The effects of dietary creatine supplements on the contractile properties of rat soleus and extensor digitorum longus muscles

Transcription

1 The effects of dietary creatine supplements on the contractile properties of rat soleus and extensor digitorum longus muscles M. McGuire, A. Bradford and M. MacDermott* Department of Physiology, Royal College of Surgeons in Ireland, St Stephen s Green, Dublin 2, Ireland (Manuscript received 14 September 2000; accepted 16 January 2001) Daily creatine supplements (0.258 g kg _1 ) were administered to adult male Wistar rats (n = 7) in the drinking water. Age matched rats (n = 6) acted as controls. After 5 6 days, contractile properties were examined in soleus and extensor digitorum longus (EDL) muscle strips in vitro at 30 C. In soleus muscles, creatine supplements decreased the halfrelaxation time of the isometric twitch from 53.6 ± 4.3 ms in control muscles to 48.4 ± 5.5 ms but had no effect on twitch or tetanic tension or on twitch contraction time. In EDL muscles twitch tension, tetanic tension, twitch contraction and half-relaxation times were all unaffected by creatine supplements. Creatine supplements increased the fatigue resistance of the soleus muscles but had no effect on that of the EDL muscles. After a 5 min low-frequency fatigue test, tension (expressed as a percentage of initial tension) was 56 ± 3% in control soleus muscles, whereas that in the creatinesupplemented muscles was 78 ± 6% (P<0.01). In the EDL muscles, the corresponding values were 40 ± 2% and 41 ± 9%, respectively. The force potentiation which occurred in the EDL muscles during the initial s of the fatigue test was 170 ± 10% of initial tension in the control muscles 24 s after the initial stimulus train but was reduced (P<0.01) to 130 ± 20% in the creatine-supplemented muscles. In conclusion, soleus muscle endurance was increased by creatine supplements. EDL endurance was unaffected but force potentiation during repetitive stimulation was decreased. Experimental Physiology (2001) 86.2, Creatine occurs naturally in the body, with 95% found in skeletal muscle where it exists as free creatine (Cr) and the high energy compound phosphocreatine (PCr). PCr can be hydrolysed with the donation of the phosphate group to ADP to form ATP. Consequently, PCr is important as a rapidly available energy source for skeletal muscle contraction. In recent years, interest has focused on the role of short-term, high-dose dietary Cr supplements in exercise performance. Controlled laboratory studies have shown that daily Cr supplements of 20 g (approximately 0.25 g (kg body weight) _1 ) given to healthy adults for 5 days resulted in increased Cr and PCr in skeletal muscle (Balsom et al. 1995; Hultman et al. 1996) and in an improvement in the capacity to perform intermittent high-intensity exercise tasks (e.g. Greenhaff et al. 1993; Balsom et al. 1995). However, other studies have reported no effect of Cr on intermittent high-intensity exercise (Cooke et al. 1995; McKenna et al. 1999). Rats respond to Cr supplements but in common with the human studies, those in rats have also produced inconclusive results. When Cr supplements were given to rats such that the daily dose was similar to that used in human studies, Cr and PCr increased in the skeletal muscles and running performance improved (Brannon et al. 1997). The Cr supplements were also shown to influence the metabolic capacity of rat skeletal muscles. This effect was fibre type dependent. Supplements had no effect on the activity of the oxidative enzyme citrate synthase in the fast twitch plantaris muscles whereas, in the predominantly slow twitch soleus muscles, the activity of citrate synthase increased following creatine supplements (Brannon et al. 1997), indicating an increase in the oxidative capacity of the muscle. This suggests that the fatigue resistance of soleus muscles may be increased by Cr supplements. However, a study in which fatigue resistance was actually evaluated reported that Cr supplements decreased the fatigue resistance of rat soleus muscles (Wataksuki et al. 1994) and another study reported that Cr supplements were without effect on exercise performance (Tanaka et al. 1997). The daily Cr consumption of the rats in both of these studies was almost three times higher than that in the study of Brannon et al. (1997). This may have contributed to the conflicting results. However, the failure of the supplements to increase the fatigue resistance of the soleus muscles despite the report that oxidative capacity was increased in these muscles (Brannon et al. 1997) was probably mainly due to the indirect muscle stimulation employed to induce fatigue (Wataksuki et al. 1994). This would have resulted in fatigue originating not only in the muscle itself but also in the motoneuron (Kuei et al. 1990) Publication of The Physiological Society *Corresponding author: mmacderm@rcsi.ie

