- Email: [email protected]
Radioimmunoassay of creatine kinase
Journal of the Neurological Sciences, 1978, 38: 235-247 © Elsevier/North-Holland Biomedical Press
235
R A D I O I M M U N O A S S A Y OF C R E A T I N E K I N A S E Studies on the Skeletal Muscle and Cardiac Muscle Enzyme ( M M and MB Isoenzymes) in Serum and Neuromuscular Tissues
G. A. NICHOLSON* and W. J. O'SULLIVAN** Department of Medicine, University of Sydney, Sydney, NS W (Australia) (Received 15 March, 1978) (Accepted 15 May, 1978)
SUMMARY A radioimmunoassay, specific for the isoenzymes of creatine kinase containing the M subunit of the enzyme (MM and MB creatine kinase), was employed to determine total creatine kinase concentrations in serum and neuromuscular tissues independently of the state of activity of the enzyme. This technique provides a method for the detection of inactive enzyme, which could be produced by inhibitors of the enzyme or by mutations involving the active site of the enzyme. A series of experiments were carried out to compare the amount of creatine kinase in various samples as assessed by normal enzyme kinetic procedures and by radioimmunoassay. The two techniques yielded equivalent results in all situations tested. Samples included serum f r o m normal subjects and subjects with genetic and acquired diseases of muscle and also extracts from skeletal and cardiac muscle. Small quantities of immunoreactive enzyme were found in nervous tissue and assessed in terms of the incidence of the M subunit.
INTRODUCTION Creatine kinase (ATP: creatine phosphotransferase, EC 2.7.3.2) is a dimeric molecule composed of M and/or B subunits. There are 3 isoenzymes, the M M isoenzyme of skeletal muscle, the BB isoenzyme of nervous tissue, and a hybrid MB isoenzyme This work was supported by grants from the Nina Annie Campbell Postgraduate Medical Fellowship of the University of Sydney and the National Health and Medical Research Council of Australia. * Present address: Regional Neurological Centre, Newcastle General Hospital, Westgate Road, Newcastle upon Tyne NE4 6BE, Great Britain. ** School of Biochemistry, The University of New South Wales, Sydney, NSW, Australia.
236 found in cardiac muscle and smooth muscle together with MM isoenzyme (Dawson, Eppenberger and Kaplan 1965). The isoenzymes found in serum both in normal and pathological conditions are almost exclusively MM and MB isoenzymes, even when a disorder of the nervous system is the cause of the elevated serum levels of the enzyme (Dubo, Park, Pennington et al. 1967). There have been suggestions that, in some circumstances, measurement of the biochemical activity of the enzyme may underestimate the amount released by the tissue (Dalal, Cilley and Winston 1972; Snehalatha, Valmikinathan, Srinivas et al. 1973). This difference could be due to inactive enzyme in serum, arising either from inactivation in the circulation or from the inherent disease process. Precedents for the presence of inactive enzymes have been reported in at least two circumstances. Mutants of hypoxanthineguanine phosphoribosyltransferase with different Km values have been reported in some Lesch-Nyhan patients (Ghangas and Milman 1975; Sutton and Wagner 1975). Secondly, accumulation of inactive fructose-l,6-diphosphate aldotase was observed by immunoassay in the livers of senescent mice, as compared to young mice (Gershon and Gershon 1973). Inactive BB creatine kinase has also been reported in mice (Armstrong, Lowden and Sherwin 1975). Radioimmunoassay of human creatine kinase has been employed to assay the B subunit isoenzymes (MM and MB) in serum (Roberts, Sobel and Parker 1976). The clinical applications of such assays as an index of myocardial damage have depended upon the assumption that MB isoenzyme in serum arises exclusively from cardiac muscle and that BB enzyme is rarely present in serum. Recent evidence indicates that significant quantities of MB isoenzyme may be found in normal muscle (Wilhelm and Todd 1977). We have developed a radioimmunoassay for human creatine kinase, using antisera to the MM isoenzyme in order to assess the efficiency of enzymatic activity methods in measuring total MM and MB creatine kinase. Preliminary results have been previously reported (Nicholson and O'Sullivan 1973). This paper gives the results of further experiments on the properties of the immunoassay, results of application of the assay to situations where inactive enzyme has been reported, i.e. in aging tissues (Gershon et al. 1973), the serum of patients with disorders of skeletal muscle (Dalai et al. 1972; Snehalatha et al. 1973) and results of application of the immunoassay to detect the M subunit isoenzymes of creatine kinase in human neuromuscular tissues. MATERIALSAND METHODS
Enzyme Human creatine kinase isoenzymes were prepared from skeletal muscle by a modification (Nicholson 1975) of the method of Keutel, Okabe, Jacobs et al. (1972). The MM isoenzyme preparation used for development of the immunoassay had a specific activity of 125 U/mg. Enzyme protein was estimated by the method of Kuby and Noltman (1962). The enzyme produced a single band of protein on polyacrylamide gel electrophoresis and a single precipitin line on immunoelectrophoresis using rabbit antiserum to the enzyme.
237
Antisera Antisera to MM creatine kinase were prepared by immunization of rabbits (Nicholson et al. 1973) using 1-2 mg of enzyme in complete Freund's adjuvant followed by a second injection using incomplete Freund's adjuvant and a final injection of enzyme in 20 mM sodium phosphate buffer pH 7.5. All antisera produced single immunoprecipitin lines to samples which contained MM or MB isoenzymes. There was no cross-reaction of the antisera to a preparation of BB isoenzyme in the immunoassay (see below and Fig. 3).
Samples Serum samples were tested for creatine kinase activity before and after storage a t --20 °C; the frozen samples were thawed in the dark as suggested by Thomson (1969). Tissue samples were obtained from cadavers less than 4 hr after death. Brain tissue was obtained from frontal cortex; peripheral nervous tissue was from the cauda equina. Heart tissue was obtained from the left ventricular wall and skeletal muscle was from the quadriceps muscle. Skeletal muscle from young and older subjects was obtained, with informed consent, from the rectus abdominus muscle of patients undergoing elective operations. The tissue was homogenised at 0 °C with 1.5 volumes of cold 20 mM sodium phosphate buffer, pH 7.5, containing 0.1 mM sodium EDTA and 1 mM mercaptoethanol, using a teflon-glass homogenizer. The homogenate was centrifuged at 3,000 x g for 20 min at 5 °C and the supernatant was retained and used immediately.
Enzyme assay Creatine kinase activity determinations were by the coupled enzyme technique (Calbiochem, U.S.A.) based on the method of Oliver (1955) as modified by Rosalki (1967). As noted in previous reports (Berndt and Bergmeyer 1965; Dinovo, Miyada and Nakamura 1973), enzyme activity determined by the coupled enzyme reaction was nonlinear at high enzyme concentrations. For this reason all comparisons of enzyme activity with immunoreactive protein were made relative to a pure enzyme standard used in all assays. Enzyme activity determinations of both the pure enzyme and enzyme containing samples were performed concurrently with the immunoassay. The linear relationship between enzyme activity and immunoreactive protein was described in a previous communication (Nicholson and O'Sullivan 1975). Samples used for comparison of biochemical activity determinations with immunoassay were stored not longer than one month before assay. However, no loss of immunoreactive enzyme by freezing and storage over a 6-month period was detected. Enzyme activity determination was performed on fresh samples and was repeated when samples were thawed for immunoassay. No loss of activity was noted over this period of storage.
Inhibition of creatine kinase activity Dilutions of human creatine kinase in 0.02 M phosphate buffer, pH 7.5, 0.1 mM
238 EDTA and 1 mM mercaptoethanol were incubated at room temperature with rabbit antiserum. After various time periods, from 30 sec to 10 min, 25/~1 aliquots were added to 1 ml of substrate solution and the enzyme activity estimated. As controls, dilutions of the enzyme were incubated under identical conditions with added buffer or with preimmune rabbit serum.
