Main

Carnosine (β-alanyl-L-histidine) is a dipeptide which is abundant in the skeletal muscle of most animals. Anserine(β-alanyl-1-methyl-L-histidine) is a component of skeletal muscle in birds, but is absent from human tissue(1, 2). Both of these dipeptides are hydrolyzed by an enzyme present in serum, carnosinase(EC 3.4.13.3).

Carnosinase exists in two isoforms(3). One form(tissue carnosinase; Mr = 90 000) is present in various tissues including liver, kidney, and spleen, but not skeletal muscle. The other form (serum carnosinase; Mr = 155 000) is found primarily in serum, but also in brain and spinal fluid(4). These enzymes differ not only in their distribution and molecular weight, but also in specificity. One very important difference is that serum carnosinase is able to hydrolyze homocarnosine(GABA-L-histidine), whereas tissue carnosinase is not. Consequently, profound elevations in cerebrospinal fluid homocarnosine have been found in children with serum carnosinase deficiency(5).

Deficiency of serum carnosinase has been described in several sibships in conjunction with tremor, myoclonic seizures, hypotonia, and profound psychomotor retardation(6–10). The clinical presentation of this disorder is quite variable, however, and there exists at least one detailed report of low enzyme activity in an apparently normal sibling of an index case(11). We describe the case of a 30-mo-old white female patient referred for evaluation of developmental delays with hypotonia and a tremor. The coexistence of this rare enzymatic deficiency and a gross deletion on the long arm of chromosome 18 suggests that the gene for serum carnosinase resides in this portion of the genome.

METHODS

Case report. P.E.Z. is a 30-mo-old white girl with tremor, hypotonia, and global developmental delays. She was born at term by vaginal delivery and weighed 6 pounds and 10 ounces. Her development appeared normal until 6 mo of age, but began to plateau by 12 mo. At that time, she developed hypotonia and tremor and was found to have a urinary tract infection. The family's diet is strictly lacto-ovo-vegetarian, and the child had not consumed any meats until this time. The family introduced meats into her diet in hopes of improving her symptoms and growth. Shortly thereafter, she developed a tremor, which worsened for approximately 3-4 mo but seemed to improve after the elimination of meats from her diet. Her development has improved, but has not returned to normal for age. Physical examination reveals her height, weight, and head circumference are at the fifth percentile. A neurologic examination was notable for a tremor (accentuated by intent), ataxia, and mild generalized hypotonia.

Magnetic resonance imaging revealed normal brain anatomy. Blood chemical analyses and hematologic profiles were essentially normal. Extensive metabolic testing demonstrated no abnormalities except for quantitative urine amino acids that revealed striking hypercarnosinuria and anserinuria. As these data suggested the diagnosis of serum carnosinase deficiency, this child has been maintained on a low carnosine diet with little change in her clinical condition. Urinary carnosine has remained modestly elevated in the absence of detectable hypercarnosinemia.

Biochemical analysis. Twenty-four-hour urine collections from the patient, family members, and 18 pediatric control subjects were assayed for amino acids and creatinine using standard techniques(12). Serum carnosine quantitation was performed using a Beckman 6300 high performance amino acid analyzer. Serum carnosinase activity was assayed in the proband's family, as well as 69 normal control subjects varying in age from 6 mo to 44 y. The stability of serum carnosinase is not well studied, but it has been found to be stable for several weeks, if samples are kept frozen(13). For this reason, serum samples were frozen and stored at -70°C until the morning of assay. Interassay reliability for frozen samples assayed 2 mo apart was 6.7%. In contrast, carnosinase activity decreased by an average of 16.8% when samples were stored for a period of 14 d at 4°C before assay.

The serum carnosinase assay was a modification of the method of Lenny et al.(13). Twenty-five microliters of serum were added to 0.15 mL of a solution containing 2.5 mmol/L CdCl2 and 15 mmol/L sodium citrate. To this 0.225 mL of 125 mmol/L NH4OH-HCl buffer(pH 8.5) was added. After 10 min at 30°C, 0.10 mL of 100 mmol/L carnosine in NH4OH-HCl buffer (pH 8.5) was added, and the reaction was allowed to proceed for 20 min at 30°C. The reaction was terminated with 0.50 mL of 0.6 N trichloroacetic acid, and insoluble material was precipitated by centrifugation at 800 × g for 10 min. Histidine content was measured in 0.50 mL of the trichloroacetic acid supernatant using the method of Ambrose et al.(14). The sensitivity of our measurements was increased 5-fold over existing techniques(13), using a very sensitive fluorescence spectrophotometer (Perkin-Elmer LS-5).

