Metabolic disorders of fetal life: Glycogenoses and mitochondrial defects of the mitochondrial respiratory chain

Seminars in Fetal & Neonatal Medicine 16 (2011) 181e189

Contents lists available at ScienceDirect

Seminars in Fetal & Neonatal Medicine journal homepage: www.elsevier.com/locate/siny

Metabolic disorders of fetal life: Glycogenoses and mitochondrial defects of the mitochondrial respiratory chain S. DiMauro*, C. Garone Department of Neurology, Columbia University Medical Center, New York, NY, USA

s u m m a r y Keywords: Fetal presentation Glycogen Glycogen storage diseases Mitochondria Mitochondrial encephalomyopathies Mitochondrial respiratory chain Neonatal presentation

Two major groups of inborn errors of energy metabolism are reviewed eglycogenoses and defects of the mitochondrial respiratory chain e to see how often these disorders present in fetal life or neonatally. After some general considerations on energy metabolism in the pre- and postnatal development of the human infant, different glycogen storage diseases and mitochondrial encephalomyopathies are surveyed. General conclusions are that: (i) disorders of glycogen metabolism are more likely to cause ‘fetal disease’ than defects of the respiratory chain; (ii) mitochondrial encephalomyopathies, especially those due to defects of the nuclear genome, are frequent causes of neonatal or infantile diseases, typically Leigh syndrome, but usually do not cause fetal distress; (iii) notable exceptions include mutations in the complex III assembly gene BCS1L resulting in the GRACILE syndrome (growth retardation, aminoaciduria, cholestasis, iron overload, lactic acidosis, and early death), and defects of mitochondrial protein synthesis, which are the ‘new frontier’ in mitochondrial translational research. Ó 2011 Published by Elsevier Ltd.

1. Introduction Sensu stricto, all mendelian or maternally inherited inborn errors of metabolism are fetal disorders because most genetic defects are expressed prenatally and only very few are developmentally regulated, such that the mutated mature enzyme has a wild-type fetal counterpart. However, the expression of a mutated gene in utero does not necessarily mean that the fetus is ‘clinically’, i.e. noticeably, affected. Which are the clinical signs that should alert us of an ongoing fetal disease? Not too many: decreased fetal movements (usually compared by the mother to a previous normal pregnancy); intrauterine growth retardation (IUGR); abnormal heart sounds. Various instrumental studies allow us to monitor fetal development more closely: ultrasonography may show polyhydramnios and enlarged cerebral ventricles and magnetic image resonance (MRI) can detect micro- or macrocephaly and agenesis of the corpus callosum. At times, the mother acts as a ‘detoxifier’ by metabolizing or excreting potentially toxic water-soluble compounds produced by the fetus and transferred through the placenta into the maternal circulation.1 More often, however, it is the mother whose symptoms alert us to a fetal metabolic disorder, such as pre-eclampsia

* Corresponding author. P&S, Room 4-424B, 630 West 168th Street, New York, NY 10032, USA. Tel.: þ1 212 305 1662; fax: þ1 212 305 3986. E-mail address: [email protected] (S. DiMauro). 1744-165X/$ e see front matter Ó 2011 Published by Elsevier Ltd. doi:10.1016/j.siny.2011.04.010

(hypertension, edema, proteinuria), eclampsia (severe hypertension, encephalopathy, seizures), HELLP (hemolysis, elevated liver enzymes, low platelets) syndrome, and acute fatty liver of pregnancy (AFLP). This spectrum of disorders is often seen in mothers carrying a fetus with a fatty acid oxidation disorder (FAOD).1 The ultimate fetal disorder is the one that causes intrauterine lethality: although fetal wasting in metabolic diseases has not been investigated systematically, anecdotal experience from mitochondrial encephalomyopathies suggests that early spontaneous abortions or neonatal deaths are common in both nuclear gene defects, such as the cardio-encephalomyopathy caused by SCO2 mutations,2 and in mtDNA mutations.3 It was also suggested 16 years ago by Wayne Fenton that genetic defects of the general mitochondrial protein importation machinery would be incompatible with life.4 His foresight seems largely validated by the handful of patients reported, most of whom in fact died in infancy or early childhood.5e8 Thus, while some inborn errors of metabolism are manifest in utero, most present at or soon after birth. The spectrum of clinical phenotypes is very wide and depends on multiple factors, including the consequence of the genetic defect (intoxication, storage, or defective energy production)9 and the type and severity of the mutation.10 Because it is obviously impossible to cover all metabolic disease in a brief review, we will confine ourselves to the two subjects with which we are more familiar, the glycogenoses and the mitochondrial diseases. Fetal disorders of fatty acid oxidation have been reviewed thoroughly and relatively recently.1,9

182

S. DiMauro, C. Garone / Seminars in Fetal & Neonatal Medicine 16 (2011) 181e189

2. Disorders of glycogen metabolism 2.1. Infantile glycogen storage disease type II (GSD II, Pompe disease) This is the most severe of the three clinical forms of GSD II. The discriminator between the infantile and later onset (juvenile or adult) forms of GSD II is that the infantile form is multisystemic whereas both later onset forms are exclusively or predominantly myopathies.11,12 Infantile acid maltase deficiency (Pompe disease) manifests in the first weeks or months of life with diffuse hypotonia and weakness, giving these infants a ‘rag doll’ appearance (floppy baby syndrome). Muscle bulk may be increased, however, and macroglossia is occasionally seen. There is massive cardiomegaly and less severe hepatomegaly. Despite their extreme weakness, these infants are usually alert and interested in their environment. In part at least, the weakness is neurogenic, due to the severe involvent of motor neurons in the spinal cord.13e15 Respiratory muscle weakness increases susceptibility to pulmonary infections, and death due to cardiac or respiratory failure occurs invariably before 2 years of age and usually within the first year. A retrospective, multinational, and multicenter study of 168 patients has defined the natural history of the infantile variant.12 Median ages were: onset, 2.0 months; diagnosis, 4.7 months; ventilator dependency, 5.9 months; death, 8.7 months. The main clinical features, in order of frequency, were: cardiomegaly (92%), hypotonia (88%), cardiomyopathy (88%), respiratory insufficiency (78%), weakness (63%), feeding difficulties (57%), and failure to thrive (53%). Despite the profound weakness often present at birth or soon thereafter, this does not appear to be a ‘fetal’ disease and arthrogryposis multiplex congenita (AMC) is not reported. The difference in clinical expression and pathology between infantile and later onset forms of acid maltase deficiency has been attributed to the presence of a small but crucial amount of residual acid maltase activity in childhood and adult cases but not in the infantile form. The difference in residual activity, first observed in muscle specimens,16 is more evident in fibroblast and muscle cultures from patients with the different variants.17 There is a generally good correlation between ‘molecular severity’ of mutations (e.g. nonsense or frame shift mutations on one side and missense mutations on the other) and severity of the clinical presentation.15,18,19 Until recently, prognosis was dismal in infantile acid maltase deficiency. However, enzyme replacement therapy (ERT) has changed considerably prognosis and life expectancy.20 Four infants with Pompe disease were treated with spectacular results: although one patient died of an intercurrent infection at 4 years of age (well beyond the limit of 2 years that characterizes the natural history of the disease), all four patients showed remarkable clinical improvement in motor and cardiac function and parallel improvement in muscle morphology.21,22 Two main factors seem determinant in the success of ERT in infantile Pompe disease: (i) early onset of therapy; and (ii) presence of cross-reacting immunological material (CRIM), that is, of small amounts of enzyme protein that protect patients from immunological reaction to the recombinant human enzyme.23

