Long-term hyperglycaemia decreases gastrocnemius susceptibility to permeability transition


António Ascensão, Research Centre in Physical Activity, Health and Leisure, Faculty of Sport Sciences, University of Porto, Rua Dr. Plácido Costa, 91, 4200-450 Porto, Portugal. Tel.: +351 225074774; fax: +351 225500689; e-mail: aascensao@fcdef.up.pt


Eur J Clin Invest 2010; 40 (4): 319–329


Background  Hyperglycaemia-resulting in mitochondrial bioenergetics’ complications is associated with skeletal muscle dysfunction. The aim of this work was to analyse the effect of long-term severe hyperglycaemia on gastrocnemius mitochondrial bioenergetics, with special relevance on the susceptibility to mitochondrial permeability transition pore (MPTP) opening.

Methods  Sixteen adult (6- to 8-week-old) male Wistar rats were randomly divided into two groups (n = 8/group): control and diabetic. A single dose (50 mg kg−1) of streptozotocin (STZ) was administrated i.p. to induce hyperglycaemia. In vitro mitochondrial oxygen consumption rates, membrane potential (Δψ) fluctuations, MPTP induction as followed by osmotic swelling and extramitochondrial calcium movements and caspase 9-like activity were evaluated 18 weeks after STZ treatment.

Results  STZ treatment induced an increase in state 4 and a decrease in the respiratory control ratio with complex I substrates (P < 0·05), whereas no differences were observed using complex II substrates. In both conditions, no significant differences were observed when measuring maximal Δψ, although STZ treatment increased Δψ during ADP-induced depolarization when succinate was used. The most critical result was that muscle mitochondria isolated from STZ-treated rats showed a decrease susceptibility to MPTP induction by calcium, as followed by two different experimental protocols. Interestingly, the protection was accompanied by a decrease in muscle caspase 9-like activity.

Conclusions  These data demonstrate that 18 weeks of STZ treatment lead to a decrease in gastrocnemius mitochondrial respiratory control ratio and to decreased calcium-dependent mitochondrial MPTP. Results from this and other works suggest that mitochondrial effects of hyperglycaemia are time and organ specific.


Diabetes mellitus is a group of common metabolic and degenerative diseases clinically characterized by a phenotype of increased blood glucose, decreased insulin secretion and/or resistance and impairments in carbohydrate, fat and protein metabolism, associated with a variety of pathophysiological complications [1], responsible for the decreased quality of life of millions of people all over the world and for 5·2% of the total number of deaths worldwide in 2000 [2]. Type I diabetes is characterized by an absolute insulin deficiency, requiring daily insulin replacement. To understand the physiological and pathological changes of the disease, animal models of diabetes of hyperglycaemia are important research tools, allowing a tight control over experimental conditions, which is difficult with humans [3]. In this study, we used streptozotocin (STZ)-induced rats as a model for human type I diabetes mellitus. STZ-induced hyperglycaemia is obtained after selective destruction of β-cell by STZ, a broad spectrum antibiotic with diabetogenic effects. STZ-treated rats present many characteristics seen in insulin-dependent diabetic human patients such as hypoinsulinaemia, hyperglycaemia, ketonuria and hyperlipidaemia. Therefore, this model is of great use to evaluate the alterations promoted by uncontrolled type I diabetes. Recent and early studies have reported morphological, biochemical and functional data suggestive of mitochondrial dysfunction in experimental diabetes induced by STZ administration [4–9].

Besides the energy producing process, mitochondria are also critical organelles in cell osmotic regulation, modulation of redox status and pH control, signal transduction and in the maintenance of calcium (Ca2+) homeostasis [10]. Mitochondria also provide important contributions for the regulation of cell fate through its role in several cell death signalling pathways [11]. It has been described that induction of cell death is often associated with disrupted mitochondrial membrane structure. In fact, the low permeability of inner mitochondrial membrane to ions and solutes is critical to the mechanism of oxidative phosphorylation [10]. When mitochondrial Ca2+ accumulates over a threshold, mitochondria can undergo increased membrane permeability to molecules with masses up to 1500 Da, in an event called mitochondrial permeability transition (MPT). The MPT is caused by the opening of a multiprotein pore (MPTP) composed by several proteins of the outer and inner mitochondrial membranes [12]. It is well established that the susceptibility to MPTP opening occurs under conditions of increased oxidative stress and Ca2+ overload [12,13]. Despite some debate on the role of MPTP in apoptosis, the majority of the literature published to date supports a role for the MPTP in the release of proapoptotic proteins during apoptosis [12,14].

Skeletal muscle, which makes up 40% of the body mass of humans and other mammalian species, is the primary tissue responsible for the peripheral disposal of glucose in response to a glucose or insulin challenge [15]. By nature of its metabolic processes, the skeletal muscle has a high reliance on mitochondrial oxidative phosphorylation.

