A high proportion (82%) of Doberman Pinschers with DCM reportedly have a 16-bp deletion in a splice-site region of a gene (PDK4) encoding for the mitochondrial protein PDK4, which is located on chromosome 14 and involved in energy metabolism.1 However, a study2 involving Doberman Pinschers of European origin revealed no evidence for involvement of PDK4 in the etiology of DCM, given that the mutation was also identified in other canine breeds. A genetic test has been developed to identify patients that are homozygous or heterozygous for the mutation. This test is applied to serum samples to evaluate whether a reduction exists in expression of PDK4 relative to expression in healthy dogs for dogs that are homozygous and heterozygous for the PDK4 mutation, with that reduction being more severe for homozygous dogs (which also had mitochondrial alterations).a Doberman Pinschers with DCM have a marked decrease in mitochondrial electron transport activity and impaired oxidative production of ATP, compared with the activity and production in healthy dogs.3 Therefore, an association likely exists between mitochondrial dysfunction and DCM in Doberman Pinschers, but a more specific link between the effect of the PDK4 mutation on mitochondrial function and the phenotype of DCM identified in affected dogs is important for understanding the pathophysiologic characteristics of the disease.
Energy in cardiac muscle cells is produced through 2 major pathways: glycolysis and oxidative phosphorylation.4 The PDC regulates which pathway is used. When PDC is active, metabolism is shifted toward glycolysis. When PDC is inactivated by pyruvate dehydrogenase kinase, metabolism is shifted toward fat utilization. Pyruvate dehydrogenase kinase isozyme 4 is a component of the pyruvate dehydrogenase kinase complex. In healthy hearts, almost all ATP produced is a result of oxidative phosphorylation in the mitochondria via β-oxidation of fatty acids.5 We believe that fat is not being utilized efficiently in mitochondria of Doberman Pinschers with DCM because of a mutation in PDK4.
The development of extracellular flux analyzers has enabled detection of mitochondrial dysfunction.6 The rates at which oxygen concentration and pH change are defined as mitochondrial OCR and ECAR, respectively. As the production of ATP increases secondary to mitochondrial respiration, OCR will also increase. In turn, if ATP production increases as a result of glycolysis, pH will decrease and ECAR will increase. Basal respiration and maximum respiratory capacity are attributes of mitochondrial function that can be determined by the addition of different reagents to fluorometric assays of extracellular flux.7 To assess specific pathways, such as fatty acid oxidation, changes to the substrates of the assay medium (eg, addition of carnitine or withholding of carbohydrates as a source of energy) can be made before the assay is performed.8 Healthy cells are expected to have a higher basal mitochondrial OCR than dysfunctional cells have,9 and a decrease in OCR and respiratory capacity has been detected for several diseases caused by mitochondrial mutations.10
Conventional palliative treatment has assisted in the management of DCM in Doberman Pinschers but does not correct the underlying dysfunction. Use of a vector to replace a dysfunctional gene (ie, gene therapy) may allow correction of a genetic problem at its source. The purpose of the study reported here was to compare basal mitochondrial OCR and maximal respiratory capacity in skin fibroblasts obtained from Doberman Pinschers with and without DCM and the PDK4 mutation and to assess in vitro the effect of gene therapy on mitochondrial OCR in such dogs. We hypothesized that basal OCR and maximal respiratory capacity would be lower in skin fibroblasts of Doberman Pinschers with DCM and the PDK4 mutation. We also hypothesized that transduction with tmAAV-PDK4 would result in an increase in the metabolism of mitochondria with defective PDK4.
Materials and Methods
Animals
Doberman Pinschers brought to the Small Animal Hospital of the University of Florida or the Doberman Pinscher Club Association shows at Pomona, Calif, in 2014 were eligible for inclusion in the study. Dogs undergoing treatment for heart disease or with clinical signs of DCM were excluded from the study. Owner consent was obtained for eligible dogs prior to inclusion.