2.2. Glycogen storage disease type III (GSD III, debrancher deficiency, CorieForbes disease) GSD III does not manifest before birth and, during infancy and childhood, is characterized by hepatomegaly, hypoglycemia, hyperlipidemia, and growth retardation. Both hepatomegaly and

hepatic symptoms tend to improve with age and usually resolve after puberty. 2.3. Glycogen storage disease type IV (GSD IV, branching enzyme deficiency, Andersen disease) Although traditionally considered a hepatic disease of infancy and childhood, with hepatosplenomegaly, progressive cirrhosis, and chronic hepatic failure, in fact GSD IV has been associated with a wide spectrum of clinical phenotypes, affecting, in varying combination, liver, heart, skeletal muscle, and brain. Onset varies from fetal life to late adult years. This is surprising because the glycogen branching enzyme (GBE) is a single polypeptide encoded by one gene (GBE1). GBE deficiency results in the deposit of an amylopectin-like polysaccharide that has fewer branching points and longer outer chains than normal glycogen and is known as polyglucosan. Polyglucosan is periodate/Schiff (PAS) positive and only partially digested by diastase, which makes it easily recognizable in various tissues and offers an important clue to the correct diagnosis. We will focus on the fatal infantile neuromuscular form of GSD IV, which has been underdiagnosed, judging from the flurry of recent papers. As recognized in a seminal paper published in 2004,10 there are two main infantile presentations. The first is a perinatal (meaning both pre- and post-natal) disorder dubbed ‘fetal akinesia deformation sequence’ or FADS, characterized by multiple congenital contractures (arthrogryposis multiplex congenital), hydrops fetalis, pulmonary hypoplasia, craniofacial abnormalities, IURG, abnormal amniotic fluid volume, and perinatal death. The second, labeled rather generically ‘congenital’, should probably be called ‘fatal infantile’, as it presents at or soon after birth with hypotonia, muscle wasting, neuronal involvement, inconsistent cardiomyopathy, and early death. FADS is a prototypical ‘fetal syndrome’ and has heterogeneous etiology, including neurogenic or myopathic disorders, restrictive dermopathy, teratogen exposure, and intrauterine constraint.24 Of the eight patients with GSD IV reported by Bruno et al.,10 three had FADS, three had the congenital form, and two had childhood myopathy. Interestingly, there was a good correlation between ‘molecular severity’ and clinical severity, which has been confirmed in several subsequent patients. It is becoming increasingly clear that patients with congenital GSD IV present a clinical continuum from FADS to a rapidly fatal congenital multisystem disorder dominated by profound hypotonia, respiratory failure, and inconsistent cardiomyopathy.10,25e31 All these patients showed signs of fetal distress, including decreased fetal movements, polyhydramnios, bradycardia, and arthrogryposis. Detailed neuropathology was performed in a girl who died at 3 months, during which she depended on mechanical ventilation and nasogastric feeding.28 She had prenatal symptoms (bradycardia) requiring cesarean section at 33 weeks of gestation. Her postnatal symptoms included hypotonia, arthrogryposis of knees and ankles, bilateral ptosis, and roving eye movements. Echocardiogram showed cardiomyopathy and computed tomography of the brain at 11 days of age showed diffuse brain atrophy and normal ventricles. PAS-positive polyglucosan inclusions were evident in neurons of basal ganglia and thalamus, oculomotor and pontine nuclei, and periaqueductal neurons. In the medulla, polyglucosan deposits were noted in the hypoglossal nucleus, the dorsal motor nucleus of the vagus, and the nucleus ambiguous. Similar findings were reported in two more infants.25,27 The motor neurons of the spinal cord are also severely affected,32 explaining how one of the patients we studied was initially diagnosed as spinal muscular atrophy type I (SMA I) until mutations in the SMN1 gene were ruled out.25

S. DiMauro, C. Garone / Seminars in Fetal & Neonatal Medicine 16 (2011) 181e189

Prenatal diagnosis is obviously of great importance to avoid repeated infantile deaths and can be established by ‘traditional’ gene sequencing in amniocytes or chorionic villi.30 However, an alternative diagnostic method e and further evidence of the generalized prenatal nature of the disease e was based on histochemical and ultrastructural studies of the placenta in a fetus at risk.33 There is no effective therapy for any of the clinical forms of GSD IV. However, it is notable that a child with the infantile presentation (congenital hypotonia and arthrogryposis multiplex) developed severe liver disease and received a liver transplantation at 3 years of age: at the time of the latest report, he was 6 years old and had mild hypotonia and joint contractures.10,34 2.4. Glycogenosis type V (GSD V, McArdle disease) Although McArdle disease is typically a disorder of adolescents or young adults, an extremely rare variant was described in detail in two sisters who were born after normal pregnancy and had normal perinatal periods, but presented in early infancy with respiratory distress, apneic episodes, and cyanosis. They were alert and appeared to have normal hearing and vision. Their serum creatine kinase was slightly elevated and electromyography showed fibrillations and increased number of polyphasic potentials in all muscles tested. The first sister died at 2 months and the second at 4 months of age. Muscle morphology and biochemistry and acrylamide slab-gel electrophoresis of phosphorylase isozymes in the heart of the second sister documented unequivocally an isolated defect of myophosphorylase.35,36 The second report of fatal infantile myophosphorylase deficiency concerns an infant girl who had evidence of intrauterine onset because she had congenital hypotonia and arthrogryposis multiplex congenita. Despite ventilatory support, this child lived only 16 days.37 Although neither biochemistry nor molecular genetic studies were performed, histochemistry in a muscle biopsy and in multiple postmortem tissues was diagnostic and nonmuscle tissues were not affected. A third case had the features of sudden infant death (SIDS) in a 3-month-old girl who was born at term after a normal pregnancy and had developed normally until the ‘crib death’ event.38 The different clinical presentations in these children are unexplained and molecular genetic studies in the first two sisters and in the infant with SIDS showed that they were homozygous for the mutation (R50X) that is prevalent among Caucasian patients with typical McArdle disease.38,39 2.5. Glycogenosis type VII (GSD VII, phosphofructokinase deficiency, Tarui disease) Typical GSD VII is a myopathy, often associated with compensated hemolytic anemia, presenting in adolescence or young adulthood with exercise intolerance, cramps, and myoglobinuria. However, an infantile form has been reported in a dozen patients between 1987 and 2008. All infants were severely hypotonic at birth and a few developed joint contractures either in utero40e42 or postnatally.43,44 Decreased fetal movements were noted in two pregnancies41,42 and polyhydramnios in one.42 In all but two cases,42,45 death occurred in infancy or early childhood due to pulmonary failure. Most children showed evidence of multisystem involvement, including seizures, cortical blindness, developmental delay, dysmorphic features, and corneal ulcers. The encephalopathy was documented by neuroradiology or neuropathology, which showed dilated ventricles and cortical or cerebellar atrophy.40,44e47 Because of the early onset, multisystem involvement, and lack of any molecular evidence of mutations in the PFKM gene, the

