Phosphocreatine plays a key role in energy provision to muscle (Jenkins, 1998). It has been administered for therapeutic benefits in patients with AIDS, muscle diseases, neuropathies and post-surgery (Mihic et al., 2000). Creatine supplementation has been used mainly to increase muscle performance during exercise (Balsom et al., 1994; Greenhaff, 1995; Volek and Kraemer, 1996; Volek et al., 1997b; Tarnopolski and Martin, 1999). Phosphocreatine content in type II human muscle fiber is 5 - 15% higher than type I (Greenhaff et al., 1994). Following creatine supplementation the total creatine and phosphocreatine content raises in both fiber types, although there is a tendency for more increase in type II fibers (Casey et al., 1996).

Creatine supplementation studies using different protocols and animal species, including humans, have been shown to increase body weight and change in the body composition (Viru et al., 1994; Balsom et al., 1995; Mujika et al., 1996; Kreider et al., 1998; Managaris and Maughan, 1998; Engelhardt et al., 1998; Volek et al., 1999). The underlying basis of this weight gain is still unclear. It may be due to stimulation of muscle protein synthesis or water retention in the initial days of creatine supplementation. Other authors, however, reported contradictory results (Thompson et al., 1996; Terrillion et al., 1997; Stout et al., 1999; Rico-Sanz and Marco, 2000). The explanation for these discrepancies still remains to be clarified. In this study, therefore, we investigated whether the sole creatine supplementation changes the cross-sectional area of skeletal muscle fibers or it is associated with the addition of training in Wistar rats.

All chemicals and enzymes used were obtained from Sigma Chemical Co., St. Louis, USA.

The National Animal Ethics Committee approved this study. Male Wistar rats (2-3 months old) were kept in standard individual vivarium cages, for food intake and body weight determination, under a controlled light/dark (12/12h) cycle and temperature (23° C ± 1) with free access to food and water. The rats were randomly divided into four groups: Sedentary with no supplementation (CON), Sedentary with creatine supplementation (CR), Exercised with no supplementation (EX) and Exercised with supplementation (CREX). The creatine-enriched diet consisted of normal rat chow supplemented with 3.3 mg creatine per g of diet (Advanced Nutrition Ltda, Brazil) according to Brannon et al. (1997).

EX and CREX rats exercised in swimming pool chambers at a water temparature of 32°C. The protocol used was similar to that described by Rombaldi (1996). Briefly, the animals swam daily for 1 hour during 5 weeks carriying an extra-weight of lead corresponding to 15% of their body weight. During weekdays, except Wednesday, the rats swam with the extra-weight. This protocol has been considered as a predominantly anaerobic supra-maximal workload (Kokobum, 1990). The rats swam for 15 seconds and rested 15 seconds for another for 15 seconds consecutively during the first 30 minutes. Following 10 minutes interval of resting, the same protocol was repeated for another 30-minutes period. At Wednesdays, the rats swam for 1 hour long with no break.

Following last exercise session, the animals were killed by cervical dislocation and the gastrocnemius (white portion), soleus and EDL (extensor digitorum longus) muscles were dissected. Each muscle was cut in the middle, transversally and one part was quickly frozen in tissue freezing medium (Jung-Germany). Serial slices of 10 µm were allocated in a cryostat at -20 °C and mounted on glass slides and was let to dry at room temperature. Then, histochemical procedure of myofibril ATPase staining was carried out to identify the fiber types I and II. This method enables differential staining of muscle fibers by utilizing the varying sensitivities of their mATPase to acid or alkaline pH (Brooke and Kaiser, 1970). Utilizing adjustment of pH in the pre-incubation to 4.6 produces a separation of different staining intensities: dark (type I), light (type IIA) and intermediate (type IIB) (Hämäläinen and Pette, 1993). The fiber type proportions from each muscle group were calculated by the formula suggested by Sullivan and Armstrong (1978), Armstrong and Phelps (1984). The slide images were obtained in a Zeiss (Germany) microscope connected to a computer; 5 to 20 fibers were randomly chosen for each section for measurement of cross sectional areas using the Image Tool software.

Statistical analysis was performed by one-way ANOVA followed by a post-hoc Tukey test. The level of p<0.05 was taken to indicate statistical significance.