Labelled enzyme [125I]creatine kinase was prepared by the chloramine-T method (Berndt et al. 1965) to a specific activity of 5/zCi/mg, as described previously (Nicholson et al. 1973). Labelled enzyme was separated from free 12~I by Sephadex G-50 chromatography and yielded a single labelled peak on paper electrophoresis. High concentrations of rabbit antisera bound greater than 90 ~ of the labelled enzyme suggesting no significant loss of immunoreactivity of the enzyme by iodination, though some variable loss (10-30 ~ ) of enzyme activity occurred. Further evidence that the immunoreactivity of the labelled enzyme was similar to that of the native enzyme was obtained by serial dilution of a serum sample with an initially high creatine kinase activity. A curve parallel to the standard curve was obtained (Fig. 1). Radioimmunoassay The creatine kinase radioimmunoassay was carried out as described previously (Nicholson et al. 1973; Nicholson et al. 1975). [125I]Creatine kinase (10,000 cpm) and rabbit antiserum sufficient to produce 50 ~o binding of the labelled enzyme (1/25,000), Serum I,$~
1,256
iq211
1,64
Dilution 1,32
I*11
I*11
1,4
1,2
I,I
80
=1 O m
%
20-
0
?o
~o
~obo
foooo
C K (nglml)
Fig. 1. Comparison of creatine kinase standard curve (O--O) with a serially diluted serum specimen with an initial activity of 600 mU/ml (F1 - --[:3). Percent bound refers to the percentage of [l~5I]creatine kinase precipitated in the immunoassay.
239 40
9
30'
20-
I0"
O
16
~o CK
~'o
;,o
ADDED (n 9)
Fig. 2. Recovery (CK assayed) of pure enzyme (CK added) to serum samples with low immunoreactive enzyme levels.Recovery was calculated by subtracting the quantity of immunoreactive enzyme assayed in serum samples with no added enzyme, from the total amount of immunoreactive enzyme assayed. was diluted in 20 mM sodium phosphate pH 7.5, containing 0.15 M NaCI, 0.1 mM sodium EDTA, 1 mM mercaptoethan01, bovine serum albumin, 5 g/l, and non-immune rabbit serum 1 ml/1. Dilutions of the pure MM isoenzyme preparation were used to prepare a standard curve. Samples for immunoassay were diluted in the same buffer to yield activities between 20 m U / m l and 200 mU/ml and protein concentrations within the working range of the immunoassay. After incubation at 4 °C for 16 hr, donkey antirabbitw-globulin (Burroughs-Wellcome) was added. After further incubation at 4 °C for 16 hr, antibody-bound [~5I]creatine kinase was collected by centrifugation and the pellet was washed twice in the same buffer. The pellet was then counted in a Wallac automatic 7-spectrometer. The lower detection limit of the assay was 10 ng/ml and full recovery of creatine kinase in serum was obtained (Fig. 2). RESULTS
Antisera to creatine kinase The results reported in this paper are based on experiments with antisera obtained from 7 rabbits. The respective antisera varied somewhat in their ability to bind [lzsI]creatine kinase, viz. the dilution of rabbit serum required in the assay to produce 50 binding of the labelled enzyme varied from 1/15,000 to 1/275,000. In all antisera there was complete cross-reaction with the MB isoenzyme but no cross-reaction with a preparation of BB isoenzyme (specific activity 68 U/mg) over a range of concentrations from 20 ng/ml to 0.2 mg/ml (Fig. 3). Conversely antisera prepared to BB enzyme had no [125I]MM creatine kinase binding activity in the immunoassay. The species specificity of the radioimmunoassay was examined by testing purified rabbit muscle creatine kinase and lobster arginine kinase, and homogenates from goat and monkey skeletal muscle. There was no cross-reaction with any of these species except for monkey (Macaca nemistrina) where both the creatine kinase activity and the immunoreactivity of material in skeletal muscle were very similar to that of the pure human M M isoen-
240 CK I0
I
(ng/ml) I0 ~
103
I04
7C
60
~
.....
.
~
.
.
~
r-
gr~ m
cauda equina
~
heort
0 I10 s
Ii10+
III0~
I:I0a
1,10
II
Supernatant Dilution Fig. 3. Immunoassay of tissue homogenates. The standard curve for the muscle isoenzyme (MM) is compared to serial dilution of semi-purifiedbrain creatine kinase (BB)plotted against the upper axis for MM creatine kinase. On a separate axis serial dilutions of supernatant from homogenates of heart, peripheral nerve (cauda equina) and brain are compared.
zyme. These results are in accord with the observations of Jockers-Wretou and Pfleiderer (1975) who prepared antibodies to human M M and BB isoenzymes in rabbits. These workers did not find any evidence for isoenzyme cross-reactivity of the respective antibodies. Different antisera produced maximum inhibition of the MM isoenzyme varying from 8 to 80 ~ but there was no effect on the activity of a preparation of BB enzyme. There was no correlation of ability to inhibit enzyme activity and ability to bind pure M M enzyme in the immunoassay. Under the conditions used, with enzyme at 50/~g/ml and a 1-10 final dilution of the respective antisera, the inhibition was effectively instantaneous and was not affected by long periods of incubation up to 4 hr. The degree of enzyme inhibition caused by the antibody-antisera reaction was not affected by any of the substrates of the reaction, either alone or in combination, including the full reaction mixture which would give a "working" enzyme. These results are in contrast to those of Samuels who reported protection by substrate (MgADP- and creatine) by a chicken antiserum to the rabbit M M enzyme (Samuels 1961). The inactivation of the native enzyme with dithiobisnitrobenzoic acid, by substitution at the two reactive sulfhydryl groups (O'Sullivan 1971), did not affect its cross reactivity with any of the antisera, indicating no loss of immunoreactivity by inactivation of the active site. Sulphydryl groups can be oxidised by chloramine-T (Butt 1969), so that inactivation of the enzyme could occur during the iodination reaction. As the maximum inactivation that occurred was 30 ~ with no loss of immunoreactivity, it would appear that the antibodies responsible for enzyme inactivation are distinct from the antibodies binding to [125I]creatine kinase utilised in the immunoassay. Evidence for the integrity of the immunoreactivity of the enzyme after the labelling procedure was
241
A
B
10
d'
o"
Q
Q
8 6 4 t,/) LU .-I
2
Q,.
=E (/) ii
O
14 12
I.U
=E
10
:3 Z
8 6
r
1
I
4A 2
20
40
60
CK ACTIVITY
80
100 120
(mU/ml)
20
40
60
80 100 120 140160
IMMUNOREACTIVE
180
CK ~ng/ml)
Fig. 4. A : frequency histograms of creatine kinase activity in serum samples from 29 normal males (upper figure)and 29 normal females(lower figure). B: frequency histograms of immunoreactive creatine kinase in the same serum samples as those in A. Serum from 29 normal males (upper figure) and 29 normal females (lower figure).
provided by the high level of maximum [125I]creatine kinase binding achieved (93 %). The antisera to human MM isoenzyme had no effect on the enzyme activity of the BB isoenzyme. It was concluded that there was no overlap between the active site of the enzyme and the antibody binding site in the series of antisera produced in this study. Any situations where the two sites appear to have areas in common would appear to be serendipitous.