183

infantile variant of phosphofructokinase deficiency appears to be a separate entity from GSD VII, and its genetic basis (or bases) remain to be clarified, despite evidence that a transgenic PFKM-null mouse mimics the infantile more than the typical muscular form of the human disease.48 2.6. Phosphorylase b kinase (PHK) and AMP-activated protein kinase (AMPK) deficiencies These two enzyme defects can be associated in the context of fetal disorders because most e and probably all e reported cases of fatal infantile cardiopathy due to PHK deficiency were, in fact, due to mutations in the g2-subunit of AMPK.49,50 In this exquisitely fetal disease, prenatal bradychardia led to cesarean section in five of six patients, while the sixth patient developed cardiorespiratory failure within 30 min of life. All infants had massive cardiomegaly and needed ventilatory support during their brief lives (age at death varied between 11 days and 5 months). Additional signs in two children included macroglossia, renal abnormalities, dysmorphic features, and hydrocephalus. In all children, there was massive glycogen storage in cardiac muscle whereas skeletal muscle was more moderately affected only in three infants. All had variably severe PHK deficiency in the heart and other tissues, but no molecular defect in any of the genes encoding the different PHK isozymes. On the other hand, two novel heterozygous mutations (R531Q and R384T) were identified in the gene (PRKAG2) encoding the g2-subunit of AMPK, a key regulatory enzyme of energy metabolism. Both mutations impair the binding of AMP and ATP to AMPK and are functionally severe. Heterozygous mutations in this same gene are known to cause autosomal dominant cardiopathy with WolffeParkinsoneWhite syndrome in adults: the sporadic nature of the cardiopathy in these children suggests that they had severe de-novo mutations.49,50 A number of questions remain unanswered, including the variable involvement of extracardiac tissues and the pathogenic mechanism of the PHK pseudodeficiency. 3. Mitochondrial disorders This is an extremely heterogeneous group of diseases sharing in common the fact that they all ultimately impair the function of the mitochondrial respiratory chain, the ‘business end’ of oxidative metabolism, thus presumably decreasing ATP production and causing excessive accumulation of reactive oxygen species (ROS).51 The approximately 80 subunits of the respiratory chain are under dual genetic control, 13 of them being encoded by mitochondrial DNA (mtDNA) and the rest by nuclear DNA (nDNA). Hence, a simple genetic classification distinguishes disorders due to mutations in mtDNA (usually maternally inherited) from those due to mutations in nDNA (mendelian traits). Before reviewing specific disorders in the two groups, a few general considerations are in order regarding the likelihood of mitochondrial disorders to manifest in utero or in the perinatal period. First, in fetal tissues anaerobic glycolysis is the major source of cellular ATP,52 thus possibly explaining why fetal wastage due to respiratory chain disorders is not noted as a major feature in the three retrospective studies of large cohorts of pediatric patients with well-established mitochondrial diseases.53e55 By contrast, the rapidly increasing energy requirements of the growing neonate can explain the frequent neonatal presentation of these disorders.56 Second, the all-or-none nature of mendelian disorders as opposed to the generally heteroplasmic nature of mtDNA mutations may explain why severe mutations in nuclear genes cause

184

S. DiMauro, C. Garone / Seminars in Fetal & Neonatal Medicine 16 (2011) 181e189

early onset and often rapidly fatal diseases whereas mtDNA-related disorders tend to manifest later in life. This concept has been documented by a relative small study of 16 families57 and is illustrated by the severity of complex I deficiencies due to nDNA mutations.58 The best example of how heteroplasmy can determine age at onset and severity is to compare the clinical presentation of NARP (neuropathy, ataxia, retinitis pigmentosa) and MILS (maternally inherited Leigh syndrome), which are due to the same mutation (m.8993T/G in ATPase 6) but cause young adult onset at 70e80% heteroplasmy or infantile Leigh syndrome at higher mutation loads.59 Third, the ubiquitous distribution of mitochondria would suggest that mtDNA-related diseases ought to be more frequently multisystemic and clinically heterogeneous than mendelian disorders. This concept, however, is contradicted by the relative stereotypical presentations of ‘classical’ mtDNA-related syndromes, such as MELAS (mitochondrial myopathy, encephalopathy, lactic acidosis, and stroke), MERRF (myoclonic epilepsy with ragged red fibers), and LHON (Leber’s hereditary optic neuropathy), which still elude rational explanation.60 It is also difficult to explain why single large-scale mtDNA deletions should result in three different clinical phenotypes, fatal infantile sideroblastic anemia (Pearson syndrome, PS), isolated late onset myopathy (chronic progressive external ophthalmoplegia, CPEO), or a severe multisystemic disease starting after 20 years of age (KearnseSayre syndrome, KSS). The most likely explanation for this conundrum refers to the ‘bottleneck’ that exists between ovum and embryo, such that only a minority of maternal mtDNA repopulates the fetus. On rare occasions, one partially deleted mtDNA present in the ovum may slip through the bottleneck and reach the blastocyst, whence the few mutated mtDNA can enter all three germ layers and result in KSS, segregate to muscle and result in CPEO, or populate e at least predominantly e the hematopoietic lineage and result in PS. We say ‘predominantly’ because the few infants that survive PS later develop KSS, showing that a few mutant mtDNA must have ‘contaminated’ the other cell layers and later expanded in postmitotic tissues. By the way, PS and KSS are another example of how the same mtDNA change can give rise to a neonatal or adult onset disease: in this case, however, the difference is not due to the degree of heteroplasmy but rather to negative selection of deleted mtDNA in highly replicating hematopoietic cells followed by selective expansion of deleted mtDNA in post-mitotic tissues. Fourth, the differential vulnerability of different tissues to defects of oxidative metabolism must play a role in the ‘hierarchy’ of tissue involvement e and consequent symptomatology e in infants with mitochondrial diseases. Thus, in their review of 75 patients, Skladal et al.54 found that skeletal muscle and the central nervous system were most commonly affected, which validates the general term ‘mitochondrial encephalomyopathies’ proposed by astute pediatricians almost 35 years ago based on clinical experience.61 Fifth, we have to consider homoplasmic mutations of mtDNA separately because, although they often affect infants, they cannot be the only culprits and an active search is underway for nuclear modifier genes that might explain the variable penetrance, tissue specificity, and e in at least one disorder e reversible course. Reversible cytochrome c oxidase (COX)-deficient infantile myopathy had been described in 1983 in a child who was floppy at birth, had severe lactic acidosis and virtually absent muscle COX activity both histochemically and biochemically.62 With appropriate medical support, the child improved spontaneously, his lactic acidosis disappeared and his muscle biopsy reverted to normal. No molecular defect was found until two years ago, when Rita Horvath noted that 17 patients from 12 families with

the same condition all harbored a homoplasmic m.14674T/C mutation in the tRNAGlu gene of mtDNA,63 an observation confirmed by Japanese investigators in eight patients with the same presentation, who harbored either the m.14674T/C or an m.14674T/G mutation.64 Another striking example of a maternally inherited homoplasmic mutation (m.1624C/T in the tRNAVal gene) is that of a woman with migraine headaches and mild weakness, who had seven children with lactic acidosis: six died within days from birth and a seventh child had Leigh syndrome. She also had a spontaneous abortion and an ectopic pregnancy.3 These case of fetal or neonatal diseases can be attributed to, but not explained by, the mtDNA mutations. Let us now review the different genetic causes of mitochondrial diseases (Table 1), focusing on those that have been associated with fetal or infantile clinical phenotypes. 3.1. Mutations in mtDNA There are two types, those that affect mitochondrial protein synthesis in toto (single deletions, tRNA or rRNA point mutations), and those that affect genes encoding subunits of the respiratory chain. 3.1.1. Defects in protein synthesis genes We have already discussed single large-scale deletions. The most important disease manifesting at birth or soon thereafter is PS, with refractory sideroblastic anemia, pancytopenia, and exocrine pancreas dysfunction. One relatively late onset child with PS developed the clinical and neuropathological features of Leigh syndrome and died at 3 years of age.65 Another child with PS developed fetal hydrops secondary to anemia.66 There are numerous point mutations in tRNA genes, many affecting children but less frequently infants. Some notable exceptions from our own experience include a fatal infantile cardiomyopathy due to a mutation (m.3303C/T) in the tRNALeu(UUR) gene, which was homoplasmic in the affected infants but heteroplasmic in their asymptomatic or oligosymptomatic maternal relatives, another example of heteroplasmy as a determinant of age at onset.67 Similarly, we reported two maternal half-sisters with early presentation of Leigh syndrome, who died before 2 years of age and whose mother developed typical MERRF in her late 20s. Different levels of a heteroplasmic mutation (m.8363G/A) in tRNALys correlated with the different clinical phenotypes.68 3.1.2. Defects in protein-coding genes These genetic defects are relatively common and important causes of infantile diseases, especially Leigh syndrome. We have already discussed above the NARP/MILS syndromes due to mutations in the ATPase 6 gene. Two of the seven mtDNA genes encoding complex I subunits, MTND3 and MTND5, are hotspots of mutations. Those in ND3 have been associated with Leigh syndrome in infants or children,69e73 whereas mutations in the MTND5 gene have been described both in adults with typical MELAS syndrome and in children with Leigh syndrome (for review, see Shanske et al.74). A sporadic 7 bp intragenic inversion in the MTND1 gene, previously described in an adult, caused fatal infantile cardiomyopathy and lactic acidosis in a girl who died at 1 month.75 A mutation in the cytochrome b gene was identified in an infant who had succumbed at 4 weeks of life to histiocytoid cardiomyopathy and had isolated complex III deficiency of the heart.76 Mutations in the three mtDNA genes of COX have been rarely associated with infantile presentations, except for a girl who had early onset of a multisystem disorder resembling Alpers syndrome and harbored a mutation (m.7706G/A) in COX II.77

