SEARCH

SEARCH BY CITATION

Keywords:

  • aging;
  • electroretinogram;
  • glaucoma;
  • mitochondria;
  • polymerase gamma retina

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

Mouse models that accumulate high levels of mitochondrial DNA (mtDNA) mutations owing to impairments in mitochondrial polymerase γ (PolG) proofreading function have been shown to develop phenotypes consistent with accelerated aging. As increase in mtDNA mutations and aging are risk factors for neurodegenerative diseases, we sought to determine whether increase in mtDNA mutations renders neurons more vulnerable to injury. We therefore examined the in vivo functional activity of retinal neurons and their ability to cope with stress in transgenic mice harboring a neural-targeted mutant PolG gene with an impaired proofreading capability (Kasahara, et al. (2006) Mol Psychiatry11(6):577–93, 523). We confirmed that the retina of these transgenic mice have increased mtDNA deletions and point mutations and decreased expression of mitochondrial oxidative phosphorylation enzymes. Associated with these changes, the PolG transgenic mice demonstrated accelerated age-related loss in retinal function as measured by dark-adapted electroretinogram, particularly in the inner and middle retina. Furthermore, the retinal ganglion cell–dominant inner retinal function in PolG transgenic mice showed greater vulnerability to injury induced by raised intraocular pressure, an insult known to produce mechanical, metabolic, and oxidative stress in the retina. These findings indicate that an accumulation of mtDNA mutations is associated with impairment in neural function and reduced capacity of neurons to resist external stress in vivo, suggesting a potential mechanism whereby aging central nervous system can become more vulnerable to neurodegeneration.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

Aging is associated with an increase in mitochondrial DNA (mtDNA) mutations in the form of both deletions (Cortopassi & Arnheim, 1990; Arnheim & Cortopassi, 1992; Corral-Debrinski et al., 1992; Simonetti et al., 1992; Khrapko & Vijg, 2007) and point mutations (Simon & Lin, 2004; Cantuti-Castelvetri et al., 2005). This is particularly pronounced in postmitotic tissues with high levels of energy requirement such as the brain and heart (Cortopassi et al., 1992). Impaired mitochondrial function consequent to mtDNA abnormalities has been determined in age-related neurodegenerative disorders, such as Parkinsons’s disease (Schnopp et al., 1996), Alzheimer’s disease (Coskun et al., 2004), and optic neuropathies (Carelli et al., 2004) including glaucoma (Abu-Amero et al., 2006). However, the in vivo effects of age-related increases in mitochondrial DNA mutations on neuronal activity and the ability for neurons to cope with injury have not been studied in detail.

Mitochondrial DNA mutations can arise from errors in DNA replication. One enzyme central to the process of mtDNA replication is polymerase γ (PolG; Kaguni, 2004; Graziewicz et al., 2006), which consists of a catalytic subunit with polymerase and exonuclease activity and a small accessory subunit that enhances binding (Lim et al., 1999). The exonuclease activity proofreads nascent mtDNA and ensures faithful replication of the mitochondrial genome (Longley et al., 2001). In humans, mutations in PolG have been shown to result in mtDNA instability causing chronic progressive external ophthalmoplegia and other neurodegenerative diseases (Graziewicz et al., 2006; Horvath et al., 2006). Mice carrying systemic mutations in the exonucleolytic proofreading domain of PolG (Kujoth et al., 2005; Trifunovic et al., 2005) exhibit early senescence, including signs of osteopenia, alopecia, infertility, cardiomyopathy, and early death. These mice also demonstrate auditory system dysfunctions consistent with age-related hearing loss (Niu et al., 2007). The phenotype is associated with defects in the mitochondrial oxidative phosphorylation (OXPHOS) pathway (Edgar et al., 2009) and increased apoptosis (Kujoth et al., 2005; Dai et al., 2010). These models have increased levels of both mtDNA deletions and point mutations, and controversy continues as to which class of mutations may be most responsible for the observed pathology (Khrapko et al., 2006; Vermulst et al., 2007, 2009; Kraytsberg et al., 2009; Edgar et al., 2010).

To examine the effects of increased mt DNA mutations on a target organ, mice have been generated with tissue-specific expression of mutant PolG gene that has impaired 3′–5′ exonuclease activity (D181A mutation; Zhang et al., 2000; Kasahara et al., 2006). Mice with cardiac-targeted mutant PolG driven by the α-myosin heavy chain promoter showed selective increase in cardiomyocyte apoptosis and development of cardiomyopathy (Zhang et al., 2005a,b). Neuro-targeted PolG transgenic mice driven by the type II calcium/calmodulin-dependent protein kinase α (CaMKIIα) promoter showed changes suggestive of neuronal dysfunction including a distorted day–night rhythm, enhanced startle response, and altered pattern of monoamine secretion in the brain (Kasahara et al., 2006). As CaMIIKα is also highly expressed in neural retina, including a strong expression in retinal ganglion cells (Liu et al., 2000), we hypothesize that retinal dysfunction could be found in neuro-targeted PolG transgenic mice, thus providing a model for exploring the impact of increased mt DNA mutations on neural function and the response to injury.

