A POPUlATION swam . * HUMAN BIOCHEMICAL GENETICS: * ‘ THE RELATIONSHIPS 0535me : METABOLm-zs m EPILEPSY AND" ; . ; ;, * MENTALRETARDATMN a. — “ DisSertation for the Degreeot Ph; D. ' ‘ : mcmm STATE umvaasm ‘ macaw mum ' ‘1974 . __.& , 800K BINDERY mc. xiitmmi’fii‘f‘. ' \ ABSTRACT A POPULATION STUDY IN HUMAN BIOCHEMICAL GENETICS: THE RELATIONSHIPS OF SEVERAL METABOLITES IN EPILEPSY AND MENTAL RETARDATION BY Habibollah Fakhrai Some mentally retarded individuals are presumed to be suffering from physiological abnormalities induced by their genetics, environment or both. If a number of these individuals have some particular condition in common, they might all have an elevated or a diminished level of a particular metabolite, leading to the discovery of a new syndrome. Over 1,700 mentally retarded individuals with and without epilepsy and'a normal control group of 230 individ- uals were tested for several metabolites: glucose, blood urea nitrogen, serum glutamate oxalacetate transaminase, uric acid, calcium, inorganic phosphate, total protein, albumin, and alkaline phosphatase. Means, standard devia- tions, and the number of high and low outliers have been determined for the control and for the retarded and epileptic populations. Habibollah Fakhrai Retarded sib pairs in an institution were also tested for the above metabolites in 1971. In 1973 the sib pairs and some individuals with idiOpathic epilepsy in the institution were also tested for these serum metabolites. In addition, these individuals were tested for serum ammonia, intestinal and liver alkaline phosphatases, ABC and Lewis blood groups and their secretor status. Alkaline phosphatase was found to be high and blood urea nitrogen was found to be low in the epileptics when compared with either the other mentally retarded or the normal pOpulation. The level of alkaline phosphatase was higher in all retarded populations than in the normal popu- lation. The evidence suggests that many mentally retarded individuals classified as epileptics may actually be suffer- ing from liver damage which has induced ammonia intoxication of the brain. This finding indicates that a more thorough evaluation of epileptics may be necessary. A POPULATION STUDY IN HUMAN BIOCHEMICAL GENETICS: THE RELATIONSHIPS OF SEVERAL METABOLITES IN EPILEPSY AND MENTAL RETARDATION By Habibollah Fakhrai A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Zoology 1974 F§)Copyright by HABIBOLLAH FAKHRAI 1 974 To: my mother, Gohar Ghotb my father, Mr. Mahmud Fakhrai my wife, Farideh my professor, Dr. Herman M. Slatis and to Tonya T. who is very special to me. iii ACKNOWLEDGMENTS I would sincerely like to thank my professors, Dr. Herman M. Slatis and Dr. James V. Higgins, for advising me during this experiment and for their many helpful suggestions and criticisms. Many thanks are due to Dr. Emanuel Hackel for con- structive criticism and providing some of the needed materials to carry out this experiment, and also to Dr. John Boezi for serving in my committee. I would like to thank my wife, Farideh, for her enormous help and patience throughout the course of my study. I would also like to sincerely thank my very good friend, Vahid Movahed, for drafting the figures in this work. Thanks are also due to the Society of Sigma Xi for a partial grant toward the fulfillment of this study. iv TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . . . . . . . . . . vii LIST OF FIGURES . . . . . . . . . . . . . . . . . . x INTRODUCTION . . . . . . . . . . . . . . . . . 1 LITERATURE REVIEW . 2 Alkaline phosphatase . . . . . . 2 Function of alkaline phosphatase 2 Activation of alkaline phosphatase . 4 Electrophoresis of alkaline phosphatase 4 Inheritance of alkaline phosphatase . . . 5 Relationship of alkaline phosphatase and other blood markers . . . . . . . . . . . 7 Alkaline phosphatase and diet . . . . 11 The relationship of alkaline phosphatase with sex and age . . . . . . . . 12 Sources and levels of alkaline phosphatase . . . 14 The inhibition of alkaline phosphatase . . . . . 15 Storage of alkaline phosphatase . . . . . . . . . 19 Alkaline phos hatase polymerization . . . . 19 The relationsfiip of alkaline phosphatase and different diseases . . . . 21 Alkaline phosphatase, Km and the effect of pH . . 24 Regan isoenzyme . . . . . . . . . . 25 Epilepsy, genetics and enzymes . . . . . 26 Causes of abnormalities in blood metabolites . . 28 Effects of high blood ammonia . . . . . . . . . . 29 Blood urea nitrogen . . . . . . . . . . . . . . . 31 MATERIALS AND METHODS . . . . . . . . . . . . . . . 32 Glucose . . . . . . . . . . . . . . . . . . . . . 33 BUN . . . . . . . . . . . . . . . . . . . . . . . 33 SGOT . . . . . . . . . . . . . . . . . . . . . . 34 Uric acid . . . . . . . . . . . . . . . . . . . . 34 Calcium . . . . . . . . . . . . . . . . . 3S Inorganic phosphate . . . . . . . . . . . . . . . 36 Total protein . . . . . . . . . . . . . . . . . . 36 Albumin . . . . . . . . . . . . . . . . . . . . . 37 Page Alkaline phosphatase . . . . . . . . . . . . . . 37 Ammonia . . . . . . . . . . . . . . . . . . . 40 Reagents . . . . . . . . . . . . . . . . . . . . 40 Blood typing . . . . . . . . . . . . . . . . . . 41 Secretor status . . . . . . . . . . . . . . . . 42 Substrate specificity . . . . . . . . . . . . . 43 RESULTS . . . . . . . . . . . . . . . . . . . . . . 44 Glucose . . . . . . . . . . . . . . . . . . . . S8 BUN . . . . . . . . . . . . . . . . . . . . . . 61 SGOT . . . . . . . . . . . . . . . . . . . . . . 62 Uric acid . . . . . . . . . . . . . . 64 Alkaline phosphatase . . . . . . . . . . . . . . 65 Sib studies . . . . . . . . . . . . . 66 Epileptic and retarded studies in 1973 . . . . . 74 DISCUSSION . . . . . . . . . . . . . . . . . . . . 102 BUN . . . . . . . . . . . . . . . . 104 Alkaline phosphatase . . . . . . . . . . . . . . 106 SUMMARY . . . . . . . . . . . . . . . . . . . . . . 114 LITERATURE CITED . . . . . . . . . . . . . . . . . 116 vi LIST OF TABLES Table Page 1. The means and standard deviation for the control group and the normal ranges . . . . . 4S 2. Analysis of the test for serum glucose . . . 47 3. The analysis of the test for blood urea nitrogen (BUN) O O O O O O O O O O O O 0 O O 48 4. The analysis of the test for SGOT . . . . . . 49 5. The analysis of the test for uric acid . . . 50 6. The analysis of the test for serum calcium . 51 7. The analysis of the test for serum phos- phate O O O O O O I O I O I O O O O O O 0 O O 52 8. The analysis of the test for serum total protein I I O O O O O O O O O O O O O O O O O 53 9. The analysis of the test for serum albumin . 54 10. The analysis of the test for serum alkaline phosphatase . . . . . . . . . . . . . . . . . SS 11. The distribution of high outliers for the different tests performed . . . . . . . . . . S6 12. The distribution of low outliers for the different tests performed . . . . . . . . . . 57 13. The chi-square test for high and low out- liers in the different populations . . . . . 59 14. Cumulative percentage of individuals for different levels of AP in IAZ and IBl populations . . . . . . . . . . . . . . . . . 67 vii Table Page 15. Analyses of intra- and inter-family differ- ences of institutionalized sib pairs observed in 1971 and 1973 . . . . . . . . . . 72 16. The correlation between age and the various metabolites tested in the IAE and IAS popu- lations . . . . . . . . . . . . . . . . . . . 75 17. The analysis of the test for serum ammonia . 77 18. The analysis of the test for serum glucose . 78 19. The analysis of the test for serum protein . 79 20. The analysis of the test for serum albumin . 80 21. The analysis of the test for serum calcium . 81 22. The analysis of the test for serum phos- phate . . . . . . . . . . . . . . . . . . . . 82 23. The analysis of the test for serum SGOT . . . 83 24. The analysis of the test for the age of the individuals . . . . . . . . . . . . . . . . . 84 25. The analysis of the test for blood urea nitrogen . . . . . . . . . . . . . . . . . . 86 26. The analysis of the test for serum uric acid . . . . . . . . . . . . . . . . . . . . 88 27. The analysis of the test for serum alkaline phosphatase . . . . . . . . . . . . . . . . . 89 28. The analysis of the test for serum intes- tinal alkaline phosphatase . . . . . . . . . 92 29. The analysis of the test for serum liver alkaline phosphatase . . . . . . . . . . . . 93 30. The analysis of the test Vmax v for serum alkaline phOSphatase . . . ./.KT . . . . . . 94 31. The distribution of persons with high blood ammonia/epilepsy in groups with high AP or total protein . . . . . . . . . . . . . . . 96 32. The distribution of ABO, and Lewis blood groups and secretor status in the IAE and IAS populations . . . . . . . . . . . . . . . 97 viii Table 33. 34. 35. Page The chi-square analysis of ABO group dis- tributions in the IAE and IAS populations . . 98 The comparison of the distribution of the B and O secretor, A and AB secretors and the non-secretors of all blood phenotypes in the IAE and IAS populations . . . . . . . 100 Mean API in different blood groups and secretor status in IAE and IAS populations . 