2 186 M. McGuire, A. Bradford and M. MacDermott Exp. Physiol Table 1. The effect of creatine supplementation on the contractile properties of soleus and extensor digitorum longus (EDL) muscles Tetanic tension Contraction Half-relaxation n Twitch tension (100 Hz) time time (N cm _2 ) (N cm _2 ) (ms) (ms) Soleus Control ± ± ± ± 4.3 Creatine supplemented ± ± ± ± 5.5* EDL Control ± ± ± ± 2.9 Creatine supplemented ± ± ± ± 2.7 Rats (n =7) consumed ± g kg _1 day _1 of added creatine monohydrate (0.258 g creatine kg _1 day _1 ). All values are expressed as means ± S.D. n=number of muscles. *Control vs. creatine supplemented: P <0.05. The effect of a Cr supplementation regimen, which has been shown to improve exercise performance, has not been determined on the function of isolated skeletal muscles. The objective of the present study was to examine the effects of dietary Cr supplements on the contractile properties and the fatigue resistance of the slow twitch soleus muscles and the fast twitch EDL muscle of rats. METHODS Dietary treatment Male Wistar rats ( g) were randomly divided into two groups. The rats were housed in a temperature (20 22 C) and light (12 h light 12 h dark cycle) controlled animal facility. Food and water were freely available. One group of rats (n =7) was given creatine supplements. These were administered via the drinking water as a solution of g of creatine monohydrate per 100 ml of water. The other group of rats (n =6) acted as controls. After 5 6 days, during which time the weight gain and food and water intakes were monitored, rats were anaesthetised (with pentobarbitone sodium 60 mg kg _1 I.P.), intubated and artificially ventilated. A femoral artery was cannulated in order to record arterial blood pressure. Body temperature was maintained at 37 C using a heating blanket and radiant heat. The soleus muscle was exposed in one hindlimb by sectioning the tendons connecting the plantaris and gastrocnemius muscles to the heel and reflecting the muscles back. The EDL muscle was exposed in the opposite hindlimb by reflection of the anterior tibialis muscle. Contractile studies The muscles were rapidly removed from the rat and the rat was then killed by an overdose of anaesthetic. Longitudinal strips of muscle (mean weight of 29 mg for the soleus muscles and 33 mg for the EDL muscles) were suspended vertically between a pair of platinum plate electrodes (5 mm w 30 mm) in a thermostatically controlled, water-jacketed muscle bath containing Krebs solution. The Krebs solution had the following composition (mmol l _1 ): 120 NaCl, 25 NaHCO 3, 1.2 NaH 2 PO 4. 2H 2 O, 1.2 MgSO 4. 7H 2 O, 5.0 KCl, 2.5 calcium gluconate, 11.5 glucose. The solution was continuously oxygenated (95 % O 2 5 %CO 2 ) and maintained at 30 C. One end of the muscle strip was anchored to the base of the bath and the other end was attached by thread to an isometric force transducer (Piodon Controls Ltd, (Canterbury, Kent, UK), unloaded resonant frequency of 600 Hz) and mounted on a micropositioner. The platinum electrodes were connected to a Square Pulse Stimulator (Grass, model S48, Astro-Med, Inc.). Isometric twitch tension, tetanic tension, contraction time, half-relaxation time, the tension frequency relationship and fatigue were measured using field stimulation (supramaximal voltage, 1 ms duration) and recorded using an analog digital converter and computer. Optimal length, i.e. the length producing maximal twitch tension, was determined by varying the muscle length in increments of 1 mm using the micropositioner. All subsequent measurements were made at optimal length. Following an equilibration period of 30 min, a single twitch was elicited. The tension frequency response was then determined using stimulation trains of 300 ms at frequencies of Hz in increments of 10 Hz. Stimulus trains were separated by a 3 min interval. Ten minutes after the tension frequency determination, fatigue resistance was evaluated using a low-frequency fatigue protocol of 30 Hz trains of 300 ms delivered every 2 s for 5 min. Data analysis Specific tension was calculated in Newtons cm _2 of strip crosssectional area. To calculate the cross-sectional area, the muscle strip was removed from the bath, blotted dry and weighed. The weight was then divided by the product of the optimal length and muscle density, assumed to be 1.06 mg mm _3. For fatigue in the soleus muscle strips, values were normalised during the first 30 s by expressing the tension generated by the stimulus trains at 4, 8, 12, 16, 20, 24 and 28 s as a percentage of that generated by the first stimulus train. Values were normalised during the remaining 4.5 min of the fatigue protocol by expressing the tension generated by the stimulus trains at 1, 2, 3, 4 and 5 min as a percentage of that generated by the first stimulus train. For fatigue in the EDL muscle strips, values were normalised during the first 60 s by expressing the tension generated by the stimulus trains at 8, 16, 24, 32, 40, 48 and 56 s as a percentage of that generated by the first stimulus train. Values were normalised during the remaining 4 min of the fatigue protocol by expressing the tension generated by the stimulus trains at 1, 2, 3, 4 and 5 min as a percentage of that generated by the first stimulus train. Values for the specific twitch and tetanic tensions, contraction and half-relaxation times and fatigue were expressed as mean ± S.D.and used to compare statistically the control and Cr-supplemented groups using ANOVA and Student s t test. P 0.05 was taken as significant.