Blood samples were collected from all enrolled dogs via 1-inch, 22-gauge needles into evacuated serum-collection tubes and submitted for detection of the PDK4 mutation.a After genetic status was identified, dogs underwent cardiographic evaluation by means of echocardiography and 24-hour Holterb monitoring. Transthoracic 2-D, M-mode, and Doppler echocardiographic examinations were performed with a digital ultrasonography systemc (nominal frequency of transducer array, 2 to 5 MHz). For this evaluation, dogs were positioned in right lateral recumbency, an ECG was simultaneously recorded during the echocardiograms, and all variables were measured from 3 consecutive beats. Standard echocardiographic M-mode measurements of the left ventricle (LVID in diastole and systole and fractional shortening) were obtained from the right parasternal short-axis view at the level of the papillary muscles.
To be considered free of DCM, dogs had to have weight-adjusted LVIDs lower than values previously published for the diagnosis of DCM in Doberman Pinschers11 and < 100 VPCs detected during the 24 hours of Holter monitoring.12 Dogs with high weight-adjusted LVIDs, with or without Holter-detected arrhythmias suggestive of DCM, were classified as having occult DCM. Echocardiographic, Holter, and genetic data were used to classify dogs as without DCM and the PDK4 mutation (group 1), with occult DCM and with only 1 copy (heterozygous) of the mutated PDK4 (group 2), and with occult DCM and with 2 copies (homozygous) of the mutated PDK4 (group 3).
Skin biopsy and fibroblast culture
In preparation for collection of samples for measurement of mitochondrial OCR, dogs were positioned in right lateral recumbency and hair was clipped from the inguinal region. The clipped area was aseptically prepared, and 2% lidocaine solutiond was administered SC close to the anticipated site of biopsy. Three-millimeter skin biopsy specimens were obtained by use of a sterile punch biopsy tool, and wounds were closed with surgical tissue glue.
Collected skin specimens were immediately transferred to a sterile flask containing Dulbecco modified Eagle mediume plus 20% fetal calf serume and transported to the laboratory.f Fibroblasts were obtained from skin specimens by use of a protocol in which fibroblasts are first released from skin specimens by enzymatic digestion and then placed in culture.13 Secondary fibroblast culture was performed with Dulbecco modified Eagle medium containing 10% calf serum, penicillin-streptomycin, and amphotericin B.e After 3 passages, fibroblasts were frozen and stored at −80°C until all samples had been collected.
AAV transduction in skin fibroblasts
To evaluate the effects of gene therapy in vitro, tmAAV2-PDK4 was constructed by substituting the wild-type AAV genes (rep and cap) with PDK4. The wild-type canine PDK4 cDNA was subcloned from a pooled canine RNA sample by addition of primers to the 5′ and 3′ ends of the RNA sample, followed by use of a PCR assay. The sequenced clone was then cloned into the AAV construct driven by the chicken β-actin promoter. Frozen fibroblasts were thawed and induced to proliferate under sterile conditions. Each sample of fibroblasts from each dog was cultured until enough cells were available to create 2 groups: fibroblasts free of AAV (untreated) and fibroblasts transduced with tmAAV2-PDK4 (treated). For this portion of the study, fibroblasts were harvested, counted by use of an automated cell counterg and trypan blue cell viability assay, and transduced with tmAAV-PDK4 at a multiplicity of infection of 1,000. Fibroblasts were then seeded and incubated at 37°C and 5% CO2. To identify the point of maximal expression of PDK4 protein, a marker gene was used. For this, a different sample of fibroblasts was transduced with tyrosine mutant AAV carrying the green fluorescein gene. Fluorescein production was evaluated with a fluorescence microscope at 24 hours, 48 hours, 72 hours, and 4 days after transduction. At the time of maximal fluorescein expression, untreated fibroblasts and fibroblasts transduced with tmAAV-PDK4 vector were harvested for a mitochondrial stress assay.
Mitochondrial stress assay
Mitochondrial stress assays were performed to evaluate effects of dog group and gene therapy on mitochondrial OCR and ECAR. Twenty-four hours prior to the assay, fibroblasts were harvested, counted, and assessed for viability. Fibroblasts were then seeded into a 96-well assay plate at a cell density of 5 × 104 cells/well, with the same growth medium that was used for the initial culture. Fibroblasts from each of the 3 groups of Doberman Pinschers and in each of the 2 AAV groups (treated and untreated) were evaluated in quadruplicate. The same day, the cartridge containing the sensors was hydrated with the appropriate buffer hydrating solution.h
The day of the assay, assay medium was prepared from Krebs-Henseleit buffer medium plus carnitinei (50mM), and the fatty acid palmitatej was added (200μM final concentration) as the only source of energy for the assay. Growth medium of the fibroblasts was exchanged with the assay medium. Fibroblasts were then incubated for 1 hour at 37°C without CO2.