Comparison of immunoassay with enzyme assay serum samples Similar frequency distribution patterns were obtained for creatine kinase in sera of normal males and females using radioimmunoassay and enzymatic activity (Fig. 4). A linear regression between immunoreactive creatine kinase and activity of the pure human muscle enzyme, over a one hundred-fold concentration, 10-1,000 ng/ml, has been reported previously (Nicholson and O'Sullivan 1975). Analysis of the intercepts and variances of regression lines also established no significant differences (P < 0.05) between immunologically determined enzyme and that estimated by standard assay procedures in sera from: (a) a group of normal subjects, 24 males and 24 females; (b) a group of subjects, 21 males and 24 females suffering from myopathies; (c) 8 females
242
B
A
j
1,000
>I=,
100
I= v,
10
C /
t
J
s"
0 lO
I
I
~oo
~ooo
~o
I
I
~oo
~ooo
~o
1
I
~oo
~ooo
IMMUNOREACTIVE CK (ng/ml) Fig. 5. A : regression comparing creatine kinase activity and immunoreactive creatine kinase protein obtained from estimations on serial dilutions of pure human creatine kinase used in the standard curves of six immunoassays. The unbroken line is the calculated regression line (r = 0.98). IogeCKA = 0.85 logelCK q- 2.23. Broken lines are the 95 % tolerance limits of the regression. B: regression comparing creatine kinase activity and immunoreactiveprotein in supernatant from human skeletal musclehomogenates. Muscle samples were obtained from 3 young subjects (two aged 21 and one aged 26) shown as (0) and 2 older subjects (aged 65 and 73) shown as (A). The calculated regression line (r -- 0.99) is shown by the solid line. logeCKA = 0.90 logelCK ÷ 2.11. Broken lines indicate 95 % tolerance limits. C: regression comparing creatine kinase activity and immunoreactive protein estimations in serial dilutions of serum from a patient with Duchenne muscular dystrophy. The calculated regression (r = 1.00) is shown by the solid line. IogeCKA = 0.89 logelCK ~ 2.71. The 95 ~ tolerance limits of the regression are indicated by broken lines.
suspected as carriers of the gene for Duchenne muscular dystrophy; and (d) 7 females known to he carriers of the gene for Duchenne muscular dystrophy. The regressions obtained by serial dilution of serum containing large amounts of the enzyme (from 5 patients with Duchenne muscular dystrophy) and by dilution of skeletal muscle homogenates were not significantly different to the regression for pure M M enzyme, an example is shown in Fig. 5, indicating full activity of the enzyme in the serum samples. Serial determinations of enzyme activity and immunoreactive protein in serum from two patients recovering from myocardial infarction demonstrated a very good correlation between the two methods (Fig. 6).
Tissue samples Radioimmunoassay of serial dilutions of skeletal and heart muscle homogenates produced parallel curves to that of the standard curve indicating the presence of a crossreacting material (Fig. 3). A small amount of cross-reacting material was apparent in central and peripheral nervous tissue. This is further demonstrated by regressions for enzyme activity and immunoreactivity of these tissues (Fig. 7). Comparison of these regressions showed no significant difference between the activity of immunoreactive protein in both heart and skeletal muscle homogenates. Both were equivalent to the regression obtained for the pure M M isoenzyme, indicating that immunoreactive protein accounts for the total enzyme activity of these tissues. Comparison of enzyme activity with immunoreactive enzyme in nervous tissue
243 I0,000"
1,000
tO01
..---e
5
I0
DAYS
17
I0,000-
1,000'
'
'
'
'
5'
'
r
,
,
10
'
37
DAYS Fig. 6. Creatine kinase activity and immunoreactive protein in serial serum specimens obtained from 2
patients with myocardial infarction. Both creatine kinase activity and immunoreactive protein are shown plotted against the same axis. A-- ~ = creatine kinase activity (mU/ml); O O = immunoreactive protein (ng/ml) The legend on the abscissa refers to days after the infarction. homogenates showed the presence of some immunoreactive protein (either M M or MB isoenzyme) (Fig. 7). If this amount of immunoreactive material was active M M or MB creatine kinase, it would represent 15 % of total peripheral nervous tissue creatine kinase activity and 8 % of total central nervous tissue creatine kinase activity. Similar quantities of active enzyme were detected in peripheral and central nervous tissue homogenates (10 % and 5 % respectively of M M and < 1% of MB) using agarose gel electrophoresis and a coupled enzyme fluorescent technique (Corning-ACI, Palo Alto, CA). The relatively elevated enzyme activities of these tissues reflect the predominance of BB isoenzyme in nervous tissue, compared to muscle. No differences were detected with age in the activity of enzyme protein in skeletal muscle from male subjects in the range from 21 to 73 years (two subjects aged 21, one 26, one 65, one 73) (Fig. 5B). DISCUSSION
The isoenzyme specificity of antibodies to creatine kinase was first demonstrated
244
,/
,dC"
. o°'°°° s >FhiOL
jo 2 IMMUNOREACTIVE
;o~ CK ( n g / m l )
Fig. 7. Creatine kinase activity and immunoreactive protein estimation in homogenates from human heart muscle (©), brain ('k') and peripheral nerve (A) compared to the regression for skeletal muscle homogenate (0).
by Bulcke and Sherwin (1969). It has been adapted to the quantitation of isoenzymes in human serum by radioimmunoassay (Nicholson et al. 1973, 1975) and by the use of specific antibodies (Jockers,Wretou et al. 1975) to remove the respective isoenzyme. Recently Roberts et al. (1976) have made use of the rare occurrence of BB isoenzyme in serum to apply a radioimmunoassay for MB and BB isoenzymes in order to assay MB isoenzyme in serum and to assay canine M M and MB isoenzymes (Roberts and Painter 1977). Our results for comparison of enzyme activity and immunoreactive enzyme in serum following myocardial infarction are similar to those reported by this group. Absence of BB isoenzyme in the serum samples tested in the present study can be inferred from the results of the regressions for serum samples, where all enzyme activity can be accounted for by the concentration of M immunoreactive protein present in the samples. The presence of small amounts of BB enzyme cannot be excluded by this method; a specific immunoassay would be required to detect BB enzyme in serum. The results are also in agreement with those of Somer, Dubowitz and Donner (1976) who report that the BB isoenzyme is rarely detectable in the serum of patients with neuromuscular disorders. The close correlation between the activity of enzyme protein in serum and that of purified MM isoenzyme or that of the enzyme protein in muscle suggests that no significant inhibition of enzyme activity occurs in serum samples. It confirms previous suggestions that apparent increases in activity on dilution of serum samples are artifactual and are most probably due to the non-linearity of the coupled en-
245 zyme reactions used in creatine kinase activity estimations (Nicholson 1975; Berndt et al. 1965; Dinovo et al. 1973). Previous reports of the tissue distribution of creatine kinase isoenzymes using methods dependent upon enzyme activity have not detected MM or MB isoenzyme in nervous tissue (Dawson and Fine 1967; Allard and Cabrol 1970; Smith 1972). The results presented in this paper do not allow us to establish whether the immunoreactive material detected in nervous tissue was MB or MM isoenzyme. A possible source of MB or MM isoenzymes in nervous tissue is smooth muscle in blood vessels, but the contribution of blood vessel smooth muscle to the total tissue volume and the amount of MM and MB isoenzyme in smooth muscle (Jockers-Wretou and Pfleiderer 1975) is insufficient to account for the amount of M subunit isoenzyme detected, indicating that the M subunit isoenzymes may be present in neurones and/or glial cells. Active MM isoenzyme has previously been detected in brain (Jacobs, Heldt and Klingenberg 1964; Mercer 1974; Friedhoffand Lerner 1977). The finding ofimmunoreactive M protein in nervous tissue could represent inactive MB or MM isoenzyme in these tissues. There is a precedent for this possibility as Armstrong et al. (1975) reported evidence for the presence of inactive BB isoenzymes in skeletal muscle. However, no evidence was obtained for inactive enzyme in brain in this study as equivalent amounts of enzymatically and immunologically active M subunit isoenzymes were detected in nervous tissue homogenates. Small amounts of the isoenzyme which are not the predominant form are found in muscle during development and trace amounts of MB and BB have been described in adult skeletal muscle (Wilhelm and Todd 1977; Kloosterboer, Stoker-deVries and Hommes 1976; Morris, Piper and Cole 1976). The presence of M subunit in nervous tissue and B subunit in normal skeletal muscle may indicate that all three creatine kinase isoenzymes are present in all neuromuscular tissues but expressed to different relative proportions. Interpretations of serum creatine kinase isoenzyme determinations in terms of organ specificity should therefore be made with caution. Creatine kinase radioimmunoassay is a highly sensitive means of quantitation of total MM and MB isoenzymes and if used in conjunction with immunoassay for the B subunit (Roberts et al. 1977), a highly sensitive method of quantitation of all three isoenzymes independent of enzyme activity is available. This could prove useful in studies of tissue differentiation and dedifferentiation. In Duchenne muscular dystrophy disturbance of the cellular metabolism of MM creatine kinase is one of the most outstanding biochemical abnormalities in the disease and may be linked to the fundamental biochemical defect. The presence of MM enzyme in nervous tissue may similarly provide a biochemical link with the known disturbances of neuronal function in the disease process (Duchenne 1868). Addendum Since this paper was written Fang, Cho and Meltzer (1977) have reported the clinical use of a creatine kinase M subunit radioimmunoassay. Their results confirm our radioimmunoassay and enzymatic activity findings for creatine kinase M subunit isoenzymes in serum samples from normal subjects and patients with Duchenne muscular dystrophy (Nicholson and O'Sullivan 1975).
