Author + information
- Received September 30, 2009
- Revision received January 13, 2010
- Accepted January 19, 2010
- Published online October 19, 2010.
- Hanqiao Zheng, MD, PhD,
- Mingxin Tang, MD,
- Qingwen Zheng, MD,
- Asangi R.K. Kumarapeli, MD, PhD,
- Kathleen M. Horak, BS,
- Zongwen Tian, PhD and
- Xuejun Wang, MD, PhD⁎ ()
- ↵⁎Reprints requests and correspondence:
Dr. Xuejun Wang, Division of Basic Biomedical Sciences, Sanford School of Medicine, University of South Dakota, 414 East Clark Street, Vermillion, South Dakota 57069
Objectives The goal of this pre-clinical study was to assess the therapeutic efficacy of doxycycline (Doxy) for desmin-related cardiomyopathy (DRC) and to elucidate the potential mechanisms involved.
Background DRC, exemplifying cardiac proteinopathy, is characterized by intrasarcoplasmic protein aggregation and cardiac insufficiency. No effective treatment for DRC is available presently. Doxy was shown to attenuate aberrant intranuclear aggregation and toxicity of misfolded proteins in noncardiac cells and animal models of other proteinopathies.
Methods Mice and cultured neonatal rat cardiomyocytes with transgenic (TG) expression of a human DRC-linked missense mutation R120G of αB-crystallin (CryABR120G) were used for testing the effect of Doxy. Doxy was administered via drinking water (6 mg/ml) initiated at 8 or 16 weeks of age.
Results Doxy treatment initiated at 16 weeks of age significantly delayed the premature death of CryABR120GTG mice, with a median lifespan of 30.4 weeks (placebo group, 25 weeks; p < 0.01). In another cohort of CryABR120GTG mice, Doxy treatment initiated at 8 weeks of age significantly attenuated cardiac hypertrophy in 1 month. Further investigation revealed that Doxy significantly reduced the abundance of CryAB-positive microscopic aggregates, detergent-resistant CryAB oligomers, and total ubiquitinated proteins in CryABR120GTG hearts. In cell culture, Doxy treatment dose-dependently suppressed the formation of both microscopic protein aggregates and detergent-resistant soluble CryABR120Goligomers and reversed the up-regulation of p62 protein induced by adenovirus-mediated CryABR120Gexpression.
Conclusions Doxy suppresses CryABR120G-induced aberrant protein aggregation in cardiomyocytes and prolongs CryABR120G-based DRC mouse survival.
Desmin-related myopathy, a well-characterized example of proteinopathy, features the presence of desmin-positive protein aggregates in myocytes. Genetic studies linked this disease to mutations in desmin, αB-crystallin (CryAB), and myotilin genes (1). Among these mutations, missense mutation R120G of CryAB (CryABR120G) is the best studied. Transgenic (TG) overexpression of either mouse or human CryABR120Gin mouse hearts causes aberrant protein aggregation and cardiomyopathy, recapitulating key features of human desmin-related cardiomyopathy (DRC) (2,3). Recent studies show that intrasarcoplasmic amyloidosis, a major type of aberrant protein aggregation in DRC and associated cardiomyopathy, are reversible on suppression of CryABR120Gexpression and, more remarkably, significantly attenuated by voluntary exercise (4,5). Proteasome proteolytic function is severely impaired in CryABR120GTG mouse hearts, and aberrant protein aggregation seems to be both responsible for and further exacerbated by proteasome functional insufficiency, forming a vicious cycle (1,6).
It is believed that abnormal protein aggregation and accumulation are deleterious in all proteinopathy, regardless of the primary cause. Notably, aberrant protein aggregation in the form of pre-amyloid oligomers has been observed in most failing human hearts, resulting from either dilated or hypertrophic cardiomyopathy (4). Moreover, aberrant protein aggregation recently was shown to trigger autophagic activation in pressure-overloaded hearts (7). To this end, the well-documented CryABR120GDRC mice represent a useful animal model for the investigation into the pathogenic role of cardiac aberrant protein aggregation as well as for therapeutic targeting of aberrant protein aggregation in congestive heart failure.