All groups gained weight by the end of the experiment. At the 33rd day, in relation to day zero, CON group increased body weight by 21.6%, CR by 17.6%, EX by 21.4% and CREX by 17.6% (Figure 1), those values were not statistically different among the groups.

Food intake (Figure 2) was increased in the CON by 19.6%, in the CR by 25.3%, in the EX by 32.6% and in the CREX by 35.8%. The daily food intake was statistically different between EX and CREX as compared to CON and CR groups, respectively (P<0.05). However, there was no statistical difference between CON and CR (P>0.05) as well as between EX and CREX (P>0.05). Creatine intake was higher (41%) in CREX as compared to CR in the 33^th day (Figure 3). The differences stated for food and creatine intake started at the second week and was maintained until the end of the experiment.

In the soleus muscle the proportion of type I and IIA muscle fibers was, in average, 90.5% and 9.5%, respectively. In the EDL muscle, type I was 9.5%, type IIA 21% and type IIB 69.4%. In the white portion of the gastrocnemius muscle, the proportion of type IIA and IIB was 70.6% and 29.4%, respectively (Figure 4C). A simple observation of the dye staining in the soleus muscle indicates the presence of types I and IIA muscle fibers with predominance of type I (Figure 4A). In the EDL muscle (4B), it was mainly type IIB followed by IIA and a lower proportion of type I muscle fibers. In the white portion of the gastrocnemius muscle, it was observed that type IIA fibers were more predominant than type IIB fibers (4C).

The data for cross-sectional areas of these muscle fibers are presented in the Table 1. The magnitude of the areas of type I and IIA fibers in soleus muscle were not different between the CR and CON groups. However, in the EX group, the muscle fiber areas for type I and type IIA were increased by 10.8% and 16.9% as compared to CON and by 16.0% and 20.8% as compared to CR, respectively. The combination of exercise and creatine supplementation (CREX) increased the areas of type I and IIA fibers by 25.5% and 29.4% as compared to CON, by 31.5% and 33.8% as compared to CR, and by 13.3% and 11.4% as compared to EX, respectively.

In EDL muscle the magnitude of the cross-sectional areas of type IIA and IIB fibers were not different between CR and CON. In the EX group, these muscle fibers areas were increased by 27.7% and 19.3% as compared to CON and by 20.3% and 9.7% as compared to CR, respectively. The combination of exercise and creatine supplementation (CREX) increased the areas of type IIA and IIB muscle fibers by 41.9% and by 36.6% as compared to CON, by 33.7% and by 25.6% as compared to CR, and by 11.1% and 14.4% as compared to EX, respectively. The magnitude of the type I fibers in EDL muscle were not altered by either exercise or creatine supplementation (Table 1).

Gastrocnemius muscle type IIA and IIB fibers areas were not different between CR and CON. Exercise enlarged the magnitude of the type IIA and IIB fibers areas by 13.2% and 20.6% as compared to CON and by 9.80% and 19.70% as compared to CR, respectively. The combination of exercise and creatine supplementation (CREX) increased the areas by 30.4% and 33.5% as compared to CON, by 26.4% and 32.5% as compared to CR and by 15.1% and 10.7% as compared to EX, respectively (Table 1).

Evidence is presented herein that both creatine supplementation and exercise training enlarges the magnitude of the cross-sectional areas of skeletal muscle fibers. This effect, however, was not related to an increase in body weight. Although several groups reported that creatine induces body weight gain in humans (Harris, 1992; Volek et al., 1997a; Engelhardt et al., 1998; Peeters et al., 1999; Kelly and Jenkins, 1998; Mihic et al., 2000; Volek et al., 1999; Mujika et al., 2000; Volek et al., 2001), others did not show any significant change in body mass in humans (Redondo et al., 1996) or Sprague-Dawley rats (Brannon et al., 1997; McKenna et al., 1999; McMillen, 2001). These discrepancies may result from protocol differences, doses of creatine, type and duration of the exercise. The administration of creatine may cause an alteration in the proportion of the lean and fat mass only. The hind limb corresponds to 8% of the total body weight of which 71% of that amount composed of muscle tissue (Armstrong and Phelps, 1984). Hence, an increase of 30% in the cross-sectional area of these muscles is not enough to provoke an increment in the total body weight. The increases in food intake of the EX and CREX groups might compensate for the energy demand required during a physical activity (Figure 2). Since creatine was added to the diet, the increase in creatine intake by the CREX (Figure 3) was due to the higher food intake.