3 Exp. Physiol Creatine supplementation and muscle contraction 187 RESULTS Dietary treatment The weight gain and food and water intake in control rats during the 5 6 day supplementation period were 5 ± 1 %, 80±4gkg _1 day _1 and 119±9mlkg _1 day _1, respectively, and were not significantly different from the corresponding values for the Cr-supplemented rats of 4 ± 3 %, 87±8mgkg _1 day _1 and 115±7mlkg _1 day _1, respectively. Based on the water intake of the Cr-supplemented rats, ± g (kg body weight) _1 of added Cr monohydrate (equivalent to g Cr kg _1 ) was consumed by each rat per day. Contractile studies Supplementation with Cr had no effect on twitch and tetanic tension in the soleus muscle strips. It caused a decrease in the half-relaxation time of the isometric twitch but had no effect on twitch contraction time. In the EDL muscle strips twitch tension, tetanic tension, twitch contraction and halfrelaxation times were all unaffected by Cr supplementation (Table 1). The relationship between tension and stimulation frequency in both the soleus and EDL muscle strips was also unaffected by Cr supplements. Supplementation with Cr significantly increased the fatigue resistance of the soleus muscles as shown in Fig.1.After the 5 min fatigue protocol, tension in the control soleus muscle strips had decreased to 56 ± 3% of the initial tension whereas the corresponding value in the Cr-supplemented muscles strips was 78 ± 6% (Fig.1A). The effect of Cr on fatigue was evident only during the final 3 min of stimulation. After the first 2 min of stimulation, tension had decreased to 76 ± 2% and to 80 ± 6% of initial tension in control and Cr-supplemented muscle strips, respectively, but during the remaining 3 min, tension continued to decrease in the control muscle strips whereas no further decrease occurred in the Cr-supplemented muscle strips. Typical recordings of the decline in tension in control and Cr-supplemented soleus muscle strips resulting from the 5 min fatigue protocol are shown in Fig.1B. Supplementation with Cr had no effect on the fatigue resistance of the EDL muscles as shown in Fig.2.In the control muscle strips tension had decreased to 40 ± 2% of initial tension after the 5 min fatigue protocol. In the Crsupplemented muscle strips, the corresponding value was 41 ± 9% (Fig.2A). Figure 2B shows typical recordings of the decline in tension in control and Cr-supplemented EDL muscle strips resulting from the fatigue protocol. In both control and Cr-supplemented muscle strips, a progressive increase in force was evident following the initial stimulus train of the fatigue protocol. This continued for s after which tension started to decline. Tension had returned to the initial tension level after 90 s of stimulation. The magnitude of the potentiation, normalised to initial train tension, was 170 ± 10 % in the control muscle strips after 24 s of Figure 1 Fatigue in control and Cr-supplemented soleus muscles strips. Fatigue was induced using a protocol of 30 Hz trains for 300 ms every 2 s for 5 min. A, values for tension during the fatigue protocol expressed as a percentage of initial tension in control ( ) and Cr-supplemented (ª) soleus muscle strips. Values are mean ± S.D. n =6 for control muscle strips and n =7 for Cr-supplemented muscle strips. * Significant difference from control (P <0.001). B, recordings show the decrease in tension resulting from the fatigue protocol in a control muscle strip in which tension at 5 min had decreased to 58 % of initial tension (trace 1) and in a Cr-supplemented muscle strip in which tension at 5 min had decreased to 76 % of initial tension (trace 2).

