Europ. J. Pediat. 126,53--59(1977)
F.urol:~andoumalof p~d~l~tlt~eS
9 by Springer-Verlag 1977
Serum Alkaline Phosphatase Isoenzymes in Childhood* K. Kruse 1**, H. Bartels ~, and H. Gfinther 2.** 1University of Kiel Medical School, Department of Pediatrics (Prof. Dr. H.-R. Wiedemann), Fr6belstr. 15/17, D-2300 Kiel, Federal Republic of Germany 2Department of Pediatrics (Prof. Dr. H. G. Hansen), Medical School, D-2400 Ltiheck, Federal Republic of Germany
Abstract. Using a combination of L-phenylalanine inhibition and heat inactivation, the serum alkaline phosphatase (AP) in 2 to 13 year old children without evidence of hepatobiliary, osseous, or intestinal disease was separated in three fractions; i.e. L-phenylalanine sensitive AP (LPSAP), heat-stable non-L-phenylalanine sensitive AP (heat-stable non-LPSAP) and heat-sensitive non-L-phenylalanine sensitive AP (heat-sensitive non-LPSAP). The activities of total AP and the different fractions were measured using optimized test conditions. LPSAP, (mainly intestinal AP), accounts for approximately 12% of the total serum AP activity, heat-stable non-LPSAP (mainly hepatobiliary AP) for approximately 9%, and heat-sensitive nonLPSAP (mainly bone AP) for approximately 77%. To give a better differentiation between bone and liver AP, the percentage ratios of heat-stable n o n - L P S A P / n o n - L P S A P (Q value), and heat-stable n o n - L P S A P / t o t a l AP, were determined. Both quotients showed a significant negative correlation with total AP, which has to be taken into account in the interpretation of the results of isoenzyme determinations of serum AP activity. The above semiquantitative separation of AP isoenzymes can be readily done in a routine clinical laboratory. Key words: Serum alkaline phosphatase isoenzymes - L-phenylalanine inhibition - H e a t inactivation - Normal values in children.
Studies on alkaline phosphatase isoenzymes (API) using techniques such as electrophoresis on a variety of media, treatment with inhibitors like L-phenylalanine and urea, heat-inactivation and specific antisera suggest that h u m a n serum alkaline phosphatase (AP) is derived from liver, bone, intestine and, in pregnant women, placenta (Fishman and Gosh, 1967). An enhancement of serum * Supported in part by Deutsche Forschungsgemeinschaft (Ba 246/11) ** Corresponding author *** The data form part of the thesis of H. Giinther, Medical School of Liibeck. Present address: Department of Anaesthesiology, Medical School, D-2400 Labeck
54
K. Kruse et al.
A P is usually due to an increase o f either the bone or the hepatobiliary isoenzyme (Kaplan, 1972). Since an elevation of A P is sometimes the only a b n o r m a l biochemical finding in a patient, the hepatobiliary or skeletal origin of the enzyme has to be established by isoenzyme investigation. Moreover, total serum A P activity within the n o r m a l range m a y be a c c o m p a n i e d by a clinically significant alteration o f one isoenzyme fraction (Ohlen et al., 1971; Kattwinkel et al., 1973; D o m i n i c k et al., 1975; Siede and Seifert, 1977). Using a c o m b i n a t i o n of chemicalinhibition and heat-inactivation Ohlen et al. (1971) achieved a semiquantitative separation o f A P I in adults which is sufficiently exact for clinical purposes and can be performed in routine clinical laboratories. Cao et al. (1972), employing a similar technique, have published reference values for n o r m a l infants and children. However, in the studies o f C a o et al. (1972) and Ohlen et al. (1971) A P activities were measured utilizing n o n - o p t i m a l test conditions. Since 1970 an optimized kinetic test, using the m e t h o d of H a u s a m e n et al. (1967) and recommended, for example by the G e r m a n Society for Clinical Chemistry (1970), has been in widespread use. The aim o f this study was to establish reference values in children for serum A P I differentiation according to Ohlen et al. (1971), using the r e c o m m e n d e d conditions for measuring A P activity.
Material and Methods Sera from 83 children (38 girls and 45 boys aged 2 to 13 years), who were in- and outpatients of the Department of Pediatrics, Liibeck, were investigated. None of the children suffered from a disease known to alter serum AP activity. In particular, there was no clinical or chemical evidence of hepatobiliary, osseous, or intestinal disease in any case. AP activity was measured according to the optimized method of Hausamen et al. (1967), utilizing a commercially available test kit (Monotest, AP optimized, Boehringer Mannheim). Isoenzyme Determinations (s, Fig. 1). After estimation of total serum AP activity (total AP), AP was measured in the presence of 5raM L-phenylalanine, representing non-L-phenylalanine sensitive AP (non-LPSAP). The decline in activity caused by L-phenylalanine corresponds to the L-phenylalanine sensitive fraction of total activity (LPSAP) and mainly reflects intestinal AP (Fishman and Gosh, 1967). Thereafter, the serum was heated for exactly 10min at 56.0+0.1~ and then cooled immediately in an icebath. AP activity measured after heatinactivation in the presence of 5 mM L-phenylalanine, representing heat-stable non-LPSAP, is mainly due to the hepatobiliary isoenzyme, whereas the activity inactivated by heat (heatsensitive non-LPSAP) is mainly related to the skeletal isoenzyme (Posen et al., 1965).