In neurons, mtDNA mutations have been hypothesized to have an impact on a wide range of cellular functions as a result of defective energy production (Kann & Kovacs, 2007). However, there are currently no studies examining the in vivo activity of retinal neurons in models with increased somatic mitochondrial DNA mutations. The retina is a specialized extension of the central nervous system where it is possible to accurately measure neuronal function and its response to external stressors using electrophysiological methods, such as electroretinogram (Saszik et al., 2002; Bui & Fortune, 2004; Fortune et al., 2004; Weymouth & Vingrys, 2008; Kong et al., 2009). By applying the scotopic full-field electroretinogram to neuro-targeted PolG transgenic mice, we test the hypothesis that accelerated accumulation of mtDNA mutations in retinal neurons can affect their ability to maintain normal neuronal function, and whether this is exacerbated during stress.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

Mutant PolG gene expression and mtDNA mutations in PolG transgenic mice retina

Mutant and wild-type PolG gene expressions were examined using quantitative reverse transcription polymerase chain reaction (RT-PCR). As shown in Fig. 1A, mutant PolG gene mRNA was found to be expressed in the retina and cerebral cortex of 12-month-old PolG transgenic (Tg) mice. Expression of the mutant PolG gene in non-neural somatic tissues (heart, liver, kidney and muscle) was minimal. Expression of the mutant transcript was low in the cerebellum compared with the cortex, consistent with that reported by Kasahara et al., 2006; which may reflect the low levels of CaMKIIα expression in the brain stem region (Burgin et al., 1990). CaMKIIα expression as identified by immunohistochemistry in PolG transgenic mice retina was most intense in the inner retina (retinal ganglion cell and inner plexiform layers), with minimal expression in outer plexiform layer, consistent with that previously reported in mice (Liu et al., 2000; Fig. 1B).

image

Figure 1.  Neural-targeted mutant polymerase γ expression in the retina is associated with increase in mtDNA mutations. (A) Mutant and wild-type gene mRNA expression in retina and in various somatic tissues of 12-month-old polymerase γ (PolG) transgenic (Tg) mice (n = 2) and wild-type (WT) mice (n = 2). (B) Immunofluorescence labeling for CaMKIIα in a 12-month-old Tg mice retina (Alexa Fluor Green); nuclear counterstained with Hoechst (blue); white bar represents 50 μm. (C) Representative gel of long extension PCR of mitochondrial DNA from DNA extracted from inner retina of 12-month-old WT (n = 4) and 12-month-old Tg (n = 6) mice; full-length PCR product is at approximately 10 kbp; positive control is whole retinal DNA extract from a 12-month-old Tg mice (D) Positions of point mutations detected from vector cloning experiment of a segment of mtDNA (bases 14080–14680) using retinal DNA from 12-month-old PolG Tg mice (pool of n = 5 retina, six in 10 219 bases) and 12-month-old WT mice (pool of n = 4 retina, one in 7525 bases).

Download figure to PowerPoint

Neural-targeted mutant PolG has been shown to result in increased frequency of mtDNA deletions and random point mutations in the brain (Kasahara et al., 2006). Whether an increase in mtDNA mutations is also present in the retina has not been previously reported. To detect mtDNA deletions in Tg mice retina, long extension (LX)-PCR was used to amplify mitochondrial DNA extracted from the inner retina of PolG transgenic and wild-type mice obtained from laser micro-dissection (Fig. 1C). Using this method, large-scale mtDNA deletions were found in four of six 12-month-old PolG transgenic mice examined, while there were no detectable deletions in 12-month-old wild-type mice (n = 4). Next, we examined the levels of mtDNA point mutations in the PolG transgenic retina. A segment of mtDNA covering part of the cytochrome B gene (14080–14680) was amplified by PCR using pooled DNA from the retina of 12-month-old PolG transgenic (n = 5) and wild-type (n = 4) mice and then cloned into plasmid vector. Sequencing of clones showed 6 point mutations in 10 219 bases screened (5.9/10 000) for PolG transgenic (A to G at locations 14136, 14285, 14178, G to A at 14257, C to T at 14666 and T to C at 14201), and 1 point mutation in 7525 bases (1.3/10 000) for wild-type mice (A to C at 14664). Repeat cloning of a selected clone showed no mutations introduced by the PCR process (0 in 7229 bases; Fig. 1D).

Neural-targeted PolG mice retina show impaired oxidative phosphorylation but not increased oxidative stress

The mitochondrial OXPHOS pathway is the major mechanism by which mitochondria produce adenosine triphosphate (ATP; Hatefi, 1985; Wallace, 1999). To determine whether OXPHOS pathway is affected in retina of PolG transgenic mice, we first analyzed protein expressions of respiratory chain enzyme complexes in whole retinal homogenates (Fig. 2A). At 3 months of age, there were no significant differences between PolG transgenic and wild-type mice in retinal OXPHOS complex expression. At 12 months of age, PolG transgenic mice showed reduced protein expression of OXPHOS complexes I (NDUFB8), II (CII-30), III (CIII-core2), and IV-subunit I as compared with 12-month-old wild-type mice. This finding is similar to previous reports in liver and heart mitochondria in the systemic mutant PolG mouse model (Edgar et al., 2009). The expression levels of ATP synthase (CV-alpha) and porin (VDAC1) were not significantly reduced in 12-month-old PolG transgenic mice, although both showed a trend for decreased expression compared with age-matched controls.

image

Figure 2.  Retinal mitochondrial oxidative phosphorylation enzymes are reduced in polymerase γ transgenic mouse. (A) Immunoblot of mitochondrial oxidative phosphorylation enzyme subunits and porin from retinal and optic nerve homogenates. Protein densitometry is expressed relative to actin protein loading control (n ≥ 5 retina per group) *< 0.05. (B) Specific enzyme activities of mitochondrial enzyme complex IV and citrate synthase measured from whole retinal homogenates (n = 5 retina per group, *< 0.05). (C) Immunoblot of key regulators of mitochondrial biogenesis in whole retinal and optic nerve homogenates (n ≥ 5 retina per group).

