Main

The pathway of mitochondrial fatty acid β-oxidation represents an important source of energy, during periods of prolonged fasting or when there is increased energy demand due to febrile illness. Genetic disorders of the pathway have recently been recognized, and many of the specific enzyme and transport protein defects have been characterized(13). Typically, these disorders present as fasting intolerance with either hepatic, myopathic, or cardiomyopathic disease of childhood. Biochemically, most patients present with a hypoketotic hypoglycemic illness due to the failure of hepatic ketogenesis. L-3-Hydroxyacyl-CoA dehydrogenase or SCHAD is the penultimate enzyme in the pathway, and its deficiency has not been previously documented. We present the clinical features and biochemical studies of two patients with this newly recognized disorder of the β-oxidation pathway.

These cases are of particular significance as they are biochemically uncharacteristic of other β-oxidation defects in that they presented with fasting-induced vomiting and ketosis. In addition, one case presented with hypoglycemia. Fastingrelated ketotic hypoglycemia is a well documented idiopathic event in childhood(4), and SCHAD deficiency may represent the underlying biochemical defect in an undetermined number of patients with this condition.

CASE REPORTS

Patient 1. This boy was born at term to healthy unrelated Caucasian parents. Pregnancy was complicated by maternal hypertension the week before delivery, which was by cesarean section. Birth parameters were normal, and he showed normal growth and development during the 1st y of life. At 13 mo of age he had an episode of vomiting and poor feeding, becoming lethargic and dehydrated. He also had three seizures associated with hyponatremia. Hypoglycemia was not documented. He recovered rapidly with parenteral fluids and promethazine and was discharged after 2 d. One month later he had a similar episode again requiring hospitalization for 2 d. Work-up was remarkable for a significant alkalosis (venous pH of 7.7) and hyperammonemia(275 μmol/L). A third episode of lethargy while in the clinic did not require hospitalization. He has subsequently had multiple further episodes; most biochemical tests performed at the time of his third presentation were within normal limits including complete blood count, glucose, electrolytes, bicarbonate, anion gap, calcium, bilirubin, liver enzymes, ammonia, lactate, pyruvate, and plasma amino acids. Plasma total carnitine and acylcarnitines were normal, but there was elevated urine carnitine at 701 nmol/mmol creatinine (normal 50-200) most of which was conjugated (552). Urine organic acids collected during a period of illness revealed highly elevated 3-hydroxybutyrate and acetoacetate, indicating ketosis and modest elevation of dicarboxylic and 3-hydroxydicarboxylic acids of chain length C6-C14 (Fig. 1). These abnormalities were detected in numerous other urines collected when there was recurrence of the episodes of vomiting. The pattern of recurrent long-chain 3-hydroxydicarboxylic aciduria is not commonly associated with pure ketotic disorders (for example, diabetic ketoacidosis). In view of the history of fasting intolerance, and the biochemical findings of apparently impaired metabolism of 3-hydroxy fatty acids, we considered that a study of the pathway of β-oxidation was indicated. Initially, he did well on a high carbohydrate, low fat diet with advice to avoid prolonged fasting. Presently, at the age of 3.5 y he has had multiple recurrences of acute illness.

Figure 1
figure 1

Total ion chromatogram of trimethylsilyl derivatives of urinary organic acids from patient with SCHAD deficiency in sample collected during illness. Peak identities: 1, lactic acid; 2, 3-hydroxybutyric acid; 3, urea; 4, acetoacetic acid(peak1); 5, ethylmalonic acid; 6, acetoacetic acid(peak 2); 7, succinic acid; 8, fumaric acid; 9, internal standard; 10, glutaric acid; 11, 3-hydroxyadipic lactone acid; 12, adipic acid; 13, 3-methyladipic acid;14, 3-hydroxyadipic acid; 15, suberic plus unsaturated suberic acids; 16, 3-hydroxysuberic acid; 17, sebacic acid; 18, monounsaturated 3-hydroxysebacic acid; 19, 3-hydroxysebacic acid; 20, diunsaturated 3-hydroxydodecanedioic acid; 21 and 22, monounsaturated 3-hydroxydodecanedioic acid; 23, 3-hydroxydodecanedioic acid; 24, diunsaturated 3-hydroxytetradecanedioic acid; and 25 and 26, monounsaturated 3-hydroxytetradecanedioic acids.

