JOURNAL OF BIOLUMINESCENCE AND CHEMILUMINESCENCE VOL 6 123-129 (1991)

A Luminometric Method for t h e Determination of ATP and Phosphocreatine in Single Human Skeletal Muscle Fibres R . Wibom. K. Soderlund," A. Lundint and E. Hultman Department of Clinical Chemistry II and ?Department of Medicine and Clinical Research Centre, Huddinge University Hospital, S-141 86 Huddinge, Sweden

A sensitive method f o r the analysis o f ATP and phosphocreatine (PCr) in single human skeletal muscle fibres is described. Muscle tissue was freeze-dried and single fibres were dissected free w i t h the aid of low-power microscopy. The fibres were then extracted in trichloroacetic acid and neutralized with KHCO,. The assay is based o n the continuous monitoring o f light produced as a result o f ATP degradation in the firefly luciferase reaction. PCr is measured as the amount o f ATP formed in the creatine kinase reaction. The coefficient o f variation was less than 4 % f o r both ATP and PCr determination. The amount o f tissue required f o r the assay is approximately 0 . 5 ~ (dry ~ weight). The assay showed good agreement w i t h spectrophotometric and highperformance liquid chromatographic (HPLC) measurements made upon extracts o f whole muscle tissue. Keywords: ATP; luminescence; phosphocreatine; single fibres

INTRO DU CTlO N

Mammalian skeletal muscle is generally considered to consist of slow- and fast-twitch fibres. Compared to fast-twitch fibres, slow-twitch fibres are characterized as having slower shortening velocities, lower peak power outputs and a greater resistance to fatigue (Faulkner et al., 1986). These differing responses are due to inherent differences in the energy metabolism of the two fibre types. ATP and phosphocreatine (PCr) are compounds principally involved in the liberation of chemically bound energy to fuel muscle contraction and relaxation. Muscle tissue rich in fast-twitch fibres, has a higher ATP and PCr content compared to muscle rich in

slow-twitch fibres (Edstrom et al., 1982). During contraction the depletion of ATP and PCr is more rapid in fast- compared to slow-twitch fibres (Hintz et al., 1982). To date, the investigation of skeletal muscle energy metabolism in man has been based principally upon the analysis of mixed-fibred muscle tissue, which at best will only represent the average changes occurring in individual fibre types. The determination of ATP and PCr in single muscle fibres has been accomplished in the past using methods based upon enzymatic cycling (Hintz et al., 1982; Rehunen and Harkonen, 1980) and highperformance liquid chromatography (Jansson et al., 1987). However, these methods have been shown to be complicated and time-consuming to

*Author for correspondence.

0884-3996/91/0200123-07$05.00 0 1991 by John Wiley & Sons, Ltd.

Received 9 August 1990 Revised 12 December 1990

R. WIBOM,

124

K. SODERLUND, A. LUNDIN AND E. HULTMAN

were dissected free from each biopsy sample for single fibre analysis with the aid of low-power microscopy. The ends of each fragment were then cut off and the remainder was weighed using a quartz fibre fish-pole balance (Lowry and Passonneau, 1972). Using the pieces cut from each fragment, the fibre fragments were characterized (slowor fast-twitch) according to the method of Brooke and Kaiser (1970). Individual fibre fragments were extracted in 1OOpl of 2.5% trichloroacetic acid (TCA; w/v) followed by neutralization with 10 p1 of KHCO, (2.2 mol/l). TCA as extraction medium Principle of the Assay has previously been shown to give a high recovery The assay is based upon the quantitative measure- of adenosine nucleotides (Lundin, 1984). All neument of light produced as a result of two coupled tralized extracts were then frozen and stored at enzyme reactions. The first reaction is catalysed by -80°C until analysed. The 12 biopsy samples for the enzyme firefly luciferase (FL) (EC 1.13.12.7), whole muscle analysis were freeze-dried and exand the second by creatine kinase (CK) tracted in a perchloric acid-KHCO, mixture, as previously described (Harris et al., 1974). (EC 2.7.3.2).

use. Luminescence has previously been employed in the determination of ATP in single horse muscle fibres (Foster et al., 1986). The present paper describes a rapid and accurate luminometric method for the simultaneous determination of ATP and PCr in single muscle fibres. The aims of the study were, first, to evaluate the luminometric method and, secondly, to compare it with other techniques currently available.