101 ix LIST OF FIGURES Figure Adaptation of the method of Morgenstern at al. from measurement of alkaline phos- phatase to both channels of AutoAnalyzer . Cumulative curve of AP values in IA2 and IBl populations . . . . . . . . . . . Distribution of ammonia in the IAE and IAS populations . . . . . . . . . . . . The distribution of alkaline phosphatase in different age groups of the IAS and IAE populations . . . . . . . . . . . . . . . Page 39 70 85 91 INTRODUCTION Some mentally retarded individuals are presumed to be suffering from physiological abnormalities induced by their genetics, environment, or both. If a number of these individuals have some particular condition in common, they might all have an elevated or a diminished level of a particular metabolite, leading to the discovery of a new syndrome. Several metabolites were tested in retarded p0pu1a- tions with or without epilepsy, including alkaline phos- phates and also several blood markers. Any abnormality related to these metabolites or the blood gene markers in any of the observed populations might manifest itself when compared with a normal population. The results of this study could be useful in better understanding of the genetics of mental retardation and epilepsy, and perhaps would be helpful in working toward their prevention or treatment. LITERATURE REVIEW Alkaline phosphatase The term alkaline phosphatase (International enzyme classification 3:1:3:l) is applied to a group of enzymes sharing the capacity to hydrolyze phosphate mono-esters in an alkaline medium [41,72,73]. The enzyme has been proven to be a glycoprotein because it releases its sialic acid when treated with crystaline neuraminidase [36]. In 1930, Kay [50] stated that he measured phosphatase of blood serum for the first time in humans by using B-glycerophosphate as the substrate. Many authors have credited him as the first man to discover alkaline phosphatase. However, in 1967 Posen [72] credited Suzuki, Yashimura, and Takaishi (Bull. Coll. Agriculture, Tokyo Imp. University 7:503, 1907) as having discovered alkaline phosphatase. Function of alkaline phosphatase Despite the many years since alkaline phosphatase was discovered, we still do not understand its function. Kay [50] observed that the enzyme will hydrolyze all the phos- phoric esters that were presented to it, namely hexosedi- phosphate, a and B-glycerophosphate, and guanine nucleotide. Still others report transphosphorylation activity for alkaline phosphates [1,60,91]. Two types of reaction, therefore, are possible. In one the enzyme acts as a hydro- lase, with cleavage of the P-O bond [50,60,91]: RO - PO H + HOH Alk. PO4 ase ROH + H P04 3 2 3 The second type of reaction, a transphosphorylation in which the enzyme acts as a phosphotransferase, does not involve the intermediate formation of inorganic phosphate in the transmission of the phosphoryl group to the acceptor [1,60, 91]: R0 - PO3 H2 + R' OH-—9’ROH e R'O PO3H2 Which of these two reactions the enzyme catalyzes depends upon the competition between water and other hydroxyl- containing.compounds for sites at the surface of the enzyme donor complex [62]. Skillen and Harrison [91] reported significant transphosphorylation at pH 9.5 as measured by the difference_in the apparent p-nitrophenyl phosphate activ- ity in bicarbonate and'methyl aminoprOpanol-HCI buffers. At pH 10.5, the degree of transphosphorylation was dependent on substrate concentration. Amador [1] reported that for a given buffer, alkaline phosphatase activation and transphos- phorylation paralleled buffer molarity. In testing 23 buffer systems, McComb and Bowers [60] observed that transphosphor- ylation was demonstrated in the two buffers in which the enzyme was most active. They also observed that for signif- icant transphosphorylation to occur the phosphate acceptor must contain a hydroxyl group, and either a second hydroxyl group or an amino group. Activation of alkaline phosphatase Kay [50] observed that Mg++ ions are a powerful stimulant to alkaline phosphatase activity. He also observed that Ca++ ions act as a mild inhibitor. Mg++ ions are known to activate many preparations of alkaline phosphatase but this effect differs with the enzyme source [28]. Moss and King [63], as well as many others, have used Mg++ ions as a stimulant in the measurement of alkaline phosphatase activity. Amador [I] observed that aminated alcohols will stimulate alkaline phosphatase. However, this activation is due to transphosphorylation of alcohol by alkaline phosphatase [1, 60]. Electrophoresis of alkaline phosphatase Of the many hundreds of intracellular enzymes which have been identified, only a handful can be detected in the blood plasma of normal subjects, and fewer still are in appreciable concentration [41]. Little is known of the process by which enzymes escape from tissue into the blood in vivo and, while the characteristics of enzymes demonstrated in tissue extracts are in general retained by the same enzymes after their passage into the serum, some degree of alteration in properties is nevertheless possible, particularly in those enzymes which are normally attached to intracellular structures [64]. The isoenzymes of alkaline phosphatase which so far have been identified in human zymograms are of liver, bone, intestinal and placental origin. To separate and identify different isoenzymes and tissue sources of alkaline phos- phatase, many different electrOphoretic techniques as well as colorimetric techniques have been employed, including cellulose acetate [32, 52, 77, 81], agarose thin film [22], polyacrylamide gel [23, 47], paper electrophoresis [5], isoelectric focusing [93], agar gel [74, 98, 114], and starch gel [13, S7, 63, 82, 109, 111]. However, regardless of which technique is ‘being used, other tests must be performed to identify the source of alkaline phosphatase. Inheritance of alkaline phosphatase When human serum alkaline phosphatase is examined by starch gel electrophoresis, it is found that some sera show only a single zone of activity, while others show in addi- tion a slower moving zone of variable intensity [53]. The site of origin of the main band which occurs in all subjects is the liver. The site of the second band is the jejunal mucosa [24]. Genetic control of organ-specific alkaline phosphatase has been observed in many organisms: Drosophila, fowl, sheep and cattle [28]. Ghane [33], in studying alkaline phosphatase isoenzymes of cattle, found no differ- ence in zymograms within 48 monozygous twin pairs, while a pronounced variation existed between pairs. He observed the presence of a particular second band, A, which seemed to be genetically controlled. Those which were of genotype FOF0 lacked band A, while those which were of genotype FAFA 0 had a dense second band, and genotypes FAF were intermediate between the FAFA and FOFO genotype. Cunningham and Rimer [21], in a p0pulation study of alkaline phosphatase, found that 60% of their population contained two alkaline phos- phatase bands, a fast moving A and a slow moving B band. Their investigation indicated that the presence of the second band may be genetically controlled. In 1963, Arfors et al. [2], in a human twin study, observed that 28 of 89 probably monozygotic twin pairs (alike with respect to sex, AlAZBO, MN, Rh (CcDEe) blood groups and Hp and gm serum groups) showed 2 AP bands (sz), in both twins, whereas in the remaining 61 monozygous pairs both twins had only one band (Ppl). Out of 111 dizygotic twin pairs, 9 were concordant for 2 bands, 63 were concordant for 1 band, and 39 were discordant. Shreffler [88], in a study of a Caucasian population classified the 2 AP isoenzymes into 5 classes: (0) no detectable phosphatase activity in the position of the slow moving band, (1) very weak or questionable band, (2) definite but weak band, (3) band of moderate intensity, and (4) with a strong slow band. In all classes band A = liver was present. In some, but not all, band B = intestine was visual- ized. Both bands A and B were variable in intensity from person to person. Shreffler found that 46% of the samples fell in class 0, having only the liver band. In 1965, Robson and Harris [80] found that the great majority of samples of placental alkaline phosphatase can be classified into one or another of six distinct phenotypes, namely Plf, P11, and P15. To identify all the possible phenotypes they had to have two electrOphoretic runs, one at pH 8.6 and another at pH 6.0. By sib and identical twin studies they also demonstrated that the placental alkaline phosphatase phenotype depends on the phenotype of the child. Relationship of alkaline phosphatase and other blood markers' In 1963, Arfors et al. [2] found that the frequency of blood group A was extremely low in individuals who had two alkaline phosphatase bands on their zymogram (sz), and that there is a non-random relationship between the alkaline phosphatase isoenzymes and Lewis blood groups. The indi- b+ glood group, and not a single individual with 2 bands was found in the Lea+b' group. viduals who had 2 bands all had Le Beckman [7] observed that the frequency of sz was 30.4% in individuals of blood group O, 30.5% in group B individuals and 2.3% in group A individuals. There was no significant difference between A1 and AZ individuals, and AB individuals were intermediate (15.2%) between groups A and B. Beckman confirmed that there was not a single case aIb' blood group individuals. However. of sz among the Le he observed that the appearance of 2 alkaline phosphatase bands in the zymogram is related to the ABH secretor status of the individual and not to the Lewis group per se. Out of 271 sz individuals observed in Beckman's study, not a single one was