CONCEPT AND CONTEXT OF NEUROMETABOLIC DISEASES

Metabolic disorders constitute an expanding group of diseases comprising such heterogeneous conditions that a uniform introductory definition becomes necessary. Strictly speaking, neurometabolic diseases arise from genetic deficiencies of intermediary metabolism enzymes. Thus, mutation of genes encoding cytostructural or regulatory proteins or proteins involved in cell division, immunity, excitability, cell-to-cell communication, secretion, or movement do not give rise to metabolic diseases sensu stricto. Nevertheless, careful reflection on the molecular mechanisms of these latter disease categories leads to the recognition of intermediary metabolism abnormalities virtually in all of them, allowing at least some of them to be included with neurometabolic diseases.

Whether involving carbohydrate, lipid, or protein metabolism, the manifestations of neurometabolic diseases are pleomorphic and can manifest at any stage during a life span. Regardless of their time and mode of manifestation, the approach to the potential patient with a neurometabolic disease includes a customized but systematic series of evaluations as well as a thorough assessment of the ancestry and family structure aimed at identifying all relatives at risk for a heritable trait. At the conclusion of the clinical interview, it should be possible to suspect the pattern of inheritance of a familial disease or, alternatively, the probability of a de novo, that is, noninherited, mutation. In the case of a potentially heritable trait, apparently unaffected relatives should be questioned and examined for the presence of specific abnormalities indicative of an incompletely penetrant trait. Then, a series of analytical investigations are tailored to confirm the diagnosis or, at least, to circumscribe the metabolic abnormality as much as possible according to the patient’s clinical syndrome. The biochemical analysis of banked patient tissues, such as biopsied muscle, and cellular elements, such as cultured fibroblasts, is often necessary to confirm a specific enzyme deficiency and should be followed, whenever possible, by genotyping. A variety of genetic test batteries and panels are available to screen genes associated with diseases that share a similar phenotype. When successful, genotyping allows for fully informed genetic counseling, for screening of at-risk relatives, and, in an increasing number of instances, for prenatal diagnosis via amniocentesis.

MANIFESTATIONS UNFOLD IN TIME

The apparent age at onset of metabolic disorders is variable, as the manifestations of neurometabolic diseases are intimately linked to the development of the nervous system. During infancy and childhood, several genetic expression programs come into play and become quiescent as the organism grows and matures. Thus, the effects of a pathogenic mutation may not be noticeable until the mutant gene is activated, and this may occur well after birth. For example, the fetal brain consumes preferentially lipid byproducts such as ketone bodies. At birth, the cerebral metabolic rate for glucose is minimal; it increases gradually during childhood, when it exceeds the neonatal rate by three-fold. By early adolescence, glucose consumption decreases and reaches the adult level, which is about twice the newborn rate. It is not surprising, then, that fetal and neonatal brains tolerate hypoglycemia relatively well. This is exemplified by the manifestations of glucose transporter type 1 deficiency, caused by mutation of the glucose carrier of the blood-brain barrier, which is usually unnoticeable in the newborn and becomes most pronounced during childhood; furthermore, they can be circumvented, to some degree, by a diet rich in ketogenic substrates, as the brain always retains the capacity to metabolize ketone bodies. In some occasions, the converse phenomenon is true, allowing for the restoration of a normal phenotype as development progresses and new genes replace the function of abnormal ones. This is illustrated by a transient or reversible form of cytochrome c oxidase–deficient myopathy that manifests in infancy with hypotonia and profound weakness, followed by a return of enzyme activity in muscle and normal strength later in childhood. In disorders causing an accumulation of a metabolite, symptoms may remain latent until the stored metabolite interferes with cellular organelles or forms deleterious aggregates. Yet, in other occasions, a precipitant factor triggers the sudden decompensation of a precarious cellular machinery. For example, a nutritional excess of fat may aggravate a fatty acid oxidation defect, inducing severe hypoglycemia and coma, or a protein load may result in hyperammonemia and mental disturbances in urea cycle defects (see Chapter 110). In some cases, well exemplified by Leigh syndrome, a “free interval” of several years may lapse before a trivial intercurrent respiratory or infectious illness precipitates cerebral necrosis, leading to fulminant disability. In such cases, meticulous questioning and review of developmental milestones often uncover a preceding history of subtle but lifelong neurological dysfunction.

