The cellular energy metabolism in human musculus gluteus medius (MGM) under normal conditions and hip osteoarthritis (OA) was explored. The functions of oxidative phosphorylation and energy transport systems were analyzed in permeabilized (skinned) muscle fibers by oxygraphy, in relation to myosin heavy chain (MHC) isoform distribution profile analyzed by SDS-PAGE, and to creatine kinase (CK) and adenylate kinase (AK) activities measured spectrophotometrically in the intact muscle. The results revealed high apparent Km for ADP in regulation of respiration that decreased after addition of creatine in MGM of traumatic patients (controls). OA was associated with increased sensitivity of mitochondrial respiration to ADP, decreased total activities of AK and CK with major reduction in mi-CK fraction, and attenuated effect of creatine on apparent Km for ADP compared with control group. It also included a complete loss of type II fibers in a subgroup of patients with the severest disease grade. It is concluded that energy metabolism in MGM cells is organized into functional complexes of mitochondria and ATPases. It is suggested that because of degenerative remodeling occurring during development of OA, these complexes become structurally and functionally impaired, which results in increased access of exogenous ADP to mitochondria and dysfunction of CK-phosphotransfer system.
- musculus gluteus medius
- myosin heavy chains
- mitochondrial respiration
- energy transfer
hip osteoarthritis (oa) characterized by pain and stiffness of the hip joint is one of the most common chronic diseases in the elderly (10). The disease is linked to neuromuscular deficit, because the patients exhibit reduced abduction, adduction, and flexion muscle strength (3). These changes are largely related to dysfunction of musculus gluteus medius (MGM) (9, 12, 17), for this muscle plays an important role in hip abduction, gait, and stabilization of the joint and pelvis (9, 16). On the other hand, functionally normal MGM is vital to prevent the implant from loosening and hasten the recovery after total hip arthroplasty in OA patients (6, 11, 21).
Notwithstanding the potentially significant role of dysfunctional MGM in the pathogenesis of OA, only a few studies have explored the underlying cellular mechanisms. The results show selective type II (fast-twitch) muscle fiber atrophy and increase in relative content of type I (slow-twitch) fibers (23, 31, 36). Because the fast-to-slow transition of muscle cells is generally associated with changes in muscle oxidative capacity and regulation of oxidative phosphorylation (22, 35, 41), one may expect altered mitochondrial function in diseased MGM, as well. However, this issue has not yet been addressed.
The objective of this study was to assess the parameters of oxidative phosphorylation and energy transfer systems mediated by creatine and adenylate kinases (CK and AK) in MGM in relation to the grade of OA. The respiratory parameters were studied in saponin-permeabilized (skinned) muscle fibers, which allows the avoidance of artifacts due to isolation of mitochondria, enables analysis of mitochondria in the small specimens of human muscle biopsy, and preserves the natural interactions of mitochondria with other cellular structures in situ (25, 29).
MATERIALS AND METHODS
Sixty sedentary subjects (31 males and 29 females, 65 ± 2.4 and 66 ± 2.2 years old, respectively) participated in this study. All subjects voluntarily gave informed, written consent, and the study was undertaken in accordance with the Declaration of Helsinki (39) and approved by the Tartu University Ethics Committee. The patients were divided into three groups. The control group (n = 15, 10/5 male/female ratio, age 68 ± 5.2) comprised the patients undergoing surgical correction of traumatic hip fracture. The two other groups included the patients with unilateral or bilateral hip replacement for OA of radiographic grade 3 (n = 11, 8/3, age 66 ± 4.1) and grade 4 (n = 34, 13/21, age 65 ± 1.2), estimated according to Kellgren and Lawrence (15). The muscle specimens (50–100 mg) were taken during surgery from the middle portion of the MGM. A part of each specimen was rapidly frozen in liquid nitrogen and stored at −70°C for enzyme and myosin heavy chain (MHC) analyses, whereas the rest was permeabilized (skinned) by saponin (5, 25, 29, 30) and used for oxygraphical studies. Because of the limited sample availability the number of patients participating in different experiments varied (see figures).
Analysis of the system of oxidative phosphorylation and its coupling to mi-CK.