It is described that diverse diabetic tissues have distinct capacities to respond to stress stimuli. For instance, microarray-based studies revealed tissue-specific transcriptome response to STZ-induced diabetes, suggesting that tissue-specific mechanisms are involved in the adaptation to diabetes complications [16]. In addition, Najemnikova et al. [17] recently reported that heat shock factor 1 response to heat stress was increased in liver, heart and kidney from STZ treated rats but not in gastrocnemius and plantaris skeletal muscles. Moreover, despite absence of differences in the constitutive levels of heat shock proteins (HSP) 25 and 72 between diabetic and non-diabetic tissues, all with the exception of the skeletal muscles overexpressed HSP25 after heat stress. The data published so far clearly indicate that differences exist regarding organ response to hyperglycaemia, which are further modulated by the time and type of treatment used to produce hyperglycaemia.

There is an absence of literature regarding the effects of hyperglycaemia, namely induced by STZ, on skeletal muscle mitochondrial bioenergetics and MPTP regulation. The hypothesis behind this study is that STZ-induced 18 weeks hyperglycaemia causes alterations in mitochondrial bioenergetics in gastrocnemius muscle, as well as an increase of susceptibility to MPTP induction, as previously described for the heart [18]. Decreased mitochondrial function and increased susceptibility to MPTP opening during uncontrolled type I diabetes may explain some of the skeletal muscle phenotype observed in patients harbouring that condition [19].



Streptozotocin [N-(Methylnitrosocarbamoyl)-a-d-glucosamine] was obtained from Sigma Chemical Co. (St Louis, MO, USA), and prepared prior to use in 100 mM citrate, pH 4·5. Calcium Green 5-N was obtained from Molecular Probes (Eugene, OR, USA). All other reagents and chemicals used were of the highest grade of purity commercially available.


Sixteen Wistar male rats (aged 6–8 weeks, weighing 200 g at the beginning of the experiments) were used. During the experimental protocol, animals were housed in collective cages (two rats per cage) and were maintained in a room at normal atmosphere (21–22 °C; ∼50–60% humidity) receiving food and water ad libitum in 12 h light/dark cycles. Male Wistar rats weighing about 200 g were randomly divided into two groups (n = 8 per group): control (C) and diabetic (STZ). The Ethics Committee of the Scientific Board of Faculty of Sport Sciences, University of Porto, Portugal, approved the experimental protocol, which followed the Guidelines for Care and Use of Laboratory Animals in research.

Induction and characterization of STZ-induced hyperglycaemia

To induce hyperglycaemia, one group was injected intraperitoneally with a single injection of STZ (50 mg kg−1), after a 16-h fasting period. The volume used was always 0·5 mL per 200 g body weight. Control animals were injected with the same volume of citrate solution. In the following 24 h, animals were orally fed with glycosilated serum to avoid hypoglycaemia resulting from massive destruction of β cells and consequent release of intracellular insulin. During this period, weight was measured and glycaemia was determined from the tail vein. Values were taken in fasting conditions just before STZ administration and in non-fasting conditions in the weeks after. If feeding blood glucose in the tail vein exceeded 250 mg dL−1, animals were considered hyperglycaemic.

Glycaemia and HbA1C evaluation

Blood glucose concentration was determined immediately after animal sacrifice (Glucometer-Elite; Bayer, Berlin, Germany). The glycated haemoglobin (HbA1C) values were determined in blood collected at the time of animal sacrifice through ionic exchange chromatography (Abbott Imx Glicohemoglobin; Abbott Laboratories, Amadora, Portugal).

Tissue preparation and skeletal muscle mitochondria isolation

After exposed to isoflurane inhalation in a glass container to perform anaesthesia, the animals were sacrificed by cervical dislocation and gastrocnemius muscles were extracted for the preparation of isolated mitochondria. Before mitochondrial isolation, a portion of approximately 20–25 mg of gastrocnemius muscle was separated, homogenized in homogenization buffer (20 mM Tris, 137 mM NaCl, 0·2 mM EDTA, 0·5 mM EGTA, 1% Triton X-100, 10% Glycerol, pH 7·6) using a Teflon pestle on a motor-driven Potter-Elvehjem glass homogenizer at 0–4 °C 3–5 times for 5 s at speed low setting, with a final burst at a higher speed setting. Homogenates were centrifuged (2 min at 2000 g, 4 °C, to eliminate cellular debris) and the resulting supernatant was stored at −80 °C for later determination of caspase 9 activity. Protein content from muscle homogenates were assayed using bovine serum albumin as standard according to Lowry et al. [20].