The proportion, size and identification of type I and IIA fibers in the soleus (Figure 4A) and EDL muscles (Figure 4B) were similar to those reported by others (Brooke and Kaiser, 1970; Sullivan and Armstrong, 1978; Armstrong and Phelps, 1984; Eddinger et al., 1985; Desypris and Parry (1990). In the white portion of the gastrocnemius muscle, type IIA (71%) and IIB (29%) fibers were most frequent than type IIA (Figure 4C). The proportion and size of these fibers, however, are inconsistent with those studies in which greater proportion of type IIB (87.2%) than IIA (12.8%) muscle fibers are reported (Sullivan and Armstrong 1978; Armstrong and Phelps 1984). In view of these results two hypotheses can be formulated. One, this could be typical for Wistar rats, considering that most studies were performed using Sprague-Dawley rats. Second, the portion of the gastrocnemius muscle used here may have a different fibers distribution, which could not be representative of the whole muscle. This hypothesis remains to be tested.

Creatine itself was not able to promote significant modification of the cross-sectional areas. This result was similar to that reported by Brannon et al. (1997). Exercise, on the other hand, raised fiber cross-sectional area significantly, as it is well known that regular and forced muscle contraction promotes hypertrophy whereas little or no activity leads to atrophy (Van Der Meulen et al., 1974; Booth and Gollnick, 1983). The applied exercise protocol assumed predominantly anaerobic due to the extra-weight of 15% of the body weight. Under this physical activity, type II muscle fibers are mostly recruited (Saltin and Gollnick, 1983). In fact, as a response to the applied training protocol the cross-sectional areas of type IIA and IIB muscle fibers enlarged. Type I fibers from soleus muscle was slightly increased whereas in the EDL muscle did not change. This could be due to the fact that these muscles are differently recruited. Soleus muscle, which is mostly involved in maintaining the posture, is more efficiently recruited during swimming than EDL muscle. In addition, the proportion of type I fibers in the soleus muscle is higher than in the EDL muscle, which is richer in type II and poorer in type I fibers. The applied exercise protocol is anaerobic, which may also help to explain these observations.

The combination of creatine and exercise had an additive effect on cross-sectional areas of all muscle fibers studied, except for type I fibers from EDL muscle. Our findings corroborate the work of Volek et al. (1999) who did a study using heavy resistance exercise. Also, Brannon et al. (1997) reported similar results in plantaris muscle of rats by determination of the dry weight. The physiological and biochemical mechanisms by which creatine and exercise combination induce additional increase in the cross-sectional area is not fully known (Bessman and Savabi, 1988) and it was not investigated in this study. Some studies have shown an increase in the heavy chain myosin and actin synthesis either in vitro and in vivo as well as in the total RNA (Ingwall et al., 1972; 1974; Ingwall and Wildenthal 1976). Bessman and Savabi (1988) suggested that changes in metabolism also should be considered to explain these observations. These authors argue that an increase in the phosphocreatine provides energy to protein synthesis resulting in muscle hypertrophy. Another hypothesis is focused on cell volume. Since creatine is an active osmotic compound (Volek et al., 1997b), increasing content of it inside the cell causes water influx and, leads to swelling of the cell. This enlargement in cell volume inhibits protein breakdown and stimulates glycogen synthesis (Häusinger et al., 1994; Hue, 1994; Lang, 1995). In addition, cell swelling also inhibits glycolysis and stimulates the flux of substrates through the pentose phosphate pathway (Häussinger et al., 1994). This metabolic pathway provides NADPH, which is important to protect the cell against oxidative stress and also substrates for lipogenesis. Furthermore, there is formation of ribose 5-phosphate, that is a component of purines and pyrimidines, which are required for cell proliferation. Finally, cell volume modifies the cytoskeleton, whereby cell swelling stabilizes the microtubule network, stimulates actin polymerization, and increases mRNA for β-actin and tubulin (Häussinger et al., 1994). All these possible hypothesis remains to be tested.

In our study, sole supplementation of creatin did not modify cross-sectional area of the hind limb muscle fibers. However, the combination of creatine supplementation with exercise training had an additive effect on increasing the cross-sectional areas of type I, IIA and IIB fibers of hind limb muscles in Wistar rats.