[ total AP incLctivated
5 mM L-pheny|alanine
1.
I
(mainly intestine AP
J non-LPSAP~ I inactivated [ 10 rnin 56~
I
[ heatsensitiVenon -LPSAP I
]heatnon_LPsApStable I
(mainly bon.__..eAP} e
(mainlyliver AP}
Fig. 1. Flow diagram of semiquantitative isoenzyme determination of serum AP
Serum Alkaline Phosphatase Isoenzymes in Childhood
55
According to Ohlen et al. (1971), the percentage ratio of heat-stable non-LPSAP/nonLPSAP, represented as the Q value, is a reliable parameter for the hepatobiliary/hone isoenzyme ratio, enhancement of the Q value above the upper normal limit indicating an increase in the hepatobiliary isoenzyme and a reduction below normal suggesting an increase in the osseous isoenzyme. Statistical Analysis. The mode of distribution was ascertained graphically. Mean values, variances, regression equations, correlation coefficients and tolerance ellipses were calculated by standard methods using a DEC PDP 11/40 computer unit.
Results The values for total A P , L P S A P , n o n - L P S A P , heat-sensitive a n d heat-stable n o n L P S A P , a n d ' Q ' followed a p p r o x i m a t e l y l o g - n o r m a l distributions. All the p a r a m e t e r s studied were i n d e p e n d e n t of age a n d sex. I n Table 1 m e a n values a n d reference ranges ( ~ + 2 SD) are given for total serum A P activity a n d for the A P fractions separated by L - p h e n y l a l a n i n e i n h i b i t i o n a n d heat-inactivation. L P S A P a c c o u n t e d for a b o u t 12% of the total A P activity, heat-stable n o n - L P S A P for a b o u t 9% a n d heat-sensitive n o n - L P S A P for a b o u t 77%. A b s o l u t e activities ( U / l ) in the different fractions showed a positive correlation with the level of total A P
Table 1. Serum alkaline phosphatase (U/l) in 83 children aged 2--13 years before and after inhibition with 5 mM L-phenylalanine and heat inactivation (10min at 56~
Total AP~ LPSAP % of total AP Non-LPSAP Heat-sensitive non-LPSAP % of total AP Heat-stable non-LPSAP % of total AP Q value a
334 38.9 11.5 288 257 77.3 28.8 8.7 10.0
~-2s
~+2s
172 10.2 3.3 146 125 63.1 13.8 4.2 4.8
647 148 39.8 566 526 91.6 60.3 18.2 20.9
Abbreviations:
Total alkaline phosphatase activity in native serum LPSAP: AP activity inhibited by 5 mM L-phenylalanine = L-phenylalanine sensitive AP Non-LPSAP: AP activity in the presence of 5 mM L-phenylalanine Heat-sensitive non-LPSAP: Non-LPSAP activity inactivated by heat (10rain at 56~ Heat-stable non-LPSAP: Residual non-LPSAP activity after heat inactivation Q value: Percentage ratio of heat-stable non-LPSAP/non-LPSAP Total AP:
56
K. Kruse et al.
30[
__
2O
~~ '
B
30
A
9
r = 0.435
~"
r = 0./,39
i O.
~
P
20
9
10
-r
O
s
s tY
9~
o
3
I
100
I
I
200 300 500 TOTAL AP (Ult)
I
1000
3
I
100
I
I
I
20O 3O0 50O TOTAL AP ( U / t )
Fig. 2. Relationship of (A) the Q value and (B) the percentage ratio heat-stable non-LPSAP/ total AP to total serum AP activity in 83 children aged 2--13 years. Individual values, regression lines and tolerance ellipses (95% range) activity. LPSAP and non-LPSAP correlated significantly with total AP (r = 0.512 and 0.967, respectively). Heat-stable and heat-sensitive non-LPSAP were significantly correlated to non-LPSAP (r=0.489 and 0.992, respectively). In contrast, the percentage ratio of LPSAP/total AP activity remained constant, independent of the level of total AP activity. Though there is a tendency for the heat-sensitive non-LPSAP/total AP ratio to increase with increasing total AP activity, no significant correlation could be established due to the marked variation of individual values. However, a significant negative correlation exists between the level of total AP activity and the Q value (percentage ratio for heatstable non-LPSAP to non-LPSAP). This is shown in Figure 2A, where individual values, regression lines and the tolerance ellipse enclosing the 95% area of tolerance are presented on a logarithmic scale, and indicate a decrease of the Q value with increasing total AP activity. As demonstrated by Figure 2B, a very similar correlation also exists between total AP activity and the percentage ratio for heat stable non-LPSAP to total AP. From the data in Figure 2 the following has to be considered: No uniform reference range can be drawn, either for the Q value or for the percentage ratio for heat-stable non-LPSAP to total AP, as these reference ranges are dependent on the level of total AP activity. Thus, to interpret data achieved by the methods used in this study, the interdependency of various parameters has to be taken into account. Discussion In the present study the reference range for semiquantitative separation of API, achieved by a combination of chemical-inhibition and heat-inactivation, was determined in children aged 2 to 13 years. This age group can be presumed to be
I 1000
Serum Alkaline Phosphatase Isoenzymesin Childhood
57