Skeletal muscle mitochondria were prepared by conventional methods involving differential centrifugation, as previously described [21]. Briefly, muscles were immediately excised and minced in ice-cold isolation medium containing 100 mM sucrose, 0·1 mM EGTA, 50 mM Tris–HCl, 100 mM KCl, 1 mM KH2PO4 and 0·2% BSA, pH 7·4. Minced blood-free tissue was rinsed and suspended in 10 mL of fresh medium containing 0·2 mg mL−1 bacterial proteinase (Nagarse E.C.3·4·21·62, Type XXVII; Sigma) and stirred for 2 min. The sample was then carefully homogenized with a tightly fitted Potter-Elvehjen homogenizer and a Teflon pestle. After homogenization, three volumes of Nagarse-free isolation medium were added to the homogenate, which was then fractionated by centrifugation at 700 g for 10 min. The resulting pellet was removed and the supernatant suspension centrifuged at 10 000 g during 10 min. The supernatant was decanted and the pellet was gently resuspended in isolation medium (1·3 mL per 100 mg of the initial tissue) and centrifuged at 7000 g for 3 min. The supernatant was discarded and the final pellet, containing the mitochondrial fraction, was gently resuspended (0·4 μL mg−1 initial tissue) in a medium containing 225 mM mannitol, 75 mM sucrose, 10 mM Tris and 0·1 mM EDTA, pH 7·4. All mitochondrial isolation procedures were performed at 0–4 °C. Mitochondrial protein concentration was spectrophotometrically estimated according to Lowry et al. [20] using bovine serum albumin as standard. The mitochondrial suspensions were used within 4 h after the excision of the muscles and were maintained on ice (0–4 °C) throughout this period. There was no significant alteration of the mitochondrial RCR between the first and the last measurements in the same animal.

Measurement of mitochondrial respiratory activity

Mitochondrial respiratory function was polarographically measured using a Clark-type oxygen electrode (Yellow Springs Instruments, Yellow Springs, OH, USA) connected to a recorder (BD 112; Kipp & Zonen, Delft, Netherlands) in a thermostated water-jacketed sealed glass chamber. Reactions were conducted in a 1 mL closed thermostated (25 °C) and magnetically stirred glass chamber containing 0·5 mg of mitochondrial protein in a reaction buffer containing 225 mM mannitol, 75 mM sucrose, 10 mM Tris, 10 mM KCl, 10 mM K2HPO4 and 0·1 mM EDTA, pH 7·5 in accordance with Tonkonogi et al. [22]. After 1-min equilibration period, mitochondrial respiration was initiated by adding pyruvate (5 mM) plus malate (2 mM) or succinate (10 mM) plus rotenone (4 μM). State 3 respiration was determined after adding 800 nmol ADP; state 4 respiration was measured as the rate of oxygen consumption which follows total ADP phosphorylation. The respiratory control ratio (RCR, the ratio between state 3 and state 4) and the ratio between the amount of ADP phosphorylated per oxygen consumed (ADP/O) were calculated according to Estabrook [23] and by using 235 nmolO2 per mL as the value for the solubility of oxygen at 25 °C. In an independent experiment, oligomycin (final concentration 1·5 μg mL−1) and carbonyl cyanide m-chlorophenylhydrazone (CCCP, 2 μM) were added during state 3 respiration promoted by adding saturated (2 mmol) amount of ADP. An estimate of the level of inner membrane integrity was made by measuring the ratio of uncoupled respiration induced by CCCP addition versus the rate in state oligomycin (CCCP/oligomycin) using succinate as substrate, according to Benamar et al. [24] with adaptations. It has been described that the ratio increases as the inner membrane impermeability to protons increases [24].

Mitochondrial membrane potential

Mitochondrial transmembrane potential (Δψ) was monitored indirectly based on the activity of the lipophilic cation tetraphenylphosphonium (TPP+) using a TPP+ selective electrode prepared in our laboratory as described by Kamo et al. [25] using a AgCL reference electrode (Tacussel, Model MI 402) and as previously described [26]. No correction factor was used to correct the passive binding contribution of TPP+ to membrane potential as the purpose of this study was to show the relative changes in the potential, rather than absolute values. As a consequence, a slight overestimation of the Δψ values is anticipated.

A mitochondrial matrix volume of 1·1 μL mg−1 of protein was assumed. Reactions were carried out in 1 mL of reaction buffer containing 225 mM mannitol, 75 mM sucrose, 10 mM Tris, 10 mM KCl, 10 mM K2HPO4 and 0·1 mM EDTA, pH 7·5, supplemented with 3 μM TPP+ and 0·5 mg mL−1 of protein with the temperature maintained at 25 °C. For measurements of Δψ with complex I-linked substrates, energization was carry out with 2 mM of malate and 5 mM of pyruvate and ADP-induced phosphorylation was achieved by adding 800 nmol ADP. For measurements of Δψ with complex II-linked substrates, 10 mM succinate supplemented with 4 μM rotenone were added to the medium containing 3 μM TPP+ and mitochondria. The lag phase, which reflects the time needed to phosphorylate the added ADP, was also measured for both substrates.