FL

1. ATP + Luciferin + 0, Oxyluciferin + C O , + light

2. ADP + phosphocreatine (PCr) creatine

AMP CK

+ PPi + ATP

+

In the firefly luciferase reaction the light produced is directly proportional to the ATP concentration (linear range 10-"-10-6 mol/l, with the reagent used). In the first step of the assay the ATP content of the muscle extract is determined. In the second step PCr determination is achieved by the addition of ADP and CK to the reagent mixture. The endpoint in the CK reaction is ascertained by continuously monitoring the reaction using a high concentration of ADP. When the endpoint is reached, the amount of ATP formed is determined by adding a known amount of ATP for internal calibration. MATERIAL AND METHODS Preparation of samples

Twenty biopsy samples were obtained from the quadriceps femoris muscle of 15 individuals at rest, and after mild and heavy exercise using the percutaneous needle biopsy technique (Bergstrom, 1962). Eight biopsy samples were used for single fibre analysis and 12 for whole muscle analysis. Following freeze-drying, fragments of single muscle fibres

Reagents and solutions for the ATP and PCr determinations in single fibres

Lyophilized ATP monitoring reagent (AMR), ADP substrate and ATP standard were purchased from Bio Orbit Oy (Turku, Finland). The AMR (firefly luciferase; D-luciferin, 5 mg; L-luciferin, 0.2 mg; bovine serum albumin, 50 mg; magnesium acetate, 0.5 mmol and inorganic pyrophosphate, 50 nmol) was dissolved in 47.5 ml of a solution containing sucrose (180 mmol/l), KH,PO, (20 mmol/l), K,HPO, (15 mmol/l) and EDTA (1 mmol/l). The pH was adjusted to 6.7 with KOH. The ADP substrate containing 0.05 mmol ADP, 0.1 mmol AMP and 20nmol P',P5-di(adenosine-5') pentaphosphate (DAPP) was dissolved in 1.5 ml of water. The ATP standard (100 nmol) was dissolved in 10 ml of water. Creatine kinase (CK) 350 U/mg (BoehringerMannheim, no. 127566) was dissolved in water (10 mg/ml). The CK preparation used was low in levels of contaminating enzymes (in particular ATPase and adenylate kinase) and was free of salts. All other chemicals used were obtained from Merck (Darmstadt, FRG). Luminometric determination of standard solutions of ATP and PCr

Two series of standard solutions were prepared in neutralized extraction medium with ATP and PCr concentrations in the range of 0.006-3 and

ATP AND PHOSPHOCREATINE IN SINGLE MUSCLE FIBRES

0.02-10pmol/l, respectively. To each of the ATP standard solutions 1 pmol/l of PCr was added, and correspondingly to each of the PCr standard solutions 0.3 pmol/l of ATP was added. These additions were made in order to determine whether there was any analytical interference between the two compounds. In the assay, both ATP and PCr were determined in the same cuvette of each standard solution. Assay procedure for the luminometric determination of single fibre ATP and PCr

10

20

containing 950 pl of reconstituted AMR were incubated at 25°C for 10min to allow temperature equilibration. In step 1 of the assay sample (25 pl) was added to the AMR, and the light emission corresponding to the sample ATP concentration was measured. In steps 2 and 3, ADP substrate (15 pl) and CK (10 pl) were added. The light emission was measured before the addition of CK and after the completion of the CK reaction. Finally, in step 4,ATP (10 p1) for internal standardization was added, and the subsequent increase in signal recorded. Procedural notes

(Fig. l(A)). The assay was carried out using a 1251 luminometer (Bio Orbit Oy, Turku, Finland), fitted with a temperature-controlled 25 position sample carousel, three automatic dispensing units and a computer. Prior to beginning the assay, cuvettes

0

125

30

Time (min)

Figure 1. (A) The analytical procedure for determination of the lurntnometrtc response t o ATP and PCr in a sample extract In the end of the assay ATP standard IS added (B) Determination of the correction factor used in the calculation of the ATP content in the samples The correction factor is calculated as the ratio between AStl and ASt2