Doxycycline (Doxy) is a Food and Drug Administration-approved second-generation antibiotic of the tetracycline family. It is suitable for long-term use because of its favorable safety profile. It was demonstrated that Doxy has other important pharmacologic actions besides its antibiotic properties. For example, Doxy has been shown to be an inhibitor for matrix metalloproteinases (MMPs) (8). It also was reported that Doxy inhibits the formation of amyloid aggregates both in vitro and in vivo (9). Furthermore, some investigators (9–11) have reported that Doxy attenuated and delayed toxicity of oculopharyngeal muscular dystrophy possibly by reducing aggregation and inhibiting cell death pathways. These important recent discoveries prompted us to test whether Doxy has therapeutic value in cardiac proteinopathies. Our study revealed that Doxy significantly attenuates CryABR120G-induced aberrant protein aggregation in cardiomyocytes and prolongs the survival of CryABR120GDRC mice, providing compelling evidence that Doxy is a promising candidate for a clinical trial to treat cardiac proteinopathies.
The FVB/N inbred strain stable TG mice with cardiomyocyte-restricted overexpression of the mouse CryABR120Gwere used in this study (2). Animal use and care protocols used in this study were approved by the Institutional Committee for the Use and Care of Animals of the University of South Dakota.
Administration of Doxy and echocardiography
Doxy (Sigma-Aldrich Corp., St. Louis, Missouri) was given in drinking water (6 mg/ml) containing 5% sucrose, starting at 8 or 16 weeks of age. The control group was given drinking water containing 5% sucrose without Doxy. Echocardiography was performed as described (12).
Neonatal rat cardiomyocyte cultures and adenovirus infection
Neonatal rat cardiomyocytes (NRCMs) were isolated and cultured as described (6,12). Recombinant adenoviruses expressing hemagglutinin (HA)-tagged CryABR102G(Ad-HA-CryABR120G) or recombinant adenoviruses expressing β-galactosidase (Ad-β-Gal) were created as described (6). The viruses were used at a multiplicity of infection of 10 to infect the cultured NRCMs.
Immunofluorescence confocal microscopy and Western blot analyses
Sample preparation, immunofluorescence staining, and Western blot analyses were performed as described (6,12). The antibodies used include the rabbit polyclonal antibodies against CryAB (Stressgen, Victoria, British Columbia, Canada), Atg5 (Novus Biologicals, Littleton, Colorado), ubiquitin, Atg7 (Sigma-Aldrich Corp.), and the mouse antibodies against HA-tag (Santa Cruz Biotechnology, Santa Cruz, California) and sarcomeric a-actinin (Sigma-Aldrich Corp.), LC3 (MBL International, Woburn, Massachusetts), beclin-1 (Santa Cruz Biotechnology), Alexa Fluor 488 antirabbit Ig, Alex-Fluor 568 antimouse Ig (Invitrogen, Eugene, Oregon), and horse-radish peroxidase-conjugated antimouse or antirabbit secondary antibodies (Santa Cruz Biotechnology). Alexa Fluor 568 conjugated phalloidin (Invitrogen, Eugene, Oregon) was used to stain F-actin.
This assay was performed as previously described (4,6).
The log-rank test was used for the Kaplan-Meier survival analysis. All quantitative data are presented as mean ± SD and were analyzed by 1-factor or multiple-factor analyses of variance using SigmaStat software version 3.0 (Systat, Point Richmond, California), where applicable. The Holm-Sidak test was used for post-hoc pairwise comparisons. A p value <0.05 was considered statistically significant.
Doxy treatment significantly prolongs survival of mice with CryABR120GDRC
CryABR120GTG mice (line 134) develop concentric cardiac hypertrophy and diastolic malfunction at 3 months, display overt congestive heart failure between 5 and 6 months, and die shortly afterward (2). Because of the expedited course of disease progression in this TG line, we used it to perform a Kaplan-Meier survival analysis on chronic Doxy treatment.