The analyzer sensor cartridge was loaded with oligomycink in port A (1μM final concentration), FCCPk in port B (1.5μM final concentration), and a combination of rotenone and antimycin Ak in port C (1μM final concentration). The assay plate and sensor cartridge were loaded on the extracellular flux analyzer,l and the mitochondrial stress test was performed. After automatic calibration of the analyzer, the assay medium was mixed, and mitochondrial OCR and ECAR were measured to evaluate basal respiration. This step was repeated 3 times. Next, oligomycin was injected through port A, and the mixing and measuring step was repeated 3 times. Then FCCP was injected through port B, and the mixing and measuring step was repeated 3 times to evaluate maximal respiratory capacity. Finally, rotenone-antimycin A was injected through port C, and the mixing and measuring step was again repeated 3 times.
Determination of study sample size
Because no data were available regarding mitochondrial OCR in healthy dogs, the OCR in skin fibroblasts from healthy Doberman Pinschers was established as described so that an adequate sample size for the study could be calculated and cell density and FCCP concentration could be optimized in subsequent assays. In that preliminary analysis, fibroblasts had a mean ± SD OCR of 106 ± 5 pmol of O2/min. That analysis revealed there would need to be 3 dogs/study group to detect a difference of 15 pmol of O2/min with 95% confidence and power of 80%. To compensate for possible attrition and to make use of the entire 96-well analyzer plate, 4 dogs were included in each study group.
Statistical analysis
Values of OCR for each group of dogs, with each dog having 4 measurements, were averaged to obtain the group mean OCR. Data were evaluated for normality of distribution by use of the Shapiro-Wilk test and a statistical software program.m Results are reported as mean ± SD for normally distributed data and as median (range) for nonnormally distributed data. The hypothesis that basal mitochondrial OCR would be lower in skin fibroblasts from Doberman Pinschers with occult DCM and the PDK4 mutation than in dogs without DCM or the mutation was tested by comparing results for untreated fibroblasts from groups 1, 2, and 3. The second hypothesis that gene therapy would result in an increase in mitochondrial OCR in skin fibroblasts obtained from Doberman Pinschers was tested by comparing results for untreated and treated fibroblasts for each group. To evaluate changes in OCR over time, the area under the curve was measured by an algorithm executed by the analyzer software, and these values were compared among the groups via ANOVA, with a Tukey test for post hoc comparisons. Values of P < 0.05 were considered significant.
Results
Animals
Twelve Doberman Pinschers were recruited initially and classified as having no DCM or PDK4 mutation (ie, with wild-type alleles; group 1), having occult DCM and being heterozygous for the PDK4 mutation (group 2), and having occult DCM and being homozygous for the PDK4 mutation (group 3). Isolation of fibroblasts from skin biopsy specimens was possible for all dogs. During the cell culture, freezing, or thawing process, fibroblasts of 2 dogs had slower growth than fibroblasts of the other dogs and poor viability; therefore, data for those 2 dogs were not included in the statistical analysis. The final study sample size was 10 dogs (group 1, 3 sexually intact females; group 2, 1 each of sexually intact male and female and 2 spayed females; group 3, 1 sexually intact female and 2 sexually intact males; Table 1).
Median (range) age, body weight, and LVID values for Doberman Pinschers that had no DCM or PDK4 mutation (group 1; n = 3), had occult DCM and were heterozygous for the PDK4 mutation (group 2; 4), and had occult DCM and were homozygous for the PDK4 mutation (group 3; 3).
| Characteristic | Group 1 | Group 2 | Group 3 |
|---|---|---|---|
| Age (y) | 5 (5–9) | 7 (5–8) | 7 (6–9) |
| Body weight (kg) | 34.0 (31.0–35.9) | 33.1 (29.5–38.6) | 38.0 (33.0–40.8) |
| LVIDDi | 3.34 (3.19–3.60) | 4.33 (4.32–4.98) | 4.11 (3.96–5.19) |
| LVIDSi | 2.64 (2.46–2.83) | 3.80 (3.51–4.00) | 3.62 (3.40–4.71) |
LVIDDi = Left ventricular internal diameter in diastole indexed to body surface area. LVIDSi = Left ventricular internal diameter in systole indexed to body surface area.
