Keeping the Heart Fitm2 during Chemotherapy

Main Text

With increasing cancer survival rates due to improved treatment options, attention to the chronic sequelae of chemotherapy is becoming increasingly more relevant. Cardiotoxicity is an important complication of several cancer therapeutic agents, but, so far, approaches to reduce cardiotoxicity have shown limited success. In this issue of Molecular Therapy, Gupta et al.¹ report a potential new strategy to counteract doxorubicin-induced myocardial atrophy and apoptosis: treatment with the pro-hypertrophic miR-132-212 cluster. The results unveil a new pathophysiological aspect of doxorubicin-induced cardiotoxicity and may provide a promising starting point for the development of new therapies.

Anthracycline drugs, such as doxorubicin, are highly effective, broad-spectrum anti-cancer agents, but cardiac side effects pose a major limitation to their use. These agents can cause severe cardiomyocyte damage, leading to degradation of the sarcomere, swelling of mitochondria, vacuolar degradation of the sarcoplasmic reticulum, and, eventually, cardiomyocyte death. Cell death may also occur in cell types other than cardiomyocytes, potentially limiting the adaptive capacity of the heart and thereby increasing the risk of future cardiovascular disease and toxicity with future administration of cardiotoxic drugs.² Together, these effects result in symptomatic heart failure in 2%–4% of patients treated with anthracyclines and an asymptomatic reduction of left ventricular ejection fraction (LVEF) in 9%–11% of patients.³ While anthracycline-induced cardiotoxicity (AIC) was initially believed to comprise both an early and a late component, recent data by Cardinale et al.⁴ indicated that 98% of patients with an asymptomatic decline in LVEF already showed a decrease at 12 months after chemotherapy. This suggests that most of the cardiotoxicity caused by anthracyclines is established rather soon after the onset of treatment.

Despite extensive research, the mechanisms underlying AIC have not been fully elucidated. A potential explanation for the cardiotocixity is linked to the known function of anthracyclines. Anthracyclines cause their intended cytotoxic effect in cancer cells by binding to topoisomerase 2 (Top2). Top2 is an enzyme that is important for the organization of DNA and exists in an α- and β-isoform in humans. Top2α expression is co-regulated with the cell cycle and peaks in G2/M phase. Binding of anthracyclines to Top2α inhibits DNA replication, thereby arresting the cell cycle in G1/G2 and leading to apoptosis.³ However, Top2β is also expressed in non-proliferating cells, such as cardiomyocytes, where binding of anthracyclines leads to the suppression of peroxisome proliferator-activated receptors (PPARs), resulting in mitochondrial dysfunction and eventually cell death.⁵ Additionally, it has been shown that anthracyclines can downregulate transcription factors, such as GATA4, that are essential for sarcomere maintenance.⁶ Another potential underlying mechanism for AIC is through the production of reactive oxygen species (ROS). Anthracyclines interact with ferric iron to convert molecular oxygen into superoxide radicals, but can additionally interfere with enzymes in the mitochondrial respiratory chain, leading to the production of ROS. The ROS then damage cellular components that can ultimately lead to apoptosis.²

So far, no effective prevention or therapy has been found for AIC. A 2016 position paper by the European Society of Cardiology⁷ suggested limiting cardiotoxicity by minimizing the total dose of anthracyclines, using less cardiotoxic doxorubicin analogs or formulations, or employing a non-anthracycline regimen. A cardiac protective treatment option, the use of the iron chelator dexrazoxane, has shown some therapeutic benefit.⁸ Some clinical trials also suggest potential benefits of the use of ACE-inhibitors and/or beta-blockers,9, 10 but further studies are currently ongoing.