Determination of mitochondrial swelling

Mitochondrial osmotic volume changes were followed by monitoring the classic decrease of absorbance at 540 nm with a Jasco V-560 spectrophotometer. The reaction was continuously stirred and the temperature was maintained at 25 °C. The assays were performed in 2 mL of reaction medium containing 200 mM sucrose, 10 mM Tris, 10 μM EGTA, 5 mM KH2PO4, pH 7·4, supplemented with 4 μM rotenone, 10 mM succinate and 240 μM of calcium with 0·5 mg mL−1 protein. Control trials were made by using 1 μM of cyclosporin-A, the MPT inhibitor [27].

Mitochondrial calcium loading capacity

Extramitochondrial free calcium was measured with the hexapotassium salt of the fluorescent probe Calcium Green 5-N. Muscle mitochondria (0·1 mg mL−1) were resuspended in 2 mL reaction buffer containing 200 mM Sucrose, 10 mM Tris, 10 μM EGTA, 5 mM KH2PO4 (pH 7·4) supplemented with 4 μM rotenone. Prior calibration was made using subsequent calcium pulses to a final concentration of 300 nmol of calcium per mg prot. Calcium uptake was initiated by adding 10 mM succinate to the reaction chamber. Free calcium was monitored with 100 nM Calcium Green 5-N. Fluorescence was continuously recorded in a water-jacketed cuvet holder at 25 °C with a Perkin-Elmer LS-50B fluorescence spectrometer with excitation and emission wavelengths of 506 and 531 nm respectively. From the fluorescence decrease (i.e. calcium entry to mitochondria) and with proper calibration (as previously described), the rate of calcium uptake was determined. After calcium accumulation, fluorescence increased (calcium release from mitochondria). At the end of each experiment, a calibration was performed for calculating the maximum and minimum fluorescence signal, with an excess of calcium and EGTA respectively. Control trial was made using 1 μM cyclosporin-A. The used dissociation constant for the complex calcium probe was 4 μM [28].

Caspase-like activity assay

To measure caspase 9-like activity, aliquots of gastrocnemius tissue homogenate containing 50 μg (for caspase 9) were incubated in a reaction buffer containing 25 mM Hepes (pH 7·4), 10% sucrose; 10 mM DTT, 0·1% CHAPS and 100 μM caspase substrate Ac-LEHD-pNA (235400, Calbiochem, Darmstadt, Germany) for 2 h at 37 °C. Caspase-like activity was determined by following the detection of the chromophore p-nitroanilide after cleavage from the labelled substrate Ac-LEHD-p-nitroanilide. The method was calibrated with known concentrations of p-nitroanilide (Calbiochem).

Statistical analysis

Mean and standard error of the mean were calculated for all variables in each group. Students’t-test for independent measures was used to compare differences between groups. Statistical Package for the Social Sciences (spss Inc, Chicago, IL, USA; version 12·0) was used for all analyses. The significance level was set at 5%.


Characterization of the animals

Body and gastrocnemius muscle weights, heart to body weight ratio, blood glucose and glycated haemoglobin are shown in Table 1. After 18 weeks of STZ treatment, whole animal and gastrocnemius muscle weights decreased significantly (P < 0·05). However, as muscle/body weight was not affected by hyperglicaemic conditions, gastrocnemius weight decrease was not tissue specific but rather a result of the whole animal body weight decrease.

Table 1.    Characterization of the animals used in this study
GroupsBody weight (g)Muscle weight (g)Muscle/body ratio (mg g−1)Glycaemia (mg dL−1)HbA1c (% total Hb)
  1. Data are mean ± SEM; * vs. control (P < 0·05). Blood glucose and glycated haemoglobin (HbA1c) were determined as described in methods.

Control452·0 ± 11·45·5 ± 0·311·2 ± 0·2105·5 ± 1·84·2 ± 0·5
STZ255·5 ± 3·3*3·2 ± 0·6*12·4 ± 0·4580·5 ± 12·8*12·6 ± 0·5*

Blood glucose concentration, assessed immediately after the sacrifice of the animals, was significantly higher in STZ-diabetic rats when compared with controls. To estimate the severity of diabetes, glycated haemoglobin (HbA1c) was also evaluated since HbA1c is a very helpful marker to determine the severity of diabetes, indicating the average glycaemia presented 2–3 months prior to the analysis. The measured content of HbA1c confirmed that blood glucose levels were significantly increased in STZ-treated rats (Table 1).

Mitochondrial oxygen consumption

To elucidate the effects of STZ treatment on skeletal muscle mitochondrial respiration, we determined oxygen consumption using both NADH and succinate-linked substrate oxidation through mitochondrial complex I and complex II respectively (Table 2). As it can be observed in Table 2, STZ treatment decreased the RCR (49%) and increased state 4 respiration (72%) of malate-pyruvate-energized muscle mitochondria. No changes were noted regarding state 3 and ADP/O when comparing control and STZ groups.