The assay was run simultaneously in all cuvettes in the sample carousel, i.e. each step in the assay was carried out in all cuvettes before the next step was started. The automatic assay of 25 cuvettes takes about 35 min. Computer programs and a complete description of the calculations for the automatic assay of ATP and PCr can be obtained from the authors. To correct for any possible contamination by ATP in the reagent solutions, two blanks containing 25 pl neutralized extraction medium (TCA-KHCO,), instead of sample, were included in each carousel. Contaminating ATP in the ADP substrate produces an increase in light emission when ADP substrate is added to the AMR, as indicated by Fig. 1. Furthermore, the firefly luciferase reagent contains a small amount of adenylate kinase, which, in the presence of ADP, will form ATP. This is partially inhibited by the presence of AMP and DAPP in the ADP substrate solution (Lundin et al., 1982). A low rate of ATP formation, however, still occurs, and must to be taken into account in the calculations. The intensity of the light emission is slightly affected by the addition of ADP substrate to the reaction mixture. Thus the value of the ATP standard used in each individual assay is valid only for the determination of PCr. The ATP standard used for the determination of the sample ATP content must be calculated using the measured standard value and a correction factor. This factor is determined by running the assay with blanks to which ATP standard is added before the addition of ADP substrate, as well as after the addition of CK (Fig. l(B)). The mean ratio between the two ATP standard values, obtained by performing this procedure on at least three occasions, is used as the correction factor.

R. WIBOM, K. SODERLUND, A. LUNDIN AND E. HULTMAN

126

Luminometric ATP and PCr estimations in whole muscle samples compared with other methods

ATP and PCr levels determined using our own routine spectrophotometric method (Harris et al., 1974),and the high-performance liquid chromatography (HPLC) methods of Ingebretsen et al. (1982) and Dunnett et al. (1 990), respectively, were compared with those obtained using the luminometric method described in this paper. Because the above methods were not sufficiently sensitive enough to detect the ATP and PCr contents of individual fibres, extracts from whole muscle were analysed. The extracts used were undiluted for photometry, diluted 1:lO for HPLC and 1:lOOO for luminometry. The TCA-KHCO, mixture was used as diluent. Statistics

The variation in the methods was calculated with analysis of variance (Snedecor and Cochran, 1976). R ES U LTS

Thus determinations of ATP and PCr were linear over a wide range of concentrations. The concentration of PCr added to all ATP standard solutions (see the section ‘Material and methods’) was 1.03 -t 0.01 pmol/l (mean & SD; n = 9). The corresponding value for the ATP added to the PCr standard solutions was 0.303 f 0.005 pmol/l ( n = 9). It can be concluded therefore that there was no analytical interference between the two compounds. Single muscle fibres

The ATP and PCr content was determined in single muscle fibres obtained at rest and after exercise. Fibres from resting muscle showed an ATP and PCr content of 22-28 and 80-107mmol/kg dry weight, respectively. PCr was not detectable in some of the fibres obtained after exercise while the ATP content ranged from 4 to 20mmol/kg. The coefficient of variation (CV) was lower for duplicate determinations of the same fibre extract (4% for ATP and 3% for PCr; n = 44, 28 fast-twitch and 16 slow-twitch fibres) when compared with the CV obtained using two separately extracted pieces of the same fibre (5% for ATP and 7 % for PCr; n = 16, 9 fast-twitch and 7 slow-twitch fibres).

Standard solutions Whole muscle samples

Linear regression analyses calculated using logarithmic values of concentrations (added versus determined) resulted in slopes close to 1 and intercepts close to 0 for both ATP and PCr (Fig. 2).

v =

-0.001

f

1.004~

r=1.00

A

The contents of ATP and PCr determined in 12 muscle biopsy samples with luminometry, spectrophotometry and HPLC are shown in Fig. 3. The

y

i

-0.006

+

r;1.00

1.006.

n c 4:

1 0 - ~ i o ” PCr(pmol

loo ‘

1.