STATIC, EPISODIC, OR PROGRESSIVE DISEASES?

Generally speaking, metabolic diseases are lifelong, permanent diseases, like their causal genetic mutations. However, they follow any imaginable temporal course even in the absence of environmental or nutritional precipitating factors, with some conditions manifesting only periodically and others exhibiting a static or apparently immutable course, which is in contrast with the still common notion of metabolic diseases as unrelenting, continuously symptomatic processes. The basis for the apparently paradoxical static and episodic manifestations of neurometabolic diseases is provided by the compartmentalization of metabolism.¹ Because all cellular functions are spatially limited and regulated by membranes, metabolic reactions occur at rates governed not only by the kinetics of their corresponding enzymes but also (and often mainly) by substrate availability and product abundance. The former process, dependent on enzyme structure, is dictated by the gene; the latter, by the ability of cells and organelles to distribute and clear substrates and products, a process that is inherently dependent on the function of the membranes and membrane compartments of the various cell types that carry out the reaction in question.² Thus, for example, an enzymatic deficiency affecting a reaction that is constantly active may be associated with the accumulation of a substrate that interferes with other reactions, causing inhibition and resulting in abnormal cell function. Such a substrate may be eliminated from the cell after it reaches a certain threshold level at a rate that is dependent on its concentration. After the substrate accumulates for the first time, production and elimination proceed indefinitely, maintaining a constant (elevated) concentration in the cell. This may constitute the basis for the permanent, immutable clinical manifestations that are sometimes associated with a static metabolic abnormality. Episodic diseases can also be modeled using the same mechanistic framework: a compound may accumulate without causing any abnormalities until a threshold concentration is reached or until another slowly fluctuating cellular process becomes vulnerable to it, such that an additional reaction is triggered, causing a decompensation that may propagate into a clinical crisis, only later followed by the restoration of the original (unstable) equilibrium. In this case, rare fluctuations of cellular metabolism may coincide and compound one another to cause a crisis in the setting of a permanently abnormal enzyme function. Metabolic control analysis computations are accommodating and predictive of such fluctuations. Of course, these scenarios may be complemented with other, higher-complexity hypothetical mechanisms that need not act exclusively of one another.

A COMPLEX GENETIC LANDSCAPE

The precise pathophysiological mechanism of most neurometabolic diseases remains unknown, despite the accelerating pace of gene discovery. Among the mechanisms of metabolic diseases, accumulation of metabolic products, deprivation of substrates downstream of a metabolic blockade, negative-feedback regulation of enzymes, and induction or repression of genes by excess metabolite are a few logical, and not mutually exclusive, possibilities, although they do not fully explain many of these diseases. Additional consideration must be given to intragenomic signaling effects, such that a point mutation in one gene may result in upregulation or downregulation of other, unrelated genes. Such gene expression changes are sometimes viewed as “compensatory,” but they may well be deleterious, contributing to pathogenesis. Last, extra diagnostic and pathophysiological challenges are posed by the abundance of genetic polymorphisms of uncertain function present in all individuals, together with difficulties in separating true phenotypical characteristics from mere epiphenomena that are only tangentially related to the fundamental disease mechanism. In fact, many diseases are diagnosed by assaying metabolites generated by processes far removed from the original enzyme defect, which are only empirically found to associate with the classic form of that particular disease. The risk of errors inherent to this diagnostic approach cannot be overemphasized, particularly when confronted with unusual or partial forms of otherwise frequent diseases or with diseases so rare that the collective experience available is too sparse or insufficiently documented.