homogenous with regard to total serum AP activity, whereas children below 2 years and above 13 years of age usually present higher total serum AP activities, due to enhancement of osseous API in early life and during puberty (Clerk and Beck, 1950; Uhsemann, 1972; Kattwinkel et al., 1973; Round, 1973; Hosenfeld and Paulsen, 1976). We found that this age group is also homogenous with regard to the different isoenzyme fractions, independent of age and sex. Definitive quantitative separation of API has not yet been achieved, either by diverse electrophoretic methods or by use of chemical-inhibition and heatinactivation (Kaplan, 1972). Recently, Siede and Seiffert (1977) published a new electrophoretic technique for quantitative API determination but clear-cut separation of hepatobiliary and osseous isoenzymes was impossible, even with this improved method. Using the comparatively simple and fast method described in this study only a semiquantitative separation of API can be obtained with regard to osseous, hepatobiliary, and intestinal isoenzymes because of considerable overlap in chemical and heat-resistance between these isoenzymes. Studies on tissue-enzymes have shown, that 5--10 mM L-phenylalanine inhibits not only the intestinal isoenzyme activity, but also 8--32% of the osseous and liver isoenzyme activities. Heat incubation for 15min at 56~ results in a 90--100% inactivation of the bone AP, but in addition, 50--70% of the hepatobiliary AP and 50--60% of the intestinal AP are inactivated (Solbach, 1969; Kattwinkel et al., 1973; Siede and Seifert, 1977). Three fractions of total AP activity were obtained by the methods used: i.e. Lphenylalanine sensitive AP activity (LPSAP), heat-sensitive non-LPSAP, and heat-stable non-LPSAP. LPSAP, which mainly but not exclusively reflects intestinal AP, accounts for approximately 12% of total AP. This is consistent with data of Cao et al. (1972) in children 2--12 years of age, but exceeds the 1.7% found by Statland et al. (1972), using similar techniques. Though estimation of the intestinal API apparently fails to have any diagnostic implication in infancy and childhood, it has to be "filtered out" by L-phenylalanine, owing to the comparable heat stability a considerable overlap has to be considered with the relative heat-stable hepatobiliary isoenzyme fraction. The heat sensitive non-LPSAP mainly reflects skeletal API. Averaging 77% of the total AP activity, it represents the predominant fraction of total AP. Contrary to Ohlen et al. (1971), heat incubation in our study was limited to 10 rather than 15 rain for two reasons: (I) According to Cadeau and Malkin (1973) a more sharply defined differentiation between normal and pathological values can be achieved, and (2) in children, residual heat-stable non-LPSAP activity after 15 rain of incubation falls within the lower limit of sensitivity of the assay used for routine determination of AP. Our data on heat-sensitive non-LPSAP are consistent with data in the literature obtained by use of a similar method (Cao et al., 1972), by urea inhibition (Statland et al., 1972) or by electrophoresis (Round, 1973; Hosenfeld and Paulsen, 1976). These data provided evidence that in children the main part of total serum AP is related to osseous API. Consequently, osseous API is responsible for the well-known enhanced normal values of total serum AP in children, as compared to adults, which are due to increased activity of osteoblasts in the growing organism (Kattwinkel, 1973; Round, 1973).
58
K. Kruse et al.
In our study heat-stable non-LPSAP, reflecting mainly hepatobiliary API, accounts for not more than 9% of the total serum AP. This is in accordance with the data of Statland et al. (1972) and Round (1973) on children aged 4--12 and 7--17 years, respectively, using urea inhibition a n d electrophoresis for API separation. In children, API separation is generally performed in order to determine the relationship between osseous and hepatobiliary API. After Ohlen et al. (1971), estimation of the Q values, i.e., the heat-stable non-LPSAP/non-LPSAP percentage ratio, is more suitable for differentiation of skeletal and hepatobiliary API than are the absolute activities of the respective fractions, evaluating deviations of the isoenzyme pattern independently of total AP activity. Though in our experiments heat incubation was limited to 10rain, the Q value averaging 10% ( + 2 SD: 4.8--20.9%) falls below that of 17.7% (+2 SD: 12.1--23.3%) determined by Ohlen et al. (1971) in adults. The lower Q value in children provides additional evidence for the predominance of osseous API in children. In adults the Q value is independent of total serum AP activity (Ohlen et al., 1971). In contrast, the Q value in children has a significant negative correlation to total AP activity. This difference between children and adults is possibly due to the fact that in normal children, in contrast to adults, variations in total AP activity are mainly i f not exclusively due to variations in the osseous API activity. The significant interrelation between the Q value and total AP activity has to be taken into account for the following reason: Comparing the data in Table 1 and Figure 2A it is evident that a Q value and the accompanying total AP may fall outside the physiologic 95% area, even though both individual values are within the normal range. On the other hand, apparently slightly enhanced or diminished single values in combination may be located within the tolerance ellipse. Since in children we found a very high correlation between total AP and non-LPSAP (r=0.967) the heat-stable non-LPSAP/total AP percentage ratio was also correlated to total serum AP activity. As demonstrated in Figure 2B, the correlation is very close to that in Figure 2A, and provides an alternative for differentiating osseous from hepatobiliary API. This has the advantage of limiting the necessary procedure to two measurements of AP activity: (1) Determination of total AP, (2) Determination of heat-stable non-LPSAP in the presence of 5 mM L-phenylalanine after incubation of the serum for 10 min at 56 ~C. Thus, screening for deviations in the API pattern with regard to the relationship between osseous and hepatobiliary API can easily be done in a routine clinical laboratory if the physiologic interrelations between these parameters are taken into account. Data on patients presenting abnormal API patterns will be published in an additional paper.