S. DiMauro, C. Garone / Seminars in Fetal & Neonatal Medicine 16 (2011) 181e189

185

Table 1 Defects of the mitochondrial respiratory chain. DNA affected

Mechanism

Mutations/complexes

Syndromes

Fetal involvement

mtDNA

Protein synthesis

Single mtDNA D

nDNA

Individual proteins Direct hits

PS KSS CPEO Various NARP/MILS LS/LEU LS LS/EM LS/LEU LS GRACILE LS; CM; HE LS/LEU

 e e e e e e

Indirect hits

Intergenomic communication

tRNA/rRNA mutations ATPase 6; MTND3; MTND5 Complex I Complex II CoQ10 Complex I Complex II Complex III Complex IV Complex V Multiple mtDNA D mtDNA depletion Defective mtDNA translation

Myopathy/EM HE/EM; NNH

 e þþ   e  þþ/e

CPEO, chronic progressive external ophthalmoplegia; EM, encephalomyopathy; GRACILE, growth retardation, aminoaciduria, cholestasis, iron overload, lactic acidosis, and early death; HE, hepatopathy; KSS, KearnseSayre syndrome; LEU, leukoencephalopathy; NARP/MILS, neuropathy, ataxia, retinitis pigmentosa/maternally inherited Leigh syndrome; NNH, Navajo neurohepatopathy; PS, Pearson syndrome.

3.2. Mutations in nuclear DNA 3.2.1. ‘Direct hits’ These are mutations in genes encoding subunits of the respiratory chain complexes. Complex I of the respiratory chain contains 45 subunits. Fourteen of these (seven hydrophobic and encoded by mtDNA and seven hydrophilic and encoded by nDNA) are catalytic; the remaining subunits probably contribute to assembly, stabilization, and functional regulation of the complex. Mutations in the seven nuclear genes have been associated with human disease, almost invariably affecting infants, and are therefore of crucial importance for this review. Describing individual cases would be tedious. Fortunately, a recent lucid review by Jan Smeitink’s group provides both references and critical clinical and pathogenic analyses.58 Based on their experience with 15 patients and on published reports of 26 more, these authors came to several remarkable conclusions. First, onset was soon after birth, generally with hypotonia, respiratory abnormalities, seizures, visual loss, and psychomotor regression. These clinical features of Leigh syndrome were confirmed by neuroradiological evidence of symmetrical lesions in the basal ganglia and brain stem. Less frequently, brain MRI would show extensive white matter lesions, sometimes with cavitation. All children died, most of them before a year of age. Second, despite the neonatal onset, there was no evidence of prenatal involvement or excessive fetal wasting. Pregnancies were mostly uneventful and the perinatal period normal, stressing that fetal tissues are less vulnerable to defects of oxidative metabolism, possibly due to their greater dependence on anaerobic glycolysis. Thus, complex I deficiency is an important entity for neonatal but not for fetal medicine. Third, clinical pictures were rather homogeneous, certainly much more so than those caused by defects of the seven mtDNAencoded genes, due to lack of heteroplasmy. However, it should be noted that different nuclear defects resulted in different levels of residual complex I activity, which did not seem to influence clinical expression. In fact, there was no obvious correlation between genotype and clinical phenotype in these patients. Fourth, it is notable that two complex I subunits (NDUFA1 and NFUFB11) are encoded by genes on the X chromosome. Hemizygous mutations in NDFUA1 were associated with disease in two patients, one of whom had Leigh syndrome.78 Possibly due to the small size of complex II (four subunits, all encoded by nDNA), ‘direct hit’ mutations have been reported

infrequently. However, the very first genetic defect of a nuclear gene encoding a respiratory chain was reported in two sisters with the MRI lesions of Leigh syndrome together with diffuse white matter abnormalities.79 Leigh syndrome (with or without white matter involvement) has been reported in a few other children: there was no evidence of prenatal distress and survival was longer than in typical Leigh syndrome.80e83 The only mutation in a nuclear gene encoding a structural subunit of complex III was reported in a girl who, at 8 months of age, developed hepatomegaly and recurrent crises of hypoglycemia and lactic acidosis. At 2.5 years of age, she was essentially normal with occasional hypoglycemic episodes. The mutant gene (UQCRB) encodes a ubiquinone-binding protein (QP-C subunit) of complex III.84 Primary coenzyme Q10 (CoQ10) deficiencies cause five major syndromes, encephalomyopathy, cerebellar ataxia, isolated myopathy, or infantile multisystemic disease.85 The latter is the only clinical phenotype manifesting in the neonatal period. Defects of the gene (COQ1) controlling the first step in CoQ10 biosynthesis (decaprenyl diphosphate synthase subunit 2, PDSS2) was identified in a child with neonatal pneumonia and hypotonia, who developed refractory seizures, episodic vomiting, and nephrosis. MRI of the brain revealed symmetrical abnormalities of the basal ganglia suggestive of Leigh syndrome. Despite daily oral supplementation with CoQ10, the child died at 8 months.86 Curiously, mutations in the gene encoding the second subunit of COQ1 (PDSS1) did not cause symptoms (deafness) until 1 year of age.87 However, mutations in the gene (COQ2) encoding the second biosynthetic enzyme were associated with infantile encephalomyopathy and nephrosis. One child died at 12 days from multiorgan failure87 whereas another child improved remarkably with CoQ10 supplementation.88 From a practical point of view, these cases stress the importance of neonatal nephrosis as a ‘red flag’ of possibly treatable CoQ10 deficiency. Until recently, no mutation had been identified in genes encoding structural subunits of COX, suggesting that they might be lethal in utero. A homozygous missense mutation in COX6B1 was fatal at birth in one member of a consanguineous kinship and may have caused a spontaneous abortion in the 3rd trimester. However, the mutation was compatible with prolonged survival in two brothers who had had some perinatal problems (low weight, poor suck, abnormal breathing) but did not develop severe neurological symptoms until 6 years of age, when they showed weakness, ataxia, visual problems, and dementia, with MRI evidence of severe leukodystrophy.89