Genotyping is also subject to special considerations when applied to neurometabolic diseases. Several well-known clinical entities can be confused with phenocopies (i.e., conditions that manifest similar phenotypes but are due to mutations of different, unrelated genes). For example, mutation of the SCO2 gene can result in a phenotype that resembles spinal muscular atrophy, which is typically due to SMN1 gene mutations. In this instance, the correct diagnosis is reached by paying attention to a clinical feature (cardiomyopathy) and a biochemical marker (lactic acidosis) that would be atypical in spinal muscular atrophy, again illustrating the value of the initial clinical and analytical assessment of the patient. Genotyping is additionally dependent on the abundance of a particular mutation in the patient’s tissues. Thus, mosaicism for MECP2 mutations in the male and skewed X chromosome inactivation in the female may account for the disparate phenotypes of Rett syndrome, encompassing male neonatal death, male mental retardation, asymptomatic female carrier status, or classic female Rett syndrome. The abundance of mitochondrial DNA mutations also varies depending on the tissue chosen for genotyping (a phenomenon known as heteroplasmy; see Chapter 88) and, in some cases, several easily accessible tissues (blood, urinary sediment, buccal smear) must be examined to detect or to exclude a low-abundance mutation. Last, carrier status should be investigated in all genetically susceptible individuals related to a patient with a low (incomplete) penetrance disease. For example, asymptomatic mothers of hypotonic infants affected by myotonic dystrophy should be examined for subtle signs of myotonia and, if appropriate, offered genotyping.

A PLEOMORPHIC PHENOMENOLOGY

The phenotypes of neurometabolic diseases are often, and sometimes predominantly, nonneurological. Thus, abnormal urine odor, hepatomegaly, cardiomyopathy, cardiac arrhythmia, facial and skeletal malformations, neutropenia or disordered coagulation, and hair and skin abnormalities, among others, are characteristic features of some diseases (Table 107-1).

TABLE 107-1 Extraneurological Manifestations of Metabolic Encephalopathies

Neurometabolic diseases may mimic other disorders. In the presence of a seemingly nonspecific constellation of nonprogressive abnormalities, neurometabolic diseases can be misdiagnosed for other, more common entities (Table 107-2). Misconceptions to be avoided include that metabolic diseases necessarily have a progressive clinical course and that terms such as “cerebral palsy” identify diseases, rather than heterogeneous syndromes defined by relatively loose criteria. In these cases, the index of clinical suspicion should remain high and metabolic screening should be applied, as some of these covert metabolic disorders are treatable.

TABLE 107-2 Manifestations of Occult Neurometabolic Disease

DIAGNOSIS AND ITS VALUE

The possibility of prenatal diagnosis through sampling of fetal tissue obtained by chorionic biopsy is a reality for numerous metabolic diseases but is limited to instances where a particular disease has been identified in the family or a suspicious malformation detected in the fetus. Biochemical or molecular genetic assays of amniocytes are available for an increasing number of conditions. Yet, the most effective mode of detection is by voluntary screening of certain populations at risk. When both members of a clinically normal couple are carriers of a recessive trait or when a dominant disease afflicts just one member, or even when the mother carries a pathogenic mitochondrial DNA mutation, several reproductive options are available. Among these, testing of an early embryo after in vitro fertilization and before implantation, in vitro fertilization by a healthy donor, nuclear transfer, abortion, or, if available, early initiation of therapy of an affected newborn are all preventive interventions. All of them are variably applied depending on technological and cultural factors.

Newborn screening is still an underdeveloped and underused methodology with the potential to detect many, if not most, metabolic disorders. Testing is usually performed between the first 24 and 48 hours of life and uses dry bloodspots obtained from the heel and placed onto a filter paper card that is sent to a referral laboratory.³ Examples of conditions universally screened for are phenylketonuria and congenital hypothyroidism. At the present time, over 50 diseases, including specific disorders of amino acid, organic acid, and fatty acid metabolism, can be commercially screened for by tandem mass spectroscopy alone. However, in the United States, for example, some states test for fewer than 10 disorders, whereas others test for more than 30.⁴ Diverse efforts are under way to make this testing uniformly regulated and available.