Ventricular dimensions as determined by echocardiography were indexed to body surface area to allow for median calculation. All dogs in groups 2 and 3 had systolic ventricular dimensions indicative of DCM (Table 1). Over the 24-hour monitoring period, Holter readings revealed ventricular arrhythmias suggestive of DCM in all dogs of groups 2 (98, 70, 484, or 427 VPCs; all dogs had couplets) and 3 (60, 205, or 67 VPCs; all dogs had couplets; 1 dog had ventricular runs). Two dogs in of group 1 had results suggestive of ventricular ectopy (2 or 4 VPCs; no couplets or ventricular runs).
AAV transduction in skin fibroblasts
Fluorescein production was observed 48 hours after transduction (ie, after gene therapy initiation). Maximal fluorescein production occurred 4 days after transduction, at which point approximately 60% of the fibroblasts had evidence of fluorescence.
Mitochondrial stress assay
Skin fibroblasts from the dogs in group 1 had a significantly (P = 0.02) higher mean ± SD basal mitochondrial OCR (115 ± 9 pmol of O2/min) than did group 2 (74 ± 8 pmol of O2/min) or group 3 (58 ± 12 pmol of O2/min). The mean OCR obtained for group 1 was similar to values obtained from the preliminary assay that had been performed to provide data for sample size calculation. No difference in OCR was detected between groups 2 and 3 (P = 0.39). Mitochondrial metabolism increased in the 3 groups after AAV transduction (gene therapy). However, this increase was not significant (ie, P ≥ 0.12) for any comparisons between treated and untreated fibroblasts within any dog group. Fibroblasts of group 1 had a mean OCR of 121 ± 26 pmol of O2/min after transduction with tmAAV-PDK4. Fibroblasts of groups 2 and 3 had mean OCRs of 88 ± 6 pmol of O2/min and 59 ± 3 pmol of O2/min, respectively.
Inspection of the stress profile suggested that fibroblasts may have been damaged after the addition of the first reagent (oligomycin; Figure 1). Therefore, changes in OCR after the use of FCCP did not provide reliable data for this study, and evaluation of maximal respiratory capacity was not possible.
Mean mitochondrial OCRs over time for skin fibroblasts obtained from Doberman Pinschers with no DCM or PDK4 mutation (n = 3; circles), with occult DCM and heterozygous for the PDK4 mutation (4; triangles), and with occult DCM and homozygous for the PDK4 mutation (3; squares). After the basal mitochondrial respiration rate was measured, compounds modulating mitochondrial function were added sequentially into the assay medium, and mitochondrial OCR was measured after each compound addition. First, oligomycin was added to inhibit ATP synthase (vertical line A). Then FCCP, an uncoupler that allows maximal respiration, was added (vertical line B). Finally, rotenone-antimycin A was added to inhibit total mitochondrial respiration (vertical line C). Although these results allowed evaluation of basal respiration, the unstable profile after the addition of oligomycin suggested that the concentration of this reagent was inadequate, preventing evaluation of maximal respiratory capacity or any other measurement of mitochondrial function. Error bars represent SD.
Citation: American Journal of Veterinary Research 77, 2; 10.2460/ajvr.77.2.156
Discussion
Results of the present study indicated that OCR in Doberman Pinschers with DCM and the PDK4 mutation was lower than in healthy Doberman Pinschers. This finding may support an association between the genetic mutation and the phenotype observed in these dogs.