Table 2.    Respiratory parameters (state 3 and state 4 rates, respiratory control rate - RCR and ADP/O ratios) of malate (2 mM)/pyruvate (5 mM) and succinate (10 mM) energized gastrocnemius muscle mitochondria isolated from control and STZ rats (18 weeks after treatment)
Malate-pyruvateState 3 (natomO min−1 mg prot−1)State 4 (natomO min−1 mg prot−1)RCRADP/O
  1. Data are mean ± SEM for skeletal muscle mitochondria (0·5 mg mL−1 protein) obtained from different mitochondrial preparations for each experimental group. Oxidative phosphorylation was polarographically measured at 25 °C in a total volume of 1 mL. Respiration medium and other experimental details are provided in the Methods section.

  2. RCR, respiratory control ratio (state 3/state 4); ADP/O, number of nmol ADP phosphorylated by natom O consumed. * vs. control (P < 0·05).

Control167·2 ± 5·715·9 ± 0·59·6 ± 0·62·5 ± 0·1
STZ136·7 ± 5·727·3 ± 2·5*4·9 ± 0·3*2·4 ± 0·2
Control173·3 ± 6·453·7 ± 2·73·3 ± 0·31·7 ± 0·1
STZ152·1 ± 3·044·7 ± 4·63·6 ± 0·11·6 ± 0·2

When succinate was used, no significant changes were detected in the rate of state 3 and state 4 respiration and on the RCR or on the ADP/O ratio (Table 2).

By using succinate as a substrate, oligomycin-inhibited state 3 respiration, uncoupled respiration (state CCCP) and the ratio CCCP/oligomycin (which is a measure of mitochondrial integrity) were unaffected by treatment, which is in accordance with the results obtained for the different respiratory rates (Table 2) using the same substrate.

Mitochondrial transmembrane electric potential

To attain further insight into the effects of STZ-induced severe hyperglycaemia on mitochondria capacity to perform oxidative metabolism, Δψ fluctuations associated with mitochondrial bioenergetics were also investigated. Considering both trials with complex I and complex II-related substrates, mitochondria isolated from control and STZ treated animals did not show significant differences regarding energization, ADP-induced depolarization and repolarization Δψ fluctuations, with the exception of depolarization when using succinate as substrate (Table 3). Upon energization with malate plus pyruvate, the two groups of muscle mitochondria (isolated from control and STZ animals) developed a Δψ around −198 and −185 mV respectively, although the difference was not statistically significant.

Table 3.    Effect of STZ treatment on gastrocnemius muscle mitochondrial transmembrane potential (Δψ) fluctuations with malate plus pyruvate or succinate as substrates
Malate-pyruvateMaximal energizationΔψ (−mV)Lag phase (s)
ADP depolarizationRepolarization
  1. Data are mean ± SEM for skeletal muscle mitochondria (0·5 mg mL−1 protein) obtained from different mitochondrial preparations for each experimental group. Table shows the average response of mitochondrial membrane potential developed, with malate (2 mM) plus pyruvate (5 mM) or succinate (10 mM) (energization), the decrease in membrane potential after ADP addition (depolarization), the repolarization value after ADP phosphorylation. Mitochondrial transmembrane potential was measured using a TPP+-selective electrode at 25 °C in a total volume of 1 mL. Reaction medium and other experimental details are provided in the Methods section. * vs. control (P < 0·05).

Control198·1 ± 2·5179·4 ± 2·5199·1 ± 3·9102·0 ± 9·2
STZ184·9 ± 2·9169·1 ± 4·9196·1 ± 4·1118·0 ± 3·5
Control209·7 ± 4·6170·1 ± 1·8205·8 ± 3·7122·8 ± 6·8
STZ215·3 ± 1·0183·7 ± 2·2*209·7 ± 2·1132·0 ± 11·4

No significant changes were observed regarding the phosphorylative lag phase with both substrates (Table 3).

Mitochondrial swelling caused by MPT induction

We hypothesize that the persistent and severe hyperglycaemic condition may favour the modulation of skeletal muscle mitochondrial function by altering the susceptibility to MPTP opening. Following this idea, the possible effects of 18 weeks STZ treatment on in vitro induction of the calcium-induced MPTP was also investigated. Figure 1a represents a typical recording from swelling experiments carried out in mitochondria from control and STZ-treated rats. As pictured, the addition of calcium results into a decrease in the mitochondrial suspension absorbance with three distinct phases. Initially, an increase in absorbance is observed, which most likely results from the formation of opaque calcium crystals inside mitochondria [29]. Upon MPTP opening, a decrease of absorbance with a slow followed by a fast kinetic rate was observed. Incubation of mitochondrial suspension with cyclosporin-A, a specific MPTP inhibitor [27], completely inhibits the absorbance decrease after calcium addition, which demonstrates the association with MPTP opening. Figure 1 (panels b–e) shows several end-points measured from the recordings obtained, namely swelling amplitude (the difference between the initial suspension absorbance and the final value upon absorbance stabilization), the average swelling rate, maximal swelling rate and time elapsed until the faster swelling kinetics starts. The results of swelling experiments demonstrate that gastrocnemius mitochondria isolated from STZ-treated rats are less susceptible to calcium-induced PTP opening when compared with controls.