186

S. DiMauro, C. Garone / Seminars in Fetal & Neonatal Medicine 16 (2011) 181e189

3.2.2. ‘Indirect hits’ These are mutations in genes encoding ancillary proteins needed for proper assembly and stabilization of respiratory chain complexes. Not surprisingly given its size, there are many assembly genes for complex I, and mutations have been identified in several consanguineous families. References to these reports can be found in a recent review.90 In general, clinical pictures were similar to those described in patients with ‘direct hits’ and included infantile onset, hypotonia, nystagmus, dystonic posturing, and psychomotor retardation or regression. In one family, there was evidence of intrauterine growth retardation, and the infant had facial dysmorphism and agenesis of the corpus callosum.91 Even the tiny complex II needs assembly factors, and mutations in the gene (SDHAF1) encoding one such factor in two patients resulted in rapidly progressive psychomotor regression starting at age 6e8 months, spastic quadriparesis, and dystonia. MRI showed leukodystrophic changes but both patients were alive past age 10 years.92 The most severe clinical consequence of mutations in BCS1L, which encodes an ATPase needed for the insertion of the Rieske protein into complex III, is the GRACILE (growth retardation, aminoaciduria, cholestasis, iron overload, lactic acidosis, and early death) syndrome, a quintessential fetal and neonatal disease.93 There is intrauterine growth retardation and affected infants are consequently small for gestational age. They develop Fanconi type aminoaciduria, cholestasis with progressive liver dysfunction, and iron overload with hemosiderosis of the liver. About half of them die in the first few days and the others within weeks after birth. The disease is dominated by liver failure and there are no dysmorphic or neurological signs (see Fellman and Kotarsky in this issue). An intriguing difference between complex I and complex IV deficiencies is the relative clinical homogeneity of the former and the remarkable tissue-specific heterogeneity of the latter. Thus, mutations in the SURF1 gene, which is essential for COX assembly, affect mostly the central nervous system and result in Leigh syndrome whereas mutations in SCO2, which controls copper metabolism and insertion, affect mostly the heart, resulting in rapidly fatal infantile cardiomyopathy.94 Although reported in the pre-molecular era, 14 children had Leigh syndrome and COX deficiency and were likely carriers of SURF1 mutations (later documented in some of them).95 There was no evidence of fetal distress, onset varied between birth and 8 months, and course could be prolonged: in fact, five children were still alive at the time of report, one of them 10 years of age. By contrast, compound heterozygous mutations in SCO2 caused severe congenital cardiomyopathy and were probably detrimental in utero.2 For reasons that are not completely clear but which may relate to differential expression of COX in different tissues,94,96 mutations in other COX assembly genes (COX10, COX15, SCO1) cause various combinations of encephalopathy, hepatopathy, and cardiomyopathy.97 Indirect hits of complex V have been associated with severe infantile multisystemic disease with lactic acidosis and, often, with 3-methylglutaconic aciduria. In a series of 23 patients, 96% had hypertrophic cardiomyopathy, 83% had hypotonia and psychomotor retardation, 39% had dysmorphic features, and 22% had hepatopathy.98,99 Thirteen children died between the ages of 1 day and 18 months; the oldest surviver was aged 17 years. 3.3. Defects of intergenomic communication As the mtDNA is the slave of nDNA, it depends on the integrity of nuclear genes for its maintenance, replication, and translation. Defects of mtDNA maintenance consist of autosomal dominant or recessive multiple mtDNA deletions: these are usually associated

with CPEO and onset is in adult life. However, defects of mtDNA replication and translation are important causes of fetal and neonatal medicine.100 3.3.1. mtDNA depletion There are two main syndromes, myopathic and hepatocerebral. The myopathic syndrome is most commonly caused by mutations in the gene encoding thymidine kinase 2 (TK2). Although a few children may show weakness and hypotonia at birth, a review of 20 patients showed that the mean age at onset was 11.4 months and the mean age at death 3.5 years.101 Skeletal muscle is the target organ, but anterior horn cells of the spinal cord may be severely affected and the clinical picture may mimic spinal muscular atrophy. A less common infantile myopathy due to mtDNA depletion, and often associated with renal tubulopathy, is associated with mutations in RRM2B, which encodes a p53-inducible ribonucleotide reductase subunit.100,102 The most common hepatocerebral presentation, known as AlperseHuttelocher syndrome, is due to mutations in the gene encoding the mitochondrial polymerase, POLG.103,104 However, this is not usually a perinatal disease, but rather a disease of children or young adults characterized by refractory seizures, psychomotor regression, and liver disease. Infantile presentation is much more common in a second form of hepatocerebral syndrome due to mutations in the gene (DGUOK) encoding deoxyguanosine kinase. A review of 11 cases revealed frequent neonatal liver failure, nystagmus, hypotonia, and developmental delay. Onset was at birth or within the first weeks of life and death in nine patients occurred between 6 and 12 months. A hint of fetal distress was that two patients had IUGR.105 Liver failure and death before the first birthday characterized the hepatocerebral syndrome due to MPV17 mutations.102 Notably, mutations in the same gene affect peripheral nerves more than the brain in the so-called Navajo neurohepatopathy (NNH), an autosomal recessive condition prevalent among Navajo children. Although onset is in infancy with hepatic dysfunction, peripheral neuropathy and cerebral leukoencephalopathy do not appear until later in childhood and these patients usually die in their teens.106 A rare but interesting hepatic mtDNA depletion is due to recessive mutations in the gene (PEO1 or Twinkle) encoding a helicase that works hand-in-hand with polymerase g. While dominant PEO1 mutations are known to cause CPEO in adults, recessive mutations cause a severe encephalopathy with infantile onset but protracted course characterized by hypotonia, athetosis, sensory neuropathy, ataxia, deafness, ophthalmoplegia, and seizures.107 3.4. Defects of mtDNA translation As pediatricians encountered increasing numbers of patients with multiple respiratory chain defects but no mtDNA tRNA mutations or evidence of mtDNA depletion, they directed their attention to the complex nuclear-encoded apparatus needed for mtDNA translation. As this is still largely work in progress, instead of reviewing clinical cases in detail, we refer the reader to the report of a recent workshop on mitochondrial protein synthesis in health and disease.108 Not surprisingly, most of these patients were affected at or soon after birth and died within weeks or months. There was evidence of prenatal involvement, especially in children with mutant elongation factors: IUGR, microcephaly, and ‘stiffness’ in an infant with mutant EFG1109; poor fetal movements, abnormal brain ultrasonography at day 4, and dysmorphic signs, including short tibial bones in patients with mutant elongation EFTs.110,111 An infant with mutant arginyl-transfer RNA synthetase (RARS) had MRI signs of cerebellar and vermian hypoplasia at 3 days, and

S. DiMauro, C. Garone / Seminars in Fetal & Neonatal Medicine 16 (2011) 181e189

another was hypotonic and lethargic since birth and died suddenly at 7 weeks.112 Fetal ultrasound documented microcephaly, dilation of the third ventricle, and hypertrophic cardiomyopathy in an infant with mutations in mitochondrial ribosomal protein MRPS22 and Cornelia de Lange dysmorphic features.113 Given the flurry of recent reports, it is somewhat surprising that systematic sequencing of candidate genes in 52 patients with evidence of generalized impairment of mitochondrial translation revealed mutations in only one patient.114 New high-throughput, pooled sequencing coupled with functional prediction and experimental validation will help discover mutated nuclear genes.115 For reasons of space, we have confined the review of mitochondrial disorders to defects of the respiratory chain. It should not be forgotten that defects of pyruvate metabolism are frequent causes of neonatal encephalopathy, often starting in utero and causing cerebral dysgenesis.116 In conclusion, it seems fair to say that both disorders of glycogen metabolism and disorders of the mitochondrial respiratory chain are important causes of neonatal or early infantile diseases. Fetal involvement is less common, and more prominent in glycogenoses than in mitochondrial disorders, possibly due to the predominantly glycolytic metabolism of fetal tissues. The next crucial steps are to improve prenatal diagnosis, to prevent the birth of affected infants, and to develop rational therapeutic strategies.

Practice points  Inborn errors of energy metabolism are likely to manifest in utero or in the neonatal period; this has been confirmed for what concerns disorders of glycogen metabolism and defects of the mitochondrial respiratory chain.  Defects of glycogen synthesis e typically brancher enzyme deficiency e cause severe fetal distress whereas defects of the mitochondrial respiratory chain, by and large and with a few exceptions, are well tolerated by the fetus.

Research directions  Among the glycogenoses, it is likely that defects of cellular fuel gauges such as AMP-dependent protein kinase (AMPK) will be increasingly associated with fetal distress; the molecular basis of the infantile form of phosphofructokinase (PFK) deficiency needs to be clarified.  Among the mitochondrial disorders, the new chapter of primary and secondary CoQ10 deficiencies is still work in progress and it has good therapeutic potential; disorders of mitochondrial dynamics, fusion, fission, and motility, remain to be further defined clinically, especially regarding their expression in fetal life or in young infants.

Conflict of interest statement None declared. Funding sources This work was supported in part by NIH grant HD32062 and by the Marriott Mitochondrial Disorders Clinical Research Fund (MMDCRF). Dr Garone is supported by the Associazione Malattie Metaboliche Congenite ereditarie (AMMeC), Italy.