an ABH non-secretor. These findings have been confirmed by later studies [6, 24, 88]. In 1964, Rendel and Stormont [76] found a strong association between alkaline phosphatase isoenzymes and blood group O in sheep. In 1966, Langman et al. [53] observed that among ABH secretors, individuals of blood groups B and 0 showed the slow moving alkaline phosphatase band much more frequently than group A individuals,wd111group AB being intermediate. They classified the genotype of the individuals according to the following scheme: P++, if the slow moving band was +0 relatively intense, P , if the band was weak but definitely present, and P00, if the band was present in trace amounts or not present. In 1967, Robinson et al. [79] observed a conspicuously 2 higher frequency of Pp in three Indian tribes, the Montag- nais, Naskapi and Sapelo, than in other populations. They attributed this to the high frequency of the blood group O and secretor genes in Indian populations. This is in accord with the proposal by Langman et al. [53] that the occurrence of the slow moving alkaline phosphatase band appears to be determined in part by the secretor locus and the ABO locus, which are separate and unlinked. Palsson et al. [69], in a study of an Irish popula- tion consisting of 295 unrelated males age 20-30, found that 72.7% of 0 individuals were sz 2 , while the frequency of Pp was 45.3% in A individuals and 67.7% in blood group B. In the same year, Walter et al. [108] found that the frequency of sz in group O was 39.5%, in B, 39%, in A, 11.7%, and in AB, 21.5% in a population of 1,145 persons in Hungary. In 1971, Walker et al. [107] studied 6 O secretors, 6 A non-secretors and 6 O non-secretors. All the O secretors showed two alkaline phosphatase bands on their zymograms, and the rest showed only one band. Fritsche and Adams-Park [32] and Sundblad et al. [98] also reported that they observed the slow moving alkaline phosphatase band in individuals of blood groups 0 and B more frequently than in A and AB indi- viduals. Beckman [7] proposed the following hypotheses: (l) H substance couples with the alkaline phosphatase enzyme and makes a slow moving band. (2) H substance complexes with a protein, then the complex binds the enzyme. 10 He also stated that the presence of blood group A tends to suppress the expression of phosphatase sz. Arfors et al. [3] had observed that incubation of alkaline phosphatase of group sz with anti-Leb serum would decrease the activity of the slow moving alkaline phosphatase band. Hence, they stated that the slow moving alkaline phosphatase band may represent a complex with which a blood group substance (Leb or H) is integrated. However, Bamford et al. [6] observed that the average level of alkaline phosphatase activity in PM sera was about 30% greater than in P00 sera, with the heterozygote intermediate, and that all differences between the three means were highly significant. If the hypothesis of Arfors et al. was correct, one would not expect a greater level of serum alkaline phosphatase in those with the slow moving band. Furthermore, when Langman et al. [54] extracted alkaline phosphatase of intestinal mucosa, they found that the mucosal alkaline phosphatase tended to be higher in individuals of blood groups 0 and B than in those of blood group A. When the secretor status was taken into account, the differences between B or 0 groups and group A became more pronounced and non-secretors had the lowest level of intestinal mucosal alkaline phosphatase. If Beckman [7] was correct in his suggestion that blood group A tends to suppress the expression of sz phosphatase in the serum, one should not observe the differences in the level of intestinal mucosal alkaline phosphatase in different blood 11 groups as reported by Langman et al. [54]. No relationship has been found between AP and MNSs, Rh, serum Hp [2, 7, 88] or Kell, Duffy, Kidd and P blood groups and Gm, and Gc serum types [88]. Beckman [7] observed that there was no correla- tion between Lutheran blood group and AP, while he may have found a correlation between AP and Duffy in O secretor females, but not any other group. Alkaline phosphatase and diet In 1965, Shreffler [88] observed that the level of alkaline phosphatase sz varied significantly in samples drawn at different times from the same individual. He suggested that this might be related to diet, seasonal variation, physiological state or diurnal variation. Lang- man et al. [53] observed that after a fatty breakfast total alkaline phosphatase rose most markedly in individuals who were 0 or B secretors, where the average rise of alkaline phosphatase 7.5 hours after the meal was 24% higher than the average concentration of the fasting sample. A similar rise was also observed in the 5 AB secretors tested. The increase in A secretors was less than half that of B and O secretors, and the increase was still less in the non-secretors. The electrophoretic findings in these various sera were consistent with the view that all the increase in the enzyme concentration was of intestinal origin and the level 12 of other phosphatases present in the sera did not change [53, 110]. Warnock [110] believes that the increase of intestinal alkaline phosphatase associated with fat intake points toward a role for alkaline phosphatase in lipid transport across the intestinal mucosal cell. Lusting [56] observed that intravenous injection of L-homoarginine in rats inhibited bone, intestinal and liver alkaline phosphatases, while the activity of the enzyme increased in kidney and lung tissue. Fishman and Gosh [28] reported that force feeding of amino acids to rats markedly increased the level of intestinal alkaline phosphatase. The relationship of alkaline phosphatase with sex and age Phosphatase type 2 (sz) occurs in equal frequency in both sexes, there is no significant difference between parents and offspring, and the equal distribution of fre- quency between the sexes holds for all blood groups [2, 7, 108]. Roberts [78] stated that some blood constituents undergo significant concentration changes as the subjects age. In 1963, Gahne [33] observed the influence of age on the alkaline phosphatase enzyme pattern in cattle. Beckman 2 group among children [7] observed a high frequency of Pp below 3 years of age, while observing only slight variations between the other age groups. He also detected an increased 13 frequency of A and AB blood types among the children with sz type alkaline phosphatase. In contrast with this, Walter et al. [108] found that the frequency of sz type phenotypes increased with increasing age values. Shreffler [88] computed a significantly higher level of alkaline phosphatase in individuals below 16 years of age when compared with an older group. However, when he removed B and O secretors from his samples the differences became non-significant. Warnock [110] observed a second broad band in the B-globulin area of zymograms in children, in patients with bone diseases, and in extracts of periosteum. Although this band has not been resolved from the liver band thus far, it is readily distinguished from it. According to Fritsche and Adams-Park [32], although the predominant isoenzyme activity in the sera of adults is of hepatic origin, the major enzyme activity of children originates in bone. Yong [114] also observed that the activity of the bone isoenzyme was highest at birth and declined with age. In 1972 Statland et al. [95] studied three age groups of normal individuals, 4 - 12 years, 13 - 17 years, and 18 - 30 years. They observed that the activities of liver and intestinal isoenzymes were independent of the age of the subjects. However, the level of total alkaline phosphatase activity was about 3-fold in the younger groups as compared 14 to the adults, and the differences among the groups are related mainly to the bone isoenzyme [38, 95]. Sources and levels of alkaline phosphatase There are many discrepancies as to the sources and levels of different alkaline phosphatase isoenzymes [83]. One reason for these discrepancies is in the variety of experimental techniques and approaches, and another is the fact that such studies involve the effects of activators and inhibitors on phosphatases of different purities. Schlamowitz [83] observed that about 27% of serum alkaline phosphatase was of intestinal origin. Schlamowitz and Bodanskey [84] reported that approximately 28% to 38% of serum phosphatase of normal individuals was not of bone or intestinal origin. However, Hodson et al. [45], using starch gel electrophoresis, concluded that most of the phosphatase enzyme in the serum was probably derived from liver. Langman et al. [53] estimated that among normal individuals the percentage of total alkaline phosphatase activity which can be attributed to the slow moving intes- tinal component was 24% in P++ sera, 16% in P+0 sera, and 4% in P00 sera. In 1967, Fishman and Gosh [28] concluded that in ABH secretors the intestine must be considered the major source of serum alkaline phosphatase. Several 15 investigators (Green et al. [38], Firtsche and Adams-Park [32] and Skillen et al. [90]) have reported that although liver, intestinal and bone alkaline phosphatases are present in the sera of all age groups, in normal children the major source of serum phosphatase is bone, while that of normal adults is the liver. The inhibition of alkaline phosphatase It has been proven