Download figure to PowerPoint

To further confirm OXPHOS impairment, specific enzymatic activity of complex IV and citrate synthase were measured by spectrophotometry using retinal homogenates. As shown in Fig. 2B, specific activity of complex IV was reduced by 36 ± 9% in 12-month-old transgenic mice compared with age-matched, wild-type mice (< 0.05). There was no difference in enzyme activity between PolG transgenic and wild-type mice at 3 months of age. Citrate synthase activity was not significantly different between cohorts. Owing to the small amounts of tissue obtainable from mouse retina, enzymatic activity of other mitochondrial complexes was not amendable to assay.

To examine mitochondrial biogenesis in the retina of transgenic mice, retinal protein expressions of key markers related to mitochondrial biogenesis, SIRT-1 (silent mating type information regulation 2 homolog 1) and NRF1 (nuclear respiratory factor 1), were assessed with Western blot (Fig. 2C). As shown, the protein expression of both SIRT-1 and NRF1 was decreased significantly in 12-month-old PolG transgenic mice compared with wild-type control.

Impairment in mitochondrial OXPHOS pathway could potentially lead to increased oxygen free radical production via electron leakage (Wei et al., 1998). However, we did not find evidence of increased oxidative stress in the PolG transgenic retina using the markers heme oxygenase-1 (HO-1) and 4-hydroxy-2-nonenal (HNE; Fig. 3). This is consistent with previous studies on mutant PolG mice, which showed that cell loss in tissues was not associated with significant increases in oxidative stress (Mott et al., 2001; Zhang et al., 2003; Kujoth et al., 2005; Trifunovic et al., 2005; Niu et al., 2007).

image

Figure 3.  Oxidative stress is not increased in the uninjured retina of polymerase γ (PolG) transgenic mice. (A) Immunoblot of oxidative stress marker heme oxygenase-1 (HO-1) in retina of PolG transgenic (Tg) and wild-type (WT) mice at 3 and 12 months of age; protein densitometry expressed relative to actin protein loading control. (n ≥ 5 retina per group) (B) Immunohistochemistry staining for the lipid peroxidation product, 4-hydroxy-2-nonenal (Alexa Fluor Red); nuclear counterstained with Hoechst (blue); white bar represents 50 μm.

Download figure to PowerPoint

Mutant PolG transgenic mice have accelerated age-related impairment in retinal neuronal function

To test whether the increase in mtDNA mutations leads to in vivo impairment in retinal function, we assessed the scotopic full-field electroretinogram (ERG) of wild-type and PolG transgenic mice at various ages. The ERG is a summation of electrical responses generated by neurons in the retina in response to light. The composite signal can be decomposed into inner retina (pSTR), ON-bipolar cell (P2), and photoreceptor (P3) components (Bui & Fortune, 2004; Weymouth & Vingrys, 2008; Kong et al., 2009). The pSTR component in mice has been shown to largely reflect retinal ganglion cell activity (Moshiri et al., 2008). Analysis of wild-type and PolG transgenic mice ERGs at 3, 12, and 18 months of age is shown in Fig. 4. As shown, PolG transgenic mice demonstrated an accelerated age-related loss of inner retinal function compared with wild-type mice (anova, = 4.52, < 0.05). Comparing 12-month-old Tg mice against age-matched controls showed reductions in the amplitude of pSTR and P2 components of the ERG by 45 ± 9% (< 0.01) and 26 ± 7% (< 0.05), respectively. The ratio of pSTR/P2 (Nguyen et al., 2008) for 12-month-old PolG transgenic mice was significantly lower than wild-type controls (less by 34 ± 8%), suggesting a significant part of the reduction in inner retinal function is not explained by the upstream reduction in ON-bipolar cell function. Further decline in inner retinal pSTR of PolG transgenic mice was observed at 18 months of age, with pSTR component decrease by 67 ± 14% compared with age-matched controls (< 0.01). There were no significant differences in the outer retinal photoreceptor P3 amplitude between wild-type and PolG transgenic mice across the three age-groups (anova, = 2.75, > 0.05). PolG transgenic mice at 3 months showed no differences in retinal response compared with age-matched, wild-type controls, indicating that the retinal function impairments in older PolG transgenic mice was not a congenital manifestation, but rather an acquired defect.

image

Figure 4.  Accelerated age-related loss of retinal function in polymerase γ (PolG) transgenic mice. (A) Left: Averaged scotopic threshold response (STR) generated by 12-month-old PolG transgenic (Tg) and wild-type (WT) mice with low luminous energy (−4.73 log cd.s.m−2). Right: Representative bright flash waveform (0.34 log cd.s.m−2) from 12-month-old Tg and WT mice demonstrating P3 (photoreceptor) and P2 (ON-bipolar cell) responses. (B) The amplitude parameter for pSTR, P2, and P3 components of the electroretinogram for 3-, 12-, and 18-month-old PolG transgenic and wild-type mice derived from fitting of parameter functions. Filled squares: PolG transgenic mice (3 months n = 7, 12 months n = 13, 18 months n = 4), unfilled squares: wild-type mice (3 months n = 10, 12 months n = 8, 18 months n = 11). *< 0.05, **< 0.01 on post hoc test.