246 REFERENCES Allard, D. and D. Cabrol (1970) Etude ~lectrophor6tique des isozymes de la cr~atine phosphokinase dans les tissus de l'homme et du lapin, Path. BioL, 18 : 847-850. Armstrong, J. B., J. A. Lowden and A. L. Sherwin (1975) Enzymatically inactive brain-type creatine kinase in mammalian heart and skeletal muscle. In: W. G. Bradley, D. Gardner-Medwin and J. N. Walton (Eds.), Recent Advances in Myology (Proceedings of the 3rd International Congress on Muscle Diseases, Newcastle upon Tyne, September, 1974) (International Congress Series, No. 360), Excerpta Medica, Amsterdam, 1975, pp. 258-266. Berndt, E. and H. U. Bergmeyer (1965) Creatine phosphokinase. In: H. H. Bergmeyer (Ed.), Methods of Enzymatic Analysis, Verlag Chemie, Weinheim, pp. 859-862. Bulcke, J. A. and A. L. Sherwin (1969) Organ specificity of creatine phosphokinase muscle isoenzyme, lmmunochemistry, 6: 681-687. Butt, W. R. (1969) Chemistry of gonadotrophins in relation to their antigenic properties, Acta endocr. (Kbh.), 63, Suppl. 142: 13-60. Dalal, R., J. Cilley and S. Winston (1972) A study of the problems of inactivation of creatine kinase in serum, Clin. Chem., 18: 330-334. Dawson, D. M. and I. H. Fine (1967) Creatine kinase in human tissues, Arch. Neurol. (Chic.), 16:175180. Dawson, D. M., H. M. Eppenberger and N. O. Kaplan (1965) Creatine kinase-- Evidence for a dimeric structure, Biochem. biophys. Res. Commun., 21 : 346-353. Dinovo, E. C., D. S. Miyada and R. M. Nakamura (1973) Evaluation of direct and indirect coupled enzyme assay systems for measurement of creatine kinase activity, Clin. Chem., 19: 994-997. Dubo, H., D. C. Park, R. J. T. Pennington et al. (1967) Serum-creatine-kinase in cases of stroke, head injury and meningitis, Lancet, 2: 743-748. Duchenne, G. B. (1868) Recherches sur le paralysie musculaire pseudohypertrophique ou paralysie myoscl6rosique, Arch. gdn. todd., 2: 5, 179, 305, 421,552. Fang, V. S., H. W. Cho and H. Y. Meltzer (1977) Radioimmunoassay for MM and BB isoenzymes substantiated by clinical application, Clin. Chem., 23 : 1898-1902. Friedhoff, A. J. and M. H. Lerner (1977) Creatine kinase isoenzyme associated with synaptosomal membrane and synaptic vesicles, Life Sci., 20: 867-874. Gershon, H. and D. Gershon (1973) Inactive enzyme molecules in aging mice, Proc. nat. Acad. Sci. U.S.A., 70: 909-913. Ghangas, G. S. and G. Milman (1975) Radioimmune determination of hypoxanthine phosphoribosyltransferase cross-reacting material in erythrocytes of Lesch-Nyhan patients, Proc. nat. Acad. Sci. U.S.A., 72: 4147-4150. Jacobs, H., H. W. Heldt and M. Klingenberg (t964) High activity of creatine kinase in mitochondria from muscle and brain and evidence for a separate mitochondrial isoenzyme of creatine kinase, Biochem. biophys. Res. Commun., 16: 516-521. Jockers-Wretou, E. and G. Pfleiderer (1975) Quantitation ofcreatine kinase isoenzymes in human tissues and sera by an immunological method, Clin. chim. Acta, 58: 223-232. Keutel, H. J., K. Okabe, H. K. Jacobs et al. (1972) Studies on adenosine triphosphate transphosphorylases, Part 11 (Isolation of the crystalline adenosine triphosphate-creatine transphosphorylases from the muscle and brain of man, calf and rabbit; and a preparation of their enzymatically active hydrids), Arch. Biochem. Biophys., 150: 648-678. Kloosterboer, H. J., S. A. Stoker-de-Vries and F. A. Hommes (1976) The development of creatine kinase in rat skeletal muscle - - Changes in isoenzyme ratio, protein, R N A and D N A during development, Enzyme, 21: 448-458. Kuby, S. A. and E. A. Noltman (1962) A T P - creatine transphosphorylase, The Enzymes, 6: 515-596. Mercer, D. W. (1974) Separation of tissue and serum creatine kinase isoenzymes by ion-exchange column chromatography, Clin. Chem., 20: 3~40. Morris, G. E., M. Piper and R. Cole (1976) Do increases in enzyme activities during muscle differentiation reflect expression of new genes ? Nature (Lond.), 263 : 76-77. Nicholson, G. A. (1975) Human Creatine Kinase - - A Comparison of lmmunoreactivity and Biochemical Activity, Ph.D. Thesis, University of Sydney. Nicholson, G. A. and W. J. O'Sullivan (1973) A radioimmunoassay for creatine kinase, Proc. Aust. Soc. Neurol., 105-108. Nicholson, G. A. and W. J. O'Sullivan (1975) Radioimmunoassay for human creatine kinase - - The role of enzyme inactivation in serum creatine kinase estimations. In: W. G. Bradley, D. Gardner-
247 Medwin and J. N. Walton (Eds.), Recent Advances in Myology (Proceedings of the 3rd International Congress on Muscle Diseases, Newcastle upon Tyne, September, 1974) (International Congress Series, No. 360), Excerpta Medica, Amsterdam, 1975, pp. 267-272. Oliver, I. T. (1955) A spectrophotometric method for the creatine phosphokinase and myokinase, Biochem. J., 61 : 116-122. O'Sullivan, W. J. (1971) The reaction of creatine kinase with dithiobisnitrobenzoic acid - - Formation of derivatives of the enzyme, Int. J. Protein Res., 3 : 139-148. Roberts, R. and A. Painter (1977) Radioimmunoassay for canine creatine kinase isoenzymes, Biochem. biophys. Acta, 480: 521-526. Roberts, R., B. H. Sobel and C. W. Parker (1976) Radioimmunoassay for creatine kinase isoenzymes, Science, 194: 855-857. Rosalki, S. B. (1967) An improved procedure for serum creatine phosphokinase determination, J. Lab. clin. Med., 69 : 696-705. Samuels, A. J. (1961) Immunocnzymological evidence suggesting a change in conformation of adenylic acid deaminase and creatine kinase during substrate combination, Biophys. J., 1 : 437 ~44. Smith, A. F. (1972) Separation of tissue and serum creatine kinase isoenzymes on polyacrylamide gel slabs, Clin. chim. Acta, 39: 351-359. S omer, H., V. Dubowitz and M. Donner (1976) Creatine kinase isoenzymes in neuromuscular diseases, J. neurol. Sci., 29: 129-136. Snehalatha, C., K. Valmikinathan, K. Srinivas et al. (1973) Creatine phosphokinase level in neuromuscular disorders - - Effect of dilution and dialysis, Clin. chim. Acta, 44: 229-235. Sutton, E. H. and R. P. Wagner (1975) Mutation and enzyme function in humans, Ann. Rev. Genet., 9: 187-212.
Thomson, W. H. S, (1969) An investigation of physical factors influencing the behaviour in vitro of serum CPK and other enzymes, Clin. chim. Acta, 23: 105-120. Wilhelm, A. H. and J. K. Todd (1977) Limited diagnostic value of CK-MB, Clin. Chem., 23 : 1509.