inadequate to have a classifica- tion of alkaline phosphatase isoenzymes based solely on starch-gel electrophoresis data. Additional biochemical studies of L-phenylalanine sensitivity, heat sensitivity, ABH blood types and secretor status can help in classifying the sources of different alkaline phosphatase isoenzymes [28]. In 1937, Bodansky [12] used bile acids to inhibit bone and kidney alkaline phosphatases. Bile acid proved to be ineffective in inhibiting intestinal alkaline phos- phatase. Schlamowitz [83] prepared antisera for bone and intestinal alkaline phosphatases. Individually or in mix- tures intestinal and bone antisera selectively precipitated their respective phosphatases. Under conditions where the enzymes are present in low concentrations, such as in serum, cross reaction of bone antisera may occur with intestinal 16 phosphatases. However, the cross reaction of intestinal antisera with bone phosphatase did not seem to occur. Schlamowitz and Bodansky [84] observed that anti-bone anti- bodies precipitated 40 - 59 percent of the alkaline phos- phatase in the serum of normal fasting individuals. In 1962, Hodson et al. [45] indicated that human liver phosphatase can be precipitated by anti-human bone phosphatase serum. This, of course, could lead to errors in identification. Boyer [13] also used immunologic means to delineate the inter- and intra-organ relationships of alkaline phos- phatase isoenzymes. Fishman et al. [27] reported that human alkaline phosphatase of intestinal origin, but not of bone, kidney and spleen, was 78% inhibited by L-phenylalanine. D- phenylalanine did not so affect alkaline phosphatase. Gosh and Fishman [35] reported that intestinal alka- line phosphatase activity in the presence of L-phenylalanine shifts its optimum pH toward the alkaline range and the inhibition is pH dependent. The extent of inhibition of the enzyme by L—phenylalanine is likewise greatly dependent on substrate concentration, and the energy of activation in the presence of the inhibitor is nearly three times greater than the corresponding value in its absence. The inhibition of the rat's intestinal alkaline phosphatase by L- phenylalanine is of the non-competitive type. 17 Horne et al. [46] tested the effect of L-phenylalanine on liver, bone and intestinal alkaline phosphatases. They found that s x 10‘3 M L-phenylalanine inhibits liver, bone and intestinal alkaline phosphatases to the extent of 37%, 30%, and 83%, respectively. The L-phenylalanine sensitive isoenzyme has been attributed to the intestinal isoenzyme [15, 77, 95]. Fishman and Sie [31] used L-homoarginine, an inhibitor of bone and liver alkaline phosphatase, in con- junction with L-phenylalanine to measure the different alka- line phosphatase isoenzymes. A 10% non-specific inhibition for both of these inhibitors has been reported [31, 38]. Many authors [27, 28, 38, 46] have reported a double specificity of L-phenylalanine. If this observation is correct, then the exclusive inhibition of intestinal alkaline phosphatase by L-phenylalanine and not D-phenylalanine sug- gests a unique difference in the catalytic site of the intestinal enzyme from that of the enzyme prepared from other sources [27]. Presant et al. [73] reported that, unlike others, their findings confirmed a significant D-phenylalanine inhibition of enzyme activity at all concentrations of D- phenylalanine employed. Many researchers (Posen et al. [71], Horne et al. [46], Cadeau and Malkin [14], Green et al. [38] and many others) have used heat treatment at 56°C or urea to denature alkaline phosphatase. 18 In serum the heat sensitivity of each enzyme source remains characteristic and independent of the influence of the others in the mixture. The resultant heat inactivation is an additive function of the heat sensitivities of members of the mixture [46]. Some bone phosphatase activity remains after urea treatment [4, 46, 95]. Urea also inactivates part of the liver, intestinal and placental phosphatase isoenzymes [4, 46, 51]. Urea inhibits bone phosphatase, but also introduces other artifacts that are distinct from those produced by heat. Heat denaturation usually causes limited inhibition of all other isoenzymes except that from placenta [47]. Bahr and Wilkinson [4] and Fennelly et al. [26] stated that urea at low concentrations might cause uncompetitive defor- mation of the enzyme molecule without interfering with its catalytic activity. At higher concentrations, however, more complete unfolding of the enzyme molecule would pro— gressively destroy its enzymatic activity. Moss et al. [64] observed that the differences in protein or urea content between different individuals are unlikely to be great enough to lead to significant variations in alkaline phosphatase heat stability. 19 Storage of alkaline phosphatase Green et al. [38] froze aliquots of a serum pool and assayed them over a 10-month period. For 42 samples, the mean and standard deviation of the heat inactivation was 74.1 i 4.0% with a coefficient of variation of 5.5%. Massion and Frankenfeld [59] observed that storage increases the activity of alkaline phosphatase, and that the lower the initial activity of a given lot, the greater was the rate of increase. The difference in rate of increase among materials from different sources varied widely. Refrigeration greatly decreased, but room temperature restored the rate of change. Ten freshly drawn sera were tested. The activity of these sera increased by an average of 0.9%, 2.7%, and 6.1% in 6, 24, and 96 hours; that of pooled serum, frozen and thawed, increased about 1% per hour. Cold storage for several months did not significantly change the level of alkaline phosphatase [71, 98]. Alkaline phosphatase polymerization In 1962, Moss and King [63] stated that if each organ does contain only one alkaline phosphatase, the main phos- phatase zone probably corresponds to the free enzyme protein. The subsidiary zone seen on starch gel electrophoresis may then represent proportions of the enzyme for which the electrophoretic mobility has been modified in some way. 20 Factors which may affect electrophoretic mobility include aggregation or disaggregation, modification of the enzyme by environmental factors (i.e., removal of charged molecules), or attachment of the enzyme protein to indifferent protein fractions [58]. A very slow-moving band of alkaline phosphatase close to the origin has been observed in some samples. Its posi- tion coincides with B-lipoproteins [63, 88]. Markert [52] stated that under suitable conditions of electr0phoresis a single LDH isoenzyme may be represented by two, three, or more closely spaced bands. The multiple bands cannot repre- sent distinct isoenzymes in terms of protein composition but may be produced by minor changes in molecular migration, perhaps due to physical-chemical environment in the cell at the site of protein synthesis. The heavier form of the enzyme can be converted to the lighter form. These two forms of enzyme have been reported to be kinetically indistinguishable from each other [36, 37, 63]. ' Beckman [8] states that the serum intestinal alkaline phosphatase, which is resistant to neuraminidase, has lost its sialic acid in the intestinal form and is combined with lipids, therefore moving slower on electrophoresis. The native form of this enzyme does not occur in serum, but is always found in extracts of fetal jejunal mucosa and some- times in biopsy specimens from the mucosa of adults. 21 Smith and Fogg [92] reported that after reconstitution a high alkaline phosphatase component predominated; during subsequent spontaneous activation this component decreased, and there was a concomitant increase in a low-molecular weight alkaline phosphatase component. The result of butanol extraction suggested that the observed change may be attrib- uted to the breakdown of a complex between alkaline phos- phatase and lipoproteins. Moss [65] observed that the antigenic determinants of the major fast electrophoretic zones are also present in both the larger and smaller slow components, and that poly- merization of the enzyme or combination with non-enzymic molecules does not mask the antigenically reactive groups. The relationship of alkaline phosphatase and different diseases Alkaline phosphatase has been observed to be elevated in hepatobilary diseases and bone disorders [16, 19, 41, 49, 52, 67, 71, 84, 114]. Gutman [41] stated that cell injury or aging may result in abnormal leakage of intracellular enzymes into extracellular fluid at the expense of the tissue enzyme content, which doubtless accounts for the increased number and quantity of circulatory enzymes en- countered in a variety of diseases. In 1959 Schlamowitz and Bodansky [84] reported that, while the ratio of bone to intestinal phosphatase is 1.4 - 22 4.6 to 1 in normal individuals, it was 6 - 190 to l for cancer patients. Schwartz [85] also reported high levels of alkaline phosphatase activity in cases of cancer tumors including bone or liver. Chiandussi et al. [19] found that Paget's disease, intrahepatic obstructive jaundice, extrahepatic obstructive jaundice, and hepatic cancer were associated with high alka- line phosphatase. Korner [52] reported that in the bone disorders with high alkaline phosphatase a large and sometimes predominant part of the increase is due to beta globulin phosphatase activity, and that alpha 2 activity is also raised. Alpha 1 activity, though usually low, may be increased. In hepa- tobiliary disorders with a high phosphatase concentration, the increase is due mainly to raised alpha 2 phosphatase activity, and there is also an increase in alpha 1 phospha- tase. Gutman [41] proposed two main hypotheses to explain the elevation of alkaline phosphatase in disease conditions: 1. Retention or impaired excretion, which assures the source of plasma alkaline phosphatase to be bone, and the liver has only an excretory function by the way of bile passage. If these passages are blocked, then the bone enzyme is retained. 