Download figure to PowerPoint

The functional loss in the retina observed in PolG transgenic mice preceded cell loss. Retinal histology showed no statistically significant decline in retinal ganglion cell count or retinal cell layer thickness between PolG transgenic and wild-type mice at 3 or 12 months of age (Supporting information Fig. S1).

Raised intraocular pressure (IOP) above normal levels could potentially cause impairments in retinal function (Fortune et al., 2004; Bui et al., 2005; Kong et al., 2009). However, IOP at baseline measured using a noninvasive rebound tonometer (iCare Finland, Espoo, Finland) showed no significant difference between PolG transgenic and wild-type mice (12-month-old Tg 12 ± 2 mmHg vs. WT 10 ± 1 mmHg, = 0.27).

Inner retinal function of PolG transgenic mice has greater vulnerability to intraocular pressure stress

To determine whether neurons in PolG transgenic mice have greater vulnerability to extracellular stress, an acute increase in IOP was applied to the eyes of PolG transgenic and wild-type mice at 12 months of age, while retinal function was monitored with ERG. By acutely increasing IOP, controllable amounts of mechanical, oxidative (Liu et al., 2007), and metabolic stress (Novack et al., 1990) can be applied to the retina. The ERG was measured concomitantly to monitor neuronal function during IOP challenge following established protocol used by our laboratory (He et al., 2006, 2008; Kong et al., 2009). The IOP was progressively increased from baseline of 12 mmHg up to 80 mmHg in 5–10 mmHg steps. As shown in Fig. 5, as IOP increases, the inner retinal function of pSTR in 12-month-old PolG transgenic mice showed greater loss compared with 12-month-old wild-type controls. The IOP level required to produce a relative reduction of pSTR by 50% in PolG transgenic mice was 31 mmHg (95% CI [27–34]), which was significantly lower compared with that required for wild-type mice (42 mmHg 95% CI [38–46]). This increase in sensitivity to IOP stress was not seen in the middle retinal ON-bipolar cell (P2) or the outer retinal photoreceptor (P3) responses.

image

Figure 5.  Susceptibility of inner retinal function to intraocular pressure challenge is increased in adult polymerase γ (PolG) transgenic mice. (A) Averaged scotopic threshold response (STR) of 12-month-old PolG transgenic (thick trace) and wild-type mice (thin trace) at baseline (12 mmHg), 35, 40, and 60 mmHg of intraocular pressure (IOP). Vertical scale: % baseline. (B) Relative waveform amplitude of retinal ganglion cell–dominant inner retinal pSTR, ON-bipolar cell–dominant P2 and photoreceptor-mediated P3 at increasing intraocular pressures. Filled squares: 12-month-old Tg mice (n = 10), unfilled squares: 12-month-old wild-type mice (n = 6). Gray areas represent 95% confidence interval for control eyes (no IOP elevation) during the experimental period.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

An increase in mitochondrial DNA (mtDNA) mutations have been associated with aging and with age-related neurodegenerative diseases (Beal, 2005). Increases in mtDNA damage have previously been reported in the aging retina (Barreau et al., 1996; Wang et al., 2010). However, whether increases in mtDNA mutations directly affect neuronal activity and alter their ability to cope with increased stress has not been previously studied. In our study, we employed a mouse model expressing a neuron-targeted mutant form of PolG to examine the hypothesis that accelerated accumulations of mtDNA mutations in retinal neurons can lead to impairment in neuronal activity. We found that the retinas of 12-month-old transgenic mice have increased levels of mtDNA mutations with associated decreases in mitochondrial OXPHOS protein expression (complexes I, II, III, and IV), complex IV enzyme activity, and mitochondrial biogenesis markers. Associated with these changes, transgenic mice demonstrated accelerated loss of retinal function, as evidenced by significant reduction in retinal ganglion cell–dominant inner retinal function (pSTR) and some reduction in ON-bipolar cell function (P2) at 12 and 18 months of age. Increased oxidative stress was not found in PolG transgenic mice retina, suggesting that increased mtDNA mutations impair neuronal function by mechanisms other than increased oxidative stress. The absence of marked oxidative stress induction is consistent with previous reports in mice with systemic PolG mutations (Kujoth et al., 2005; Trifunovic et al., 2005), although a recent report demonstrated that cardiac pathology in systemic PolG mutant mouse can be partially reversed by overexpression of mitochondrial-targeted catalase (Dai et al., 2010). Importantly, we found the inner retinal pSTR component of retinal function of PolG transgenic mice was significantly more vulnerable to stress induced by raised intraocular pressure compared with wild-type mice at 12 months of age. Taken together, our findings show that increased mtDNA mutations result in impairment of neural function and increased vulnerability to neural stress, which may be a possible mechanism contributing to the age-related neuronal loss in neurodegenerative disease of the eye such as glaucoma.