235
R A D I O I M M U N O A S S A Y OF C R E A T I N E K I N A S E Studies on the Skeletal Muscle and Cardiac Muscle Enzyme ( M M and MB Isoenzymes) in Serum and Neuromuscular Tissues
G. A. NICHOLSON* and W. J. O'SULLIVAN** Department of Medicine, University of Sydney, Sydney, NS W (Australia) (Received 15 March, 1978) (Accepted 15 May, 1978)
SUMMARY A radioimmunoassay, specific for the isoenzymes of creatine kinase containing the M subunit of the enzyme (MM and MB creatine kinase), was employed to determine total creatine kinase concentrations in serum and neuromuscular tissues independently of the state of activity of the enzyme. This technique provides a method for the detection of inactive enzyme, which could be produced by inhibitors of the enzyme or by mutations involving the active site of the enzyme. A series of experiments were carried out to compare the amount of creatine kinase in various samples as assessed by normal enzyme kinetic procedures and by radioimmunoassay. The two techniques yielded equivalent results in all situations tested. Samples included serum f r o m normal subjects and subjects with genetic and acquired diseases of muscle and also extracts from skeletal and cardiac muscle. Small quantities of immunoreactive enzyme were found in nervous tissue and assessed in terms of the incidence of the M subunit.
INTRODUCTION Creatine kinase (ATP: creatine phosphotransferase, EC 2.7.3.2) is a dimeric molecule composed of M and/or B subunits. There are 3 isoenzymes, the M M isoenzyme of skeletal muscle, the BB isoenzyme of nervous tissue, and a hybrid MB isoenzyme This work was supported by grants from the Nina Annie Campbell Postgraduate Medical Fellowship of the University of Sydney and the National Health and Medical Research Council of Australia. * Present address: Regional Neurological Centre, Newcastle General Hospital, Westgate Road, Newcastle upon Tyne NE4 6BE, Great Britain. ** School of Biochemistry, The University of New South Wales, Sydney, NSW, Australia.
236 found in cardiac muscle and smooth muscle together with MM isoenzyme (Dawson, Eppenberger and Kaplan 1965). The isoenzymes found in serum both in normal and pathological conditions are almost exclusively MM and MB isoenzymes, even when a disorder of the nervous system is the cause of the elevated serum levels of the enzyme (Dubo, Park, Pennington et al. 1967). There have been suggestions that, in some circumstances, measurement of the biochemical activity of the enzyme may underestimate the amount released by the tissue (Dalal, Cilley and Winston 1972; Snehalatha, Valmikinathan, Srinivas et al. 1973). This difference could be due to inactive enzyme in serum, arising either from inactivation in the circulation or from the inherent disease process. Precedents for the presence of inactive enzymes have been reported in at least two circumstances. Mutants of hypoxanthineguanine phosphoribosyltransferase with different Km values have been reported in some Lesch-Nyhan patients (Ghangas and Milman 1975; Sutton and Wagner 1975). Secondly, accumulation of inactive fructose-l,6-diphosphate aldotase was observed by immunoassay in the livers of senescent mice, as compared to young mice (Gershon and Gershon 1973). Inactive BB creatine kinase has also been reported in mice (Armstrong, Lowden and Sherwin 1975). Radioimmunoassay of human creatine kinase has been employed to assay the B subunit isoenzymes (MM and MB) in serum (Roberts, Sobel and Parker 1976). The clinical applications of such assays as an index of myocardial damage have depended upon the assumption that MB isoenzyme in serum arises exclusively from cardiac muscle and that BB enzyme is rarely present in serum. Recent evidence indicates that significant quantities of MB isoenzyme may be found in normal muscle (Wilhelm and Todd 1977). We have developed a radioimmunoassay for human creatine kinase, using antisera to the MM isoenzyme in order to assess the efficiency of enzymatic activity methods in measuring total MM and MB creatine kinase. Preliminary results have been previously reported (Nicholson and O'Sullivan 1973). This paper gives the results of further experiments on the properties of the immunoassay, results of application of the assay to situations where inactive enzyme has been reported, i.e. in aging tissues (Gershon et al. 1973), the serum of patients with disorders of skeletal muscle (Dalai et al. 1972; Snehalatha et al. 1973) and results of application of the immunoassay to detect the M subunit isoenzymes of creatine kinase in human neuromuscular tissues. MATERIALSAND METHODS
Enzyme Human creatine kinase isoenzymes were prepared from skeletal muscle by a modification (Nicholson 1975) of the method of Keutel, Okabe, Jacobs et al. (1972). The MM isoenzyme preparation used for development of the immunoassay had a specific activity of 125 U/mg. Enzyme protein was estimated by the method of Kuby and Noltman (1962). The enzyme produced a single band of protein on polyacrylamide gel electrophoresis and a single precipitin line on immunoelectrophoresis using rabbit antiserum to the enzyme.
237
Antisera Antisera to MM creatine kinase were prepared by immunization of rabbits (Nicholson et al. 1973) using 1-2 mg of enzyme in complete Freund's adjuvant followed by a second injection using incomplete Freund's adjuvant and a final injection of enzyme in 20 mM sodium phosphate buffer pH 7.5. All antisera produced single immunoprecipitin lines to samples which contained MM or MB isoenzymes. There was no cross-reaction of the antisera to a preparation of BB isoenzyme in the immunoassay (see below and Fig. 3).
Samples Serum samples were tested for creatine kinase activity before and after storage a t --20 °C; the frozen samples were thawed in the dark as suggested by Thomson (1969). Tissue samples were obtained from cadavers less than 4 hr after death. Brain tissue was obtained from frontal cortex; peripheral nervous tissue was from the cauda equina. Heart tissue was obtained from the left ventricular wall and skeletal muscle was from the quadriceps muscle. Skeletal muscle from young and older subjects was obtained, with informed consent, from the rectus abdominus muscle of patients undergoing elective operations. The tissue was homogenised at 0 °C with 1.5 volumes of cold 20 mM sodium phosphate buffer, pH 7.5, containing 0.1 mM sodium EDTA and 1 mM mercaptoethanol, using a teflon-glass homogenizer. The homogenate was centrifuged at 3,000 x g for 20 min at 5 °C and the supernatant was retained and used immediately.
Enzyme assay Creatine kinase activity determinations were by the coupled enzyme technique (Calbiochem, U.S.A.) based on the method of Oliver (1955) as modified by Rosalki (1967). As noted in previous reports (Berndt and Bergmeyer 1965; Dinovo, Miyada and Nakamura 1973), enzyme activity determined by the coupled enzyme reaction was nonlinear at high enzyme concentrations. For this reason all comparisons of enzyme activity with immunoreactive protein were made relative to a pure enzyme standard used in all assays. Enzyme activity determinations of both the pure enzyme and enzyme containing samples were performed concurrently with the immunoassay. The linear relationship between enzyme activity and immunoreactive protein was described in a previous communication (Nicholson and O'Sullivan 1975). Samples used for comparison of biochemical activity determinations with immunoassay were stored not longer than one month before assay. However, no loss of immunoreactive enzyme by freezing and storage over a 6-month period was detected. Enzyme activity determination was performed on fresh samples and was repeated when samples were thawed for immunoassay. No loss of activity was noted over this period of storage.
Inhibition of creatine kinase activity Dilutions of human creatine kinase in 0.02 M phosphate buffer, pH 7.5, 0.1 mM
238 EDTA and 1 mM mercaptoethanol were incubated at room temperature with rabbit antiserum. After various time periods, from 30 sec to 10 min, 25/~1 aliquots were added to 1 ml of substrate solution and the enzyme activity estimated. As controls, dilutions of the enzyme were incubated under identical conditions with added buffer or with preimmune rabbit serum.