Patient 2. This female patient was born to unrelated Caucasian parents. Pregnancy was uncomplicated. There was an obstetric history of a spontaneous abortion at 10-wk gestation. At birth the patient was noted to have dysmorphic features including micrognathia, hypertelorism, and downward slanting eyes. At d 5, she presented with hypertonia and myoclonic seizures which were related to hypocalcemia. A diagnosis of Di George syndrome was made, and chromosomal analysis confirmed a microdeletion in 22q11. At 2 mo she developed cytomegalovirus septicemia, during which she had several episodes of moderate neurologic deterioration with generalized hypotonia. At 3 mo hepatomegaly was noted with elevated transaminases. A liver biopsy was taken which excluded cytomegalovirus hepatitis, but electron microscopy revealed accumulated lipid droplets and enlarged mitochondria. At 12 mo her height and weight were below the third percentile for age.

At 15 mo and after an episode of gastroenteritis, she had an episode of hypoglycemia (1.8 mmol/L) in which ketones were detected in the urine. Urine organic acid analysis demonstrated similar findings to those in patient 1(Fig. 1). A skin biopsy was taken for fibroblast culture. Significant biochemical findings included reduced total plasma carnitine (22μmol/L). Acylcarnitine profiling using electrospray mass spectrometry/mass spectrometry revealed elevated C4 acylcarnitine. All other biochemical parameters were within normal limits. Magnetic resonance imaging of the head revealed bilateral, symmetrically demyelinating lesions of the periventricular white matter. Presently, the etiology of these abnormalities is unknown. The patient was placed on a high carbohydrate (70% total calories), low fat (20% total calories) diet, and the parents were advised to avoid prolonged fasting. She has gained 4 kg in 6 mo, has demonstrated improved muscle tone, and has caught up developmentally. Her liver function abnormalities have normalized.

METHODS

Skin fibroblasts were cultured in Dulbecco's modified eagles medium containing 10% fetal bovine serum, penicillin G (100 U/mL) streptomycin sulfate (100 U/mL), and amphotericin B (0.25 μg/mL). At confluence cells were either harvested from a 25-cm2 flask (for whole cell studies) or three 75-cm2 flasks (for mitochondrial isolation). The oxidation of[9,10-3H]palmitate and [9,10-3H]myristate was performed in intact fibroblasts according to the method of Manning et al.(5). The activities of LCHAD, SCHAD, and SKAT were assayed according to the method of Wanders et al.(6) which we have adopted for the Cobas Fara analyzer(7) (Roche, Montclair, NJ). Mitochondria were isolated from cells harvested by trypsinization. The cells were washed with Dulbecco's PBS and resuspended in homogenization buffer containing 150 mmol/L potassium chloride, 5 mmol/L Tris-HCl, pH 7.2, and homogenized in a glass homogenizer for 10 passes. The homogenate was centrifuged at 600 × g for 10 min, and the pellet was discarded. The mitochondrial fraction was isolated as the pellet obtained after centrifugation at 3300 × g for 10 min. The mitochondria were resuspended in 100 mmol/L potassium phosphate buffer, pH 6.3, 0.1 mmol/L dethiothreitol, 0.1% Triton X-100 and disrupted by ultrasonication before enzyme assay. This fraction contained no activities to peroxisomal (acyl-CoA oxidase) and lysosomal(β-N-acetylglucosaminidase) enzyme markers. Lactate dehydrogenase, a cytoplasmic enzyme marker was detected at a specific activity of approximately 10% of the whole cell preparation, indicating some cytoplasmic contamination of the final mitochondrial preparation.

RESULTS

The oxidation of tritiated palmitate and myristate was within the normal range for both patients; palmitate oxidation; picomoles/min/mg of protein; patient 1 = 33.5, patient 2 = 30.0, controls = 35.6 ± 8.2 (SD,n = 12) myristate oxidation; picomoles/min/mg of protein; patient 1= 37.5, patient 2 = 31.2, controls = 24.7 ± 5.6 (SD, n = 12).Table 1 shows the results for SCHAD and two mitochondrial control enzyme assays in disrupted whole cell fibroblast preparations, or isolated mitochondria from cultured skin fibroblasts from normal controls, the two patients described here, the mother of one of our patients (an obligate heterozygote), and from disease controls with LCHAD and SKAT deficiency. In the disrupted whole cell assay there was a reduction in SCHAD activity in cells from patient 1 to approximately 60% of normal. The LCHAD activity in this patient was also reduced but not to a level consistent with that seen in patients with LCHAD deficiency. In isolated mitochondria the reduction in SCHAD activity was pronounced with 5.3% residual activity in patient 1 and 6.6% in patient 2, confirming the diagnosis of mitochondrial SCHAD deficiency. The SCHAD activity in the obligate heterozygote was intermediate between the patient levels and the normal controls.