10’

lo2

’ ) added

Figure 2. Standard solutions of (A) ATP (0.006-3 prnol/l) and ( 6 ) phosphocreatine (PCr) (0.02-1 0 prnol/l) determined with lurninometry (logarithmic scale)

ATP AND PHOSPHOCREATINE IN SINGLE MUSCLE FIBRES

’

127

0

0 ATP (mmol

i

. 3.4

+

1.05x

r

i

30

. kg-1)

ATP (mmol

Spectrophotometric

y

20

10

. kg-1)

HPLC

1.00

80-

60-

01 0

20

40 PCr (mmol

60

80

. kg-l)

Spectrophotomelric

1

I

//i 20

40

60

80

1

0

PCr (mmol. kg-’) HPLC

Figure 3. Luminornetric determination of ATP (A, B) and phosphocreatine (PCr) (C, D) in 1 2 muscle tissue extracts related to spectrophotometric and HPLC methods

three methods show good agreement and a similar DISCUSSION precision. The mean CV for ATP (mmol/kg dry muscle) was 0.7 %, 3.4 % and 0.5 % for the luminometric, spectrophotometric and HPLC determina- Assay conditions tions, respectively (Fig. 3). The corresponding values for PCr were 1.1 %, 3.4 % and 6.2 %, respec- The ‘ADP substrate’ is a mixture of ADP, AMP and DAPP. AMP and DAPP are added in order to tively. In five of the whole muscle extracts the ATP and inhibit the adenylate kinase reaction (see the secPCr contents were determined repeatedly over a tion ‘Material and methods’). A disadvantage with five-day period using the luminometric method. the use of AMP is its interaction with the luciferase The ATP contents of the five extracts were reaction, thus when ‘ADP substrate’ is added to the (mean f SD): 24.5 & 0.7; 22.3 k 0.6; 18.0 f 0.5; reagent the intensity of the light emission is affected 10.7 f 0.3 and 5.6 f 0.1 mmol/kg dry muscle. The (Lee and McElroy, 1971; Lundin et al., 1982). corresponding values for PCr were 95.7 f 1.9; Despite this interference being small (5-10 %), the 81.4 1.3; 55.5 f 1.1; 41.6 k-1.8 and 7.8 k value obtained for the ATP standard added at the 0.6 mmol/kg dry muscle. The CVs were 1.6-2.9 % in end of the assay creates an error in the calculation all determinations, except in the sample with the of the ATP content of the sample. The problem can be overcome by adding an extra aliquot of ATP lowest PCr content, where it was 8 %.

128

R. WIBOM, K. SODERLUND, A. LUNDIN AND

standard after the measurement of the ATP content in the sample, but before the ADP substrate is added. This, however, considerably reduces the sensitivity of the PCr measurements, because a small difference in signal has to be measured above a high background signal. A more accurate method is described in the present paper. A correction factor is determined based upon the values obtained for the ATP standard added to separate blank cuvettes before and after the addition of ADP substrate (Fig. 1). The operator simply has to prepare three extra blank cuvettes, after which the additions of ADP, CK and ATP, and the subsequent calculations for the sample ATP and PCr contents will be performed by a computer program. The assay buffer was chosen because it was used for other purposes in the laboratory. It has an optimal pH for the CK reaction but conditions are far from optimal for the luciferase reaction. However, this does not influence the accuracy of the assay, since all measurements are performed with an internal ATP standard. Furthermore, the detection limit for PCr is determined by reagent contaminations rather than light detection. Single muscle fibres

Previous determinations of ATP and PCr in single muscle fibres have been performed using an enzymatic cycling assay (Hintz et al., 1982; Rehunen and Harkonen, 1980). ATP has also been determined using HPLC (Jansson et al., 1987). With the enzymatic cycling assay a whole battery of reagent solutions are needed, a complex analytical flow chart has to be followed, and very small volumes are used (less than 5 pl, sometimes fractions of 1 pl, involving a special pipetting technique). The HPLC method also requires small volumes and both methods are laborious. The precision of the present method is equal to or higher than the methods described previously. It is rapid (more than 100 fibres per day can be analysed), it requires no specialized pipetting procedures, and only four solutions are needed. Furthermore, repeated determinations can be performed on one muscle fibre extract. In the present study a total of 60 fibres were assayed. They were chosen to give a wide range of concentrations in order to estimate the accuracy of the method. The CVs were 4 % for ATP and 3 % for PCr, which is satisfactory. The CVs were higher when two pieces of the same fibre were extracted

E. HULTMAN

and analysed separately compared to duplicate determinations of the same extract. This variation originates from the extraction procedure, but may possibly also be an effect of the biological variation in ATP and PCr content in different parts of the muscle fibre (Hintz et a!., 1984).