Labelled enzyme [125I]creatine kinase was prepared by the chloramine-T method (Berndt et al. 1965) to a specific activity of 5/zCi/mg, as described previously (Nicholson et al. 1973). Labelled enzyme was separated from free 12~I by Sephadex G-50 chromatography and yielded a single labelled peak on paper electrophoresis. High concentrations of rabbit antisera bound greater than 90 ~ of the labelled enzyme suggesting no significant loss of immunoreactivity of the enzyme by iodination, though some variable loss (10-30 ~ ) of enzyme activity occurred. Further evidence that the immunoreactivity of the labelled enzyme was similar to that of the native enzyme was obtained by serial dilution of a serum sample with an initially high creatine kinase activity. A curve parallel to the standard curve was obtained (Fig. 1). Radioimmunoassay The creatine kinase radioimmunoassay was carried out as described previously (Nicholson et al. 1973; Nicholson et al. 1975). [125I]Creatine kinase (10,000 cpm) and rabbit antiserum sufficient to produce 50 ~o binding of the labelled enzyme (1/25,000), Serum I,$~
1,256
iq211
1,64
Dilution 1,32
I*11
I*11
1,4
1,2
I,I
80
=1 O m
%
20-
0
?o
~o
~obo
foooo
C K (nglml)
Fig. 1. Comparison of creatine kinase standard curve (O--O) with a serially diluted serum specimen with an initial activity of 600 mU/ml (F1 - --[:3). Percent bound refers to the percentage of [l~5I]creatine kinase precipitated in the immunoassay.
239 40
9
30'
20-
I0"
O
16
~o CK
~'o
;,o
ADDED (n 9)
Fig. 2. Recovery (CK assayed) of pure enzyme (CK added) to serum samples with low immunoreactive enzyme levels.Recovery was calculated by subtracting the quantity of immunoreactive enzyme assayed in serum samples with no added enzyme, from the total amount of immunoreactive enzyme assayed. was diluted in 20 mM sodium phosphate pH 7.5, containing 0.15 M NaCI, 0.1 mM sodium EDTA, 1 mM mercaptoethan01, bovine serum albumin, 5 g/l, and non-immune rabbit serum 1 ml/1. Dilutions of the pure MM isoenzyme preparation were used to prepare a standard curve. Samples for immunoassay were diluted in the same buffer to yield activities between 20 m U / m l and 200 mU/ml and protein concentrations within the working range of the immunoassay. After incubation at 4 °C for 16 hr, donkey antirabbitw-globulin (Burroughs-Wellcome) was added. After further incubation at 4 °C for 16 hr, antibody-bound [~5I]creatine kinase was collected by centrifugation and the pellet was washed twice in the same buffer. The pellet was then counted in a Wallac automatic 7-spectrometer. The lower detection limit of the assay was 10 ng/ml and full recovery of creatine kinase in serum was obtained (Fig. 2). RESULTS
Antisera to creatine kinase The results reported in this paper are based on experiments with antisera obtained from 7 rabbits. The respective antisera varied somewhat in their ability to bind [lzsI]creatine kinase, viz. the dilution of rabbit serum required in the assay to produce 50 binding of the labelled enzyme varied from 1/15,000 to 1/275,000. In all antisera there was complete cross-reaction with the MB isoenzyme but no cross-reaction with a preparation of BB isoenzyme (specific activity 68 U/mg) over a range of concentrations from 20 ng/ml to 0.2 mg/ml (Fig. 3). Conversely antisera prepared to BB enzyme had no [125I]MM creatine kinase binding activity in the immunoassay. The species specificity of the radioimmunoassay was examined by testing purified rabbit muscle creatine kinase and lobster arginine kinase, and homogenates from goat and monkey skeletal muscle. There was no cross-reaction with any of these species except for monkey (Macaca nemistrina) where both the creatine kinase activity and the immunoreactivity of material in skeletal muscle were very similar to that of the pure human M M isoen-
240 CK I0
I
(ng/ml) I0 ~
103
I04
7C
60
~
.....
.
~
.
.
~
r-
gr~ m
cauda equina
~
heort
0 I10 s
Ii10+
III0~
I:I0a
1,10
II
Supernatant Dilution Fig. 3. Immunoassay of tissue homogenates. The standard curve for the muscle isoenzyme (MM) is compared to serial dilution of semi-purifiedbrain creatine kinase (BB)plotted against the upper axis for MM creatine kinase. On a separate axis serial dilutions of supernatant from homogenates of heart, peripheral nerve (cauda equina) and brain are compared.
zyme. These results are in accord with the observations of Jockers-Wretou and Pfleiderer (1975) who prepared antibodies to human M M and BB isoenzymes in rabbits. These workers did not find any evidence for isoenzyme cross-reactivity of the respective antibodies. Different antisera produced maximum inhibition of the MM isoenzyme varying from 8 to 80 ~ but there was no effect on the activity of a preparation of BB enzyme. There was no correlation of ability to inhibit enzyme activity and ability to bind pure M M enzyme in the immunoassay. Under the conditions used, with enzyme at 50/~g/ml and a 1-10 final dilution of the respective antisera, the inhibition was effectively instantaneous and was not affected by long periods of incubation up to 4 hr. The degree of enzyme inhibition caused by the antibody-antisera reaction was not affected by any of the substrates of the reaction, either alone or in combination, including the full reaction mixture which would give a "working" enzyme. These results are in contrast to those of Samuels who reported protection by substrate (MgADP- and creatine) by a chicken antiserum to the rabbit M M enzyme (Samuels 1961). The inactivation of the native enzyme with dithiobisnitrobenzoic acid, by substitution at the two reactive sulfhydryl groups (O'Sullivan 1971), did not affect its cross reactivity with any of the antisera, indicating no loss of immunoreactivity by inactivation of the active site. Sulphydryl groups can be oxidised by chloramine-T (Butt 1969), so that inactivation of the enzyme could occur during the iodination reaction. As the maximum inactivation that occurred was 30 ~ with no loss of immunoreactivity, it would appear that the antibodies responsible for enzyme inactivation are distinct from the antibodies binding to [125I]creatine kinase utilised in the immunoassay. Evidence for the integrity of the immunoreactivity of the enzyme after the labelling procedure was
241
A
B
10
d'
o"
Q
Q
8 6 4 t,/) LU .-I
2
Q,.
=E (/) ii
O
14 12
I.U
=E
10
:3 Z
8 6
r
1
I
4A 2
20
40
60
CK ACTIVITY
80
100 120
(mU/ml)
20
40
60
80 100 120 140160
IMMUNOREACTIVE
180
CK ~ng/ml)
Fig. 4. A : frequency histograms of creatine kinase activity in serum samples from 29 normal males (upper figure)and 29 normal females(lower figure). B: frequency histograms of immunoreactive creatine kinase in the same serum samples as those in A. Serum from 29 normal males (upper figure) and 29 normal females (lower figure).
provided by the high level of maximum [125I]creatine kinase binding achieved (93 %). The antisera to human MM isoenzyme had no effect on the enzyme activity of the BB isoenzyme. It was concluded that there was no overlap between the active site of the enzyme and the antibody binding site in the series of antisera produced in this study. Any situations where the two sites appear to have areas in common would appear to be serendipitous.