Our finding for a decrease in mitochondrial OXPHOS enzymes in the retina is consistent with previous studies in the systemic PolG mutant mouse (Edgar et al., 2009). It has been suggested that increase in mtDNA mutations can lead to decreased stability of mitochondrial OXPHOS complexes owing to amino acid substitutions. Mutations in the control region of mtDNA may also impair replication, and tRNA mutations interfere with translation of mtDNA-encoded proteins. Whether mt DNA mutation load in the form of point mutations or deletions have a greater influence on the function of OXPHOS enzymes is still a matter of debate (Kraytsberg et al., 2009; Vermulst et al., 2009; Edgar et al., 2010). We also found markers of mitochondrial biogenesis to be decreased in the retina of transgenic mice. This finding is consistent with a report of reduced mitochondrial biogenesis in the cardiac muscle of systemic PolG mutant mice (Dai et al., 2010), suggesting that some somatic tissues may react to the presence of mt DNA mutations by dampening mitochondrial biogenesis.

The neural-targeted mutant PolG model has some potential limitations. First, while CaMKIIα expression in the retina is highest in the inner retina in mice and rat (Ochiishi et al., 1994; Liu et al., 2000), it is also expressed, albeit to a lesser extent, in the retinal pigment epithelium (RPE; Ochiishi et al., 1994). Mitochondrial dysfunction in the RPE has been implicated in a number of retinal diseases (Nordgaard et al., 2008), in particular age-related macular degeneration. The ERG recording method used in this study does not assess RPE activity directly, although we would have expected RPE dysfunction to affect photopigment recycling and photoreceptor outer segment lengths. Our findings for normal retinal histology coupled with normal photoreceptoral P3 function in PolG mice suggest that the effects on the RPE may not be significant. Secondly, retinal CaMKIIα expression has been shown to change with age and development (Xue et al., 2002); hence, the expression of mutant PolG and the rate of accumulation of mtDNA mutations may not occur in a constant fashion throughout the life of the animal. Finally, the CaMKIIα regulation has been implicated in mediating stress responses in retinal ganglion cells (Fan et al., 2007). Therefore, it is possible that retinal ganglion cells under cellular stress could increase the expression of mutant PolG leading to further cellular impairment. Despite these limitations, this model is valuable in studying the effect of mtDNA mutations on in vivo neuronal responses.

In summary, our study shows that a mutation in PolG, which accelerates age-related accumulation of mtDNA mutations in neurons in the retina, produces impaired mitochondrial respiratory chain function, resulting in impaired neural function and increased neuronal vulnerability to external stress.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

Animals

All experimental methods and animal care conformed to the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research and were approved by our Institutional Animal Experiment Ethics Committee (The Royal Victorian Eye and Ear Hospital and The University of Melbourne). Neuron-targeted PolG mutant transgenic (Tg) mice (C57Bl/6 background) were purchased from RIKEN (Rikagaku Kenkyusho, Wako-shi, Saitama, Japan) at generations 3–4. Mice were maintained in a 22 ± 1 °C, 12-h light (approximately 40 lux, on at 8 am)/12-h dark environment with normal murine chow (WEHI chow, Barastoc, VIC, Australia) and water available ad libitum. Transgenic animals used in all experiments were heterozygotes. Female transgenic animals were not used in breeding to avoid the possibility of germline transmission of mitochondrial mutations. Genotyping was performed as per the protocol outlined by Kasahara et al. of RIKEN (Saitama, Japan). Controls were wild-type littermates. Experiments were performed on transgenic and control mice at 3, 12, and 18 months of age. Consistent with previous studies (Kasahara et al., 2006), PolG transgenic mice weighed approximately 10% less than wild-type littermates (12-month-old WT = 31 ± 1 g, Tg = 28 ± 1 g, = 0.10). Transgenic mice have normal tolerance to anesthesia (ketamine/xylazine 70:7 mg kg−1), with less than approximately 5% mortality to anesthesia in our hands.

Tissue collection

Mice were euthanized by cervical dislocation. For protein analysis, eyes were enucleated and retina and optic nerve stump were dissected with a dissecting microscope (12.5× magnification; Leica, Wetzlar, Germany). Tissues were snap-frozen on dry ice and stored at −80 °C until use. For RNA analysis, tissues were first immersed in RNAlater® (Sigma-Aldrich, St. Louis, MO) prior to storage at −80 °C. For histology, eyes were enucleated and immersion-fixed in 4% paraformaldehyde in 0.1 m phosphate-buffered saline (PBS) at pH 7.4 for 3 h. Eyes were rinsed twice in PBS, immersed in 30% sucrose solution 4 °C overnight, and then embedded in Tissue-Tek OCT Compound (Sakura Finetek, Tokyo, Japan).