The skinned fibers were incubated in solution B containing (in mM): 2.77 CaK2EGTA, 7.23 K2EGTA, 1.38 MgCl2, 0.5 DTT, 100 K-Mes, 20 imidazole, 20 taurine, 3 K2HPO4, 10 pyruvate, 2 malate, and 5 mg/ml BSA, with pH 7.1 at 25°C, in a chamber (volume 1.5–2.5 ml) of oxygraph (Rank Brothers, Cambridge, England or Oroboros, Paar KG, Graz, Austria) equipped with Clark electrode, assuming the solubility of oxygen in the medium to be 215 nmol O2/ml (18). After registration of the basal respiration rate (v0), 1 mM ADP was added to monitor the NADH-linked ADP-dependent respiration rate in the presence of pyruvate (VPyr), followed by subsequent additions of 10 μM rotenone to inhibit the complex I, 10 mM succinate to activate FADH2-linked, ADP-dependent respiration (VSucc), 0.1 mM atractyloside to assess the respiratory control by adenine nucleotide translocase (ANT), and 10 μM antimycin A to inhibit the electron flow from complex II to cytochrome c and to measure the proton leak as the antimycin-sensitive respiration in the presence of atractyloside. The VPyr/v0 ratio was used as the respiratory control index (RCIPyr) for NADH-linked respiration, whereas the VSucc/VAtr ratio, where VAtr represents the respiration rate with atractyloside, was taken as the respiratory control index for FADH2-linked respiration (RCISucc). After measurements, the fibers were removed from the chamber and dried overnight at 105°C. The rates of oxygen consumption were normalized per milligram of dry muscle weight. Coupling between oxidative phosphorylation and mi-CK was estimated from the V̇o2 vs. [ADP] relationships in solution B supplemented with 10 mM pyruvate and 2 mM malate in the presence or absence of 20 mM creatine; the corresponding apparent Km values were calculated by fitting to Michaelis-Menten equation and the interaction between mi-CK and ANT was expressed as [KmADP(–Cr)/KmADP(+Cr)] (creatine index).
Determination of the activities of AK and CK and MHC isoform distribution.
The activities of CK and AK were determined spectrophotometrically by using the coupled enzyme systems (28). The MHC and CK isoenzyme profiles were assayed by 7.2% SDS-PAGE (27, 32, 33) and 1% agarose gel electrophoresis (37), respectively.
All reagents were purchased from Sigma (St. Louis, MO).
One-way ANOVA with Newman-Keuls post hoc test and paired one-tailed Student's t-test were performed when appropriate to test the differences between groups. The means ± SE are presented.
Figure 1 demonstrates that the mean values of the relative contents of MHC were not statistically different between the MGM groups studied. However, a grade 4 group (n = 29) contained a subgroup of nine patients characterized by the lack of either one (MHCIIX, see third panel, inset) or both fast isoforms (MHCIIX and MHCIIA, see fourth panel, inset), the latter change observed in seven patients out of nine. Such a fiber-type dropout was not observed in the other two patient groups. Thus progression of the disease from grade 3 to 4 was associated with a complete loss of type II fibers in a subgroup of patients with the severest grade of disease, this change being in line with predominantly type II fiber atrophy observed earlier (23, 31, 36).
To assess whether OA is associated with altered mitochondrial functions, the parameters of oxidative phosphorylation and regulation of respiration by ADP and creatine in skinned MGM fibers were studied. Figure 2 shows that in control MGM fibers ADP (1 mM) produced an increase in the respiration rate (VPyr) by 7.8 times (RCIPyr) compared with the levels of v0 in the presence of pyruvate and malate, which points to highly coupled state between oxidation and phosphorylation in situ. The respiratory parameters with succinate (VSucc and RCISucc) were not statistically different (ANOVA) from their counterparts with NADH-related substrates. Atractyloside, an inhibitor of ANT, effectively suppressed the succinate oxidation, and the atractyloside-insensitive respiration (VAtr) was mostly caused by a proton leak; these data indicate preserved intactness of the inner mitochondrial membrane in skinned muscle fibers. A comparison of the respiratory data revealed no significant difference between the controls (grade 0) and diseased patients (grades 3 and 4). Thus OA did not affect the systems of oxidative phosphorylation in MGM cells.
Figure 3A shows that in control patients, the kinetics of activation of mitochondrial respiration by exogenous ADP followed the Michaelis-Menten equation, with the apparent Km for ADP of 331 μM in the absence of creatine. A mean value of that parameter for the control group (282 ± 56 μM, Fig. 3B) corresponds well to that in oxidative muscles such as heart and m. soleus but largely exceeds that in glycolytic muscles and isolated mitochondria (10–20 μM; Refs. 18 and 38). To test whether the CK-phosphotransfer system functions in the MGM cells, the effect of creatine on V̇o2 vs. [ADP] relationship was determined. Creatine significantly shifted this relationship toward lower [ADP] (Fig. 3A), thus producing a fourfold decrease in the mean apparent Km for ADP (Fig. 3B) but exerting no effect on Vmax (Fig. 3A) or v0 of respiration (not shown). This effect of creatine on apparent Km strongly indicates the coupling of mi-CK to ANT, a key process in the CK-phosphotransfer system (26).