Acknowledgement. We wish to thank Dr. K. Moldenhauer and V. Malerczyk, Department of Pediatric Cardiology and Bioengineering, Kiel, who kindly performed the statistical analysis. References
Cadeau, B. J., Malkin, A.: A relative heat stability test for the identification of serum alkaline phosphatase isoenzymes. Clin. Chim. Acta 45, 235--242 (1973)
Serum Alkaline Phosphatase Isoenzymes in Childhood
59
Cao, A., Coppa, G., Trabalza, N., Marcucci, F., de Virgilis, S., Lungarotti, A., Furbetta, A.: Characterization of serum alkaline phosphatase in infancy and childhood. Z. Kinderheilk. 113, 289--296 (1972) Clark, L. C., Jr., Beck, E.: Plasma "alkaline" phosphate activity. I. Normative data for growing children. J. Pediat. 36, 335--341 (1950) Conapa-Anson, R., Rowe, D. J. F.: Electroph0retic separation of tissue-specific serum alkaline phosphatases. J. clin. Path. 23, 499--508 (1970) Dominick, H. C., van Husen, N., HOsemann, R., Gerlach, U.: Isoenzymes of alkaline phosphatase in the serum of patients with cystic fibrosis. Z. Kinderheilk. 119, 261--267 (1975) Fishman, W. H., Gosh, N. K.: Isoenzymes of human alkaline phosphatase. Adv. clin. Chem. 10, 254--370 (1967) Gerlach, U., Hiby, W., Paul, L.: Uber den diagnostischen Wert der Bestimmungyon Isoenzymen der alkalischen Phosphatase im Serum. Med. Welt 21, 1252--1258 (1970) Hausamen, T.-U., Helger, R., Rick, W., Gross, W.: Optimal conditions for the determination of serum alkaline phosphatase by a new kinetic method. Clin. Chim. Acta 15, 241--245 (1967) Hosenfeld, D., Paulsen, H.: Die Isoenzyme der aklalischen Serumphosphatase im Neugeborenen- und Kleinkindesalter. Klin. Pgdiat. 188, 55--61 (1976) Kaplan, M. M.: Alkaline phosphatase. Gastroenterology 62, 452--468 (1972) Kattwinkel, J., Taussig, L. M., Statland, B. E., Verter, K. I.: The effects of age on alkaline phosphatase and other serological liver function tests in normal subjects and patients with cystic fibrosis. J. Pediat. 82, 234--242 (1973) Ohlen, J., Pause, H., Richter, J.: Alkaline phosphatase: Diagnostic value of isoenzyme determination. Europ. J. Clin. Invest. 1, 445--451 (1971) Posen, S., Neale, F. C., Clubb, J. S.: Heat inactivation in the study of human alkaline phosphatase. Ann. intern. Med. 62, 1234--1243 (1965) Recommendations of the German Society for Clinical Chemistry: Z. Klin. Chem. Klin. Biochem. 10, 659--660 (1970) Richter, J., Ohlen, J.: Vergleichende Untersuchungen zur katalytischen Aktivit~it von Isoenzymen der alkalischen Phosphatase unter ,,konventionellen" und ,,optimierten" Testbedingungen. Z. Kiln. Chem. Klin. Biochem. 12, 432--436 (1974) Round, J. M.: Plasma calcium, magnesium, phosphorus and alkaline phosphatase levels in normal british schoolchildren. Brit. reed. J. 1973 III, 137--140 Siede, H., Seiffert, U. B.: Quantitative alkaline phosphatase isoenzyme determination by electrophoresis on cellulose acetate membranes. Clin. Chem. 23, 28--34 (1977) Statland, B. E., Nishi, H. H., Young, D. S.: Serum alkaline phosphatase: total activity and isoenzyme determinations made by use of the centrifugal fast analyzer. Clin. Chem. 18, 1468--1474 (1972) Stolbach, L. L.: Clinical application of alkaline phosphatase isoenzyme analysis. Ann. N.Y. Acad. Sci. 166, 760--774 (1969) Uhsemann, T.: Untersuchungen tiber die Altersabhgngigkeit des Normalbereiches von Calcium, anorganischem Phosphat und alkalischer Phosphatase im menscblichenSerum. Thesis 1972, Kiel
Received March 31, 1977
F.urol:~andoumalof p~d~l~tlt~eS
9 by Springer-Verlag 1977
Serum Alkaline Phosphatase Isoenzymes in Childhood* K. Kruse 1**, H. Bartels ~, and H. Gfinther 2.** 1University of Kiel Medical School, Department of Pediatrics (Prof. Dr. H.-R. Wiedemann), Fr6belstr. 15/17, D-2300 Kiel, Federal Republic of Germany 2Department of Pediatrics (Prof. Dr. H. G. Hansen), Medical School, D-2400 Ltiheck, Federal Republic of Germany