In recent years, research has focused on the function of non-coding RNAs (ncRNAs) in AIC. microRNAs (miRs) are short strands of RNA that influence the translation of mRNA into protein by causing either translational repression or degradation of mRNA.¹¹ miR-208a was shown to be upregulated by doxorubicin, and inhibition of this miR resulted in a reduction in apoptosis through the derepression of Gata4 and Bcl2.¹² In a similar fashion, the miR-30 family was found to be downregulated in response to doxorubicin, while overexpression of miR-30e proved beneficial by reducing apoptosis and ROS production.¹³

In the current study, Gupta et al.¹ hypothesized that overexpression of pro-hypertrophic miRs would be able to counteract AIC-induced atrophy. They show that doxorubicin causes apoptosis and a reduction in cell size in isolated neonatal rat ventricular myocytes (NRVMs). While miR-132 and miR-212 levels do not change during doxorubicin-induced cardiomyocyte toxicity, cotreatment with pre-miR-132 or pre-miR-212 protected the cells against these effects. And while in vivo AAV9-mediated overexpression of the cluster, as expected, resulted in cardiac hypertrophy in doxorubicin-naive animals,¹⁴ viral overexpression of miR-132-212 during doxorubicin treatment actually ameliorated the doxorubicin-induced reduction in ejection fraction, remodeling, myofibril loss, and apoptosis.

To elucidate the mechanism behind the cardioprotective effect of the miR-132-212 on doxorubicin-treated hearts, the authors analyzed the cardiac transcriptomes of doxorubicin-treated mice with or without viral overexpression of miR-132-212. Among the differentially regulated genes, cell-death related processes were enriched and the gene fat storage-inducing transmembrane protein 2 (Fitm2) was identified as a potential direct target of miR-132 and miR-212. Follow-up in vitro experiments showed that overexpression of Fitm2 was able to block the protective effects of miR-132 or miR-212 during doxorubicin-induced cardiomyocyte toxicity. Based on these data, the authors conclude that increasing the miR-132-212 cluster might be a good therapeutic means for reducing AIC and that these effects are at least partially due to a decrease in Fitm2.

Fitm2 is known to be involved in the formation of cytosolic lipid droplets and has been shown to have an important role in skeletal muscle energy metabolism.¹⁵ Since it has not been studied for its function in cardiomyocyte biology, future studies are required to examine how its decrease would trigger hypertrophy or block AIC. As this gene was not found to be upregulated by the miR-132-212 cluster during doxorubicin treatment in vivo, a logical explanation might be that additional mechanisms are at play, which can explain the observed cardioprotective effects.

Another important biological question that remains relates to the long-term effect of blocking doxorubicin-induced cardiomyocyte apoptosis. Since a major cause of cardiomyocyte loss in AIC is apoptosis in response to intracellular damage, inhibition of this process by a transient early increase in miR-132/212 might be beneficial, but long-term expression might be detrimental since damaged cells are not removed. In addition, the fact that long-term overexpression of the miR-132/212 cluster under baseline conditions previously was shown to result in heart failure could underscore a level of caution regarding a long-term increase in the miRs.¹⁴

From a therapeutic perspective, some hurdles must be overcome before increasing levels of miR-132 and -212 can be considered a viable option for treating AIC. Increasing miR levels is possible through oligonucleotide therapy (mimicry), but targeting a specific cell type or organ is difficult. and the stability of these compounds is limited because stabilizing modifications generally preclude incorporation into RNA-induced silencing complex (RISC).¹⁶ Viral gene therapy circumvents the stability and delivery issues, but has the risk of the patient having or developing immunity to the virus being used.¹⁷ Additionally, even when local delivery could be achieved, sustained overexpression of miRs could have all types of unwanted off-target effects, since each miR is predicted to regulate multiple (sometimes unrelated) mRNA targets.

Currently, to avoid AIC, clinicians limit anthracycline dose, use less toxic formulations, or avoid anthracyclines altogether, even though they are very potent drugs. Therefore, there is a dire need for approaches to circumvent cardiotoxicity during anthracycline regimens. The use of a prohypertrophic miR cluster to counter the atrophy in AIC is an intriguing choice. It will be very interesting to see what future studies will reveal regarding the durability of this protective effect as well as the mechanism underlying these effects. Will it turn out that Fitm2 fits in with the ROS paradigm because it influences ROS production by influencing cardiomyocyte energy metabolism or will a completely new mechanism of cardioprotection be revealed? Regardless of what the final outcomes will be, the work of Gupta et al.¹ presented here paves the way for a new avenue of promising research into AIC.

Conflicts of Interest

E.v.R. is a scientific co-founder and member of the Scientific Advisory Board of miRagen Therapeutics, Inc.