Transgene mRNA expression analysis

Total RNA was extracted from whole dissected retina, two areas of the brain (cerebral cortex and cerebellum), and other somatic tissues (liver, kidney, quadriceps muscle, heart) using a total RNA extraction kit (Promega, Madison, WI, USA). RNA concentration was determined using a Nanodrop spectrophotometer (ND-100 Nanodrop, Wilmington, DE, USA). The DNase-I-treated total RNA (30 ng) was reverse-transcribed using an anchor oligo15(dT) primer (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and M-MLV reverse transcriptase (Sigma-Aldrich, St. Louis, MO, USA). Two pairs of PCR primers that anneal at site-directed mutagenesis positions were used to specifically amplify either mutant (transgenic) or wild-type (endogenous) PolG cDNA. Primers used were transgenic mutant PolG primers, 5′-CTG CCT TAC TTG GAG GCT-3′ (forward) and 5′-CAA GCA GAC CTC CAC GG-3′ (reverse); endogenous wild-type PolG primers, 5′-CTG CCT TAC TTG GAG GCG-3′ (forward) and 5′-CAA GCA GAC CTC CAC GT-3′ (reverse); and GAPDH primers 5′-TGC ACC ACC AAC TGC TTA G-3′(forward) and 5′-GGA TGC AGG GAT GAT GTT C-3′ (reverse). Reverse transcription PCR products were visualized on 1.0% TBE–Agrose Gel with DNA visualization dye (GelRed; Biotium, Haywood, CA, USA). The absence of significant nuclear DNA contamination was verified by performing the same RT-PCR procedure in the absence of reverse transcriptase step.

Immunohistochemistry

Eyes were cryosectioned at 12 μm (Leica CM1850 Cryostat; Leica) on glass slides (Superfrost plus®; Menzel-Gläser, Braunschweig, Germany) and dried overnight at 37 °C. Sections were then labeled overnight at 4 °C with primary rabbit polyclonal antibodies to 4-hydroxy-2-nonenal (HNE, 1:100; Alpha Diagnostics, San Antonio, TX, USA; Liu et al., 2007; Chrysostomou et al., 2009), rabbit monoclonal antibodies to BRN3A (Abcam, Cambridge, UK), or mouse monoclonal antibodies to calmodulin-dependent protein kinase α-type II (CaMKII α, 1:100, Clone 6G9, Cayman Chemical, Ann Arbor, MI, USA). Sections were then subsequently incubated with biotin-conjugated secondary antibody (1:200; Sigma-Aldrich, St. Louis, MO, USA) for 2 h at room temperature and then with Alexa Fluor® Streptavidin conjugate (1:1000, Invitrogen; Invitrogen, Eugene, OR, USA) for 1.5 h. Sections were nuclear counterstained with Hoechst. Florescence label was visualized under florescence microscope (Nikon Instruments Inc., Melville, NY, USA).

Laser micro-dissection of inner retinal layer

Frozen eye sections were dehydrated through serial ethanol solutions of 50%, 75%, 95%, and 100% concentration. Laser micro-dissection of the retinal ganglion cell layer was performed using PALM® MicroBeam System (PALM Microlaser Technologies, Carl Zeiss, Hennigsdorf, Germany). Inner retinal tissues were lifted into PCR cup containing 30 μL buffer solution from DNAeasy® Blood & Tissue Kit (Qiagen, Valencia, CA, USA).

Long extension PCR amplification of mtDNA

Long extension (LX)-PCR was performed as previously described (Santos et al., 2002). LX-PCR can be used qualitatively as a sensitive method of detecting large deletions of mtDNA (Edgar et al., 2009) and mtDNA damage (Santos et al., 2002; Wang et al., 2010). Total genomic DNA of inner retina obtained from laser micro-dissection was isolated with DNAeasy® Blood & Tissue Kit (Qiagen). The quantification of purified DNA was performed using Nanodrop spectrophotometer (ND-100; Nanodrop, Wilmington, DE, USA). PCR primers used were 5′-GCC AGC CTG ACC CAT AGC CAT AAT AT-3′ (forward) and 5′-GAG AGA TTT TAT GGG TGT AAT GCG G-3′ (reverse). LX-PCR was performed using long-range polymerase supplied in GeneAmp® XL PCR Kit (Applied Biosystems, Branchburg, NJ, USA). The thermocycler program included an initial denaturation at 94 °C for 1 min, 35 cycles of 94 °C for 30 s, annealing at 55 °C for 30 s, and extension at 65 °C for 10 min, with final extension at 72 °C for 10 min. Template DNA used was 10 ng for each reaction. The primer set was used at 10 pm to amplify an expected product of approximately 10 kbp.

Assessment of point mutations a region mtDNA from retinal tissue

Pooled retinal DNA samples were prepared from equal quantities of DNA from five 12-month-old transgenic mice and four 12-month-old wild-type mice. The region spanning part of the cytochrome B (Cyt b) gene was amplified using PCR with Taq DNA polymerase (Invitrogen, Eugene, OR, USA) and the primers 5′-ACA CAG CAT TCA ACT GCG ACC AA-3′ (forward) and 5′-TCG GGT CAA GGT GGC TTT GTC T-3′ (reverse). The PCR products were then ligated to pCR2.1-TOPO® plasmid vector (Invitrogen) and cloned in TOP10® (Invitrogen) chemocompetent cells. The plasmid DNA from colonies was extracted using PureLink Quick Plasmid Miniprep Kit (Invitrogen), and DNA was sequenced (Australian Genome Research Facility, Melbourne, Australia). Cloning was repeated on DNA of a clone from Tg retinal DNA to assess any background mutations introduced by PCR. Approximately 7000–8000 bases were sequenced per group. Sequence data were analyzed using ClustalX (University College Dublin, Dublin, Ireland) and FinchTV (Geospiza Inc., Seattle, WA, USA) by comparing with published mitochondrial DNA sequences from C57BL/6J (GenBank EF108336.1, National Center for Biotechnology Information).