Table 1 Mitochondrial enzyme studies in cultured skin fibroblasts

DISCUSSION

We present the clinical and biochemical findings in two patients with SCHAD deficiency, a newly recognized defect of mitochondrial β-oxidation of fatty acids. Both patients presented in infancy with episodes of starvation-related ketosis. Hypoglycemia was documented in one patient making the history compatible with a well documented idiopathic syndrome of recurrent illness in children. In patients with ketotic hypoglycemia, starvation or the introduction of a high fat, low carbohydrate diet results in ketonemia and hypoglycemia(4, 8, 9). Both of our patients responded rapidly to treatment with glucose as do patients with ketotic hypoglycemia and other β-oxidation defects, and patient 2 is now thriving on a low fat, high carbohydrate diet. The diagnosis may have been missed in patient 1, as the working diagnosis included Munchausen's disease by proxy. Toxicologic investigations were made at the local hospital despite ongoing studies by a specialized metabolic clinic. Di George syndrome in patient 2 may be unrelated to this particular biochemical abnormality, as hypoglycemia and ketosis are not recognized features of Di George syndrome, which is a dominantly inherited disorder with absence of thymus and parathyroids, dysmorphic focal features, and with linkage to chromosome 22(10). The chromosomal location for SCHAD is at present unknown.

MCAD deficiency, another disorder of mitochondrial β-oxidation has also been implicated as a possible cause of some cases of ketotic hypoglycemia(2, 3, 11). It is reasonable to make this clinical association as SCHAD and MCAD defects are both sufficiently far down the β-oxidation pathway to be able to generate some ketones from long-chain fatty acids(11). Clinically, SCHAD deficiency should be associated with fasting intolerance, as is well documented in the many reported cases of MCAD deficiency(2, 3).

Important clues which led us to the diagnosis of SCHAD deficiency in our patients came from the abnormal organic aciduria, with evidence of partial fatty acid oxidation in samples collected during metabolic decompensation(Fig. 1). These organic acid profiles demonstrated ketonuria, with 3-hydroxybutyric acid being the most prominent peak, and saturated and unsaturated 3-hydroxydicarboxylic aciduria of chain lengths C6-C14 with relatively larger amounts of the shorter chain length intermediates. In our experience, this pattern is not seen in other ketotic conditions, such as diabetic ketoacidosis or starvation-induced ketosis, and is essentially a mirror-image of the pattern seen in LCHAD deficiency(12). Enzyme analysis in whole skin fibroblast preparations revealed reduced SCHAD activity, but with an unusual high residual activity. It was necessary to isolate mitochondria to demonstrate the enzyme defect. We assume that the high level of residual enzyme activity in whole cell preparations relates to the presence of extramitochondrial enzyme(s) capable of converting acetoacetyl-CoA to 3-hydroxybutyryl-CoA.

High residual SCHAD activity was also observed in a patient recently reported to have muscle-specific SCHAD deficiency(13). This was a 16-y-old girl who presented with recurrent myoglobinuria and encephalopathy and who had normal SCHAD activity in skin fibroblasts. To date, there is no strong evidence to suggest tissue-specific isoforms of SCHAD, and it is not possible to confirm the diagnosis in that patient in the absence of data from isolated muscle mitochondria. Our patients presented at a much younger age and do not have evidence of significant muscle involvement. The demonstration of intermediate enzyme activity in the mother of one of our patients, an obligate SCHAD heterozygote, supports the expectation that SCHAD deficiency, as all other known fatty acid oxidation disorders, is an autosomal recessive disease.

The lengthy investigations to exclude Munchausen's disease by proxy in patient 1 serve as an important reminder that physicians need to consider metabolic causes in the differential diagnosis of hypoglycemia and/or atypical illnesses in children(14).

Ketotic hypoglycemia is the most common syndrome of idiopathic hypoglycemia in infancy. Once recognized, the treatment directed at preventing the starvation-induced ketoacidosis is relatively simple and effective. These cases indicate that SCHAD deficiency is responsible for at least some cases of ketotic hypoglycemia and advocate appropriate metabolic and enzymatic studies for confirmatory purposes. We hope that awareness of this disorder and routine organic acid screening of children presenting with recurrent ketotic hypoglycemia could lead to determine the true incidence of SCHAD deficiency in this population.