187

References 1. Shekhawat PS, Matern D, Strauss AM. Fetal fatty acid oxidation disorders, their effect on maternal health and neonatal outcome: impact of expanded newborn screening on their diagnosis and management. Pediatr Res 2005;57:78Re86R. 2. Tay SKH, Shanske S, Kaplan P, DiMauro S. Association of mutations in SCO2, a cytochrome c oxidase assembly gene, with early fetal lethality. Arch Neurol 2004;61:950e2. 3. McFarland R, Clark KM, Morris AAM, et al. Multiple neonatal deaths due to homoplasmic mitochondrial DNA mutation. Nat Genet 2002;30:145e6. 4. Fenton WA. Mitochondrial protein transport e a system in search of mutations. Am J Hum Genet 1995;57:235e8. 5. Agsteribbe E, Huckriede A, Veenhuis M, et al. A fatal, systemic mitochondrial disease with decreased mitochondrial enzyme activities, abnormal ultrastructure of the mitochondria and deficiency of heat shock protein 60. Biochem Biophys Res Commun 1993;193:146e54. 6. Briones P, Vilaseca MA, Ribes A, et al. A new case of multiple mitochondrial enzyme deficiencies with decreased amount of heat shock protein 60. J Inherit Metab Dis 1997;20:569e77. 7. Magen D, Georgopoulos C, Bross P, et al. Mitochondrial Hsp60 chaperonopathy causes an autosomal-recessive neurodegenerative disorder linked to brain hypomyelination and leukodystrophy. Am J Hum Genet 2008;83:30e42. 8. Davey KM, Parboosingh JS, McLeaod DR, et al. Mutation of DNAJC19, a human homologue of yeast inner mitochondrial membrane co-chaperones, causes DCMA syndrome, a novel autosomal recessive Barth syndrome-like condition. J Med Genet 2006;43:385e93. 9. Illsinger S, Das AM. Impact of selected inborn errors of metabolism on prenatal and neonatal development. IUBMB Life 2010;62:403e13. 10. Bruno C, van Diggelen OP, Cassandrini D, et al. Clinical and genetic heterogeneity of branching enzyme deficiency (glycogenosis type IV). Neurology 2004;63:1053e8. 11. Engel AG, Gomez MR, Seybold ME, Lambert EH. The spectrum and diagnosis of acid maltase deficiency. Neurology 1973;23:95e105. 12. Kishnani PS, Hwu WL, Mandel H, et al. A retrospective, multinational, multicenter study on the natural history of infantile-onset Pompe disease. J Pediatr 2006;148:671e6. 13. Gambetti PL, DiMauro S, Baker L. Nervous system in Pompe’s disease. Ultrastructure and biochemistry. J Neuropath Exp Neurol 1971;30:412e30. 14. Martin JJ, DeBarsy T, Van Hoof F. Pompe’s disease: an inborn lysosomal disorder with storage of glycogen: a study of brain and striated muscle. Acta Neuropathol 1973;23:229e44. 15. Engel AG, Hirschhorn R, Huie MI. Acid maltase deficiency. In: Engel AG, FranziniArmstrong C, editors. Myology, vol. 2. New York: McGraw-Hill; 2004. p. 1559e86. 16. Mehler M, DiMauro S. Residual acid maltase activity in late-onset acid maltase deficiency. Neurology 1977;27:178e84. 17. Reuser AJ, Kroos M, Willemsen R, et al. Clinical diversity in glycogenosis type II. Biosynthesis and in situ localization of acid alpha-glucosidase in mutant fibroblasts. J Clin Invest 1987;79:1689e99. 18. Hermans MM, van Leenen D, Kroos M, et al. Twenty-two novel mutations in the lysosomal alpha-glucosidase gene (GAA) underscore the genotypeephenotype correlation in glycogen storage disease type II. Hum Mutat 2004;23:47e56. 19. Nascimbeni AC, Fanin M, Tasca E, Angelini C. Molecular pathology and enzyme processing in various phenotypes of acid maltase deficiency. Neurology 2008;70:617e26. 20. Klinge L, Straub V, Neudorf U, Voit T. Enzyme replacement therapy in classical infantile Pompe disease: results of a ten-month follow-up study. Neuropediatr 2005;36:6e11. 21. Van den Hout JMP, Reuser AJJ, de Klerk JBC, et al. Enzyme therapy for Pompe disease with recombinant human alpha-glucoside from rabbit milk. J Inherit Metab Dis 2001;24:266e74. 22. Van den Hout J, Kamphoven JHJ, Winkel LPF, et al. Long-term intravenous treatment of Pompe disease with recombinant human alpha-glucosidase from milk. Pediatrics 2004;113:e448e57. 23. Kishnani PS, Goldenberg PC, DeArmey SL, et al. Cross-reacting immunological material status affects treatment outcomes in Pompe disease infants. Molec Genet Metab 2010;99:26e33. 24. Witters I, Moerman P, Fryns J- P. Fetal akinesia deformation sequence: a study of 30 consecutive in utero diagnoses. Am J Med Genet 2002;113:23e8. 25. Tay SKH, Akman HO, Chung WK, et al. Fatal infantile neuromuscular presentation of glycogen storage disease type IV. Neuromusc Disord 2004;14:253e60. 26. Janecke AR, Dertinger S, Ketelsen UP, et al. Neonatal type IV glycogen storage disease associated with “null” mutations in glycogen branching enzyme 1. J Pediatr 2004;145:705e9. 27. Konstantinidou AE, Anninos H, Gyftodimou Y, et al. Neonatal neuromuscular variant of glycogen storage disease type IV: histopathological findings leading to the diagnosis. Histopathology 2006;48:878e80. 28. Taratuto AL, Akman HO, Saccoliti M, et al. Branching enzyme deficiency/glycogenosis storage disease type IV presenting as a severe congenital hypotonia: muscle biopsy and autopsy findings, biochemical and molecular studies. Neuromusc Dis 2010;20:783e90. 29. Alegria A, Martins E, Dias M, et al. Glycogen storage disease type IV presenting as hydrops fetalis. J Inherit Metab Dis 1999;22:330e2. 30. Akman HO, Karadimas CL, Gyftodimou Y, et al. Prenatal diagnosis of glycogen storage disease type IV. Prenat Diagn 2006;26:951e5.