A cohort of 37 CryABR120GTG mice was divided randomly into 2 groups: the Doxy (TG Doxy group; 19 mice) and the placebo (TG control [CTL] group; 18 mice) groups. A parallel cohort of nontransgenic (NTG) littermates was treated similarly to detect any potential adverse effects of Doxy or placebo on normal animals. Echocardiography was performed the day before the initiation of treatment. The treatment was initiated at 16 weeks of age when concentric cardiac hypertrophy and significantly decreased cardiac output were evident (Table 1).Mice of the TG CTL group showed a median lifespan of 25 weeks, similar to what was observed previously in the untreated TG mice (2). However, the premature death was significantly delayed in the TG Doxy group, with 60% of them still alive by the time all TG CTL mice had died. Their median lifespan was 30.4 weeks, 20.16% longer than that of the TG CTL group (p < 0.01) (Fig. 1).
The treatment to the NTG cohort was terminated when all mice in the TG cohort died. No difference in animal death was observed between the Doxy- and placebo-treated NTG mice (data not shown).
Doxy treatment attenuates cardiac hypertrophy in DRC mice
To evaluate the effects of Doxy on cardiac hypertrophy and left ventricular function in DRC mice at an earlier time, another cohort of NTG and TG mice was subjected to 4 weeks of Doxy treatment starting at 8 weeks of age. Previous characterization of this mouse line revealed that cardiac hypertrophy started between 1 and 3 months of age and the down-regulation of α-myosin heavy chain and CryABR120Gexpression was observed at 6 months (2). Hence, choosing the period between 2 and 3 months avoids the potential impact from the down-regulation of TG expression. At 4 weeks after Doxy treatment, echocardiography assessments revealed significantly increased thickness in the left ventricle posterior wall at the end of both diastole and systole in the TG CTL mice, but the increases were prevented or were attenuated significantly in the TG Doxy group (Table 2).These indicate that a 4-week Doxy treatment is sufficient to suppress cardiac hypertrophy. The increase in ventricular wall thickness in CryABR120GTG CTL mice was accompanied by decreased internal diameter at end-diastole and -systole and elevated ejection fraction and fractional shortening. Doxy treatment did not significantly alter ejection fraction, fractional shortening, or the end-diastolic left ventricle internal diameter in the TG mice (Table 2). The changes in cardiac mass assessed by echocardiography were confirmed by gravimetric measurements at terminal experiments. The heart weight-to-tibial length ratio and the ventricular weight-to-tibial length ratio were significantly lower in the TG Doxy group than in the TG CTL group (Fig. 2).
Doxy treatment reduces aberrant protein aggregation in DRC mouse hearts
Because protein aggregation is an important pathogenic process in DRC and Doxy has been shown to suppress intranuclear protein aggregation in nonmyocytes, we further tested whether Doxy's protection against DRC was associated with any effect on aberrant protein aggregation in the heart.
Compared with the TG CTL group, TG Doxy hearts showed substantially less CryAB-positive protein aggregates by immunofluorescence confocal microscopy (Fig. 3),significantly less sodium dodecyl sulfate (SDS)-resistant CryAB oligomers by filter-trap assays (Fig. 4B),and markedly less ubiquitinated proteins (Fig. 4C). Doxy treatment did not change total CryAB protein levels (Fig. 4A).
Doxy dose-dependently inhibits aberrant protein aggregation induced by CryABR120Gin cultured NRCMs
To determine whether Doxy-induced suppression of aberrant protein aggregation in the heart is cardiomyocyte-autonomous, we further tested the effect of Doxy on CryABR120G-induced aberrant protein aggregation in cultured NRCMs. One day after the Ad-HA-CryABR120Ginfection, different doses (0.5 to 10 mM) of Doxy were administered daily to the culture media, and the cells were collected at 3 or 11 days after treatment. As revealed previously (4,6,13), CryAB-positive protein aggregates were formed in the cytoplasm of cardiomyocytes infected by Ad-HA-CryABR120G, but not by Ad-β-Gal. The extent of the aggregates was reduced markedly by Doxy treatment (Fig. 5).Western blot analyses showed that Doxy treatments did not alter the HA-CryABR120Gprotein level discernibly in either the soluble or the insoluble fractions (Fig. 6A),but the filter-trap assays revealed that the SDS-resistant oligomeric forms of HA-CryABR120Gsignificantly reduced by Doxy in a dose-dependent manner (Figs. 6B and 6C), indicating inhibition of aberrant protein aggregation by Doxy. Moreover, this inhibitory effect seems to be more pronounced at 11 days than at 3 days.