Comparison of immunoassay with enzyme assay serum samples Similar frequency distribution patterns were obtained for creatine kinase in sera of normal males and females using radioimmunoassay and enzymatic activity (Fig. 4). A linear regression between immunoreactive creatine kinase and activity of the pure human muscle enzyme, over a one hundred-fold concentration, 10-1,000 ng/ml, has been reported previously (Nicholson and O'Sullivan 1975). Analysis of the intercepts and variances of regression lines also established no significant differences (P < 0.05) between immunologically determined enzyme and that estimated by standard assay procedures in sera from: (a) a group of normal subjects, 24 males and 24 females; (b) a group of subjects, 21 males and 24 females suffering from myopathies; (c) 8 females
242
B
A
j
1,000
>I=,
100
I= v,
10
C /
t
J
s"
0 lO
I
I
~oo
~ooo
~o
I
I
~oo
~ooo
~o
1
I
~oo
~ooo
IMMUNOREACTIVE CK (ng/ml) Fig. 5. A : regression comparing creatine kinase activity and immunoreactive creatine kinase protein obtained from estimations on serial dilutions of pure human creatine kinase used in the standard curves of six immunoassays. The unbroken line is the calculated regression line (r = 0.98). IogeCKA = 0.85 logelCK q- 2.23. Broken lines are the 95 % tolerance limits of the regression. B: regression comparing creatine kinase activity and immunoreactiveprotein in supernatant from human skeletal musclehomogenates. Muscle samples were obtained from 3 young subjects (two aged 21 and one aged 26) shown as (0) and 2 older subjects (aged 65 and 73) shown as (A). The calculated regression line (r -- 0.99) is shown by the solid line. logeCKA = 0.90 logelCK ÷ 2.11. Broken lines indicate 95 % tolerance limits. C: regression comparing creatine kinase activity and immunoreactive protein estimations in serial dilutions of serum from a patient with Duchenne muscular dystrophy. The calculated regression (r = 1.00) is shown by the solid line. IogeCKA = 0.89 logelCK ~ 2.71. The 95 ~ tolerance limits of the regression are indicated by broken lines.
suspected as carriers of the gene for Duchenne muscular dystrophy; and (d) 7 females known to he carriers of the gene for Duchenne muscular dystrophy. The regressions obtained by serial dilution of serum containing large amounts of the enzyme (from 5 patients with Duchenne muscular dystrophy) and by dilution of skeletal muscle homogenates were not significantly different to the regression for pure M M enzyme, an example is shown in Fig. 5, indicating full activity of the enzyme in the serum samples. Serial determinations of enzyme activity and immunoreactive protein in serum from two patients recovering from myocardial infarction demonstrated a very good correlation between the two methods (Fig. 6).
Tissue samples Radioimmunoassay of serial dilutions of skeletal and heart muscle homogenates produced parallel curves to that of the standard curve indicating the presence of a crossreacting material (Fig. 3). A small amount of cross-reacting material was apparent in central and peripheral nervous tissue. This is further demonstrated by regressions for enzyme activity and immunoreactivity of these tissues (Fig. 7). Comparison of these regressions showed no significant difference between the activity of immunoreactive protein in both heart and skeletal muscle homogenates. Both were equivalent to the regression obtained for the pure M M isoenzyme, indicating that immunoreactive protein accounts for the total enzyme activity of these tissues. Comparison of enzyme activity with immunoreactive enzyme in nervous tissue
243 I0,000"
1,000
tO01
..---e
5
I0
DAYS
17
I0,000-
1,000'
'
'
'
'
5'
'
r
,
,
10
'
37
DAYS Fig. 6. Creatine kinase activity and immunoreactive protein in serial serum specimens obtained from 2
patients with myocardial infarction. Both creatine kinase activity and immunoreactive protein are shown plotted against the same axis. A-- ~ = creatine kinase activity (mU/ml); O O = immunoreactive protein (ng/ml) The legend on the abscissa refers to days after the infarction. homogenates showed the presence of some immunoreactive protein (either M M or MB isoenzyme) (Fig. 7). If this amount of immunoreactive material was active M M or MB creatine kinase, it would represent 15 % of total peripheral nervous tissue creatine kinase activity and 8 % of total central nervous tissue creatine kinase activity. Similar quantities of active enzyme were detected in peripheral and central nervous tissue homogenates (10 % and 5 % respectively of M M and < 1% of MB) using agarose gel electrophoresis and a coupled enzyme fluorescent technique (Corning-ACI, Palo Alto, CA). The relatively elevated enzyme activities of these tissues reflect the predominance of BB isoenzyme in nervous tissue, compared to muscle. No differences were detected with age in the activity of enzyme protein in skeletal muscle from male subjects in the range from 21 to 73 years (two subjects aged 21, one 26, one 65, one 73) (Fig. 5B). DISCUSSION
The isoenzyme specificity of antibodies to creatine kinase was first demonstrated
244
,/
,dC"
. o°'°°° s >FhiOL
jo 2 IMMUNOREACTIVE
;o~ CK ( n g / m l )
Fig. 7. Creatine kinase activity and immunoreactive protein estimation in homogenates from human heart muscle (©), brain ('k') and peripheral nerve (A) compared to the regression for skeletal muscle homogenate (0).
by Bulcke and Sherwin (1969). It has been adapted to the quantitation of isoenzymes in human serum by radioimmunoassay (Nicholson et al. 1973, 1975) and by the use of specific antibodies (Jockers,Wretou et al. 1975) to remove the respective isoenzyme. Recently Roberts et al. (1976) have made use of the rare occurrence of BB isoenzyme in serum to apply a radioimmunoassay for MB and BB isoenzymes in order to assay MB isoenzyme in serum and to assay canine M M and MB isoenzymes (Roberts and Painter 1977). Our results for comparison of enzyme activity and immunoreactive enzyme in serum following myocardial infarction are similar to those reported by this group. Absence of BB isoenzyme in the serum samples tested in the present study can be inferred from the results of the regressions for serum samples, where all enzyme activity can be accounted for by the concentration of M immunoreactive protein present in the samples. The presence of small amounts of BB enzyme cannot be excluded by this method; a specific immunoassay would be required to detect BB enzyme in serum. The results are also in agreement with those of Somer, Dubowitz and Donner (1976) who report that the BB isoenzyme is rarely detectable in the serum of patients with neuromuscular disorders. The close correlation between the activity of enzyme protein in serum and that of purified MM isoenzyme or that of the enzyme protein in muscle suggests that no significant inhibition of enzyme activity occurs in serum samples. It confirms previous suggestions that apparent increases in activity on dilution of serum samples are artifactual and are most probably due to the non-linearity of the coupled en-
245 zyme reactions used in creatine kinase activity estimations (Nicholson 1975; Berndt et al. 1965; Dinovo et al. 1973). Previous reports of the tissue distribution of creatine kinase isoenzymes using methods dependent upon enzyme activity have not detected MM or MB isoenzyme in nervous tissue (Dawson and Fine 1967; Allard and Cabrol 1970; Smith 1972). The results presented in this paper do not allow us to establish whether the immunoreactive material detected in nervous tissue was MB or MM isoenzyme. A possible source of MB or MM isoenzymes in nervous tissue is smooth muscle in blood vessels, but the contribution of blood vessel smooth muscle to the total tissue volume and the amount of MM and MB isoenzyme in smooth muscle (Jockers-Wretou and Pfleiderer 1975) is insufficient to account for the amount of M subunit isoenzyme detected, indicating that the M subunit isoenzymes may be present in neurones and/or glial cells. Active MM isoenzyme has previously been detected in brain (Jacobs, Heldt and Klingenberg 1964; Mercer 1974; Friedhoffand Lerner 1977). The finding ofimmunoreactive M protein in nervous tissue could represent inactive MB or MM isoenzyme in these tissues. There is a precedent for this possibility as Armstrong et al. (1975) reported evidence for the presence of inactive BB isoenzymes in skeletal muscle. However, no evidence was obtained for inactive enzyme in brain in this study as equivalent amounts of enzymatically and immunologically active M subunit isoenzymes were detected in nervous tissue homogenates. Small amounts of the isoenzyme which are not the predominant form are found in muscle during development and trace amounts of MB and BB have been described in adult skeletal muscle (Wilhelm and Todd 1977; Kloosterboer, Stoker-deVries and Hommes 1976; Morris, Piper and Cole 1976). The presence of M subunit in nervous tissue and B subunit in normal skeletal muscle may indicate that all three creatine kinase isoenzymes are present in all neuromuscular tissues but expressed to different relative proportions. Interpretations of serum creatine kinase isoenzyme determinations in terms of organ specificity should therefore be made with caution. Creatine kinase radioimmunoassay is a highly sensitive means of quantitation of total MM and MB isoenzymes and if used in conjunction with immunoassay for the B subunit (Roberts et al. 1977), a highly sensitive method of quantitation of all three isoenzymes independent of enzyme activity is available. This could prove useful in studies of tissue differentiation and dedifferentiation. In Duchenne muscular dystrophy disturbance of the cellular metabolism of MM creatine kinase is one of the most outstanding biochemical abnormalities in the disease and may be linked to the fundamental biochemical defect. The presence of MM enzyme in nervous tissue may similarly provide a biochemical link with the known disturbances of neuronal function in the disease process (Duchenne 1868). Addendum Since this paper was written Fang, Cho and Meltzer (1977) have reported the clinical use of a creatine kinase M subunit radioimmunoassay. Their results confirm our radioimmunoassay and enzymatic activity findings for creatine kinase M subunit isoenzymes in serum samples from normal subjects and patients with Duchenne muscular dystrophy (Nicholson and O'Sullivan 1975).