Whole muscle samples

The samples used in the whole muscle study were obtained from a variety of experiments investigating exercise metabolism and therefore gave a wide range of ATP and PCr contents. The analytical results obtained using the photometric and HPLC methods (extracts undiluted and diluted 1:10, respectively) were in good agreement with the results obtained using the bioluminometric method (dilution 1:lOOO). The three methods show a linear relationship over the wide range of ATP and PCr contents (Fig. 3). Control values for ATP and PCr in resting human quadriceps muscle are 24.0 & 2.6 and 75.5 k 7.6 mmol/kg dry muscle, respectively (Harris et al., 1974). The reproducibility of the bioluminometric method, determined in five of the whole muscle extracts was found to be good (CV < 3%). Only in the sample with the lowest PCr content was the reproducibility somewhat lower.

CONCLUSION

From the first determinations of ATP and PCr in whole muscle tissue (Hultman et al., 1967), a vast number of studies have been undertaken to investigate the functional role of these compounds in normal muscle (Hultman and Harris, 1988) and in a variety of pathological conditions (Edwards et al., 1981) Little however, is known about the breakdown and resynthesis of ATP and PCr at the single fibre level. The method described here is well-suited for the study of exercise metabolism in single muscle fibres. There are also, however, clinical applications for the method, for example in the study of patients with low muscle ATP and PCr contents (Bergstrom et al., 1976; Nordemar et al., 1974). The technique could also be adapted for analysis of microsamples of other biological materials such as liver (Hultman et al., 1975), kidney (Collste et al., 1971) and spermatozoa (Pousette et al., 1986).

ATP AND PHOSPHOCREATINE IN SINGLE MUSCLE FIBRES

Acknowledgements This work was supported by grants from the Swedish Medical Research Council (02647) and the Swedish Work Environment Fund (81 -0173). The authors especially wish to thank Dr P. Greenhaff for constructive criticism of this manuscript. We also thank K. Bodin, K. W%hlen and H. Ahlman for their technical assistance, and A. Adestam for preparation of the manuscript.

REFERENCES Bergstrom, J. (1962). Muscle electrolytes in man. Determination by neutron activation analysis on needle biopsy specimens. A study on normal subjects, kidney patients and patients with chronic diarrhoea. Scand. J . Clin. Lab. Invest., 14, suppl. 68, 1-110. Bergstrom, J., Bostrom, H., Fiirst, P., Hultman, E. and Vinnars, E. (1976). Preliminary studies of energy-rich phosphagens in muscle from severely ill patients. Crit. Care Med., 4, 197-204. Brooke, M. H. and Kaiser, K. K. (1970). Muscle fiber types: how many and what kind? Arch. Neurol., 23, 369-379. Collste, H., Bergstrom, J., Hultman, E. and Melin, B. (1971). ATP in the cortex of canine kidneys undergoing hypothermic storage. Lije Sci., 10, 1201-1206. Dunnett, M., Harris, R. C. and Orme, C. E. (1991). Reverse phase ion-pairing high performance liquid chromatography of phosphocreatine, creatine and creatinine in equine muscle. Scand. J . Clin. Lab. Invest., 51(2). Edstrom, L., Hultman. E., Sahlin, K. and Sjoholm, H. (1982). The contents of high-energy phosphates in different fibre types in skeletal muscle from rat, guinea-pig and man. J . Physiol., 332, 47-58. Edwards, R.H.T., Harris, R.C. and Jones D.A. (1981). The biochemistry of muscle biopsy in man: clinical applications. Rec. Adv. Clin. Biochem., 2, 243-269. Faulkner, J. A,, Claflin, D. R. and McCully, K. K. (1986). Power output of fast and slow fibers from human skeletal muscles. In Human Muscle Power, Jones, N. L., McCartney, N. and McComas, A. (Eds), Human Kinetics Publishers Inc., Champaign, IL, pp. 81-94. Foster, C. V. L., Harman, J., Harris, R. C. and Snow, D. H. (1986). ATP distribution in single muscle fibres before and after maximal exercise in the thoroughbred horse. J . Physiol., 378, 64P. Harris, R. C., Hultman, E. and Nordesjo, L.-0. (1974). Glycogen, glycolytic intermediates and high-energy phosphates determined in biopsy samples of musculus quadriceps femor-