To determine the potential mechanism of the inhibition of protein aggregation by Doxy, we examined the protein expression of heat shock protein 25 (Hsp25) and p62/SQSTM1. Hsps play a critical role in preventing protein aggregation (13,14). p62 was shown to promote the formation of ubiquitinated protein inclusion bodies (15,16). As shown in Figures 6D and 6E, p62 in both the soluble and the insoluble fractions and Hsp25 in the insoluble fraction were significantly up-regulated by CryABR120Gexpression, and the up-regulation of p62 but not Hsp25 was significantly less in Doxy-treated cells versus vehicle-treated cells.
Activation of autophagy in DRC hearts is not enhanced by Doxy
Autophagy plays an important role in protein quality control (17). Autophagic activation was shown to protect against DRC in mice (18). Hence, we examined the conversion of the native form of microtubule-associated protein light chain 3 (LC3-I) to the lipidated form of LC3 (LC3-II) (a commonly used marker of autophagic activation) and several other autophagy-related proteins (19). The protein level of LC3-II, the LC3-II-to-LC3-I ratio, and a cleaved form of Atg5 (autophagy-related gene 5) in TG CTL hearts were significantly greater than those in the NTG CTL group, but no statistically significant difference in these parameters was detected between TG CTL and TG Doxy groups (Figs. 7Aand 7B) (data not shown). In cultured NRCMs overexpressing HA-CryABR120G, Doxy at 1 and 5 mM, but not 0.5 mM, significantly reduced LC3-II protein levels, but did not alter the LC3-II-to-LC3-I ratio. Doxy treatment did not change Atg7 and beclin1 protein expression (Figs. 7C and 7D). These results indicate that autophagy is activated in the TG heart, but that Doxy-elicited cardioprotection is independent of autophagy.
Collectively, our in vivo and in vitro experiments demonstrate that Doxy can effectively inhibit aberrant protein aggregation induced by CryABR120G, which likely contributes to its protection against DRC.
Despite recent advances in understanding the genetic basis of DRC (1,3), no effective therapy is available to treat this devastating disease. Using both cell culture and a DRC mouse model, the present study revealed for the first time that Doxy can inhibit aberrant protein aggregation in cardiomyocytes, can attenuate significantly a DRC-linked misfolded protein-induced adverse cardiac remodeling, and effectively can prolong the lifespan of a well-documented TG mouse model of DRC. These results provide compelling evidence that Doxy is a promising drug candidate to treat DRC.
The dosage and route chosen here for Doxy administration were previously proven effective in treating a mouse model of oculopharyngeal muscular dystrophy (10). It should be noted that Doxy concentration used in the drinking water (6 mg/ml) for this study was 6-fold higher than what is used commonly to manipulate transgene expression in the tetracycline-inducible transgenic system. We tested only Doxy here, but other tetracycline derivatives, especially those with better tissue permeability (e.g., minocycline), could be as effective or even more effective, as demonstrated in neural proteinopathies (20).
Notably, Doxy treatment in our survival study was initiated at a relatively late stage when DRC pathologic features and clinical signs are readily detectable (Table 1). The rationale behind this experimental design is to maximize its clinical relevance. It remains to be tested, but it is very likely that the survival improvement by Doxy would be much greater should the treatment be started earlier. Supporting this prediction, we have observed that a significant attenuation of cardiac hypertrophy without deteriorating cardiac function was detected 1 month after Doxy treatment was initiated at 8 weeks of age (Fig. 2, Table 2), 8 weeks earlier than the starting point of the survival study.