Figure 1.

 Effect of 18-week STZ treatment on mitochondrial MPTP induction in succinate-energized gastrocnemius muscle mitochondria. The absorbance of mitochondrial suspension was followed at 540 nm. Mitochondria were incubated as described in the methods section. A 240 μM calcium pulse, corresponding to 960 nmol mg protein−1 was added to 0·5 mg of mitochondrial protein to attain the cyclosporin A-sensitive swelling, because of MPTP opening. Cyclosporin-A curve corresponds to both the control and STZ groups. Panel a – Typical recording of mitochondrial swelling (decrease of optical density) of different mitochondrial preparations for each experimental group is shown; Panel b – swelling amplitude; Panel c – average swelling rate; Panel d – maximal swelling rate and Panel e – Time to Vmax, i.e. the lag phase elapsed until large scale swelling starts. Data are mean ± SEM; * vs. control (P < 0·05). 258 × 139 mm (96 × 96 DPI).

Calcium loading capacity

The susceptibility of gastrocnemius mitochondria to MPTP opening was further analysed by following calcium fluxes through the mitochondrial membranes. Figure 2a represents a typical recording of the fluorimetric monitoring of extramitochondrial calcium movements through the fluorescent probe Calcium Green 5-N of both groups. Surprisingly, gastrocnemius mitochondria isolated from STZ rats were not only able to accumulate a higher amount of calcium (Fig. 2b) when compared with their control counterparts, but also released a significantly lower amount of calcium than controls (Fig. 2c). Cyclosporin-A was able to prevent calcium release to the external buffer (Fig. 2a).

Figure 2.

  Effect of 18 weeks of STZ treatment on extramitochondrial calcium fluxes in succinate energized gastrocnemius muscle mitochondria. Panel a – Typical measurements of calcium movements using the fluorescent calcium sensitive probe Calcium Green 5-N as described in the Methods section. The previous calibration curve is not shown. The reactions were carried out in 2 mL of reaction medium supplemented with 0·1 mg protein, 4 μM rotenone, 100 nM Calcium Green 5-N. Calcium entry into mitochondria (indicated by fluorescence decay) was initiated by 10 mM succinate addition. Fluorescence was monitored continuously and the increase in signal indicates calcium release. The release of calcium from mitochondria of both experimental groups was inhibited by 1 μM cyclosporin-A, indicating that the observed release was resulting from MPTP opening. Calcium fluxes are representatively expressed as arbitrary units of fluorescence (AFU); Panel b – mitochondrial calcium uptake; and Panel c – mitochondrial calcium release. Data are mean ± SEM; * vs. control (P < 0·05). 244 × 173 mm (96 × 96 DPI).

Caspase 9 activity

One indication of intrinsic mitochondrial-driven activation of apoptotic cell death is the activity of caspase 9. The analysis revealed that STZ-induced severe and prolonged hyperglycaemia decreased the basal activity of caspase 9 in gastrocnemius muscle (Fig. 3).

Figure 3.

 Effect of 18 weeks of STZ treatment on caspase 9 activity in gastrocnemius muscle. Caspase- 9-like activity was measured by following the cleavage of the colorimetric substrates Ac-LEHD-pNA. The caspase-like activity was expressed by [pNA] released nM per Dg prot. Data are expressed as mean ± SEM; * vs. STZ (P < 0·05).


Overview of principal findings

The present investigation provides new insights into the possible mechanisms by which persistent (18 weeks) and severe hyperglycaemia induced by STZ treatment alters mitochondrial bioenergetics in rat gastrocnemius muscle.

Data from this study suggest that the respiratory control ratio (RCR) of mitochondria isolated from the gastrocnemius muscle of STZ-treated rats was affected when complex I substrates were used. However, the present data showed for the first time that gastrocnemius mitochondria isolated from STZ-treated animals had an enhanced tolerance to permeability transition pore opening in a calcium/phosphate medium. Somewhat unexpectedly, gastrocnemius mitochondria isolated from STZ rats were able to accumulate higher amounts of calcium and were demonstrated to be less susceptible to PTP opening than the control group, as mitochondrial swelling amplitude and rates as well as calcium release were lower in the STZ group.