Understanding the role of the PDK4 enzyme is essential to interpreting the results of the mitochondrial stress test. Acetyl coenzyme A is the molecule from which energy can be obtained in the mitochondria, and the sources of this enzyme are mainly carbohydrates, fatty acids, and amino acids.14 The choice of pathway between glycolysis and fatty acid oxidation is regulated by the PDC (the function of which is inhibited by PDK4) as well as other factors such as energy starvation and energy requirements.4 We presumed that in Doberman Pinschers with the PDK4 mutation, reduction of PDK4 expression leads to deficient control of the PDC, shifting metabolism away from fatty acid oxidation. As a consequence, glycolysis serves as the main pathway used for energy production. This may appear intuitively to be a more efficient way of obtaining energy, but over time, unregulated glycolysis may lead to impaired mitochondrial electron transport15 and predispose cells to mitochondrial damage, as has been observed in Doberman Pinschers with DCM and the PDK4 mutation.1
When a mitochondrial stress assay is performed with a medium containing fatty acids as the only source of energy, cells are forced to decrease their glycolytic activity and use fatty acids oxidation to survive. The lack of any enzyme involved in this pathway results in lower mitochondrial production of ATP, which in turn is reflected as a lower OCR, as was identified in the fibroblasts from dogs with the PDK4 mutation in the present study. However, the results of this assay indicated that the fibroblasts from dogs homozygous for that mutation had ATP production. This finding may be explained by use of endogenous substrates as a source of energy,16 although it is unknown whether one of the other isoenzymes in the PDC may have compensated for the mutation in PDK4.
The assay used in the study reported here can be used to quantify mitochondrial function and cellular bioenergetics in intact cells, avoiding many of the artifacts associated with mitochondrial isolation.16 This assay allows determination of the basal OCR as well as the mitochondrial response in various states of energy demand. Maximum respiration rate is measured by adding an uncoupler such as FCCP, given that FCCP generally stimulates respiration above basal rates. The difference between the FCCP-stimulated rate and basal OCR provides an estimate of the reserve capacity of the cells. Reserve capacity is defined as the amount of oxygen consumption when ATP demand increases or during other stress.7,16 A decrease in reserve capacity is a strong indicator of potential mitochondrial dysfunction. However, titration of the uncoupler in each new cell line and condition is critical, as is titration of the other reagents used in the assay. We performed an initial evaluation to optimize the concentration of FCCP, but this assay was performed without the interaction of the other compounds. As a consequence, results obtained with the mitochondrial stress assay were not reliable enough to provide any conclusions regarding maximum respiration capacity. Therefore, we believe that further optimization of the protocol is necessary to allow maximal use of the mitochondrial stress assay in future research.
Recombinant AAV2 vectors are being evaluated in phase I and II clinical trials of gene therapy in humans with various diseases,17–19 and several studies20–22 involving dogs have had encouraging results. In the present study, gene therapy involving canine fibroblasts in vitro did not increase mitochondrial function significantly. Only 60% of the fibroblasts infected with tyrosine mutant AAV that carried the green fluorescence gene had evidence of fluorescence, indicating a likely insufficient transduction. The amount of vector particles per cell should be considered when interpreting this result. We used a multiplicity of infection of 1,000 because, in the authors' experience, this amount of vector provides an adequate expression. However, increasing the number of vector particles per cell may provide a higher degree of transduction in future studies.
Another aspect to consider when assessing transduction efficiency is the type of vector used. Our research group has performed in vitro experiments to compare AAV8, AAV9, and tmAAV2 vectors containing a fluorescent reporter gene. Results of flow cytometry and quantitative PCR assays of vector genomes (unpublished data) confirmed results of a previous study23 that indicated tmAAV2 is superior to other AAV serotypes. By introducing mutations to substitute surface-exposed tyrosine for phenylalanine residues in the capsids of AAV2 vectors, improved resistance is achieved to intracellular degradation of the vector, and thus increased intracellular trafficking to the nucleus of the cells is achieved, enhancing transduction efficiency.24 Nevertheless, strategies in gene therapy change constantly, and adding newer techniques25 to reduce vector degradation in future studies may enhance further cell transduction.
The present study had some limitations. Although the number of specimens included in the study was sufficient for comparison of basal mitochondrial OCR between dog groups, the sample size was perhaps too small to entirely appreciate the effect of gene therapy on OCR. In addition, inclusion of dogs with DCM but without the PDK4 mutation would have been useful; healthy mitochondrial function in such dogs would have supported the hypothesis of the mutation in PDK4 as the cause of mitochondrial dysfunction. However, the presence of mitochondrial dysfunction in fibroblasts from dogs with DCM but without the PDK4 mutation might suggest that another genetic cause of DCM is also mitochondrial.