188

S. DiMauro, C. Garone / Seminars in Fetal & Neonatal Medicine 16 (2011) 181e189

31. Assereto S, van Diggelen OP, Diogo L, et al. Null mutations and lethal congenital form of glycogen storage disease type IV. Biochem Biophys Res Commun 2007;361:445e50. 32. Herrick MK, Twiss JL, Vladutiu GD, et al. Concomitant branching enzyme and phosphorylase deficiencies. An unusual glycogenosis with extensive neuronal polyglucosan storage. J Neuropath Exp Neurol 1994;53:239e46. 33. Konstantinidou A, Anninos H, Dertinger S, et al. Placental involvement in glycogen storage disease type IV. Placenta 2008;29:378e81. 34. Bruno C, DiRocco M, Doria Lama L, et al. A novel missense mutation in the glycogen branching enzyme gene in a child with myopathy and hepatopathy. Neuromusc Disord 1999;9:403e7. 35. DiMauro S, Hartlage P. Fatal infantile form of muscle phosphorylase deficiency. Neurology 1978;28:1124e9. 36. Miranda AF, Nette G, Hartlage P, DiMauro S. Phosphorylase isoenzymes in normal and myophosphorylase deficient human heart. Neurology 1979;29:1538e41. 37. Milstein JM, Herron TM, Haas JE. Fatal infantile muscle phosphorylase deficiency. J Child Neurol 1989;4:186e8. 38. El-Schahawi M, Bruno C, Tsujino S, et al. Sudden infant death syndrome (SIDS) in a family with myophosphorylase deficiency. Neuromusc Disord 1997;7:81e3. 39. Tsujino S, Shanske S, DiMauro S. Molecular genetic heterogeneity of myophosphorylase deficiency (McArdle’s disease). N Engl J Med 1993;329:241e5. 40. Moerman P, Lammens M, Fryns JP, et al. Fetal akinesia sequence caused by glycogenosis type VII. Genet Couns 1995;6:15e20. 41. Swoboda KJ, Specht L, Jones HR, et al. Infantile phosphofructokinase deficiency with arthrogryposis: clinical benefit of a ketogenic diet. J Pediatr 1997;131:932e4. 42. Spriggs EL, Marles SL, Lacson A, et al. Long-term survival and normal cognitive development in infantile phosphofructokinase-1 deficiency. Clin Genet 1999;56:235e7. 43. Guibaud P, Carrier H, Mathieu M, et al. Observation familiale de dystrophie musculaire congenitale par deficit en phosphofructokinase. Arch Fr Pediatr 1978;35:1105e15. 44. Danon MJ, Carpenter S, Manaligod JR, Schliselfeld LH. Fatal infantile glycogen storage disease: deficiency of phosphofructokinase and phosphorylase b kinase. Neurology 1981;31:1303e7. 45. Al-Hassnan ZN, Al Budhaim M, Al-Owain M, et al. Muscle phosphofructokinase deficiency with neonatal seizures and nonprogressive course. J Child Neurol 2007;22:106e8. 46. Servidei S, Bonilla E, Diedrich RG, et al. Fatal infantile form of muscle phosphofructokinase deficiency. Neurology 1986;36:1465e70. 47. Pastoris O, Dossena M, Vercesi L, et al. Muscle phosphofructokinase deficiency in a myopathic child with severe mental retardation and aplasia of cerebellar vermis. Child Nerv Syst 1992;8:237e41. 48. Garcia M, Pujol A, Ruzo A, et al. Phosphofructo-1-kinase deficiency leads to severe cardiac and hematological disorder in addition to skeletal muscle glycogenosis. PLoS Genet 2009;5:e1000615. 49. Burwinkel B, Scott JW, Buhrer C, et al. Fatal congenital heart glycogenosis caused by a recurrent activating R531Q mutation in the gamma2-subunit of AMP-activated protein kinase (PRKAG2), not by phosphorylase kinase deficiency. Am J Hum Genet 2005;76:1034e49. 50. Akman HO, Sampayo JN, Ross FA, et al. Fatal infantile cardiac glycogenosis with phosphorylase kinase deficiency and a mutation in the gamma2-subunit of AMP-activated protein kinase. Pediatr Res 2007;62:499e504. 51. DiMauro S, Schon EA. Mitochondrial respiratory-chain diseases. N Engl J Med 2003;348:2656e68. 52. Papa S. Mitochondrial oxidative phosphorylation changes in the life span. Molecular aspects and physiopathological implications. Biochem Biophys Acta 1996;1276:87e105. 53. von Kleist-Retzow JC, Cormier-Daire V, Viot G, et al. Antenatal manifestations of mitochondrial respiratory chain deficiency. J Pediatr 2003;143:208e12. 54. Skladal D, Sudmeier C, Konstantopoulou V, et al. The clinical spectrum of mitochondrial disease in 75 p.diatric patients. Clin Pediat 2003;42:703e10. 55. Gibson K, Halliday JL, Kirby DM, et al. Mitochondrial oxidative phosphorylation disorders presenting in neonates: clinical manifestations and enzymatic and molecular diagnoses. Pediatrics 2008;122:1003e8. 56. Minai L, Martinovic J, Chretien D, et al. Mitochondrial respiratory chain complex assembly and function during human fetal development. Molec Genet Metab 2008;94:120e6. 57. Rubio-Gozalbo ME, Dijkman KP, van den Heuvel LP, et al. Clinical differences in patients with mitochondriocytopathies due to nuclear versus mitochondrial DNA mutations. Hum Mutat 2000;15:522e32. 58. Distelmaier F, Koopman WJH, van den Heuvel LP, et al. Mitochondrial complex I deficiency: from organelle dysfunction to clinical disease. Brain 2009;132:833e42. 59. Tatuch Y, Christodoulou J, Feigenbaum A, et al. Heteroplasmic mtDNA mutation (T/G) at 8993 can cause Leigh disease when the percentage of abnormal mtDNA is high. Am J Hum Genet 1992;50:852e8. 60. DiMauro S, Davidzon G. Mitochondrial DNA and disease. Ann Med 2005;37:222e32. 61. Shapira Y, Harel S, Russell A. Mitochondrial encephalomyopathies: a group of neuromuscular disorders with defects in oxidative metabolism. Isr J Med Sci 1977;13:161e4. 62. DiMauro S, Nicholson JF, Hays AP, et al. Benign infantile mitochondrial myopathy due to reversible cytochrome c oxidase deficiency. Ann Neurol 1983;14:226e34.

63. Horvath R, Kemp JP, Tuppen HAL, et al. Molecular basis of infantile reversible cytochrome c oxidase deficiency. Brain 2009;132:3165e74. 64. Mimaki M, Hatakayama H, Komaki H, et al. Reversible infantile respiratory chain deficiency: a clinical and molecular study. Ann Neurol 2010;68:845e54. 65. Santorelli F, Barmada MA, Pons R, et al. Leigh-type neuropathology in Pearson syndrome associated with impaired ATP production and a novel mtDNA deletion. Neurology 1996;47:1320e3. 66. Rotig A, Cormier V, Blanche S, et al. Pearson’s marrowepancreas syndrome. A multisystem mitochondrial disorder in infancy. J Clin Invest 1990;86:1601e8. 67. Silvestri G, Santorelli FM, Shanske S, et al. A new mtDNA mutation in the tRNA(LeuUUR) gene associated with maternally inherited cardiomyopathy. Hum Mutat 1994;3:37e43. 68. Shtilbahns A, Shanske S, Goodman S, et al. G8363A mutation in the mitochondrial DNA transfer ribonucleic acid Lys gene: another cause of Leigh syndrome. J Child Neurol 2000;15:759e61. 69. Lebon S, Chol M, Benit P, et al. Recurrent de novo mitochondrial DNA mutations in respiratory chain deficiency. Am J Med Genet 2003;40:896e9. 70. Crimi M, Papadimitriou A, Galbiati S, et al. A new mitochondrial DNA mutation in ND3 gene causing severe Leigh syndrome with early lethality. Pediatr Res 2004;55:842e6. 71. McFarland R, Kirby DM, Fowler KJ, et al. De novo mutations in the mitochondrial ND3 gene as a cause of infantile mitochondrial encephalopathy and complex I deficiency. Ann Neurol 2004;55:58e64. 72. Leshinsky-Silver E, Lev D, Tzofi-Berman Z, et al. Fulminant neurological deterioration in a neonate with Leigh syndrome due to maternally transmitted missense mutation in the mitochondrial ND3 gene. Biochem Biophys Res Commun 2005;334:582e7. 73. Chae JH, Lee JS, Kim KJ, et al. A novel ND3 mitochondrial DNA mutation in three Korean children with basal ganglia lesions and complex I deficiency. Pediatr Res 2007;61(5 Pt 1):622e4. 74. Shanske S, Coku J, Lu J, et al. The G13513A mutation in the ND5 gene of mitochondrial DNA as a common cause of MELAS or Leigh syndrome. Arch Neurol 2008;65:368e72. 75. Blakeley EL, Rennie CJ, Jones L, et al. Sporadic intragenic inversion of the mitochondrial DNA MTND1 gene causing fatal infantile lactic acidosis. Pediatr Res 2006;59:440e4. 76. Andreu AL, Bruno C, Dunne TC, et al. A nonsense mutation (G15059A) in the cytochrome b gene in a patient with exercise intolerance and myoglobinuria. Ann Neurol 1999;45:127e30. 77. Uusimaa J, Finnila S, Vainionpaa L, et al. A mutation in mitochondrial DNAencoded cytochrome c oxidase II gene in a child with AlperseHuttenlocherlike disease. Pediatrics 2003;111:e262e8. 78. Fernandez-Moreira D, Ugalde C, Smeets R, et al. X-linked NDUFA1 gene mutations associated with mitochondrial encephalomyopathy. Ann Neurol 2007;61:73e83. 79. Bourgeron T, Rustin P, Chretien D, et al. Mutation of a nuclear succinate dehydrogenase gene results in mitochondrial respiratory chain deficiency. Nat Genet 1995;11:144e9. 80. Parfait B, Chretien D, Rotig A, et al. Compound heterozygous mutation in the flavoprotein gene of the respiratory chain complex II in a patient with Leigh syndrome. Hum Genet 2000;106:236e43. 81. Van Coster R, Seneca S, Smet J, et al. Homozygous Gly555Glu mutation in the nuclear-encoded 7-kDa flavoprotein gene causes instability of respiratory chain complex II. Am J Med Genet 2003;120A:13e8. 82. Bugiani M, Lamantea E, Invernizzi F, et al. Effects of riboflavin in children with complex II deficiency. Brain Dev 2006;28:576e81. 83. Horvath R, Abicht A, Holinski-Feder E, et al. Leigh syndrome caused by mutations in the flavoprotein (Fp) subunit of succinate dehydrogenase (SDHA). J Neurol Neurosurg Psychiatr 2006;77:74e6. 84. Haut S, Brivet M, Touati G, et al. A deletion in the human QP-C gene causes a complex III deficiency resulting in hypoglycemia and lactic acidosis. Hum Genet 2003;113:118e22. 85. Quinzii CM, Hirano M. Coenzyme Q and mitochondrial disease. Dev Disabil Res Rev 2010;16:183e8. 86. Lopez LC, Quinzii C, Schuelke M, et al. Leigh syndrome with nephropathy and CoQ10 deficiency due to decaproneyl diphosphate synthase subunit 2 (PDSS2) mutations. Am J Hum Genet 2006;79:1125e9. 87. Mollet J, Giurgea I, Schlemmer D, et al. Prenyldiphosphate synthase (PDSS1) and OH-benzoate prenyltransferase (COQ2) mutations in ubiquinone deficiency and oxidative phosphorylation disorders. J Clin Invest 2007;117:765e72. 88. Quinzii C, Naini A, Salviati L, et al. A mutation in para-hydoxybenzoatepolyprenyl transferase (COQ2) causes primary coenzyme Q10 deficiency. Am J Hum Genet 2006;78:345e9. 89. Massa V, Fernandez-Vizarra E, Alshahwan S, et al. Severe infantile encephalomyopathy caused by a mutation in COX6B1, a nucleus-encoded subunit of cytochrome c oxidase. Am J Hum Genet 2008;82:1281e9. 90. Valsecchi F, Koopman WJH, Manjeri GR, et al. Complex I disorders: causes, mechanisms, and development of treatment strategies at the cellular level. Dev Disabil Res Rev 2010;16:175e82. 91. Sugiana C, Pagliarini DJ, McKenzie M, et al. Mutation of C20orf7 distupts complex I assembly and causes lethal neonatal mitochondrial disease. Am J Hum Genet 2008;83:468e78. 92. Ghezzi D, Goffrini P, Uziel G, et al. SDHAF1, encoding a LYR complex II specific assembly factor, is mutated in SDH-defective infantile leukoencephalopathy. Nat Genet 2009;41:654e6.