SD–PAGE immunoblotting

Whole retina and optic nerve stump were lysed with radioimmunoprecipitation assay buffer [1% TritonX-100, 158 mm NaCl, 5 mm EDTA, 10 mm Tris (pH 7.0), 0.1% sodium orthovanadate, 0.1% aprotinin, and 2% phenylmethylsulfonyl fluoride], sonicated, and centrifuged at 18 000 g for 20 min. Equal amount of protein homogenates were electrophoresed in 10–12% acrylamide SDS/PAGE gel system (Bio-Rad Laboratories, Hercules, CA, USA). Proteins were transferred to polyvinylidene difluoride membranes (PVDF, Amersham Hybond-PTM; GE Healthcare, Buckinghamshire, UK). After blocking with 5% skim milk in PBS-T (phosphate-buffered saline, 0.05% v/v Tween 20, pH 7.4) at room temperature for 1 h, preparations were incubated with the primary antibody at 1:1000–1:10000 dilutions in PBS-T with 5% fetal bovine serum. Membranes were then probed with respective secondary horseradish peroxidase (HRP)-labeled antibodies (1:2000–1:10000 dilution). Immunolabeling was detected using chemoluminescence method (Amersham ECL; GE Healthcare). Antibodies used include heme oxygenase-1 (HO-1) polyclonal antibody (1:5000; Stressgen Bioreagents, San Diego, CA, USA), anti-actin monoclonal antibody (1:2000; Sigma-Aldrich, St. Louis, MO, USA), anti-SIRT1 polyclonal antibody (Cell Signaling Technology, Danvers, MA, USA), anti-nuclear respiratory factor 1 (NRF1) polyclonal antibody (1:2000; Abcam, Cambridge, UK), anti-OXPHOS cocktail (MitoProfile® Total OXPHOS Rodent WB Antibody Cocktail, Mitosciences®, Eugene, OR, USA), HRP-conjugated goat anti-rabbit secondary antibody (1:2000; Bio-Rad), and HRP-conjugated goat anti-mouse secondary antibody (1:2000; Bio-Rad). Band intensity was determined by densitometry using Image J software (NIH, Bethesda, MD, USA).

Mitochondrial enzyme activity

Mitochondrial enzyme activity of citrate synthase and cytochrome c oxidase (COX) was measured from whole retinal homogenate using a modification of an established technique (Trounce et al., 1996). In brief, mouse retina (n = 5 mice from each cohort with no previous anesthesia exposure) were dissected from the eye cup as described above and immediately placed into microcentrifuge tubes in 100 μl isolation buffer (210 mm mannitol, 70 mm sucrose, 5 mm HEPES, 1 mm K-EGTA and 0.5% BSA, pH 7.2 with KOH) kept cold on an ice slurry. Two retinas from each animal were then fully homogenized using a Dounce tissue homogenizer, and protein concentration was determined using Lowry protein assay (Lowry et al., 1951). Mitochondrial enzyme activity assays were performed using a single-beam spectrophotometer (Varian Cary 3000; Varian Inc, Palo Alto, CA, USA) fitted with a thermostat-regulated heating block. All assays were performed at 30 °C in 1 mL of medium in quartz cuvettes.

Mitochondrial complex IV cytochrome C oxidase (ferrocytochrome C:oxygen oxidoreductase, EC 1.9.3.1) activity was measured by adding 40 μg whole-cell homogenate to assay medium containing degassed 10 mm KH2PO4, pH 7.2, and 2 mm ferricytochrome C. Oxidation rate of reduced ferrocytochrome C was monitored at 550 nm (extinction coefficient of 19.0 mm−1 cm−1) for 60 s. Mitochondrial citrate synthase (EC 4.1.3.7) activity was measured by adding 20 μg whole-cell homogenate to a reaction mixture containing 125 mm Tris–HCl, 100 μm DTNB [5,5′-dithiobis(2-nitrobenzoic acid)], and 300 μm acetyl coenzyme A and incubated at 30 °C for 10 min. The reaction was initiated by the addition of 500 μm oxaloacetate, and DTNB reduction rate was measured at 412 nm for 120 s (extinction coefficient E = 13.6 mm−1 cm−1).

Retinal thickness measurements and retinal ganglion cell counts

For each eye, results were averaged across three sagittal retinal cross-sections (12 μm thickness) through the optic nerve and pupil, and sampling for thickness measurements and retinal ganglion cell counts was performed at ×200 magnification within the central 30 degree of the retina, with at least four samples taken per retinal cross-section (John et al., 1998). The thickness of retinal layers was measured on digital images of Hoechst-labeled cryosections. At each measurement location, the thickness of the outer nuclear layer (ONL) and the thickness of the retina, from inner to outer limiting membrane (ILM-OLM), were recorded. The ratio of ONL to ILM-OLM was used for analysis (Chrysostomou et al., 2009).