Cytogenetic analysis. Peripheral blood was obtained for chromosome analysis on P.E.Z. and her parents. Lymphocytes were cultured at 37°C for 72 h in tissue culture medium RPMI 1640 withN-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid buffer and 20% FCS. Ethidium bromide was used to achieve chromosome elongation(15). Cells were harvested, and slides were prepared and aged by standard methods. Chromosome morphology was examined by G-banding.

Molecular analysis. Microsatellite markers were amplified by the polymerase chain reaction. Each reaction contained 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 40 ng of each primer, 100 μM dATP, dGTP, and dTTP, 2.5 μM unlabeled dCTP, 16.7 nM 32P-labeled dCTP(3,000 Ci/mmol), and 0.45 U of Amplitaq DNA polymerase (Perkin-Elmer), in a total volume of 15 μL. Samples were denatured at 94°C for 5 min, then 27 cycles of 94°C for 1 min, 55°C for 2 min, and 72°C for 1 min were finally followed by a 6-min final extension period at 72°C. Polymerase chain reaction products were electrophoresed on a 6% acrylamide/7 M urea gel and visualized by autoradiography with XAR (Kodak) film.

RESULTS

Quantitative amino acid analysis of the patient's urine revealed striking hypercarnosinuria and anserinuria (Table 1). When all dietary sources of meat were eliminated, hypercarnosinuria persisted. The introduction of white poultry meat into the diet, for just 48 h, was accompanied by a marked elevation in anserine, in the presence of low 1-methylhistidine (a metabolite derived primarily from serum carnosinase action upon anserine). The patient's parents did not exhibit hypercarnosinuria on a meat-free diet. However, urine amino acid analysis of the patient's brother demonstrated moderately increased carnosine excretion on a vegetarian diet.

Table 1 Urinary excretion of carnosine, anserine, and 1-methylhistidine in 24-h urine collections from proband on various dietary regimens
Full size table

Fifteen chromosome spreads were analyzed from G-banded preparations and revealed a terminal deletion involving the long arm of chromosome 18 with most likely breakpoint 18q21.3 (Fig. 1). The karyotype was designated 46,XX, del (18)(q21.3). Parental karyotypes were normal.

Figure 1
figure 1

High resolution cytogenetic analysis of chromosome 18 in the proband and her parents. The deletion consists of all material distal to 18q21.3.

Full size image

The breakpoint of the deletion was determined to lie within an 8-centiMorgan region, between the dinucleotide repeat markers D18S38 and D18S42(16). The patient had two distinct alleles, and was therefore not deleted, for the D18S38 marker in 18q12.3 (left panel,Fig. 2), as well as the more proximal 18q markers D18S35, D18S34, and the short arm markers D18S40 and D18S59 (data not shown). These chromosome 18 markers, which displayed biparental inheritance, were all consistent with the presumed paternity of this patient. The patient had one allele at the D18S42 marker in 18q22.1, corresponding to one of the two maternal alleles seen for this marker (right panel, Fig. 2). There was no evidence of a paternal contribution at this marker, or the more distal 18q markers D18S43, MBP, and D18S70 (data not shown). These studies reveal that the deletion was present on the paternal chromosome 18. Paternity was further investigated by examining highly polymorphic markers on chromosomes 11, 13, 16, and 22. The patterns observed were consistent with paternity for all of these markers as well.

Figure 2
figure 2

Autoradiographs from the patient (P), her mother (M), and her father (F) using nucleotide repeat markers D18S38 (left) and D18S42 (right). For D18S38, the patient has inherited allele 1 from her father and allele 2 from her mother, suggesting that the allele breakpoint is distal to this marker at 18q21.31. In the case of D18S42, the patient has inherited allele 3 from her mother but has no paternal allele, indicating a deletion on the paternal chromosome 18, which includes this portion of 18q22.1.

Full size image

Carnosinase activity was detectable throughout a wide range of activities. Normal control values varied from 0.42 to 40.5 μmol/mL/h. Serum carnosinase activity in adults (23.2 ± 10.6 μmol/mL/h; mean ± SD) is somewhat variable (Fig. 3). Enzyme activity in the proband's father (19.4 ± 0.36 μmol/mL/h) is in the middle of the adult normal range. Moreover, the mother's carnosinase activity (3.7 ± 0.24 μmol/mL/h) is in the lowest fifth percentile, consistent with the carrier state for this deficiency. A log plot of carnosinase activityversus age (Fig. 4) demonstrates that the activity of this enzyme increases with age throughout childhood, approaching adult normal values at about age 14 y. Calculation of a prediction interval for enzyme activity in children reveals that the proband and, to a lesser extent, her unaffected sibling lie outside of values which would be predicted with 95% confidence. The proband's carnosinase activity (0.16 ± 0.07μmol/mL/h) is much lower than similar aged and even younger control subjects. Of note, the brother's activity (0.65 ± 0.10 μmol/mL/h) is comparable with values previously considered to be consistent with carnosinase deficiency(11).