2. Hepatogenic theory: most serum alkaline phospha- tase is of hepatic origin, and that the hepatobiliary system, 23 when disordered, contributes wholly or in large part to the increase in plasma alkaline phosphatase. In contrast, Yong [114] found that electrophoresis of bile reveals an isoenzyme having the same electrophoretic mobility as liver type alkaline phosphatase but not bone type. Furthermore, Kaplan and Righetti [49] observed that in rats the rise in the serum alkaline phosphatase in obstruc- tive jaundice was intimately related to de novo synthesis of this enzyme by liver. Griffith et al. [40] observed a generally high degree of correlation between abnormalities of serum alkaline phos— phatase and 5'-nucleotidase in hepatobiliary diseases. Phelan et al. [70] state that it is now apparent that the increase in serum alkaline phosphatase in liver disease arises largely from a de novo synthesis of enzyme protein by liver cells, and while there is as yet no evidence that a similar mechanism accounts for the elevation of serum 5'- nucleotidase, changes in the relative levels of the two phosphatases at different stages of disease may be due to alterations in the extent to whidh a single type of cell capable of producing both enzymes is stimulated to synthesize one rather than the other. Betro [11] observed that 23% of cases who had both high alkaline phosphatase and lactic dehydrogenase also had congestive heart failure. 24 Skillen et al. [90] observed that increases in the serum intestinal alkaline phosphatase are likely to be an indication of a disorder of the hepatobiliary system or chronic renal failure. Alkaline phosphatase, Km and the effect of pH In 1934 Linweaver and Burk [55] proposed the theory of the dissociation constant, Ks = (E) (S)/(ES)' On the basis of this theory, the rate of the observed reaction is directly proportional to the concentration of the enzyme substrate complex (ES) at all values of the concentration of the substrate (S); Ks is proportional to (8) only at low values of substrate. In 1962, Moss and King [63], using B-naphtylphosphate as the substrate, observed that alkaline phosphatase extract of different tissues had different affinities, while differ- ent bands of the same tissue had the same affinities for the substrate. The bands of bone phosphatase had Km values of 0.110 mM and 0.118 mM, that of liver and 0.067, 0.067, and 0.070; kidney phosphatase had Km values of 0.105, 0.103, and 0.096 for different bands and intestinal phosphatase had Km values of 0.090 and 0.098. These findings clearly point out the existence of a single alkaline phosphatase in each tissue, with pr0perties characteristic of the tissue of origin. 25 Gosh and Fishman [37] purified two interconvertible placental phosphatases. The lighter form had a molecular weight of 70,000 and the heavier form had a molecular weight above 200,000 by sucrose density gradient centrifugation. The Km of the two variants were identical (18 mM) at pH 10.6. However, when pH was raised to 10.7, Km raised to 72 mM of phenylphosphate° Skillen and Harrison [91] observed that the Optimum substrate concentration in a bicarbonate buffer, with p- nitrophenyl phosphate as substrate, was 55 mM for liver, 50 mM for bone and 14 mM for intestinal phosphatase. Al- though this observation does not clearly distinguish between bone and liver phosphatases, it clearly separates these two from intestinal phosphatase. Regan isoenzyme Regan isoenzyme is named after the cancer patient in whose serum the isoenzyme was first discovered. Fishman et al. [30] observed that the Regan phosphatase in cancer tumors resembled placental phosphatase and not those of the tissue of origin. They also observed that the isoenzyme appeared in cancer cells, as well as in the serum of patients with bronchogenic cancer. Fishman and coworkers [29, 66, 96] observed that Regan isoenzyme was indistinguishable from placental alkaline phosphatase on immunological as well as on biochemical 26 grounds. It was suggested that these findings could be an indication of derepression of the genome by the tumor and that this may be general for tumor proteins. There is a high incidence of Regan isoenzyme in carcinoma of the ovary followed by pancreatic, gastric and lung carcinoma. The lowest incidence was observed in bronchogenic carcinoma and breast cancer. Blood grouping of those patients with Regan isoenzyme showed a sharp decrease in group A and a sharp increase in group O. In 1972, Higashino et al. [44] observed the Regan isoenzyme in patients with hepatocellular carcinoma. Their biochemistry also resembled that of placental alkaline phos- phatase, although the Km values were 1.1 mM and 1.6 mM of substrate for the variant and placental alkaline phosphatase, respectively. Epilepsy, genetics and enzymes There is an elevation of glutamate oxalacetate trans- aminase (GOT) in cerebrospinal fluid of epileptics [43, 61]. Hain and Nutter [43] observed that an age factor was also involved which should be taken into consideration. In 1969 Niebroj-Dobosz and Hetnarska [68] found elevated lactic dehydrogenase in idiopathic epileptics. Other enzymes were normal (creatine phosphokinase, phosphohexosisomerase, aminotransferase, aspartate transaminase, malate dehydrogen- ase and lactate dehydrogenase). Wright and Pollitt [113], 27 in an epileptic patient, observed high levels of plasma ornithine and 240 ug/lOO ml. of ammonia (upper limits of normal are 60 ug/lOO ml.). Ornithine carbamoyl transferase was just higher than normal. There was no significant urinary excretion of ornithine, but large amounts of homo- citrulline. Plasma ornithine levels in parents and siblings were just twice that of normal. In 1970 Mabry et al. [57] reported a syndrome of very high levels of alkaline phosphatase in three siblings and a first cousin, all of whom had epilepsy and mental retarda- tion. All these individuals were from consanguineous mar- riages° They also reported normal levels of serum alkaline phosphatase in 129 severely retarded individuals, many of whom had seizures. Casey et al. [16] found 15 males and 6 females with elevated alkaline phosphatase among 18 male and 12 female epileptics. However, the number of epileptics compared to the total number of 17,431 individuals was not large enough for statistical studies. Tsairis et al. [106] observed epilepsy in two daugh- ters, a son and their mother. One daughter and their father did not have epilepsy. The mother and two daughters had elevated blood and cerebrospinal fluid pyruvate and lactate levels in a basal state, and abnormally high blood levels after glucose loading. 28 In 1972, Faed et al. [25] found a ring chromosome in many cells of a girl who was epileptic since the age of four. She had no previous head injury. Causes of abnormalities in blood metabolites It is well known that an increased rate of enzyme synthesis or a decreased rate of enzyme degradation can induce an increase in enzyme concentration. Certain sub- stances can stabilize pre-existing enzyme proteins and delay their degradation. A third mode of induction is an increase in the activity of individual enzyme molecules. Griffin and Cox [39] observed that prednisolone mediates a configura- tional change in alkaline phosphatase during its synthesis that leads to an increase in the number of catalytic sites or a lowering of energy level of the enzyme substrate transi- tion state. Using radioactive leucine, these workers observed that prednisolone did not produce an increased enzyme protein level. In 1972, Singh et al. [89] used aniline derivatives, salicylic acid derivatives, hydrazines, catechol amines, and purine derivatives to observe their effect on blood meta- bolites. These drugs did not influence the levels of in— organic phosphate, blood urea nitrogen, cholestrol, total protein, albumin and alkaline phosphatase. 29 Effects of high blood ammonia In 1954, Sherlock et al. [86], by using ammonium chloride and high protein diets, were able to cause an altered mental state, characteristic tremor, and electro- encephalographic changes indistinguishable from impending hepatic coma. Bessman and coworkers [9, 10] observed a significant arterial-venous difference of free ammonia, suggesting that the free ammonia is converted by brain and muscle to a bound form. They advanced the following hypothesis: of the many chemical reactions utilizing ammonia in the body, two are of significance in the brain, glutamine synthesis, and the reversal of glutamate oxidation, namely, reductive amination of a-ketoglutarate. They suggested that the synthesis of glutamic acid from the a-ketoglutarate generated by the Krebs cycle in the brain was the mechanism of hepatic coma when the arterial ammonia concentration rises. It was also suggested that transamination of aspartic acid to replace the a-ketoglutarate necessary for the Krebs cycle cannot be accepted, since due to impermeability of the brain to organic anions the brain cannot obtain an exogenous source of aspartic acid. Summerskill, Wolfe and Davidson [97] measured different arterial/venous blood ammonia levels in hepatic coma patients. However, they did not observe a good correlation with clini- cal status. 