
Matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs) are critical in extracellular matrix remodeling by degrading certain components, regulating cell proliferation, migration, differentiation, and apoptosis as well as angiogenesis. More specifically, studies have shown that MMP-2 promotes angiogenesis by releasing angiogenic factors from the matrix, while TIMPs −1 to −3 can inhibit vascular endothelial cell migration and/or angiogenesis through releasing anti-angiogenic factors from matrix or interacting with angiogenic growth factor receptors
[31, 32]. With regard to cardiomyocyte hypertrophic remodeling, TIMPs −1 and −3 have been shown to exhibit an inhibitory effect
[33–35]. Little is known about the role of TIMP-4 on cardiac remodeling besides its inhibitory effects on MMPs. The expression of MMPs and TIMPs following MI has a temporal and spatial pattern, where in the post-MI late stage, MMP-2 expression in the remote and border zone varied in different studies, TIMPs −1 to −3 expression was reduced in both remote and border zone, and TIMP-4 was unchanged in the remote myocardium but decreased in the border area
[36–39]. In the present study, there was a tendency for increased MMP-2 and TIMPs −1 to −3 expression in the non-infarcted myocardium (includes both remote and border zone) late after MI. With T4 treatment, there was a tendency for further increase in MMP-2 expression but a decrease in TIMPs −1 to −4 expressions in the above-mentioned area. The difference in the findings of MMP-2 and TIMPs expression in these MI studies might be due to animal model differences and different methods of tissue sampling. No report can be found regarding the effects of THs on MMPs and TIMPs expression in the non-infarcted area during late stage post-MI LV remodeling. However, Ziegelhoffer-Mihalovicova et al. reported that T3-induced cardiac hypertrophy was not accompanied by cardiac fibrosis but an increase in MMP-2 and TIMP-2 expression
. Ghose Roy et al. found a reduction in collagen I and III in cardiac tissue with an increase in MMP-1 activity and a decrease in TIMP-3 and TIMP-4 expression in T3-induced cardiac hypertrophy
. Thus, the long-term effects of T4 on myocyte remodeling and arteriolar growth in the non-infarcted area following MI might also relate to its actions on the expression of MMPs and TIMPs, which requires additional investigation to verify.
A number of animal studies in recent years have demonstrated a decline in tissue T3 levels after MI due to increased cardiac expression of the D3 deiodinase, which converts T4 to rT3 and T3 to T2
[41, 42]. A report showing worse outcomes in post-MI patients with elevated serum rT3 levels suggests a similar process occurs in humans
. Based on this finding, one might expect T3, rather than T4, to be more effective in restoring cardiac tissue T3 levels post-MI. While the increase was not statistically significant, serum T3 levels were 32% higher in T4-MI rats compared to MI or sham groups. This trend suggests that peripheral conversion of T4 to T3 occurred as a result of elevated serum T4 and likely provided additional T3 for cardiac myocyte uptake.
Regulation of intracellular T3 is a complex process involving availability of free T3 in serum, as well as thyroid hormone membrane transporters and intracellular thyroid hormone binding proteins in the target cells. There is currently no information available regarding changes in thyroid membrane transporters and intracellular thyroid binding proteins in heart disease. Nonetheless, improvements in cardiac remodeling and function observed here suggest increased availability of T3 in the myocytes of treated rats. Now that important cardiac benefits of "supraphysiological" doses of THs post-MI have been demonstrated here and previously
, a critical question remains unanswered. Can the functional and remodeling benefits of T3 or T4 treatment of MI be safely and effectively implemented in patients without adverse effects?
TH signaling is a complex and poorly understood process involving genomic and non-genomic signaling mechanisms
. Genomic signaling involves T3 binding to thyroid nuclear receptors associated with the thyroid response element of targeted genes. Involvement of many thyroid receptor isoforms, co-activators, and co-repressors contribute to the complexity of genomic signaling. Many signaling pathways have been implicated in non-genomic TH signaling, including MAP kinases, PKC, and Akt. Work from our lab has identified a particularly important role for Akt signaling in myocyte and vascular remodeling