It is also possible to diagnose a neurometabolic disease postmortem, and every effort should be made to offer an exhaustive biochemical investigation to each family in whom an unexpected neonatal or an infantile sudden death occurs.⁵ At a minimum, dry blood cards and skin punch biopsies can be obtained; the latter can be deferred for up to 18 hours after death and are used to establish live fibroblast cultures for use in biochemical and genetic assays. Other tissues may also be harvested under the guidance of the appropriate metabolic consultant. Photographs and a radiographic skeletal survey are important additional investigational tools.

GROUNDWORK FOR FUTURE THERAPIES

The principles of metabolic therapy (Table 107-3) have changed little in recent years, but their mode of application is being improved continuously. In the emergency setting, the decompensation or first manifestation of a neurometabolic disease may be accompanied by poor feeding, tachypnea, and acidosis due to accumulation of organic acids. This situation is associated with intracellular dehydration and, if too rapidly corrected by administration of fluids and/or bicarbonate, may lead to cerebral edema. Thus, gradual replacement of estimated losses is mandatory. Exchange transfusion is effective for the transient removal of soluble toxic metabolites. Peritoneal dialysis is simple but not as effective as exchange transfusion or hemodialysis in the emergent clearance of organic compounds. Hemofiltration uses an extracorporeal membrane to replace a plasmatic ultrafiltrate with electrolytes and nutrients. Hemodialysis is the most effective and rapid method for the removal of small soluble compounds but requires a significant commitment of resources and is thus not routinely performed in the emergency setting. Diets that diminish the use of a deficient metabolic pathway can be administered enterally or infused parenterally. Diets containing low protein, low carbohydrate, or high glucose sometimes with extra fat supplementation, all meeting minimum caloric, protein, and essential amino acid requirements, are available for specific diseases. Cofactors and vitamins should be administered at high doses when suspecting a potential vitamin-responsive disorder and they may later be maintained at a lower dose if the clinical response is inconclusive. Parenteral enzyme infusions are used with some success in lysosomal storage diseases such as Fabry disease, Gaucher disease, mucopolysaccharidosis type I, and Pompe disease, for lack of better methods to deliver the missing enzymes to the most affected tissues. Bone marrow transplantation corrects the enzymatic deficiencies of cells of hematopoietic origin in some mucopolysaccharidoses, and in some cases, the enzyme is partially restored in the brain. Early transplantation may prevent progression of neurological disease, but its longterm benefits are obscured by residual problems such as progression of skeletal and joint disability. Maximum safety and effectiveness are realized when the disease is in the preclinical stage and an HLA-matched sibling donor is available. Hepatic and liver/kidney transplantation has been considered in a variety of disorders with mostly anecdotal, mixed success. Liver transplantation appears to benefit patients afflicted by Wilson’s disease (see Chapter 108). Gene therapy remains an elusive ideal, as difficulties relative to targeting, maintenance, and expression of the corrected gene construct are being solved. Approaches that are under active investigation include pharmacological stimulation of residual alleles or unrelated genes using histone deacetylation inhibitors and loosening of translational fidelity by aminoglycoside antibiotic derivatives applied to mutations that result in the generation of premature DNA termination codons.

TABLE 107-3 Modalities of Metabolic Therapy

CLASSIFICATION OF THE METABOLIC ENCEPHALOPATHIES

The metabolic disorders of the nervous system can be classified according to several criteria such as age at onset (birth, infancy, childhood, adolescence, and adulthood), size of the predominant abnormal metabolite (large polypeptides or carbohydrates or small intermediary metabolism compounds), mechanism of inheritance (mendelian or mitochondrial, including the special varieties of imprinted, anticipated, polygenetic, and intergenomic signaling diseases), and loss or gain of protein function or, as preferred here, from a cellular perspective (Table 107-4). Such a classification, based on cellular structure, reflects the increased understanding of disease as a perturbation of cellular function, as well as the improved comprehension of the relationships that exist between individual cellular functions and structures. Thus, just as the cell is the reference framework in which genome, proteome, molecular function, regulation, and phenotype converge constituting the fundamental living entity, diseases and their symptoms and treatments are better understood at the level of complexity provided by a cellular point of view.

TABLE 107-4 Metabolic Encephalopathies