30 In 1963, Stahl [94] observed that elevated blood ammonia level, especially arterial, may precede coma by many days. Subsequently, the patients lapsed into coma with increasing arterial and venous ammonia, increasing arteri- ovenous difference, and later, a progressively diminishing arteriovenous difference. A favorable clinical outcome was usually preceded by a decline in blood ammonia concentration with an increasing arteriovenous difference. However, some patients remained comatose even after a normal level of blood ammonia had been reached. Stahl suggested that this finding strongly favored Bessman's hypothesis of a long active depletion by ammonia of the Krebs cycle in the brain. In 1965, Warren et al. [112] observed very little change in electroencephalograph readings before and during an ammonia tolerance test. However, after five days of high protein diet and ammonium chloride intake, the tracing was significantly lower in one patient, and marked changes con- sistent with those seen in hepatic coma were observed in four out of ten patients. Cohn and Castell [20] observed electroencephalographic changes in only 1 of 19 persons when they loaded them with ammonium acetate. This observation caused them to suggest that the ammonia which is immediately available to the brain :from the blood is not the major determinant of gross changes 111 the electroencephalography seen in patients with hepatic eIlcephalopathy. 31 Glasgow et al. [34] observed that the highest blood ammonia levels were seen in the most deeply comatose patients in Reyes syndrome. Shih et al. [87] found a similar condi- tion of ammonia intoxication of the central nervous system in patients with enzyme defects of the urea cycle. Carter et al. [18] indicated that most ammonia in the blood is in the ionized ammonium (NHZ) form and that an. increase in blood pH would increase the amount that could readily penetrate the blood brain barrier. Blood urea nitrogen A high incidence of significantly low urea nitrogen occurs in various mental disorders, namely paranoid states and personality disorders in males, and neurosis, schizo- phrenia and psychoses in males and females. No relationship was observed between low urea nitrogen and elevated leucine aminopeptidase, alkaline phosphatase, glutamic oxalacetic transaminase or lactic dehydrogenase [16, 17]. In 1973, Redman et al. [75] observed that the birth weight of neonates was, on average, inversely proportional to the maternal urea. This trend appeared at 28 weeks gesta- tion, but was most striking at 32 weeks when a urea level above 20 mg/100 ml was associated with significantly smaller babies. MATERIALS AND METHODS Blood was collected by venopuncture from a group of mentally retarded individuals in institution A in 1971 (1A1) and from another group in 1972 (1A2), from employees of institution A as a normal control (IAC), and from residents of institution B (IBl). (Over 95% of the residents in insti- tution B are epileptics.) Blood was also collected from the institutionalized sib pairs in 1971 (Sib Pairs 1971), and in 1973 (Sib Pairs 1973). In 1973 blood was collected from all the known epileptics in institution A (IAE). A control was derived by selecting the data for that sib in each pair whose first name was alphabetically last (IAS). When it was desired to have a population which was age 20 and older, persons in the sib pair sample who were less than 20 years old were, where possible, replaced in the sample by sibs of over 20. After clotting, the blood was centrifuged at a speed of 2200 RPM and the serum was capped and stored in a refrigerator for analyses. A Technicon two channel AutoAnalyzer was used to measure the level of the following metabolites: glucose (Glu), blood urea nitrogen (BUN), serum glutamate oxaloace- tate transaminase (SGOT), uric acid (UA), calcium (Ca), 32 33 inorganic phosphate (Pi), total protein (TP), albumin (Alb), and alkaline phosphatase (AP). Glucose Glucose was measured by the inverse colorimetric technique, measuring the decrease in color of potassium ferricyanide as it is being reduced by reducing sugars. The procedure, adapted to automation by Technicon from the method of Hoffman [J. Biol.(HmmL 120: 51, 1937], involves the dilution of the sample in saline followed by dialysis into air-segmented alkaline potassium ferricyanide. The dialisate is then heated to 95° in a heating bath and the color change is measured at 520 mu wave length (Technicon Pamphlet N-l6b). BUN The AutoAnalyzer urea nitrogen method is a slightly modified version of the procedure described by Marsh et al. [Clinical Chemistry 11: 624, 1965, as described in Technicon Pamphlet N-l6b]. The BUN procedure is a modification of the carbamido-diacetyl reaction as applied to the determina- tion of urea nitrogen. It is based on the direct reaction of urea and diacetyl monoxime under acid conditions. The presence of thiosemicarbazide intensified the color of the reaction product and enables the determination to be run without the need of concentrated acid reagents (Technicon 34 Pamphlet N-l6b). Salineediluted serum is dialized into an air-segmented stream of diacetyl monoxime-thiosemicarbazide mixture. The mixture is acidified by introducing into it a mixture of ferric chloride-phosphoric acid and sulfuric acid. The acidified solution is heated to 95° in a heating bath where urea reacts with diacetyl. The color product is then measured at 520 mu in a 15 mm tubular flowcell. SGOT SGOT is measured by the procedure described in Techni- con Pamphlet N-25b and it is based on the procedure of Morgenstern et al. [Clin. Chemistry 12: 95—111, 1966]. Serum samples incubated with SGOT substrate are brought to 37° in a heating bath coil and the resultant oxalacetate produced by the action of serum enzyme is dialyzed into a reagent stream of citrate buffer. The dialysate is then reacted with the diazonium salt of N-butyl-4-methoxymetani1amide (Azone Fast Red. PDC) which couples with the oxalacetate during passage through the second 37°C heating bath coil. The colored product is read at 460 mu in the colorimeter in a 15 mm. flow-cell. Blank interferences are eliminated by the use of dialysis and the specificity of the color reaction. Uric acid Uric acid is measured by the procedure described in Technicon Pamphlet N-l3b. The automated procedure is adapted 35 from the manual method described in Practical Physiological Chemistry by Hawk, Oser and Summerson [13th ed., p. 564]. The sample is diluted with physiological saline and then dialyzed. To the dialysate is added a mixture of sodium cyanide-urea followed by the addition of phosphotungstic acid reagent. The quantitative measurement of uric acid involved the reduction of phosphotungstate complex to a phosphotungstite complex. The presence of cyanide intensi- fies the color and prevents turbidity. After mixing, the blue color of the reaction product is measured at 660 mu in a 15 mm flowcell. Calcium Calcium was measured by the procedure described in Technicon Pamphlet 26a. The method is based on the proced— ures of Gitelman [Anal. Biochem. 18: 521, 1967], who incor- porates the use of 8-hydroxyquinoline with the method of Kessler and Wolfman [Clin. Chem. 10: 686-702, 1964] to virtually eliminate the interference of magnesium. The serum is diluted with 0.25 N HCl to release the protein- bound calcium, and the mixture is dialyzed. Cresolphthalein complexone, containing 8-hydroxyquinoline and diethylamine potassium cyanide base, are then added to the dialysate. A colored calcium dye complex is formed in the presence of diethylamine. The developed color was measured in a 15 mm flowcell at 580 mu. 36 Inorganic phosphate The automated inorganic phosphate procedure was adapted to the AutoAnalyzer by Kraml [Clin. Chim. Acta. 13: 442, 1966] from the procedure reported by Hurst [Can. Jour. of Biochem. 42: 287, 1964] (Technicon Pamphlet N-26a). The serum sample is diluted with 0.25 N HCl and dialyzed. An acidic solution of ammonium molybdate is added to the dia- lysate. Phosphomolybdic acid is formed and this is imme- diately reduced by stannous chloride-hydrazine sulfate which is introduced into the air—segmented stream of dialysate- ammonium molybdate. The absorption of the blue product is measured at 660 mu in a 15 mm flowcell. Total protein The procedure for determination of total protein is described in Technicon Pamphlet (N-l4b). The procedure was adapted to the AutoAnalyzer by D. L. Stevens and is a mod- ification of the biuret reaction, proposed by Weichselbaum [Amer. J. Clin. Path. 7: 40, 1946]. The sample stream is diluted with an air-segmented stream of biuret reagent. The biuret reaction depends upon the formation of a purple colored complex of c0pper in an alkaline solution, with two or more carbamyl groups (-CO-NH-) which are joined directly together or through a single atom of nitrogen or carbon. The developed color is measured at 550 mu using a 15 mm flowcell. 