Results of the present study indicated that Doberman Pinschers with DCM and PDK4 mutation had a lower basal mitochondrial OCR than did healthy Doberman Pinschers, suggesting the presence of mitochondrial dysfunction. In addition, gene therapy at the multiplicity of infection used did not increase basal OCR significantly. The effect of mitochondrial dysfunction on the contractile properties of cardiac muscle cells remains unknown, but it is intuitive to hypothesize that provision of suboptimal energy would lead to improper sarcomere contraction. Nevertheless, the results reported here will contribute to the design of future studies on sarcomere mechanics and energetics.
Acknowledgments
Funded in part by a resident research grant from the American College of Veterinary Internal Medicine, by the Doberman Pinscher Club of America, and by a resident research grant from the University of Florida College of Veterinary Medicine.
Presented as a poster at the Keystone Symposia on Molecular and Cellular Biology: Mitochondria, Metabolism and Heart Failure, Santa Fe, NM, January 2015.
ABBREVIATIONS
| AAV | Adeno-associated virus |
| DCM | Dilated cardiomyopathy |
| ECAR | Extracellular acidification rate |
| FCCP | Carbonyl cyanide-4-(trifluoromethoxy) phenylhydrazone |
| LVID | Left ventricular internal diameter |
| OCR | Oxygen consumption rate |
| PDC | Pyruvate dehydrogenase complex |
| PDK4 | Pyruvate dehydrogenase kinase isozyme 4 |
| tmAAV-PDK4 | Tyrosine mutant adeno-associated virus containing the PDK4 gene |
| VPC | Ventricular premature complex |
Footnotes
Veterinary Cardiac Genetics Laboratory, Raleigh, NC.
Trillium Holter Recorder-Analyzer, Forest Medical LLC, Syracuse, NY.
Vivid E9 GE Medical Systems, Milwaukee, Wis.
Lidocaine HCl 2%, Abbott Laboratories, North Chicago, Ill.
Fisher Scientific Inc, Hampton, NH.
Cytogenetics Laboratory, Department of Pathology, College of Medicine, University of Florida, Gainesville, Fla.
Cellometer Vision, Nexcelom Bioscience, Lawrence, Mass.
XF Calibrant Solution, Seahorse Bio, North Billerica, Mass.
Sigma Chemical Co, St Louis, Mo.
XF palmitate-BSA FAO substrate, Seahorse Bio, North Billerica, Mass.
XF Cell Mito stress test kit, Seahorse Bio, North Billerica, Mass.
XF96 extracellular flux analyzer, Seahorse Bio, North Billerica, Mass.
IBM SPSS Statistics for Windows, version 21, IBM Corp, Armonk, NY.
References
1. Meurs KM, Lahmers S, Keene BW, et al. A splice site mutation in a gene encoding for PDK4, a mitochondrial protein, is associated with the development of dilated cardiomyopathy in the Doberman Pinscher. Hum Genet 2012; 131: 1319–1325.
2. Owczarek-Lipska M, Mausberg TB, Stephenson H, et al. A 16-bp deletion in the canine PDK4 gene is not associated with dilated cardiomyopathy in a European cohort of Doberman Pinschers. Anim Genet 2013; 44: 239.
3. Lopes R, Solter PF, Sisson DD, et al. Characterization of canine mitochondrial protein expression in natural and induced forms of idiopathic dilated cardiomyopathy. Am J Vet Res 2006; 67: 963–970.
4. Lodish H, Berk A, Zipursky SL, et al. Section 12.1. First steps of glucose and fatty acid catabolism: glycolysis and the citric acid cycle. In: Molecular cell biology. 4th ed. New York: W. H. Freeman, 2000;480–491.
5. Grynberg A, Demaison L. Fatty acid oxidation in the heart. J Cardiovasc Pharmacol 1996; 28(suppl 1):S11–S17.
6. Ferrick DA, Neilson A, Beeson C. Advances in measuring cellular bioenergetics using extracellular flux. Drug Discov Today 2008; 13: 268–274.
7. Dranka BP, Benavides GA, Diers AR, et al. Assessing bioenergetic function in response to oxidative stress by metabolic profiling. Free Radic Biol Med 2011; 51: 1621–1635.