The mechanisms underlying Doxy's beneficial effects on DRC are potentially quite complex because of Doxy's versatile pharmacologic actions. Besides its antimicrobial action, Doxy is also known to inhibit MMPs. By breaking down the extracellular matrix, MMPs play important roles in tissue remodeling, cell migration, angiogenesis, and interstitial remodeling (21). Hence, Doxy's MMP inhibition property is believed to contribute to a wide range of its biological effects. Timed administration of Doxy seems to protect cardiac function by modulating post–myocardial infarction remodeling (22–25). We cannot rule out the possibility that Doxy's MMP inhibition property may contribute to its beneficial effects on DRC, but 2 lines of evidence stand against this possibility. First, previous characterization showed no significant interstitial fibrosis in the heart of the DRC mice used here (2). Second, compared with NTG, myocardial activities of MMPs were not increased in TG mice (data not shown). Notably, it was reported recently that Doxy mitigated cardiac remodeling without significantly affecting myocardial MMP activities (26).
Misfolded proteins, when failed to be repaired, are escorted by the chaperones to degradation by the ubiquitin–proteasome system (1). When chaperones and/or the ubiquitin–proteasome system are overwhelmed, misfolded proteins undergo aberrant aggregation, which produces initially soluble oligomers. If not removed in time, the oligomers will fuse to form large insoluble aggregates. The soluble oligomers generally are believed to be toxic, whereas the insoluble aggregates perhaps are not (1). Cardiac toxicity of aberrant protein aggregation was demonstrated directly by the sufficiency of expressing a mutant prion protein or polyglutamine preamyloid oligomers in cardiomyocytes to induce heart failure in mice (27,28). In the present study, we observed that not only insoluble aggregates (Figs. 3and 5), but also oligomeric CryABR120G(Figs. 4and 6), were decreased significantly by Doxy treatment in vivo and in vitro. These data suggest that Doxy may act as a pharmacologic chaperone that prevents CryABR120Gfrom inducing aberrant oligomerization of endogenous and TG CryAB, allowing the formation of normal CryAB polymers that can exert the normal chaperoning function. Indeed, both tetracycline and Doxy have been shown to interact with prion proteins and to reduce in vitro prion protein aggregation and in vivo infectivity (29). Supporting this notion, the same extent of CryABR120Gprotein overexpression, in the presence of Doxy, caused significantly less up-regulation of p62 (Fig. 6) and less accumulation of ubiquitinated proteins (Fig. 4C). Our data suggest that Doxy's inhibition of aberrant protein aggregation of misfolded CryABR120Gmay contribute to its protection against DRC.
Consistent with its presence in the protein aggregates-associated desminopathy (30), p62 was up-regulated significantly in both the soluble and insoluble fractions of cultured cardiomyocytes overexpressing CryABR120G. Interestingly, the increases in p62 protein levels were attenuated significantly by Doxy (Figs. 6D and 6E). p62 is known to interact with both ubiquitinated proteins in the aggregates and autophagosomes and target protein aggregates for selective autophagic degradation (31). p62 also mediates the formation of ubiquitin-positive inclusion bodies in hepatocytes and neurons when autophagy is impaired, with varying consequences (16). The role of p62 up-regulation on aberrant protein aggregation in cardiomyocytes has not been defined, but our results indicate that attenuation of p62 up-regulation likely is beneficial to the heart and may be an underlying mechanism of Doxy. Consistent with this postulate, the down-regulation of p62 by Doxy treatment did not seem to be caused by autophagic activation, because Doxy did not enhance autophagic activation in either the TG heart or in NRCMs expressing CryABR120G(Fig. 7).
It was reported in a TG mouse model of human CryABR120Gthat stress-inducible Hsp's were up-regulated differentially in DRC hearts, with a major increase in Hsp25 expression associated with progression to heart failure and increased mortality (3). However, Hsp's such as Hsp22, Hsp70, and HspB8 disrupt oligomer formation induced by CryABR120Gunder certain conditions (13,14). In both CryABR120GTG mouse hearts (data not shown) and CryABR120Gexpressing cultured cardiomyocytes, we observed a significant Hsp25 up-regulation, but Doxy treatment failed to alter it (Figs. 6D and 6E). Hence, the suppression of protein aggregation by Doxy is unlikely through an effect on Hsp25.
DRC, by itself, is not a common disease, but it exemplifies cardiac proteinopathies featured by intrasarcoplasmic aberrant protein aggregation. Hence, the significance of this study could be far beyond DRC because preamyloid oligomers were observed in the heart tissue of a large subset of human congestive heart failure cases resulting from hypertrophic/dilated cardiomyopathies (4). Moreover, aberrant protein aggregates also were observed in pressure-overloaded mouse hearts (7).