37 Albumin The AutoAnalyzer procedure for albumin (Technicon Pamphlet N-15c) is based on the work of Ness et al. [Clin. Chim. Acta. 12: 532, 1965]. It is based on the quantitative binding of the anionic dye, 2-(4-hydroxyazobenzene) benzoic acid (HABA), Specifically to serum albumin. Nishi and Rhodes [Automation in Analytical Chemistry, edited by Skeggs, 1966, pp. 321-23], adopted the method of Ness et al. to the AutoAnalyzer. A serum sample is introduced into an air- segmented HABA dye diluted in a phosphate buffer. The developed color is measured at 505 mu in a 15 mm flowcell. Alkaline phosphatase The automated alkaline phosphatase procedure is based on the method of Morgenstern et al. [Clin. Chem. 11: 876, 1965], who modified the manual process of Bessey et al. [J. Biol. Chem. 164: 321, 1946]. The serum is diluted in an air-segmented stream of the substrate p-nitrophenyl phosphate dissolved in 2-Amino-2-methyl-l-propanol buffer at pH 10.2510.05. The mixture of serum-substrate is incubated at 37°C for 4 minutes. The enzyme breaks p-nitrophenyl phos- phate into p-nitrophenol and phosphate. The mixture is then dialyzed into an air-segmented stream of 2-amino-2-methyl—1- propanol buffer. The dialyzed p-nitrophenol is highly colored under alkaline conditions and thereby provides its own chromagen. Dialysis eliminates the interference of 38 bilirubin and the need for blank correction (Technicon Pamphlet N-6b). The absorbance of p-nitrophenol is measured at 410 mu in a 15 mm flowcell. The blood from the sib pairs in institution A (Sib Pairs 1973) and from the epileptic patients in institution A (IAE) were studied for the level of ammonia, the levels of bone, liver and intestinal alkaline phosphatase, alkaline phosphatase substrate specificity, ABC and Lewis blood groups and their secretor status. A number of other institution- alized individuals with high alkaline phosphatase were also tested for the level of their ammonia. For determination of liver, bone and intestinal alkan line phosphatase the automated method of Morgenstern et al. described above was adapted to the two channels of the Auto- Analyzer (Figure l). The total alkaline phosphatase was first determined as described previously. Then a 1 cc ali- quot of serum was heated in a bath for 15 minutes at 56.0: 5°C (Horne et al.[46]) and immediately transferred to an ice water bath for 3 minutes. The serum samples were then brought to room temperature and assayed on the AutoAnalyzer. Bone alkaline phosphatase is more sensitive to heat than liver or intestinal alkaline phosphatase (Horne et al. [46], Warnock [109, 110] and many others). In one channel p- nitrophenyl phosphate was used. This would measure the heat-stable alkaline phosphatases. In the other channel p-nitrophenyl phosphate + .005 M L-phenylalanine was used 39 .honxfimcmv wumwcmum cam memos one .H mfinmh 46 sufficiently variable from run to run to cause wide variation in the mean results. Such run variation was observed for glucose, SGOT, Ca, Pi, total protein (TP), and albumin (Alb). In addition, the BUN values for 1972 differed from those of 1971 because of a change in standards, although there does not appear to be run variation for BUN within a given year. Run variation makes it impractical to compare popula- tion means by simple statistical tests. However, the means will be similar in all runs so that the frequency of outlying values should be comparable. The standard deviations may be compared because the variability within a sample should remain constant while the mean varies slightly. The various biochemical tests were statistically analyzed and are presented in Tables 2-10. Where appropriate, each sample of institutionalized persons was compared with the control group to determine whether the mean or standard deviations were significantly different by t or F test, respectively. Similar comparisons were made between IAl and IAZ to check for the existence of variation in method- ology over the interval of a year, and between IA2 and IBl to check for a difference due to epilepsy. Values outside of the normal ranges are termed "outliers" and their numbers are presented in Tables 11 and 12. The frequency of individuals with an unusual amount of each metabolite was analyzed by a standard 2 x 2 chi-square test which tested two p0pulations for the number of high 47 mn.HH No.~n vow HmH wn.~H mm.mw 0N5 Nop ppmwcmum :wwm z :ofiumasmom .omousfiw Eduom wow “wow map we mfimzfimc< .N macaw 48 Table 3. The analysis of the test for blood urea nitrogen (BUN). Population N ggzn Standard deviation IAC 230 14.94 3.87 IAl 668 17.03 4.33 IA2 726 15.59 4.33 IBl 364 13.61 4.55 P0pulations compared F test t test IAC and 1A1 1.25* 6.85** IAC and IA2 1.25* 2.15** IAC and IBl 1.38* 3.81** 1A1 and IA2 1.00 6.20** IA2 and IBl 1.10 7.07** * Significant at 0.05 level. ** Significant at 0.01 level. 49 Table 4. The analysis of the test for SGOT. . ~ Mean . . Populat1on N Karmen units Standard deV1ation IAC 230 28.27 7.13 IAl 668 22.89 8.22 IA2 726 22.32 11.12 IA2# 1ess outliers 663 . 22.38 7.72 IBl 364 35.51 13.78 IB1@ less high AP and SGOT individuals 297 31.09 7.56 P0pulations compared F test t test IAC and 1A1 1.32* 9.48** IAC and-IA2# 1.17* 10.56** IAC and 131° 1.12 4.35** 1A1 and IA2# 1.13 1.16 IA2# and 131° 1.04 l6.25** * Significant at 0.05 level. ** Significant at 0.01 level. #,@ Populations used in comparisons. 50 Table 5. The analysis of the test for uric acid. Population N :gzn Standard deviation IAC 230 5.73 1.47 IAl' 668 5.59 1.36 IA2 726 5.19 1.35 IBl 364 5.16 1.30 Populations compared F test t test IAC and 1A1 1.16* 1.27 IAC and IA2 1.18* 4.95** IAC and IBl 1.27* 4.81** 1A1 and IA2 1.00 5.49** IA2 and IBl 1.07 0.35 ** Significant at 0.05 level. Significant at 0.01 level. Sl mm.o Ho.OH com HmH mo.o oo.oa own ~ow pumpnmum *me new: 2 coflamfiamom .Esflofimo Esuom How umop may mo mfimxfimcm one .o oanmh 52 vo.o wu.m com HmH Ho.o mH.v own Now wnsucmum was can: 2 coflumfinnom .oumcmmozm enhom How umou ecu mo mfimxamcm 0:9 .5 manmh 53 em.o om.n com HmH wm.o em.u 0N5 Nou ppmwcmum whw new: 2 aofiumfidmom .cfiouopm Hmpou Essom sow umou ecu mo mfimxamcw one .w oHan 54 N¢.o oH.v eom HmH om.o oH.v emu Now wumwcmum wsm new: 2 cowumasmom .cfiesnam essom sow umou can mo mwmxfimcm one .m manmh 55 Table 10. The analysis of the test for serum alkaline phosphatase. Population N Mean mIU/ml Standard deviation IAC 230 72.80 24.52 IAl 668 91.75 41.85 IA2 726 91.70 46.43 IA2 1ess high AP 8 SGOT# 624 90.84 43.56 IBl 364 158.55 71.17 IBl less high AP 6 SGOT@ 297 147.30 69.59 Populations compared F test t test IAC and 1A1 ‘ 2.91** 8.28** IAC and IA2# 3.15** 7.58** IAC and 131@ 8.05** 17.13** 1A1 and IA2# 1.08 0.21 1A1 and 1A2 1.23** 0.02 IA2# and 131@ 2.55** 12.84** #,@ Populations used in comparison. * Significant at 0.05 level. ** Significant at 0.01 level. 56 NmH moa com 0 5mm 5 mom N vom o omm m mmN ea mmm HH HmH mus med - - «mm mm was as see AH son ON men mm awe mm N m< mo o>psu o>fiumaseso .N shaman HE\DHB «mousing—m 0543th no H035 oov own 02 0mm com 0.: _ o3 — _ o_m an _____________________________ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ H N _ _ _ A _ _ _ _ 1 _ _ _ _ _ r11 xx. \ 11 :3 u 5 3.53365 60333 >235... \ oHuaoZQo ama uo>o mo mcuuaHucoo ceaugqaoa Ann .\ \ [old 32 . 5 k \ .2 use .80... fion 5 :3: 3.3336,: 33 2: .III . . \ [II :3 u 5 2.53365 H6336» \\ \ xgaucoe cauaoomoncoc uo mcHuuHacou coHuuHsnoa N4: llllllll .[Il \ \a ll..|o.~| \\\ \ 1|: \. \ I46 \\ 1 4 .\\\ \ IJn \ III \\\\. \ . [flu \\\\\\. \\ 11: \\ \. \ [an \\ \\ \ III \\\ N \\ \. \ 14w \ \. N [I .\. \\\. \ \I. \\ \\ \\ [r04 \ \\ \\\.\\ ll]: \ \II.\I..\I\ .\.\\ [III “\\\H“\ ‘‘‘‘‘ \\\..\ om l I Illi.\.‘. ......... [I .A“ . 1...... llllllllllll [[3 H uoa no 71 protein, albumin, and AP. In 1973 three additional metabolites: ammonia, heat and L-phenylalanine resistant AP (APL) and heat resistant, L-phenylalanine sensitive AP, (API) were also tested The results of intra- and inter-family difference analyses for 1971 and 1973 are presented in Table 15. The intra- and inter-family mean differences were not significant for either 1971 or 1973, for metabolites glucose, SGOT, uric acid, Ca and albumin. However, except for glucose, the mean for inter-family differences exceeded the mean for intra-family differences, as expected. In 1971, the mean inter-family difference in metabolites BUN, Pi and AP were significantly higher than the mean for intra- family differences, while there were no significant differ- ences for these metabolites in 1973. Although in these cases there was a change from significant to non-significant or the reverse from one year to the next, the differences were always in the same direction. For the three additional metabolites, ammonia, APL and API tested in 1973, the mean inter-family difference was 33.6, 27.8 and 66 per cent above the intra-family difference, respectively. The differences were not signifi- cant for APL and ammonia, while that of API was significant at the 0.05 level. 72 --- --- .mm.N NN.H mN.H «.ON.4 No.H NQ.H om.H «.ow.N om.o H --- --- .mN.N oo.H «mm.H «.OH.N os.H «sm.H NN.H N4.H so.H a --- --- mm as so so so so so so so .2 --- --- mm.NH so.o oo.o mo.o wo.o 8H.o 8H.N om.o mm.H m --- --- Nm.w so.o mo.o oo.o oo.o NH.o 4N.H om.o mm.H 3 .m.m --- --- om.mm wN.o om.o No.0 N6.o 8N.H om.NH N¢.N NH.HH m --- --- om.oo wNwo Hs.o Ns.o Nm.o No.H mN.sH NN.N Hw.OH z .n.m --- --- om.om 6N.o 6m.o sm.o NN.o 6N.H NN.NH 4N.4 NN.HH m --- --- om.m4 NN.o ms.o Hm.o 6m.o mH.H Nm.w 4H.m No.9H 2 e862 HE HE HE a msHe: N N \aHe \aHe \oHe Hum New Nae Nae N e assume N a N e HNmH sHa< HHa< a< BH< as Ha so <3 Boom 22m :Hu .mNmH we. HNmH eH e6>somno mHHma cam uoNHHmcowusprmcfl mo moonshomme xHfismm-H0p:N use -mpucw mo momxamc< .mH oHan 73 .ommumzmmoca ocfimeHm o>HuHmcom ocficmHmHzcmnm-H ucmumwmou uwom @Hm< .ommuwcmmogm ocmeme ucmumwmoh ocwcwamaxcogm-q was umom «Hm< .H6>6H Ho.o Hm pemuHmHeNHm 4. .HmsoH mo.o as semuHmHemHm . «mo.