OA (grade 4) was associated with decreased total CK activity due to reduced activities of all CK isoforms (Table 1). The mi-CK activity decreased in earlier phase of the OA (stage 3) and to a greater extent than did other isoforms. Analysis of regulation of respiration with ADP and creatine revealed a different kinetics compared with normal muscles (Fig. 3B): the apparent Km for ADP was decreased in the absence of creatine, and creatine exerted less effect on that parameter, thus giving rise to a decreased creatine index. These changes suggest increased access of exogenous ADP to mitochondria and impaired coupling of mi-CK to ANT during the progression of OA (see discussion). Table 1 shows that OA was also associated with decreased total AK activity in MGM.
To our knowledge, this study is the first to characterize the cellular energy metabolism in human MGM and its alterations associated with OA. The important finding was that mitochondria in skinned fibers of MGM of control patients exhibited an apparent Km for ADP (282 ± 56 μM), which is much higher value than registered for isolated mitochondria (10–20 μM) in regulation of respiration. The principal question here is whether the phenomenon of low apparent affinity of mitochondria to exogenous ADP in situ represents an artifact due to experimental conditions and limited ADP diffusion owing to long diffusion distances within the muscle fiber, or it reflects intrinsic mechanisms of intracellular regulation of mitochondrial function. In these experiments, 5 mg/ml BSA was added to the incubation medium to protect mitochondria from uncoupling caused by free fatty acids (FFA) (19, 24) produced by the phospholipases (4). Because BSA binds not only FFA, but also adenine nucleotides (Kd = 120 μM at pH 7.4; Ref. 34), partial binding of ADP may shift the V̇o2 vs. [ADP] to the right, which gives rise to high apparent Km for ADP. Several observations contradict this interpretation, however. 1) Considering that ATP binds BSA stronger than ADP (34), the dependency of V̇o2 vs. [ATP] should have resulted in higher values of Km than that found from V̇o2 vs. [ADP], but we have registered similar Km values for ATP and ADP in skinned heart fibers (2). 2) In our experiments the incubation medium contained 3 mM inorganic phosphate, a competitive inhibitor of adenine nucleotide binding (Ki =1 mM; Ref. 34), thus hindering ADP from binding to BSA. 3) We have found that in the presence of similar concentrations of BSA, isolated mitochondria always exhibit much lower Km for ADP than mitochondria in situ, in skinned slow-twitch muscle fibers (18). 4) Liobikas et al. (20) have demonstrated that the addition of 10% BSA (i.e., 20 times more than in our experiments) exerted no effect on Km for ADP in saponin-skinned heart fibers compared with control conditions with 0.2% BSA. Altogether, these data make it highly unlikely that high apparent Km for ADP in regulation of respiration in skinned MGM results from partial binding of ADP to BSA.
A number of arguments also exclude the possibility that high Km for ADP in skinned MGM is related to long diffusion distances for ADP (Rdif) from medium into cell core. In correctly performed experiments, combined mechanical and chemical (saponin) treatment results in separation of skinned muscle fibers (cells) from each other, thus making the mean Rdif comparable to half of the fiber's diameter (29). In permeabilized cardiac cells, type I MGM fibers, and type I m. soleus fibers, the maximum Rdif is 10 μm (29), 24 μm (31), and 50 μm (40), respectively. In spite of fivefold differences in Rdif, these muscle specimens exhibit similarly high Km values for ADP in regulation of mitochondrial respiration (200–400 μm; Ref. 29 and Fig. 3). Furthermore, skinned fast twitch glycolytic muscle fibers having longer Rdif (40–50 μm) than MGM fibers display a very high affinity to ADP (Km = 8–22 μM) comparable to that in isolated mitochondria (29). From these data, it is clear that high Km for ADP in regulation of mitochondrial respiration in situ in skinned MGM fibers results not from specific composition of polarographic medium or morphological properties of the muscle but rather from specific organization of the intracellular energy metabolism, which may be similar to other oxidative muscles (18, 30, 38). It has been suggested that in oxidative muscle cell mitochondria, ATPases and part of the cellular adenine nucleotides are compartmentalized into functional complexes (also called intracellular energetic units, ICEUs), most probably by cytoskeletal proteins (25, 30). By creating localized restrictions on diffusion of ADP, these proteins limit the access of exogenous ADP to mitochondria, which underlies low apparent affinity of mitochondria to this nucleotide (25, 30) (Fig. 3).