Figure 3
figure 3

Serum carnosinase activity plotted against age in 69 controls (○) as well as the proband's family (•).

Full size image
Figure 4
figure 4

A log/log plot of serum carnosinase activityvs age clearly differentiates the proband from normal controls at all ages tested. The thin line represents the 95% prediction interval based upon this normative sample. Mildly reduced activity in the patient's brother suggests that he is heterozygous for carnosinase deficiency.

Full size image

DISCUSSION

We report the case of a 2.5-y-old child with serum carnosinase deficiency and a concomitant deletion in the terminal portion of the long arm of chromosome 18. To our knowledge, this is the first report of a chromosomal abnormality in association with this metabolic disorder. The presence of normal carnosinase activity and normal urinary carnosine excretion in the patient's father suggests he does not harbor a defective gene for this rare autosomal recessive disorder. Unfortunately, carnosinase activity in adults demonstrates considerable variability, making carrier detection difficult. However, the presence of low/normal carnosinase activity in the patient's mother would be consistent with the carrier state. Moreover, the moderate hypercarnosinuria and low carnosinase activity exhibited by the patient's brother suggest that the carrier state for this condition presents a normal phenotype despite hypercarnosinuria.

These findings are consistent with a unique genetic phenomenon leading to autosomal recessive inheritance of this disorder. We postulate that this child inherited a mutant allele from her mother and that this defect remained unbalanced by a normal paternal allele due to the gross chromosomal deletion in the long arm of chromosome 18. In favor of this assertion is the demonstration through molecular analysis that indeed the 18 q-is of paternal lineage. Furthermore, molecular analysis clearly established the paternity of this child. These data, cumulatively, suggest that the gene encoding this enzyme likely resides in this region of chromosome 18. Potential alternative explanations for these data include the rare possibilities of uniparental disomy or spontaneous mutation of the carnosinase gene in the paternal germ cell. As the paternal germ line has clearly been demonstrated to be the origin of the overt chromosomal deletion, it seems unlikely to postulate a spontaneous mutation in addition to this deletion.

Lenny et al.(13) previously reported a sensitive fluorometric assay for serum carnosinase activity, which takes advantage of the fact that cadmium is an activator of this enzyme in vitro. By altering the incubation conditions and using a very sensitive fluorescence spectrophotometer, we were able to improve on the sensitivity of this assay by approximately 5-fold. This led to reliable segregation of the proband from similarly aged, and even younger, normal children.

The significance of the improved sensitivity of this assay is magnified by a number of reports of “low” carnosinase activities in unaffected family members of index cases(8, 11). Such reports have led to the suggestion that clinical manifestations do not necessarily accompany the biochemical phenotype. We suspect that a number of carriers for this disorder were erroneously labeled as deficient, due to the inability of previous methods to clearly differentiate heterozygotes from homozygotes. Careful analysis of these reports reveals that the subjects, in fact, had only mildly reduced levels of carnosinase activity and were labeled as deficient primarily due to the presence of hypercarnosinuria. Moreover, the degree of hypercarnosinuria evident in these subjects was considerably milder than that demonstrated in our patient and the preponderance of previously reported cases.

In the present case study, a mild degree of hypercarnosinuria was evident in the patient's phenotypically normal brother. Although hypercarnosinuria on a meat-free diet was not demonstrable in the patient's mother, both brother and mother had relatively low carnosinase activity. These facts highlight the hazards of interpretation of urinary carnosine excretion and the importance of a sensitive serum assay in the definitive diagnosis of serum carnosinase deficiency. Unfortunately, normal adult carnosinase activity demonstrates considerable variability, making carrier detection difficult even with this improved sensitivity. As a result, we cannot completely rule out the possibility that the mother's and brother's marginal carnosinase activity are statistical coincidence. This possibility seems unlikely in light of the brother's hypercarnosinuria. In any case, the absence of carrier status in these individuals would not alter the basic premise, which is that the gene for serum carnosinase resides on the long arm of chromosome 18.

The presence of neurologic dysfunction in this child, and others with serum carnosinase deficiency, gains potential significance in light of the recent identification of high levels of this enzyme in the brains of higher primates(4). Interestingly, this enzyme is absent from the brains of lower primates and most other mammals. As serum carnosinase is essential to the conversion of homocarnosine to GABA and histidine, it is likely that one of this enzyme's most vital functions is in providing a source of the neurotransmitter GABA in the CNS. The fact that homocarnosine is found in high concentration (>1 mM), with a distinct pattern of localization in the CNS(17), lends further credence to this function. We postulate that homocarnosine may represent a reservoir of GABA which is unavailable to carnosinase-deficient patients, and that hypercarnosinuria merely serves as a marker for this profound neurologic disorder.