129

is of man at rest. Methods and variance of values. Scand. J . Clin. Lab. Invest., 33, 109-120. Hintz, C. S., Chi, M. M.-Y., Fell, R. D., Ivy, J. L., Kaiser, K. K., Lowry, C. V. and Lowry, 0.H. (1982). Metabolite changes in individual rat muscle fibers during stimulation. Am. J . Physiol., 242, C218-C227. Hintz, C. S., Chi, M. M.-Y. and Lowry, 0. H. (1984). Heterogeneity in regard to enzymes and metabolites within individual muscle fibers. Am. J . Physiol., 246, C288-C292. Hultman, E. and Harris, R. C. (1988). Carbohydrate metabolism. In Principles of Exercise Biochemistry. Medicine and Sport Science, Vol. 27, Poortmans, J. R. (Ed.), Karger, Basel, pp. 78- 119. Hultman, E., Bergstrom, J. and McLennan Anderson, N. (1967). Breakdown and resynthesis of phosphorylcreatine and adenosine triphosphate in connection with muscular work in man. Scand. J . Clin. Lab. Invest., 19, 56-66. Hultman, E., Nilsson H:son L. and Sahlin K. (1975). Adenine nucleotide content of human liver. Normal values and fructose-induced depletion. Scand. J . Clin. Lab. Invest., 35, 245-251. Ingebretsen, 0.C., Bakken, A. M., Segadal, L. and Farstad, M. (1982). Determination of adenine nucleotides and inosine in human myocard by ion-paired reversed-phase high-performance liquid chromatography. J . Chrornatogr., 242, 119- 126. Jansson, E., Dudley, G . A., Norman, B. and Tesch, P. A. (1987). ATP and I M P in single human muscle fibres after high intensity exercise. Clin. Physiol., 7, 337-345. Lee, R. T. and McElroy, W. D. (1971). Effects of 5’-adenylic acid on firefly luciferase. Arch. Biochem. Biophys., 145, 78-84. Lowry, 0. H. and Passonneau, J. V. (1972). The quartz fiber fishpole balance. In A Flexible System of Enzymatic Analysis, Academic Press, New York, pp. 237-249. Lundin, A. (1984). Extraction and automatic luminometric assay of ATP, ADP and AMP. In Analytical Applications of Bioluminescence and Chemiluminescence, Kricka, L. J., Stanley, P. E., Thorpe, G. H. G. and Whitehead, T. P. (Eds), Academic Press, New York, pp. 491-501. Lundin, A,, Jaderlund, B. and Lovgren, T. (1982). Optimized bioluminescence assay of creatine kinase and creatine kinase B-subunit activity. Clin. Chem., 28, 609-614. Nordemar, R., Lovgren, O., Fiirst, P., Harris, R. C. and Hultman, E. (1974). Muscle ATP content in rheumatoid arthritis-a biopsy study. Scand. J . Clin. Lab. Invest., 34, 185 191. Pousette, A,, Akerlof, E., Lundin, A,, Rosenborg, L. and Fredricsson, B. (1986). On the use of adenosine triphosphate for estimation of motility and energy status in human spermatozoa. In?. J . Andrology, 9, 331-340. Rehunen, S. and Harkonen, M. (1980). High-energy phosphate compounds in human slow-twitch and fast-twitch muscle fibres. Scand. J . Clin. Lab. Invest., 40, 45-54. Snedecor, G. W. and Cochran, W. G. (1976). Statistical Methods, 6th edn, The Iowa State University Press, Iowa.