Within the ICEUs, the mitochondria communicate with ATPases via CK- and AK-phosphotransfer systems and/or direct transfer of adenine nucleotides. In normal heart cells the CK-phosphotransfer system plays a predominant role in energy transduction (8, 26). The functional coupling between ANT and mi-CK converting ATP that is generated by the mitochondria to PCr and provides local/endogenous ADP to stimulate respiration via ANT represents a key step in this system. Owing to this mechanism, the mitochondrial respiration becomes less dependent on the limited ADP flux from the cytoplasm, this effect is seen as a decrease in Km for exogenous ADP after addition of creatine (Fig. 3). Our study revealed that MGM of control patients possesses a significant amount of mi-CK activity (4.3 ± 1.8% of total CK activity, Table 1), which is close to that in m. soleus (6.8 ± 2.8%; Ref. 1). Likewise, creatine caused a fourfold decrease in apparent Km for ADP (Fig. 3), thus showing that mi-CK is functionally coupled to mitochondrial oxidative phosphorylation (18, 26). Thus, by its type of respiratory regulation and key enzyme activities, the MGM represents a novel muscle belonging to the class of oxidative muscles. Most likely, it is the specific unitary organization of energy metabolism ensuring precise regulation of mitochondrial ATP synthesis in response to its use (25, 26) that enables chronically strong twitches of MGM for stabilization of the hip joint and pelvis during gait.
We found that Km for ADP decreased in MGM more than twice and that creatine exerted a diminished effect on that parameter after development of OA (Fig. 3). Given the independence of Km for ADP of muscle cell geometry (see above), these changes cannot be attributed to the modest decrease in MGM fiber diameter (4–32%, depending on patient's age and fiber type) accompanying the muscle atrophy in patients with OA (31). It seems more likely that OA is associated with loosening of the cytoskeletal restrictions for ADP diffusion on the borders of or inside the ICEU, because this compound added exogenously reaches mitochondria more easily than in normal muscle, thus resulting in decreased apparent Km for ADP in the absence of creatine (Fig. 3B). An excess ADP flux from the medium/cytoplasm to the intermembrane space of mitochondria may inhibit PCr synthesis in mi-CK reaction, whereas diminished mi-CK activity may result in the same effect via impaired coupling to ANT. Clearly, both mechanisms should reduce the effectiveness of CK energy transfer between mitochondria and ATPases in MGM. Because similar defects have been revealed under different diseases [e.g., ischemia/reperfusion impairment and heart failure (7, 14)] and in oxidative muscles genetically devoid of dystrophin or desmin (5, 13), they probably represent a universal type of alteration resulting from disintegration of the ICEUs. It has been suggested that in case of a failing CK system, activation of AK-phosphotransfer can play a compensatory role (8). In diseased MGM, however, this mechanism may also become limited, as suggested by decreased AK activity (Table 1).
In conclusion, the present study demonstrates that energy metabolism in MGM cells is organized similarly to that in oxidative muscles, probably in the form of complexes of mitochondria and ATPases. Pathogenesis of OA involves disintegration of these complexes, which results in dysfunction of CK-phosphotransfer system and increased access of exogenous ADP to mitochondria. In these conditions, the mitochondrial ATP synthesis becomes dependent on fluctuations of the cytoplasmic [ADP], which reduces both the mitochondrial PCr synthesis and the effectiveness of ATP usage at the ATPase sites (25) and therefore may contribute to decreased hip muscle strength (3) in OA patients. In clinical terms, this study infers that arthroplasty undertaken before development of the grade 3 OA may improve the postsurgical rehabilitation by anticipating the deterioration of the intracellular energy transfer processes in MGM cells.
This work was supported by Estonian Science Foundation's Grants 4611, 4928, 4930 5515, 6501 and 6142, and by the Grants 0182549As03 and 018178S01 from the Estonian Ministry of Education and Research.
The authors thank E. Gvozdkova, M. Kruus, Dr. L. Rips, Dr. A. Toom, and Dr. I. Kaur for technical assistance.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 the American Physiological Society