Mitochondrial oxygen consumption and transmembrane potential

The data from Tables 2 and 3 suggest that long-term experimentally induced hyperglycaemia resulted in mitochondrial respiratory dysfunction as seen by a decreased respiratory control ratio, a measure of the dependence of the respiratory rate on ATP synthesis, and alterations on state 4 respiratory rate. The alterations were more visible when complex I-linked substrates were used. Regardless of the increase in Δψ observed on repolarization after ADP phosphorylation, no significant changes were observed in the remaining Δψ end-points between groups (Table 3). The apparent contradiction between respiratory and Δψ data can be explained by two distinct reasons. Although the two methods are different, we are inclined to think that the unchanged maximal membrane potential using complex I substrates and the equivalent increased state 4 respiration agrees with a possibility of decoupling at the complex I level (see below). As for the differences observed in ADP-induced depolarization for complex II substrates, the simpler explanation is that this end-point cannot be easily accessed from respiratory data values and thus is not comparable. Thus, our data show that long-term experimental hyperglycaemia causes altered coupling between respiration and phosphorylation in gastrocnemius muscle. The lack of coupling may be attributable to several causes, although the lack of differences in uncoupled/state 3 respiration and at the ADP/O ratio suggests that no damage to the respiratory chain or to the phosphorylation system occurred. The increase in state 4 respiration would suggest an increased inner membrane permeability, leading to higher proton influx (i.e. uncoupling) [30]. Nevertheless, the increase in state 4 respiration with malate + pyruvate, but not succinate, occur without alteration in Δψ (Table 3). A new explanation may involve an intrinsic slippage of protons (decoupling) through complex I as a result of modifications of the intrinsic H+/e stoichiometry. This would lead to a decreased number of protons pumped per electron transferred, which would increase the rate of electron transfer to compensate for the decreased proton pumping [31]. It must be stressed that as intrinsic decoupling at mitochondrial complexes with exception of complex IV is still a matter of debate [32].

As stated, several studies reported decreased muscle mitochondrial metabolism and morphological disturbances in STZ-treated rats, including decreased mitochondrial mass, increased number of disarrayed cristae and mitochondrial swelling, reduced matrix electron density and compromised activity of several Krebs Cycle enzymes [5–9]. Mitochondrial alterations associated with decreased function seem to be related with increased oxidative stress and damage within the muscle from hyperglycaemic subjects [9]. Bonnard et al. [9] showed that both type I and II diabetic mice skeletal muscle presented mitochondrial morphological disturbances accompanied by increased levels of protein carbonylation, cytochrome c release, decreased mtDNA/nDNA ratio that were reverted by antioxidants and insulin. It is thus likely that sustained hyperglycaemia-induced enhanced muscle free radical generation can contribute to alterations in mitochondrial respiration. However, it must be noted that enhanced muscle oxidative stress may also lead to up-regulated antioxidant defences, including mitochondrial chaperons and low molecular weight antioxidants, which may contribute to an apparent increased resistance to MPTP induction in the STZ group.

Modulation of the permeability transition pore

Mitochondrial dysfunction can also be triggered by stress-induced opening of the MPTP, which has been involved in calcium dynamics and cell death [12,14]. A hallmark of MPTP is the cyclosporin A-inhibited large-scale swelling induced by calcium in the presence of phosphate, which occurs through increase in oxidative stress [33]. The main consequences of MPTP opening are loss or collapse of mitochondrial potential, depending on the reversibility of the phenomenon, and an efflux of calcium into the extramitochondrial space. The irreversible opening of a high conductance MPTP in cells inevitably cause mitochondrial swelling, cytochrome c release, caspase activation and apoptosis, or the fall of the mitochondrial potential, ATP depletion and energetic collapse followed by necrotic cell death [34]. However, it is not at all clear whether the pore plays any other physiological role for mitochondria, or indeed whether it even exists except under extremely stressful conditions [35]. Some studies have also shown that an increased vulnerability to skeletal muscle MPTP opening develops under various physiological and pathological conditions including ageing [36], denervation-induced atrophy [37,38], bupivacaine administration [39] and muscle dystrophies [40]. Moreover, other studies in hyperglycaemic animals showed an increased susceptibility to MPTP induction in the heart [18], brain [41] and kidney [42]. However, the present work is the first to demonstrate the effect of severe and persistent hyperglycaemia on the susceptibility to MPTP opening in gastrocnemius skeletal muscle.