Abstract. Using a combination of L-phenylalanine inhibition and heat inactivation, the serum alkaline phosphatase (AP) in 2 to 13 year old children without evidence of hepatobiliary, osseous, or intestinal disease was separated in three fractions; i.e. L-phenylalanine sensitive AP (LPSAP), heat-stable non-L-phenylalanine sensitive AP (heat-stable non-LPSAP) and heat-sensitive non-L-phenylalanine sensitive AP (heat-sensitive non-LPSAP). The activities of total AP and the different fractions were measured using optimized test conditions. LPSAP, (mainly intestinal AP), accounts for approximately 12% of the total serum AP activity, heat-stable non-LPSAP (mainly hepatobiliary AP) for approximately 9%, and heat-sensitive nonLPSAP (mainly bone AP) for approximately 77%. To give a better differentiation between bone and liver AP, the percentage ratios of heat-stable n o n - L P S A P / n o n - L P S A P (Q value), and heat-stable n o n - L P S A P / t o t a l AP, were determined. Both quotients showed a significant negative correlation with total AP, which has to be taken into account in the interpretation of the results of isoenzyme determinations of serum AP activity. The above semiquantitative separation of AP isoenzymes can be readily done in a routine clinical laboratory. Key words: Serum alkaline phosphatase isoenzymes - L-phenylalanine inhibition - H e a t inactivation - Normal values in children.
Studies on alkaline phosphatase isoenzymes (API) using techniques such as electrophoresis on a variety of media, treatment with inhibitors like L-phenylalanine and urea, heat-inactivation and specific antisera suggest that h u m a n serum alkaline phosphatase (AP) is derived from liver, bone, intestine and, in pregnant women, placenta (Fishman and Gosh, 1967). An enhancement of serum * Supported in part by Deutsche Forschungsgemeinschaft (Ba 246/11) ** Corresponding author *** The data form part of the thesis of H. Giinther, Medical School of Liibeck. Present address: Department of Anaesthesiology, Medical School, D-2400 Labeck
54
K. Kruse et al.
A P is usually due to an increase o f either the bone or the hepatobiliary isoenzyme (Kaplan, 1972). Since an elevation of A P is sometimes the only a b n o r m a l biochemical finding in a patient, the hepatobiliary or skeletal origin of the enzyme has to be established by isoenzyme investigation. Moreover, total serum A P activity within the n o r m a l range m a y be a c c o m p a n i e d by a clinically significant alteration o f one isoenzyme fraction (Ohlen et al., 1971; Kattwinkel et al., 1973; D o m i n i c k et al., 1975; Siede and Seifert, 1977). Using a c o m b i n a t i o n of chemicalinhibition and heat-inactivation Ohlen et al. (1971) achieved a semiquantitative separation o f A P I in adults which is sufficiently exact for clinical purposes and can be performed in routine clinical laboratories. Cao et al. (1972), employing a similar technique, have published reference values for n o r m a l infants and children. However, in the studies o f C a o et al. (1972) and Ohlen et al. (1971) A P activities were measured utilizing n o n - o p t i m a l test conditions. Since 1970 an optimized kinetic test, using the m e t h o d of H a u s a m e n et al. (1967) and recommended, for example by the G e r m a n Society for Clinical Chemistry (1970), has been in widespread use. The aim o f this study was to establish reference values in children for serum A P I differentiation according to Ohlen et al. (1971), using the r e c o m m e n d e d conditions for measuring A P activity.
Material and Methods Sera from 83 children (38 girls and 45 boys aged 2 to 13 years), who were in- and outpatients of the Department of Pediatrics, Liibeck, were investigated. None of the children suffered from a disease known to alter serum AP activity. In particular, there was no clinical or chemical evidence of hepatobiliary, osseous, or intestinal disease in any case. AP activity was measured according to the optimized method of Hausamen et al. (1967), utilizing a commercially available test kit (Monotest, AP optimized, Boehringer Mannheim). Isoenzyme Determinations (s, Fig. 1). After estimation of total serum AP activity (total AP), AP was measured in the presence of 5raM L-phenylalanine, representing non-L-phenylalanine sensitive AP (non-LPSAP). The decline in activity caused by L-phenylalanine corresponds to the L-phenylalanine sensitive fraction of total activity (LPSAP) and mainly reflects intestinal AP (Fishman and Gosh, 1967). Thereafter, the serum was heated for exactly 10min at 56.0+0.1~ and then cooled immediately in an icebath. AP activity measured after heatinactivation in the presence of 5 mM L-phenylalanine, representing heat-stable non-LPSAP, is mainly due to the hepatobiliary isoenzyme, whereas the activity inactivated by heat (heatsensitive non-LPSAP) is mainly related to the skeletal isoenzyme (Posen et al., 1965).