Interestingly, although chronic Doxy treatment was shown to attenuate isoproterenol and transaortic constriction-induced cardiac hypertrophy (32), Vinet et al. (26) reported that 1 month, but not 2 months, of low-dose Doxy enhanced transaortic constriction-induced hypertrophy. Protective actions of Doxy on rat diabetic cardiomyopathy also were reported recently (25).
The present study provides compelling pre-clinical evidence that Doxy is a promising drug candidate to treat DRC.
Dr. Zheng is currently at Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts. Dr. Tang is currently at the Center for Cardiovascular Research, John A. Burns School of Medicine, University of Hawaii, Manoa, Hawaii. Dr. Kumarapeli is currently at the Department of Pathology, Buffalo General Hospital, Buffalo, New York. Dr. Wang is an established investigator of the American Heart Association (AHA).
This work was supported in part by grants R01HL072166and R01HL085629from the National Institutes of Health; by grant 0740025Nfrom the AHA(to Dr. Wang); by AHApost-doctoral (to Dr. H. Zheng) and pre-doctoral (to Dr. Kumarapeli) fellowships; and by the MD/PhD Program of University of South Dakota. All other authors have reported that they have no relationships to disclose. Drs. Zheng and Tang contributed equally to this work.
- Abbreviations and Acronyms
- recombinant adenoviruses expressing β-galactosidase
- adenoviruses expressing hemagglutinin-tagged CryABR102G
- missense mutation R120G of αB-crystallin
- desmin-related cardiomyopathy
- heat shock protein
- microtubule associated protein light chain 3
- matrix metalloproteinase
- neonatal rat cardiomyocyte
- Received September 30, 2009.
- Revision received January 13, 2010.
- Accepted January 19, 2010.
- American College of Cardiology Foundation
- Wang X.,
- Robbins J.
- Wang X.,
- Osinska H.,
- Klevitsky R.,
- et al.
- Sanbe A.,
- Osinska H.,
- Saffitz J.E.,
- et al.
- Maloyan A.,
- Gulick J.,
- Glabe C.G.,
- Kayed R.,
- Robbins J.
- Chen Q.,
- Liu J.B.,
- Horak K.M.,
- et al.
- Tannous P.,
- Zhu H.,
- Nemchenko A.,
- et al.
- Peterson J.T.
- Bao Y.P.,
- Sarkar S.,
- Uyama E.,
- Rubinsztein D.C.
- Kumarapeli A.R.,
- Su H.,
- Huang W.,
- et al.
- Sanbe A.,
- Yamauchi J.,
- Miyamoto Y.,
- Fujiwara Y.,
- Murabe M.,
- Tanoue A.
- Chavez Zobel A.T.,
- Loranger A.,
- Marceau N.,
- Theriault J.R.,
- Lambert H.,
- Landry J.
- Gal J.,
- Strom A.L.,
- Kilty R.,
- Zhang F.,
- Zhu H.
- Tannous P.,
- Zhu H.,
- Johnstone J.L.,
- et al.
- Kassiotis C.,
- Ballal K.,
- Wellnitz K.,
- et al.
- Villarreal F.J.,
- Griffin M.,
- Omens J.,
- Dillmann W.,
- Nguyen J.,
- Covell J.
- Cheung P.Y.,
- Sawicki G.,
- Wozniak M.,
- Wang W.,
- Radomski M.W.,
- Schulz R.
- Vinet L.,
- Rouet-Benzineb P.,
- Marniquet X.,
- et al.
- Trifilo M.J.,
- Yajima T.,
- Gu Y.,
- et al.
- Pattison J.S.,
- Sanbe A.,
- Maloyan A.,
- Osinska H.,
- Klevitsky R.,
- Robbins J.
- Forloni G.,
- Iussich S.,
- Awan T.,
- et al.
- Pankiv S.,
- Clausen T.H.,
- Lamark T.,
- et al.
- Errami M.,
- Galindo C.L.,
- Tassa A.T.,
- Dimaio J.M.,
- Hill J.A.,
- Garner H.R.