8. Abe Y, Sakairi T, Kajiyama H, et al. Bioenergetic characterization of mouse podocytes. Am J Physiol Cell Physiol 2010; 299: C464–C476.
9. Jarrett SG, Rohrer B, Perron NR, et al. Assessment of mitochondrial damage in retinal cells and tissues using quantitative polymerase chain reaction for mitochondrial DNA damage and extracellular flux assay for mitochondrial respiration activity. In: Weber BHF, Langmann T, eds. Methods in molecular biology 935. Retinal degeneration: methods and protocols. Berlin: Springer-Verlag Inc 2013;227–243.
10. Baruffini E, Dallabona C, Invernizzi F, et al. MTO1 mutations are associated with hypertrophic cardiomyopathy and lactic acidosis and cause respiratory chain deficiency in humans and yeast. Hum Mutat 2013; 34: 1501–1509.
11. Summerfield NJ, Boswood A, O'Grady MR, et al. Efficacy of pimobendan in the prevention of congestive heart failure or sudden death in Doberman Pinschers with preclinical dilated cardiomyopathy (the PROTECT study). J Vet Intern Med 2012; 26: 1337–1349.
12. Wess G, Schulze A, Geraghty N, et al. Ability of a 5-minute electrocardiography (ECG) for predicting arrhythmias in Doberman Pinschers with cardiomyopathy in comparison with a 24-hour ambulatory ECG. J Vet Intern Med 2010; 24: 367–371.
13. Takashima A. Establishment of fibroblast cultures. Curr Protoc Cell Biol 2001; 2.1:Unit 2.1.1–2.1.12.
14. Dashty M. A quick look at biochemistry: carbohydrate metabolism. Clin Biochem 2013; 46: 1339–1352.
15. McCutcheon LJ, Cory CR, Nowack L, et al. Respiratory chain defect of myocardial mitochondria in idiopathic dilated cardiomyopathy of Doberman Pinscher dogs. Can J Physiol Pharmacol 1992; 70: 1529–1533.
16. Brand MD, Nicholls DG. Assessing mitochondrial dysfunction in cells. Biochem J 2011; 435: 297–312.
17. Brantly ML, Spencer LT, Humphries M, et al. Phase I trial of intramuscular injection of a recombinant adeno-associated virus serotype 2 α1-antitrypsin (AAT) vector in AAT-deficient adults. Hum Gene Ther 2006; 17: 1177–1186.
18. Snyder RO, Francis J. Adeno-associated viral vectors for clinical gene transfer studies. Curr Gene Ther 2005; 5: 311–321.
19. Jessup M, Greenberg B, Mancini D, et al. Calcium upregulation by percutaneous administration of gene therapy in cardiac disease (CUPID): a phase 2 trial of intracoronary gene therapy of sarcoplasmic reticulum Ca2+-ATPase in patients with advanced heart failure. Circulation 2011; 124: 304–313.
20. Callejas D, Mann CJ, Ayuso E, et al. Treatment of diabetes and long-term survival after insulin and glucokinase gene therapy. Diabetes 2013; 62: 1718–1729.
21. Bish LT, Sleeper MM, Forbes SC, et al. Long-term restoration of cardiac dystrophin expression in Golden Retriever muscular dystrophy following rAAV6-mediated exon skipping. Mol Ther 2012; 20: 580–589.
22. Sleeper MM, Fornasari B, Ellinwood NM, et al. Gene therapy ameliorates cardiovascular disease in dogs with mucopolysaccharidosis VII. Circulation 2004; 110: 815–820.
23. Li M, Jayandharan GR, Li B, et al. High-efficiency transduction of fibroblasts and mesenchymal stem cells by tyrosine-mutant AAV2 vectors for their potential use in cellular therapy. Hum Gene Ther 2010; 21: 1527–1543.
24. Zhong L, Li B, Mah CS, et al. Next generation of adeno-associated virus 2 vectors: point mutations in tyrosines lead to high-efficiency transduction at lower doses. Proc Natl Acad Sci USA 2008; 105: 7827–7832.
25. Ahmadiankia N, Neshati V, Zeinab N, et al. Generation of helper plasmids encoding mutant adeno-associated virus type 2 capsid proteins with increased resistance against proteasomal degradation. Iran J Basic Med Sci 2013; 16: 813–821.