S. DiMauro, C. Garone / Seminars in Fetal & Neonatal Medicine 16 (2011) 181e189 93. Fellman V. The GRACILE syndrome, a neonatal lethal metabolic disorder with iron overload. Blood Cells Molec Dis 2002;29:444e50. 94. Sue CM, Karadimas C, Checcarelli N, et al. Differential features of patients with mutations in two COX assembly genes, SURF-1 and SCO2. Ann Neurol 2000;47:589e95. 95. Van Coster R, Lombes A, DeVivo DC, et al. Cytochrome c oxidase-associated Leigh syndrome: phenotypic features and pathogenetic speculations. J Neurol Sci 1991;104:97e111. 96. Brosel S, Yang H, Tanji K, et al. Unexpected vascular enrichment of SCO1 over SCO2 in mammalian tissues. Am J Pathol 2010;177:2541e8. 97. Barrientos A, Barros MH, Valnot I, et al. Cytochrome oxidase in health and disease. Gene 2002;286:53e63. 98. De Meirleir L, Seneca S, Lissens W, et al. Respiratory chain complex V deficiency due to a mutation in the assembly gene. ATP12. J Med Genet 2004;41:120e4. 99. Cizkova A, Stranecky V, Mayr JA, et al. TMEM70 mutations cause isolated ATP synthase deficiency and neonatal mitochondrial encephalomyopathy. Nat Genet 2009;40:1288e90. 100. Suomalainen A, Isohanni P. Mitochondrial DNA depletion syndromes e many genes, common mechanisms. Neuromusc Dis 2010;20:429e37. 101. Oskoui M, Davidzon G, Pascual J, et al. Clinical spectrum of mitochondrial DNA depletion due to mutations in the thymidine kinase 2 gene. Arch Neurol 2006;63:1122e6. 102. Spinazzola IF, Carrara F, Lamantea E, et al. Clinical and molecular features of mitochondrial DNA depletion syndromes. J Inherit Metab Dis 2009;32:143e58. 103. Nguyen KV, Ostergard E, Ravn SH, et al. POLG mutations in Alpers syndrome. Neurology 2005;65:1493e5. 104. Wong L-JC, Naviaux RK, Brunetti-Pierri N, et al. Molecular and clinical genetics of mitochondrial diseases due to POLG mutations. Hum Mutat 2008;29:E150e72. 105. Dimmock DP, Zhang Q, Dionisi-Vici C, et al. Clinical and molecular features of mitochondrial DNA depletion due to mutations in deoxyguanosine kinase. Hum Mutat 2007;29:330e45.

189

106. Karadimas CL, Vu TH, Holve SA, et al. Navajo neurohepatopathy is caused by a mutation in the MPV17 gene. Am J Hum Genet 2006;79:544e8. 107. Hakonen AH, Isohanni P, Paetau A, et al. Recessive twinkle mutations in early onset encephalopathy with mtDNA depletion. Brain 2007; 130:3032e40. 108. Chrzanowska-Lightowlers ZMA, Horvath R, Lightowlers RN. 175th ENMC international workshop: mitochondrial protein synthesis in health and disease, 25e27th June, 2010, Naarden, The Netherlands. Neuromusc Dis 2011;21:142e7. 109. Coenen MJH, Antonicka H, Ugalde C, et al. Mutant mitochondrial elongation factor G1 and combined oxidative phosphorylation deficiency. N Engl J Med 2004;351:2080e6. 110. Smeitink JAM, Elpeleg O, Antonicka H, et al. Distinct clinical phenotypes associated with a mutation in the mitochondrial translation elongation factor EFTs. Am J Hum Genet 2006;79:869e77. 111. Valente L, Tiranti V, Marsano RM, et al. Infantile encephalopathy and defective mitochondrial DNA translation in patients with mutations of mitochondrial elongation factors EGF1 and EFTu. Am J Hum Genet 2007;80:44e58. 112. Edvardson S, Shaag A, Kolesnikova O, et al. Deleterious mutation in the mitochondrial arginyl-transfer RNA synthetase gene is associated with pontocerebellar hypoplasia. Am J Hum Genet 2007;81:857e62. 113. Smits P, Saada A, Wortmann SB, et al. Mutation in mitochondrial ribosomal protein MRPS22 leads to Cornelia de Lange phenotype, brain abnormalities and hypertrophic cardiomyopathy. Eur J Hum Genet 2010;19:394e9. 114. Kemp JP, Smith PM, Pyle A, et al. Nuclear factors involved in mitochondrial translation cause a subgroup of combined respiratory chain deficiency. Brain 2011;134:183e95. 115. Calvo TEJ, Compton A, Kirby DM, et al. High-throughput, pooled sequencing identifies mutations in NUBPL and FOXRED1 in human complex I deficiency. Nat Genet 2010;42:851e8. 116. Volpe JJ. Neurology of the newborn. 5th ed. Philadelphia: Saunders/Elsevier; 2008.