4 188 M. McGuire, A. Bradford and M. MacDermott Exp. Physiol stimulation. This was significantly (P<0.01) greater than that in the Cr-supplemented muscle strips of 130 ± 20% at the same time. DISCUSSION In this study, the effects of 5 6 days of Cr supplements on the contractile properties of soleus and EDL muscles of adult rats were examined. The supplements had no effect on the food and water intake of the rats or on their body weight gain. The absence of an effect of creatine supplements on weigh gain in rats has already been reported (Brannon et al. 1997). During the supplementation period, ± g (kg body weight) _1 of added creatine monohydrate (equivalent to g creatine (kg body weight) _1 ) was consumed by each rat per day. This is similar to the daily intake in rats in the study of Brannon et al. (1997). In that study, significant increases in total Cr and in PCr had occurred in the soleus and plantaris muscles after only 4 days of supplementation. It is fair to assume therefore that significant increases in Cr and PCr would also have occurred in the rat muscles used in the present study after the 5 6 days of Cr supplementation. Muscle contractile properties were examined in vitro at 30 C in small strips of soleus and EDL muscles. The use of small muscle strips avoided the possibility of hypoxia developing (Goldberg et al. 1975). The tension and twitch contraction and half-relaxation times measured in the strips from control soleus and EDL muscles were similar to those measured by Segal & Faulkner (1985) in intact soleus and EDL muscles, indicating that the contractile status of the muscle strips was similar to that of intact muscles. After 5 6 days of creatine supplements, the half-relaxation time of the soleus muscles had decreased whereas no change had occurred in the half-relaxation time of the EDL muscles. Creatine supplements had no effect on developed tension or contraction time in either muscle. The decreased halfrelaxation time agrees with results in humans where a Cr supplementation regimen, similar to that of the present study, produced a decrease in the half-relaxation time of the biceps muscles (Van Leemputte et al. 1999). A decrease in the halfrelaxation time of rat soleus muscles following Cr supplements has been reported previously (Wakatsuki et al. 1994) although the daily Cr intake was two to three times higher than the intake in the present study. The most likely explanation for the decreased relaxation time in the soleus muscles following Cr supplements is an alteration in the activity of the sarcoplasmic reticulum (SR) Ca 2+ pump. This pump derives its ATP mainly from PCr by the action of the SR bound creatine kinase. Consequently, an increase in muscle PCr could lead to an increase in SR Ca 2+ pump activity. In mdx mouse myotubes, increased PCr resulting from exposure of the myotubes to Cr was associated with increased Ca 2+ accumulation by the SR (Pulido et al. 1998) while studies in rat cardiac myocytes showed that the absence of PCr was associated with decreased Ca 2+ accumulation by the SR (Yang & Steele, 1999). The observation that Cr supplements decreased relaxation time in Figure 2 Fatigue in control and Cr-supplemented EDL muscles strips. Fatigue was induced using a protocol of 30 Hz trains for 300 ms every 2 s for 5 min. A, values for tension during the fatigue protocol expressed as a percentage of initial tension in control ( ) and Cr-supplemented (ª) EDL muscles strips. Values are mean ± S.D. n =6 for control muscle strips and n =7 for Cr-supplemented muscle strips. * Significant difference from control (P <0.01). B, recordings show the decrease in tension resulting from the fatigue protocol in a control muscle strip in which tension at 5 min had decreased to 41 % of initial tension (trace 1) and in a Cr-supplemented muscle strip in which tension at 5 min had decreased to 46 % of initial tension (trace 2).