[ total AP incLctivated
5 mM L-pheny|alanine
1.
I
(mainly intestine AP
J non-LPSAP~ I inactivated [ 10 rnin 56~
I
[ heatsensitiVenon -LPSAP I
]heatnon_LPsApStable I
(mainly bon.__..eAP} e
(mainlyliver AP}
Fig. 1. Flow diagram of semiquantitative isoenzyme determination of serum AP
Serum Alkaline Phosphatase Isoenzymes in Childhood
55
According to Ohlen et al. (1971), the percentage ratio of heat-stable non-LPSAP/nonLPSAP, represented as the Q value, is a reliable parameter for the hepatobiliary/hone isoenzyme ratio, enhancement of the Q value above the upper normal limit indicating an increase in the hepatobiliary isoenzyme and a reduction below normal suggesting an increase in the osseous isoenzyme. Statistical Analysis. The mode of distribution was ascertained graphically. Mean values, variances, regression equations, correlation coefficients and tolerance ellipses were calculated by standard methods using a DEC PDP 11/40 computer unit.
Results The values for total A P , L P S A P , n o n - L P S A P , heat-sensitive a n d heat-stable n o n L P S A P , a n d ' Q ' followed a p p r o x i m a t e l y l o g - n o r m a l distributions. All the p a r a m e t e r s studied were i n d e p e n d e n t of age a n d sex. I n Table 1 m e a n values a n d reference ranges ( ~ + 2 SD) are given for total serum A P activity a n d for the A P fractions separated by L - p h e n y l a l a n i n e i n h i b i t i o n a n d heat-inactivation. L P S A P a c c o u n t e d for a b o u t 12% of the total A P activity, heat-stable n o n - L P S A P for a b o u t 9% a n d heat-sensitive n o n - L P S A P for a b o u t 77%. A b s o l u t e activities ( U / l ) in the different fractions showed a positive correlation with the level of total A P
Table 1. Serum alkaline phosphatase (U/l) in 83 children aged 2--13 years before and after inhibition with 5 mM L-phenylalanine and heat inactivation (10min at 56~
Total AP~ LPSAP % of total AP Non-LPSAP Heat-sensitive non-LPSAP % of total AP Heat-stable non-LPSAP % of total AP Q value a
334 38.9 11.5 288 257 77.3 28.8 8.7 10.0
~-2s
~+2s
172 10.2 3.3 146 125 63.1 13.8 4.2 4.8
647 148 39.8 566 526 91.6 60.3 18.2 20.9
Abbreviations:
Total alkaline phosphatase activity in native serum LPSAP: AP activity inhibited by 5 mM L-phenylalanine = L-phenylalanine sensitive AP Non-LPSAP: AP activity in the presence of 5 mM L-phenylalanine Heat-sensitive non-LPSAP: Non-LPSAP activity inactivated by heat (10rain at 56~ Heat-stable non-LPSAP: Residual non-LPSAP activity after heat inactivation Q value: Percentage ratio of heat-stable non-LPSAP/non-LPSAP Total AP:
56
K. Kruse et al.
30[
__
2O
~~ '
B
30
A
9
r = 0.435
~"
r = 0./,39
i O.
~
P
20
9
10
-r
O
s
s tY
9~
o
3
I
100
I
I
200 300 500 TOTAL AP (Ult)
I
1000
3
I
100
I
I
I
20O 3O0 50O TOTAL AP ( U / t )
Fig. 2. Relationship of (A) the Q value and (B) the percentage ratio heat-stable non-LPSAP/ total AP to total serum AP activity in 83 children aged 2--13 years. Individual values, regression lines and tolerance ellipses (95% range) activity. LPSAP and non-LPSAP correlated significantly with total AP (r = 0.512 and 0.967, respectively). Heat-stable and heat-sensitive non-LPSAP were significantly correlated to non-LPSAP (r=0.489 and 0.992, respectively). In contrast, the percentage ratio of LPSAP/total AP activity remained constant, independent of the level of total AP activity. Though there is a tendency for the heat-sensitive non-LPSAP/total AP ratio to increase with increasing total AP activity, no significant correlation could be established due to the marked variation of individual values. However, a significant negative correlation exists between the level of total AP activity and the Q value (percentage ratio for heatstable non-LPSAP to non-LPSAP). This is shown in Figure 2A, where individual values, regression lines and the tolerance ellipse enclosing the 95% area of tolerance are presented on a logarithmic scale, and indicate a decrease of the Q value with increasing total AP activity. As demonstrated by Figure 2B, a very similar correlation also exists between total AP activity and the percentage ratio for heat stable non-LPSAP to total AP. From the data in Figure 2 the following has to be considered: No uniform reference range can be drawn, either for the Q value or for the percentage ratio for heat-stable non-LPSAP to total AP, as these reference ranges are dependent on the level of total AP activity. Thus, to interpret data achieved by the methods used in this study, the interdependency of various parameters has to be taken into account. Discussion In the present study the reference range for semiquantitative separation of API, achieved by a combination of chemical-inhibition and heat-inactivation, was determined in children aged 2 to 13 years. This age group can be presumed to be
I 1000
Serum Alkaline Phosphatase Isoenzymesin Childhood
57