~ hm.o mw.o Hv.H mn.H «mH.~ o~.o ow.o Nm.o OH.H Hw.o NH.o H 55.H 0N.H mH.H mo.H «Hm.m «amo.~ mo.H «anm.m «mo.~ 0H.H «eHo.N 0N.H m mm mm mm mm mm mm mm mm mm mm mm mm .2 m~.o mN.N vo.NH om.o oo.o mH.o no.o OH.o mH.o mm.H mm.o mm.v m wH.o ou.H mo.HH No.4 wo.o wo.o wo.o oo.o mH.o mN.H om.o cw.m 3 .m.m mv.H mm.NH oo.mn Nn.om mm.o mn.o mv.o No.o mo.H mo.n mm.m mm.v~ m wo.H nu.m mm.oo oo.w~ om.o mv.o N¢.o mm.o ou.o HH.N mo.~ mH.- 3 .m.m oo.H vm.~H om.Hn 5N.m¢ m¢.o mm.o om.o mm.o Hv.H em.m mm.m Nw.wH m oo.H Hw.m ~v.nm um.mm em.o em.o mm.o m¢.o m~.H em.o om.m Hm.mH 3 new: He as as *m: Nam whw was me *me mafia: wms was \DHE \DHE \DHE m nH< me Hm mu <3 coshmx 3H0 man sHa< NHa< a< :2 . Boom 23m wozcfipaou .mH oHan 74 Epileptic and retarded studies in l973 In 1973 all the known retarded individuals with constitutional epilepsy in institution A were tested for the metabolites: ammonia, glucose, BUN, SGOT, uric acid, Ca, Pi, total protein, albumin, AP, APL, and API. In addi- tion, they were also typed for their ABC and Lewis blood group systems and secretor status. Throughout the text this population will be referred to as IAE. This study was performed on the same day as that of the sib pairs, with the two groups going through the Auto- Analyzer in random order. One member of each sib pair was selected as a con- trol. The control individual was that sib whose given name was alphabetically the last in the pair. Where a teSt elim- inated persons less than 20 years old, the other sib replaced the selected sib, if possible. This group will be referred to as IAS. The correlations between the various metabolites and age are presented in Table 16. The age correlations for the various metabolites: ammonia, glucose, BUN, uric acid, Ca, total protein, APL and API are not significant in either the IAE or IAS p0pu1ations. In the IAE p0pu1ation age 10 years and older, there is a significant correlation between age and Pi, Vmax/va and AP determinations. However, when individuals in the age .HasaH Ho.o s. saauHsHaNHm .. .HasaH mo.o as samuHsHamHm . 75 em. OO. NH.- Hm. NN.- mo. NO. OH. vO. m~.- OH. MH.u Nm. OHo msmoz ON Ho>o mo mo mo m Ha< Haa a< aH< as Ha so on» can own coozson =0HsmHossoo one .OH oHnms 76 group 10-19 years old are taken out of the population, the correlation coefficient for all the metabolites tested is not significant. In the IAS population there is a significant corre- lation between age and metabolites: albumin, AP and SGOT. However, these correlations also are not significant after removal of individuals 10-19 years old. Tables 17-24 present the results of analyses of variance for the metabolites ammonia, glucose, SGOT, Ca, Pi, total protein, albumin and for age. The mean difference analyses for these metabolites show no significant differ- ences between the IAE and IAS populations, whether the ages are 10 and older or 20 and older. Figure 3 shows an upward shift in the level of ammonia in epileptics, when compared to the non-epileptic mentally retarded individuals. However, both populations have large standard deviations and the mean differences are not significant. When the many high outliers (ammonia levels over 100) in the two p0pu1ations are compared, there are 8 of 33 in the IAS and 20 out of 61 in the IAE p0pu1ation. The chi- square value for the comparison is 0.75 which is not signi- ficant at 0.05 level. Table 25 presents the results of the BUN analysis. The mean and standard deviation are 13.12 mg/100 ml and 4.24 for the IAE p0pu1ation age 10 and older, and 15.12 mg/100 ml and 3.73 for the IAS p0pu1ation age 10 and older, 77 .HosoH Ho.o .4 paauHsHaaHm 4. .HasoH mo.o pa HeauHsHaaHm . .OHo mums» ON so>o 9: was mo moO Osmwcmsm cum: 2 .choEEm Essom sow away may we mmeHmcm was .sH oHnms 78 can .HasoH mo.o a. pamoHHHaNHm 4 .OHo msmox_om so>o mo moO Osmwcmpm cum: 2 .omousHm Essen sow away was we mmeHmcm was .OH oHan .Ho>oH HO.O um semusmsewsm .4 .Ho>oH m0.0 um ucmonHcmHm 4 .OHo msmox_O~ sm>o mo m6H Ho.o .4 samuHsHaMHm .. .HosaH mo.o pa paausssamsm 4 .OHo msmos ON so>o mo moO Osmvcmsm cum: 2 .cHESOHm Essom sow smou on» mo mHmsHmcm one .ON OHOms 81 .Ho>oH H0.0 pm sQOUHmHOMHm .4 .HasaH mo.o a. sesoHsHaNHm . .OHo msmos ON sm>o mo moO Osmwcmum :moz z .EDHuHmo Essmm sow smos was we mHmuscm one .HN OHan 82 was .H6>6H HO.O p4 pesosasemsm 4. .HosaH OO.O p. HaauHaHaNHm . .OHo msmos ON so>o mo woH H0.0 so scmonscmHm «4 .Ho>oH m0.0 so scoonHcmHm 4 .OHo msoos ON so>o mo moH H0.0 so scoonscmHm «a .H6>6H m0.0 o4 seauHsHamHm 4 .OHo msoo» ON so>o mo moO Osmucosm coo: z .mHost>HO:H one mo omm ons sow smos onu mo mHmsHocm one .QN oHDOH 85 .mcoHsoH -smoa m6H H0.0 pa pesosssamsm .4 .HosaH OO.O .4 oa46HsHeNHm 4 .OHo msoos ON so>o mo moO Osowcosm coo: z .comossH: mos: OooHO sow poop ons mo mHmsHocm one .mN oHan 87 respectively. The P value for the populations is 1.29, which is not significant. The mean difference analysis with t = 2.05 is significant at the 0.05 level, indicating a lower value for the epileptic population. There is no significant difference in the BUN values without persons 10 - 19 years old. Table 26 presents the results of the uric acid determinations. The means are 5.41 and 5.97 mg/100 ml, and the standard deviations are 1.15 and 1.43 for the IAE and IAS p0pu1ations age 10 and older, respectively. The F test is not significant, but t = 2.06 for the mean difference analysis is significant at the 0.05 level. The means are not significantly different when only individuals 20 years or older are compared, although F = 1.79 is significant indicating different variances in the two p0pu1ations. Apparently the level of uric acid is lower in the epileptic compared with the non-epileptic mentally retarded popula- tions. Table 27 presents the results of serum alkaline phosphatase determinations. The means are 137.00 and 88.16 mIU and the standard deviations are 80.20 and 35.74 for the IAE and the IAS populations age 10 and older, respectively. F = 5.04 is significant at the 0.01 level indicating that the two populations have different variances. The mean difference analysis with t = 4.07, which is significant at the 0.01 level, indicates a higher level of AP in the IAE .H6>6H HO.O o. onUHOHONHm .4 .HosoH H0.0 H4 oesusssamsm 4 .OHo msooz ON so>o mo m0fl ©HMflfimHm dam: Z .OHoo oHs: Essom sow poop ons mo mHmsHoco ocs .ON oHOos 89 .HosoH HO.O 44 oesuHHHamHm 4. .Ho>oH HO.O o. oaOUHsHeNHO 4 .OHo msmos ON so>o mo m6H H0.0 o4 Haaosmsamsm 4 .OHQ osmos ON so>o moO Osowcosm coo: z .omosmnmmonm ocHHmHHo HocHumoscH Essom sow umou ons mo mHmsHmcm oak .ON oHan 93 .Ho>oH H0.0 so scmonchHm 44 .HosaH HO.O .4 oaauHHHamHm 4 .OHo osmos ON so>o mo moO Osowcosm coo: z .ommsogmmonm ocHHoxHo so>HH Edsom sow poop on» mo mHmsHmco was .ON OHOOs 94 .HosoH HO.O 44 oaauHoHaNHm 44 .H6>6H HO.O 44 OOOOHHHONHO . .OHo msmos ON so>o mo moO Osowcosm coo: z .ommuonmmonm ocHHoxHo Essom sow EM>\xoe > smos ogs mo mHmsHocm one .Om oHQOH 95 deviations 0.15 and 0.09 for the IAE and IAS populations age 10 and older, respectively. The mean difference anal- ysis, t = 2.04, is significant at the 0.05 level. Although the mean difference is only 0.05, the small standard errors make this difference significant. When the 10 - 19 year age group is taken out of the population, the mean difference drops to 0.03, and t = 1.37 which is not significant. Those individuals who were high in the level of their total protein, AP or both in the original screen were re-tested, and the level of their blood ammonia was also measured. These individuals were also checked to see if they had epilepsy. Eleven out of 49 of these tested individuals had epilepsy (22.5 per cent), and 6 out of 11 epileptics had high ammonia (55.5 per cent) (Table 31). The frequency of epileptics in institution A is Stated by the medical personnel to be about 4 per cent. Table 32 shows the number of individuals observed in each ABC and Lewis blood group phenotype and the number of secretors and non-secretors in each population. The chi-square analyses indicate that there are no significant differences with respect to blood groups and secretor status between the IAE and IAS populations. In Table 33 the IAE and IAS samples are compared with a large sample of the general population from South- east Michigan (provided by Dr. E. Hackel) with respect to 96 O N O HN a< Os. cHososm OOH: H m O NH :Hooosa Hassoa .a< ONHO N O m HH a< Hesse: .cHouosm OOH: N22 OOH; Omz H4246: usoaaHHmm OOH; UHOaOHHam .UHHaOHHam Omz OWHO z mmsosm :H somoHHmo\oH:oEEm OOOHO anc .OHososa H4464 so a< OOH: OOH: nqu chmsom mo coHsanssmHO och .Hm oHOos 97 u. . .u x u. . .n x H m O OH N H M O NO N O OH N NN _ H m O OH soOHo Ono ON omm :oHuoHsmon msomno coHsoHomom msomno coHsoHsmom mHOOH OHOOOH>HOOH -aoz OO Os. < o On. a Hu.O.O HH.O.Nx Nn.O.O MO.HuNx ON w m N. 0H HOflHO Una OH omo coHumHsmoa mHOOH mHaaussHOaH -eoz m< OO. O o On. a [[1 .mcoHuoHsuom mOH OO.O .4 .OOOHOHOOHO 4 101 ON ON OO .0.0 NO.H .NN.N OO.H mHmsHOOO .OOHO amow NO.H OO.H ON.H a ON.O ON.H O0.0 OO.O NO.H ON.N sOOHo OO. OH omo :OHsoHsmom m