5 Exp. Physiol Creatine supplementation and muscle contraction 189 the soleus muscles and not in the EDL muscles might be explained if the rate-determining process in the relaxation of the soleus muscle differed from that of the EDL muscle. It was suggested (Celio & Heizmann, 1982) that parvalbumin binding of Ca 2+ might determine the relaxation rate in the predominantly fast twitch EDL muscle but not in the slow twitch soleus muscle. This suggestion was based on the observation that the Ca 2+ binding protein parvalbumin is present in fast twitch muscle fibres but not in slow twitch fibres (Celio & Heizmann, 1982). However, over the temperature range C, the temperature sensitivity of the relaxation rate was similar in rat soleus and EDL muscles indicating that the rate-determining process underlying relaxation is similar in both muscles (Ranatunga & Wylie, 1983). The differential effect of Cr supplements on relaxation of the soleus and EDL muscles could be related to the properties of the SR Ca 2+ pump in the two muscle types. Slow and fast twitch muscle fibres express different isoforms of the SR Ca 2+ pump and these display different sensitivities to the regulatory systems which influence their activity (Hawkins et al. 1994). Consequently, although Cr supplements resulted in similar relative increases in PCr in slow and fast twitch muscles (Brannon et al. 1997), the response of the SR Ca 2+ pump to the increase may vary in the two muscle types. The differential effect of Cr supplements on relaxation of the soleus and EDL muscles was also evident in the effect on fatigue resistance. The fatigue resistance of the soleus muscles was markedly increased after Cr supplements whereas fatigue resistance of the EDL muscles was unchanged after Cr supplements. The stimulation protocol used to induce fatigue, consisting of repetitive submaximal tetanic stimulation, results in low-frequency fatigue and accumulation of inorganic phosphate and H + (Reid et al. 1992). Inorganic phosphate and H + decrease the sensitivity of the myofibrils to Ca 2+ and this contributes to the decline in tension associated with the fatigue (Fryer et al. 1995). The decline in tension resulting from this type of fatigue protocol is also related to the oxidative capacity of the muscles. Highly oxidative muscles are more fatigue-resistant than those of lower oxidative capacity (Burke et al. 1971). In the soleus muscles of the Cr-supplemented rats, the increase in fatigue resistance is consistent with the increase in the oxidative capacity reported to occur in soleus muscles after creatine supplements (Brannon et al. 1997). It is of interest that oxidative capacity increased only in the slow twitch soleus muscles and not in the fast twitch plantaris muscles after Cr supplements (Brannon et al. 1997), suggesting a fibre type-dependent effect. This probably explains why the fatigue resistance of the soleus muscles was increased whereas that of fast twitch EDL muscles was unchanged by Cr supplements in the present study. Creatine supplements did, however, decrease the force potentiation which developed in the EDL muscles at the beginning of the fatigue test. The mechanism responsible for force potentiation has not been established. Consequently, it is difficult to speculate on the reason for the reduced force potentiation in the Cr-supplemented EDL muscles. In contrast to the present findings of an increase in the fatigue resistance of Cr-supplemented rat soleus muscles, a decrease in the fatigue resistance of these muscles has also been reported (Wakatsuki et al. 1994). The daily Cr intake of the rats in the latter study was two to three times greater than that in the present study. This could have contributed to the conflicting results. However, the type of fatigue induced was different in the two studies. This is probably the main reason for the conflicting results. Fatigue was induced by direct muscle stimulation in the present study so that muscle fatigue was the sole contributor to the resultant decline in tension. In the study of Wakatsuki et al. (1994), fatigue was induced by stimulation of the motor nerve. Consequently, neuromuscular fatigue as well as fatigue of the muscle would have contributed to the resultant decline in tension in that study (Kuei et al. 1990). In conclusion, the study has shown that daily Cr supplements given to rats have differential effects on the slow twitch soleus muscle and the fast twitch EDL muscle. In the soleus muscle, relaxation time was decreased and fatigue resistance was increased. In the EDL muscle, these