homogenous with regard to total serum AP activity, whereas children below 2 years and above 13 years of age usually present higher total serum AP activities, due to enhancement of osseous API in early life and during puberty (Clerk and Beck, 1950; Uhsemann, 1972; Kattwinkel et al., 1973; Round, 1973; Hosenfeld and Paulsen, 1976). We found that this age group is also homogenous with regard to the different isoenzyme fractions, independent of age and sex. Definitive quantitative separation of API has not yet been achieved, either by diverse electrophoretic methods or by use of chemical-inhibition and heatinactivation (Kaplan, 1972). Recently, Siede and Seiffert (1977) published a new electrophoretic technique for quantitative API determination but clear-cut separation of hepatobiliary and osseous isoenzymes was impossible, even with this improved method. Using the comparatively simple and fast method described in this study only a semiquantitative separation of API can be obtained with regard to osseous, hepatobiliary, and intestinal isoenzymes because of considerable overlap in chemical and heat-resistance between these isoenzymes. Studies on tissue-enzymes have shown, that 5--10 mM L-phenylalanine inhibits not only the intestinal isoenzyme activity, but also 8--32% of the osseous and liver isoenzyme activities. Heat incubation for 15min at 56~ results in a 90--100% inactivation of the bone AP, but in addition, 50--70% of the hepatobiliary AP and 50--60% of the intestinal AP are inactivated (Solbach, 1969; Kattwinkel et al., 1973; Siede and Seifert, 1977). Three fractions of total AP activity were obtained by the methods used: i.e. Lphenylalanine sensitive AP activity (LPSAP), heat-sensitive non-LPSAP, and heat-stable non-LPSAP. LPSAP, which mainly but not exclusively reflects intestinal AP, accounts for approximately 12% of total AP. This is consistent with data of Cao et al. (1972) in children 2--12 years of age, but exceeds the 1.7% found by Statland et al. (1972), using similar techniques. Though estimation of the intestinal API apparently fails to have any diagnostic implication in infancy and childhood, it has to be "filtered out" by L-phenylalanine, owing to the comparable heat stability a considerable overlap has to be considered with the relative heat-stable hepatobiliary isoenzyme fraction. The heat sensitive non-LPSAP mainly reflects skeletal API. Averaging 77% of the total AP activity, it represents the predominant fraction of total AP. Contrary to Ohlen et al. (1971), heat incubation in our study was limited to 10 rather than 15 rain for two reasons: (I) According to Cadeau and Malkin (1973) a more sharply defined differentiation between normal and pathological values can be achieved, and (2) in children, residual heat-stable non-LPSAP activity after 15 rain of incubation falls within the lower limit of sensitivity of the assay used for routine determination of AP. Our data on heat-sensitive non-LPSAP are consistent with data in the literature obtained by use of a similar method (Cao et al., 1972), by urea inhibition (Statland et al., 1972) or by electrophoresis (Round, 1973; Hosenfeld and Paulsen, 1976). These data provided evidence that in children the main part of total serum AP is related to osseous API. Consequently, osseous API is responsible for the well-known enhanced normal values of total serum AP in children, as compared to adults, which are due to increased activity of osteoblasts in the growing organism (Kattwinkel, 1973; Round, 1973).
58
K. Kruse et al.
In our study heat-stable non-LPSAP, reflecting mainly hepatobiliary API, accounts for not more than 9% of the total serum AP. This is in accordance with the data of Statland et al. (1972) and Round (1973) on children aged 4--12 and 7--17 years, respectively, using urea inhibition a n d electrophoresis for API separation. In children, API separation is generally performed in order to determine the relationship between osseous and hepatobiliary API. After Ohlen et al. (1971), estimation of the Q values, i.e., the heat-stable non-LPSAP/non-LPSAP percentage ratio, is more suitable for differentiation of skeletal and hepatobiliary API than are the absolute activities of the respective fractions, evaluating deviations of the isoenzyme pattern independently of total AP activity. Though in our experiments heat incubation was limited to 10rain, the Q value averaging 10% ( + 2 SD: 4.8--20.9%) falls below that of 17.7% (+2 SD: 12.1--23.3%) determined by Ohlen et al. (1971) in adults. The lower Q value in children provides additional evidence for the predominance of osseous API in children. In adults the Q value is independent of total serum AP activity (Ohlen et al., 1971). In contrast, the Q value in children has a significant negative correlation to total AP activity. This difference between children and adults is possibly due to the fact that in normal children, in contrast to adults, variations in total AP activity are mainly i f not exclusively due to variations in the osseous API activity. The significant interrelation between the Q value and total AP activity has to be taken into account for the following reason: Comparing the data in Table 1 and Figure 2A it is evident that a Q value and the accompanying total AP may fall outside the physiologic 95% area, even though both individual values are within the normal range. On the other hand, apparently slightly enhanced or diminished single values in combination may be located within the tolerance ellipse. Since in children we found a very high correlation between total AP and non-LPSAP (r=0.967) the heat-stable non-LPSAP/total AP percentage ratio was also correlated to total serum AP activity. As demonstrated in Figure 2B, the correlation is very close to that in Figure 2A, and provides an alternative for differentiating osseous from hepatobiliary API. This has the advantage of limiting the necessary procedure to two measurements of AP activity: (1) Determination of total AP, (2) Determination of heat-stable non-LPSAP in the presence of 5 mM L-phenylalanine after incubation of the serum for 10 min at 56 ~C. Thus, screening for deviations in the API pattern with regard to the relationship between osseous and hepatobiliary API can easily be done in a routine clinical laboratory if the physiologic interrelations between these parameters are taken into account. Data on patients presenting abnormal API patterns will be published in an additional paper.