As mentioned, as MPTP opening can be induced by increasing amounts of calcium and by oxidative stress, and given that diabetes leads to increased free radical generation, caused by hyperglycaemia [43], our hypothesis was that hyperglycaemic animals would have an increased susceptibility of gastrocnemius mitochondria to undergo MPTP induction. However, our results showed that gastrocnemius muscle mitochondria isolated from rats after 18 weeks of severe hyperglycaemia had a higher calcium loading capacity before MPTP induction (Figs 1 and 2). The data obtained are not completely unexpected if compared with data obtained in other hyperglycaemic animal models, including the previous results obtained in heart and liver mitochondria isolated from hyperglycaemic animals, showing a decreased susceptibility to MPTP induction [44,45]. Ferreira et al. [45] tested the susceptibility of liver mitochondria to MPTP induction in both type I (STZ) and type II (lean Goto-Kakizaki) experimental diabetes models and suggested that there is an apparent positive correlation between the severity and the duration of the disease, and the susceptibility to MPTP induction, with also a clear organ specific difference present. In this regard, it is important to note that in our study, muscle mitochondria were tested 18 weeks after STZ treatment, and this long-term experimental diabetes model may contribute to some tissue and specifically mitochondrial metabolic adaptations to oxidative stress conditions caused by severe and persistent hyperglycaemia. These may include increases in mitochondria coenzyme Q and cardiolipin contents [45], up-regulation of antioxidant defences including vitamin E content, reduced (GSH) to oxidized (GSSG) glutathione ratio and GSH peroxidase and GSSG reductase activities [46]. Despite the recent reports showing that short-term STZ treatment does not induce improvements in some extramitochondrial defence systems such as muscle HSF-1, HSP72, HSP25 [17], it is likely that long-term severe hyperglycaemic condition would target other cellular protecting mechanisms, i.e. heme oxygenase 1 and catalase [47], possibly contributing to the observed additional tolerance to stress-induced MPTP. Another hypothesis which seems attractive involves a possible reduction in the levels of muscle hexokinase bound to mitochondria that may contribute to these results, as hexokinase expression is both known to be regulated by insulin [48,49] and to be involved in an inhibitory regulation of the MPTP complex [12]. Further studies are needed to determine how long-term hyperglycaemia results in decreased susceptibility to the MPTP in gastrocnemius skeletal muscle. Regardless of the mechanism involved, the decreased susceptibility of the MPTP in hyperglycaemic animals may be a countermeasure to prevent mitochondrial dysfunction, which limits damage during the alterations in ionic and redox status in muscle cells under chronic hyperglycaemic conditions.

Interestingly, no STZ effects were observed on succinate-sustained respiration despite the fact that differences exist between the two groups regarding calcium-induced MPTP in the presence of succinate. There are several possibilities for that apparent discrepancy, including STZ-induced alterations in proteins that do not participate directly in oxidative phosphorylation but that are part of the MPTP complex (e.g. cyclophilin D) or even the existence of a reserve of mitochondrial proteins that can be recruited to form the MPTP upon the presence the calcium but whose recruitment have no effect in the yield and fluxes of oxidative phosphorylation. Moreover, it must be stressed that the MPTP experiments are conducted in the mandatory presence of calcium, which is a MPTP agonist. When measuring oxidative phosphorylation, no calcium is present; which may mask alterations in oxidative phosphorylation that are secondary to MPTP opening.

We further determined the activity of caspase 9-like in the whole tissue homogenate as a marker of mitochondrial apoptotic signalling. The results of this study show that long-term severe hyperglycaemia decreased caspase 9-like activity suggesting that this condition may predispose gastrocnemius to a lower induction of the intrinsic, mitochondrial, pathway for apoptosis.

In summary, the data from this study indicate that in an animal model for prolonged severe hyperglycaemia, mimicking uncontrolled type I diabetes in humans, the mitochondrial respiratory control ratio is affected and a decreased susceptibility to MPTP induction exists in gastrocnemius skeletal muscle. Specific and differential adaptations within muscle mitochondria to the 18weeks of high blood glucose levels were likely translated into protection against MPTP induction and impairments in the respiratory system. A more accurate and organ-specific understanding of mitochondrial impairments would possibly contribute to better design preventive and/or therapeutic countermeasures against the disease.


The authors acknowledge the collaboration of Dr Marta Fraga and Dr Inês Aleixo for their technical assistance regarding animals’ care and treatment routines, and Dr Laura Pereira for glycated haemoglobin (HbA1c) assays. José Lumini-Oliveira, José Magalhães, António Ascensão and Cláudia V Pereira are supported by grants from the Portuguese Foundation for Science and Technology (FCT) (SFRH/BD/30906/2006, SFRH/BPD/66935/2009, SFRH/BPD/4225/2007 and SFRH/BD/48029/2008 respectively).


Research Centre in Physical Activity, Health and Leisure, Faculty of Sport Sciences, University of Porto, Porto, Portugal (J. Lumini-Oliveira, A. Ascensão, J. Magalhães); Department of Physical Therapy, School of Health, University of Fernando Pessoa, Porto, Portugal (J. Lumini-Oliveira); Centre of Neuroscience and Cell Biology, Department of Zoology, University of Coimbra, Coimbra, Portugal (C. V. Pereira, P. J. Oliveira); Department of Clinical Analysis, Faculty of Pharmacy, University of Porto, Porto, Portugal (S. Magalhães, F. Marques); Institute for Molecular and Cell Biology, University of Porto, Porto, Portugal (F. Marques).