246 REFERENCES Allard, D. and D. Cabrol (1970) Etude ~lectrophor6tique des isozymes de la cr~atine phosphokinase dans les tissus de l'homme et du lapin, Path. BioL, 18 : 847-850. Armstrong, J. B., J. A. Lowden and A. L. Sherwin (1975) Enzymatically inactive brain-type creatine kinase in mammalian heart and skeletal muscle. In: W. G. Bradley, D. Gardner-Medwin and J. N. Walton (Eds.), Recent Advances in Myology (Proceedings of the 3rd International Congress on Muscle Diseases, Newcastle upon Tyne, September, 1974) (International Congress Series, No. 360), Excerpta Medica, Amsterdam, 1975, pp. 258-266. Berndt, E. and H. U. Bergmeyer (1965) Creatine phosphokinase. In: H. H. Bergmeyer (Ed.), Methods of Enzymatic Analysis, Verlag Chemie, Weinheim, pp. 859-862. Bulcke, J. A. and A. L. Sherwin (1969) Organ specificity of creatine phosphokinase muscle isoenzyme, lmmunochemistry, 6: 681-687. Butt, W. R. (1969) Chemistry of gonadotrophins in relation to their antigenic properties, Acta endocr. (Kbh.), 63, Suppl. 142: 13-60. Dalal, R., J. Cilley and S. Winston (1972) A study of the problems of inactivation of creatine kinase in serum, Clin. Chem., 18: 330-334. Dawson, D. M. and I. H. Fine (1967) Creatine kinase in human tissues, Arch. Neurol. (Chic.), 16:175180. Dawson, D. M., H. M. Eppenberger and N. O. Kaplan (1965) Creatine kinase-- Evidence for a dimeric structure, Biochem. biophys. Res. Commun., 21 : 346-353. Dinovo, E. C., D. S. Miyada and R. M. Nakamura (1973) Evaluation of direct and indirect coupled enzyme assay systems for measurement of creatine kinase activity, Clin. Chem., 19: 994-997. Dubo, H., D. C. Park, R. J. T. Pennington et al. (1967) Serum-creatine-kinase in cases of stroke, head injury and meningitis, Lancet, 2: 743-748. Duchenne, G. B. (1868) Recherches sur le paralysie musculaire pseudohypertrophique ou paralysie myoscl6rosique, Arch. gdn. todd., 2: 5, 179, 305, 421,552. Fang, V. S., H. W. Cho and H. Y. Meltzer (1977) Radioimmunoassay for MM and BB isoenzymes substantiated by clinical application, Clin. Chem., 23 : 1898-1902. Friedhoff, A. J. and M. H. Lerner (1977) Creatine kinase isoenzyme associated with synaptosomal membrane and synaptic vesicles, Life Sci., 20: 867-874. Gershon, H. and D. Gershon (1973) Inactive enzyme molecules in aging mice, Proc. nat. Acad. Sci. U.S.A., 70: 909-913. Ghangas, G. S. and G. Milman (1975) Radioimmune determination of hypoxanthine phosphoribosyltransferase cross-reacting material in erythrocytes of Lesch-Nyhan patients, Proc. nat. Acad. Sci. U.S.A., 72: 4147-4150. Jacobs, H., H. W. Heldt and M. Klingenberg (t964) High activity of creatine kinase in mitochondria from muscle and brain and evidence for a separate mitochondrial isoenzyme of creatine kinase, Biochem. biophys. Res. Commun., 16: 516-521. Jockers-Wretou, E. and G. Pfleiderer (1975) Quantitation ofcreatine kinase isoenzymes in human tissues and sera by an immunological method, Clin. chim. Acta, 58: 223-232. Keutel, H. J., K. Okabe, H. K. Jacobs et al. (1972) Studies on adenosine triphosphate transphosphorylases, Part 11 (Isolation of the crystalline adenosine triphosphate-creatine transphosphorylases from the muscle and brain of man, calf and rabbit; and a preparation of their enzymatically active hydrids), Arch. Biochem. Biophys., 150: 648-678. Kloosterboer, H. J., S. A. Stoker-de-Vries and F. A. Hommes (1976) The development of creatine kinase in rat skeletal muscle - - Changes in isoenzyme ratio, protein, R N A and D N A during development, Enzyme, 21: 448-458. Kuby, S. A. and E. A. Noltman (1962) A T P - creatine transphosphorylase, The Enzymes, 6: 515-596. Mercer, D. W. (1974) Separation of tissue and serum creatine kinase isoenzymes by ion-exchange column chromatography, Clin. Chem., 20: 3~40. Morris, G. E., M. Piper and R. Cole (1976) Do increases in enzyme activities during muscle differentiation reflect expression of new genes ? Nature (Lond.), 263 : 76-77. Nicholson, G. A. (1975) Human Creatine Kinase - - A Comparison of lmmunoreactivity and Biochemical Activity, Ph.D. Thesis, University of Sydney. Nicholson, G. A. and W. J. O'Sullivan (1973) A radioimmunoassay for creatine kinase, Proc. Aust. Soc. Neurol., 105-108. Nicholson, G. A. and W. J. O'Sullivan (1975) Radioimmunoassay for human creatine kinase - - The role of enzyme inactivation in serum creatine kinase estimations. In: W. G. Bradley, D. Gardner-
247 Medwin and J. N. Walton (Eds.), Recent Advances in Myology (Proceedings of the 3rd International Congress on Muscle Diseases, Newcastle upon Tyne, September, 1974) (International Congress Series, No. 360), Excerpta Medica, Amsterdam, 1975, pp. 267-272. Oliver, I. T. (1955) A spectrophotometric method for the creatine phosphokinase and myokinase, Biochem. J., 61 : 116-122. O'Sullivan, W. J. (1971) The reaction of creatine kinase with dithiobisnitrobenzoic acid - - Formation of derivatives of the enzyme, Int. J. Protein Res., 3 : 139-148. Roberts, R. and A. Painter (1977) Radioimmunoassay for canine creatine kinase isoenzymes, Biochem. biophys. Acta, 480: 521-526. Roberts, R., B. H. Sobel and C. W. Parker (1976) Radioimmunoassay for creatine kinase isoenzymes, Science, 194: 855-857. Rosalki, S. B. (1967) An improved procedure for serum creatine phosphokinase determination, J. Lab. clin. Med., 69 : 696-705. Samuels, A. J. (1961) Immunocnzymological evidence suggesting a change in conformation of adenylic acid deaminase and creatine kinase during substrate combination, Biophys. J., 1 : 437 ~44. Smith, A. F. (1972) Separation of tissue and serum creatine kinase isoenzymes on polyacrylamide gel slabs, Clin. chim. Acta, 39: 351-359. S omer, H., V. Dubowitz and M. Donner (1976) Creatine kinase isoenzymes in neuromuscular diseases, J. neurol. Sci., 29: 129-136. Snehalatha, C., K. Valmikinathan, K. Srinivas et al. (1973) Creatine phosphokinase level in neuromuscular disorders - - Effect of dilution and dialysis, Clin. chim. Acta, 44: 229-235. Sutton, E. H. and R. P. Wagner (1975) Mutation and enzyme function in humans, Ann. Rev. Genet., 9: 187-212.
Thomson, W. H. S, (1969) An investigation of physical factors influencing the behaviour in vitro of serum CPK and other enzymes, Clin. chim. Acta, 23: 105-120. Wilhelm, A. H. and J. K. Todd (1977) Limited diagnostic value of CK-MB, Clin. Chem., 23 : 1509.