Acknowledgement. We wish to thank Dr. K. Moldenhauer and V. Malerczyk, Department of Pediatric Cardiology and Bioengineering, Kiel, who kindly performed the statistical analysis. References
Cadeau, B. J., Malkin, A.: A relative heat stability test for the identification of serum alkaline phosphatase isoenzymes. Clin. Chim. Acta 45, 235--242 (1973)
Serum Alkaline Phosphatase Isoenzymes in Childhood
59
Cao, A., Coppa, G., Trabalza, N., Marcucci, F., de Virgilis, S., Lungarotti, A., Furbetta, A.: Characterization of serum alkaline phosphatase in infancy and childhood. Z. Kinderheilk. 113, 289--296 (1972) Clark, L. C., Jr., Beck, E.: Plasma "alkaline" phosphate activity. I. Normative data for growing children. J. Pediat. 36, 335--341 (1950) Conapa-Anson, R., Rowe, D. J. F.: Electroph0retic separation of tissue-specific serum alkaline phosphatases. J. clin. Path. 23, 499--508 (1970) Dominick, H. C., van Husen, N., HOsemann, R., Gerlach, U.: Isoenzymes of alkaline phosphatase in the serum of patients with cystic fibrosis. Z. Kinderheilk. 119, 261--267 (1975) Fishman, W. H., Gosh, N. K.: Isoenzymes of human alkaline phosphatase. Adv. clin. Chem. 10, 254--370 (1967) Gerlach, U., Hiby, W., Paul, L.: Uber den diagnostischen Wert der Bestimmungyon Isoenzymen der alkalischen Phosphatase im Serum. Med. Welt 21, 1252--1258 (1970) Hausamen, T.-U., Helger, R., Rick, W., Gross, W.: Optimal conditions for the determination of serum alkaline phosphatase by a new kinetic method. Clin. Chim. Acta 15, 241--245 (1967) Hosenfeld, D., Paulsen, H.: Die Isoenzyme der aklalischen Serumphosphatase im Neugeborenen- und Kleinkindesalter. Klin. Pgdiat. 188, 55--61 (1976) Kaplan, M. M.: Alkaline phosphatase. Gastroenterology 62, 452--468 (1972) Kattwinkel, J., Taussig, L. M., Statland, B. E., Verter, K. I.: The effects of age on alkaline phosphatase and other serological liver function tests in normal subjects and patients with cystic fibrosis. J. Pediat. 82, 234--242 (1973) Ohlen, J., Pause, H., Richter, J.: Alkaline phosphatase: Diagnostic value of isoenzyme determination. Europ. J. Clin. Invest. 1, 445--451 (1971) Posen, S., Neale, F. C., Clubb, J. S.: Heat inactivation in the study of human alkaline phosphatase. Ann. intern. Med. 62, 1234--1243 (1965) Recommendations of the German Society for Clinical Chemistry: Z. Klin. Chem. Klin. Biochem. 10, 659--660 (1970) Richter, J., Ohlen, J.: Vergleichende Untersuchungen zur katalytischen Aktivit~it von Isoenzymen der alkalischen Phosphatase unter ,,konventionellen" und ,,optimierten" Testbedingungen. Z. Kiln. Chem. Klin. Biochem. 12, 432--436 (1974) Round, J. M.: Plasma calcium, magnesium, phosphorus and alkaline phosphatase levels in normal british schoolchildren. Brit. reed. J. 1973 III, 137--140 Siede, H., Seiffert, U. B.: Quantitative alkaline phosphatase isoenzyme determination by electrophoresis on cellulose acetate membranes. Clin. Chem. 23, 28--34 (1977) Statland, B. E., Nishi, H. H., Young, D. S.: Serum alkaline phosphatase: total activity and isoenzyme determinations made by use of the centrifugal fast analyzer. Clin. Chem. 18, 1468--1474 (1972) Stolbach, L. L.: Clinical application of alkaline phosphatase isoenzyme analysis. Ann. N.Y. Acad. Sci. 166, 760--774 (1969) Uhsemann, T.: Untersuchungen tiber die Altersabhgngigkeit des Normalbereiches von Calcium, anorganischem Phosphat und alkalischer Phosphatase im menscblichenSerum. Thesis 1972, Kiel
Received March 31, 1977