Electroretinography

The full-field (Ganzfeld) scotopic electroretinogram (ERG) was used to assess retinal function at baseline and during acute IOP challenge according to the established protocols (Kong et al., 2009). In brief, animals were dark-adapted overnight (> 12 h) with experimental manipulation being performed using head-mounted night vision goggles (Scout2; Trivisio Prototyping GmbH, Dreieich, Germany). Animals were anesthetized with intraperitoneal injection of ketamine/xylazine (70: 7 mg kg−1; Troy Laboratories Pty Ltd., Smithfield, NSW, Australia) followed by supplementation with 20% of the initial dose every 30 min (Saszik et al., 2002; Peachey & Ball, 2003). Mydriasis was achieved with 1 drop of tropicamide (0.5%, Alcon Laboratories Inc., Fort Worth, TX, USA) and phenylephrine (2.5%, Minims; Chauvin Pharmaceuticals, Surrey, UK). Corneal anesthesia was achieved with a single drop of proxymetacaine hydrochloride (0.5%, Alcon Laboratories Inc.). Animals were lightly secured to a platform with wire loops across the upper back and nose. A circulating warm water heating pad was used to maintain body temperature (37–38 °C). Signals were recorded using silver/silver chloride electrodes (99.99% purity, 0.329 mm ¼ 29 G A&E Metal Merchants, Sydney, NSW, Australia) with a scleral coil reference and corneal active (He et al., 2006; Weymouth & Vingrys, 2008). Both of these were referenced to a stainless steel ground (F-E2-60, Grass Technologies, West Warwick, RI, USA) inserted into the tail. Electrode placement and anterior chamber cannulation with a 50-μm-diameter borosilicate needle were performed using monocular night vision scopes (NVMT1; Yukon, Mansfield, TX, USA) fitted on the eye piece of a dissecting microscope (MZ6; Leica). Light stimuli were brief (1 ms) white flashes (5 Watt white LEDs, 5500°K; Luxeon Calgary, Alberta, Canada) delivered via a Ganzfeld integrating sphere (Photometric Solutions International, Huntingdale, VIC, Australia). Signals were amplified 1000× and recorded with band-pass setting of −3 dB at 0.3–1000 Hz (P511 AC Preamplifier, Grass Telefactor, West Warwick, RI, USA) and digitized with a 4-kHz acquisition (Powerlab 8SP, ADIntruments, Bella Vista, NSW, Australia).

Following electrode placement, the anterior chamber of the left eye was cannulated using a borosilicate needle (approximately 50 μm, 1B100-6, WPI) connected via polyethylene tubing (0.97 mm inner diameter; Microtube Extrusions, North Rocks, NSW, Australia) to a pressure transducer (Transpac IV; Abbott Critical Care Systems, Sligo, Ireland), which was connected in series with a sterile Hanks’ balanced salt solution reservoir (JRH Biosciences, Lenexa, KS, USA). By altering the height of the reservoir, the level of intraocular pressure in the eye can be altered. Patency of the needle was determined by observation for anterior chamber distension under microscope. Previous work by our laboratory showed that the needle aperture has low resistance and thus a negligible effect on pressure equilibration within the anterior chamber (Kong et al., 2009).

Following cannulation of the anterior chamber of the left eye, intraocular pressure was raised in a stepwise manner starting from 25 mmHg up to 50 mmHg at 5 mmHg increments, and from there up to 80 mmHg in 10 mmHg increments. At each step, IOP was stabilized for 10 min before ERG recording. ERG recordings were made at 10-min intervals at baseline and during IOP elevation. Stimuli to elicit the scotopic threshold response (STR, −4.54 log cd.s.m−2, 30 averages, 3-second interstimulus interval), the ON-bipolar cell response (P2, −2.23 log cd.s.m−2, single flash), and a photoreceptoral P3 response (single flash, 0.34 log cd.s.m−2) were employed. All procedures (ERG and IOP elevation) were performed with the genotype of animals masked to the investigator at the time of the procedure.

Electroretinogram analysis

Response amplitudes of the positive STR (pSTR) was measured at a fixed time of 120 ms after the stimulus flash (Kong et al., 2009), which coincides with the pSTR peak in control responses. The photoreceptoral response (P3) and ON-bipolar cell response (P2) were analyzed as previously described (Bui & Fortune, 2004; Bui et al., 2005; Weymouth & Vingrys, 2008; Kong et al., 2009).

Statistical analysis

Statistical analysis was performed using commercially available software (SPSS® v 15.00, SPSS Inc., Chicago, IL, USA). Two-tailed Student’s t-tests assuming unequal variance were used for comparison between groups with alpha of 0.05. Two-way anova was used to test the statistical difference between groups across age or intraocular pressure, and Tukey’s test was used for post hoc analysis of subcategories.

Acknowledgments

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

This research was supported by National Health and Medical Research Council Grants (475603, JGC; 400127 BVB), Ophthalmic Research Institute of Australia Grants, Glaucoma Australia Fund, Henry Greenfield Research Fund and Edols Trust Fund. Centre for Eye Research Australia receives Operational Infrastructure Support from the Victorian Government of Australia.

Author contributions

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

Research was designed by J.C., I.T., A.V., B.B., and Y.K. and was performed by Y.K., I.T., N.V., V.C., and H.W. Data analysis was carried out by Y.K., N.V., V.C., B.B., J.C., and I.T. Manuscript was written by Y.K., J.C., I.T., B.B., A.V., N.V., and V.C.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. Author contributions
  9. References
  10. Supporting Information

Fig. S1 Retinal histology of aging PolG transgenic mice.

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
ACEL_690_sm_FigS1.doc24KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.