effects on relaxation time and fatigue resistance were absent but force potentiation at the beginning of the fatigue run was reduced. The results are consistent with the report (Brannon et el. 1997) that Cr supplements improved the ability of rats to perform a running test designed to exceed their maximal aerobic capacity but did not improve their ability to perform intermittent high intensity exercise tests lasting 30 s. BALSOM, P. D., SÖDERLUND, K., SJÖDIN, B. & EKBLOM, B. (1995). Skeletal muscle metabolism during short duration high-intensity exercise: influence of creatine supplementation. Acta Physiologica Scandinavica 154, BRANNON, T. A., ADAMS, G. R., CONNIFF, C. L. & BALDWIN, K. M. (1997). Effects of creatine loading and training on running performance and biochemical properties of rat skeletal muscle. Medicine and Science in Sports and Exercise 29, BURKE, R. E., LEVINE, D. N. & ZAJAK, F. E. (1971). Mammalian motor units: physiological histochemical correlation in three types in cat gastrocnemius. Science 174, CELIO, M. R. & HEIZMANN, C. W. (1982). Calcium-binding protein parvalbumin is associated with fast contracting muscle fibres. Nature 297, COOKE, W. H., GRANDJEAN, P. W. & BARNES, W. S. (1995). Effect of oral creatine supplementation on power output and fatigue during bicycle ergometry. Journal of Applied Physiology 78, FRYER, M. W., OWEN, V. J., LAMB, G. D. & STEPHENSON, D. G. (1995). Effects of creatine phosphate and P i on Ca 2+ movements and tension development in rat skinned skeletal muscle fibres. Journal of Physiology 482, GOLDBERG, A. L., MARTEL, S. B. & KUSHMERICK, M. J. (1975). In vitro preparations of the diaphragm and other skeletal muscles. Methods in Enzymology 39, GREENHAFF, P. L., CASEY, A., SHORT, A. H., HARRIS, R., SÖDERLUND, K. & HULTMAN, E. (1993). Influence of oral creatine supplementation on muscle torque during repeated bouts of maximal voluntary exercise in man. Clinical Science 84,

6 190 M. McGuire, A. Bradford and M. MacDermott Exp. Physiol HAWKINS, C., XU, A. & NARAYANAN, N. (1994). Sarcoplasmic reticulum calcium pump in cardiac and slow twitch skeletal muscle but not fast twitch skeletal muscle undergoes phosphorylation by endogenous and exogenous Ca 2+ /cadmodulin-dependent protein kinase. Characterization of optimal conditions for calcium pump phosphorylation. Journal of Biological Chemistry 269, HULTMAN, E., SÖDERLUND, K., TIMMONS, J. A., CEDERBLAD, G. & GREENHAFF, P. L. (1996). Muscle creatine loading in men. Journal of Applied Physiology 81, KUEI, J. H., SHADMEHR, R. & SIECK, G. C. (1990). Relative contribution of neurotransmission failure to diaphragm fatigue. Journal of Applied Physiology 68, MCKENNA, M. J., MORTON, J., SELIG, S. E. & SNOW, R. J. (1999). Creatine supplementation increases muscle total creatine but not maximal intermittent exercise performance. Journal of Applied Physiology 87, PULIDO, S. M., PASSAQUIN, A. C., LEIJENDEKKER, W. J., CHALLET, C., WALLIMANN, T. & RÜEGG, U. T. (1998). Creatine supplementation improves intracellular Ca 2+ handling and survival in mdx skeletal muscle cells. FEBS Letters 439, RANATUNGA, K. W. & WYLIE, S. R. (1983). Temperature-dependent transitions in isometric contractions of rat muscle. Journal of Physiology 339, REID, B. M., HAACK, K. E., FRANCHEK, K. M., VALBERG, P. A., KOBSIK, L. & WEST, M. S. (1992). Reactive oxygen in skeletal muscle 1. Intracellular oxidant kinetics and fatigue in vitro. Journal of Applied Physiology 73, SEGAL, S. S. & FAULKNER, J. A. (1985). Temperature-dependent physiological stability of rat skeletal muscle in vitro. American Journal of Physiology. 248, C TANAKA, T., OHIRA, Y., DANDA, M., HATTA, H. & NISHI, I. (1997). Improved fatigue resistance not associated with maximum oxygen consumption in creatine-depleted rats. Journal of Applied Physiology 82, VAN LEEMPUTTE, M. K., VANDENBERGHE, K. & HESPEL, P. (1999). Shortening of muscle relaxation time after creatine loading. Journal of Applied Physiology 86, WAKATSUKI, T., OHIRA, Y., YASUI, W., NAKAMURA, K., ASAKURA, T., OHNO, H. & YAMAMOTO, M. (1994). Responses of contractile properties in rat soleus to high-energy phosphates and/or unloading. Japanese Journal of Physiology 44, YANG, Z. & STEELE, D.S. (1999). Effects of phosphocreatine on SR Ca 2+ regulation in isolated saponin-permeabilized rat cardiac myocytes. Journal of Physiology 521, 27P. Acknowledgements The authors wish to thank Mr T. Dowling and Mr J. Slattery for technical assistance. The work was supported by a grant from The Research Committee of The Royal College of Surgeons in Ireland.