Infantile cases present before the third year of life with motor delay or regression and cognitive difficulties. As the disease progresses, ataxia and hypotonia appear and mental function worsens, with death resulting within the first decade. The adult form of ARSA deficiency, defined as age of onset after 15–21 years of life, is much less common than the infantile variety and usually presents as a disturbance of mental function with subsequent evidence of diffuse CNS disease, including visual loss, ataxia, and movement disorders. Severe debilitation and death are inevitable, but may take 15 or more years to occur. Manifestations of the juvenile form are intermediate between those of the infantile and adult variants. Multiple sulfatase deficiency is clinically more severe than ARSA deficiency, with organomegaly and ichthyosis being unique to this variant (Thomas 1993).

While not ordinarily the dominant manifestation of MLD in any of its variant forms, peripheral neuropathy is usually present, albeit sometimes only electrophysiologically (Percy et al. 1977; Vos et al. 1982). Rare reports document a chronic progressive demyelinating neuropathy as the presenting manifestation of MLD in adults or children (Fressinaud et al. 1992; Vos et al. 1982; Yudell et al. 1967). The demyelinating nature of the neuropathy is evidenced by a marked slowing of conduction velocities (Vos et al. 1982; Martin et al. 1982a).

Diagnosis of MLD is usually easily made by demonstration of decreased cerebroside sulfatase activity in serum samples or cultured cells or by analysis of a gene defect. However, enzyme assays are normal in the rare cases of defective arylsulfatase A activator protein (Hahn et al. 1981; Shapiro et al. 1979; Stevens et al. 1981). Conversely, a relatively common pseudodeficiency of ARSA has been identified which can make assays of enzyme activity misleading (Hohenschutz et al. 1989; Kappler et al. 1991; Penzien et al. 1993). Without biochemical testing of family members, interpretation of low ASA activity can be difficult (Vos et al. 1982). Accumulation of sulfatides separates MLD from ARSA pseudodeficiency, is seen in all MLD variants, and is traditionally demonstrated by the finding of metachromasia in urinary sediment cells or a quantitative increase in urinary sulfatide excretion (Moser 1985). Identification of defects in arylsulfatase A ARSA, prosaposin (PSAP), or sulfatase-modifying factor–1 gene (SUMF1) now obviates some of the confusion.

The central nervous system shows diffuse demyelination sparing subcortical “U” fibers, with some loss of neurons, and numerous metachromatic-debris-filled macrophages (Gregoire et al. 1966; Haberland et al. 1973; Alves et al. 1986; Satoh et al. 1988). The ultrastructure of the inclusions seen in the CNS is similar to that of the PNS as described below (Gregoire et al. 1966). Metachromatic storage material is also detected in non-nervous tissues (Haberland et al. 1973; Gregoire et al. 1966; Satoh et al. 1988).

The common denominator of all the MLD is the accumulation of sulfatide, and this has been shown to be the major component of the material accumulated in the characteristic metachromatic inclusions (Fig. 20.6) (Suzuki et al. 1967). The storage material is seen prior to the onset of myelination and in nonmyelinating Schwann cells and is thus unlikely to reflect uncleared myelin debris (Meier and Bischoff 1976). The lack of correlation between sulfatide accumulation and the presence of segmental myelin damage (Martin et al. 1982a; Dayan 1967) suggests that Schwann cell dysfunction is not necessarily due to toxic accumulations of storage material. An alteration in the fatty acid composition and possibly the physical properties of myelin may be present in MLD, and this may somehow result in myelin damage (Thomas 1993).

The biochemical errors that cause MLD are dysfunction of arylsulfatase A, its activator protein (AB variant), or of multiple sulfatases. The ARSA gene has been fully characterized (Gieselmann et al. 1991). The fundamental difference between the various age-defined MLD subsets seems to be related to the amount of residual ARSA activity (Polten et al. 1991). In cell culture, fibroblasts from patients with adult-onset MLD are better able to clear sulfatide than those taken from infantile-onset patients (Percy et al. 1977). Polten and colleagues (1991) identified two important mutations of the ARSA gene: the “I” allele and the “A” allele. The former results in an unstable RNA that does not permit synthesis of a functional ASA, while the latter leads to synthesis of a rapidly degraded ASA protein (Polten et al. 1991). Patients with an I/I genotype have no functional ASA activity and present in infancy. Patients with the A/A genotype have sufficient enzyme activity to prevent presentation until early adult life, and patients with the I/A genotype are intermediate (Polten et al. 1991). The MLD pseudodeficiency gene is yet another ARSA allele which leaves sufficient enzyme activity to permit a normal phenotype, even if paired with an MLD allele (Penzien et al. 1993). Although combination of such alleles seem to give a satisfactory account for phenotypic variability, there are some pedigrees with phenotype variability which would be hard to explain on this basis (Clarke et al. 1989).

Pathogenetic mechanisms in MLD have been further examined in transgenic mice deficient in cerebroside sulfotransferase (CST). CST mice show impaired maintenance of Nav 1.6 sodium channels in mature nodes and migration of Kv1.2 channels from the paranodal to the juxtaparanodal region, resulting in the disappearance of Caspr and neurofascin 155 (NF155) clusters, essential components of the axo-glial junction (Eckhardt 2008; Dupree et al. 2005). In the peripheral nervous system, CST-deficient nerves extend axonal protrusions at the node of Ranvier, abnormal enlarged vesicles, and unusually short or absent Caspr and NF155 clusters (Hoshi et al. 2007). Additionally, the number of Schmidt–Lanterman incisures are increased and elevated levels of annexin II suggest structural changes at the Schmidt–Lanterman incisures (Eckhardt 2008; Hayashi et al. 2007).

A variety of treatment schemes have been developed for a number of lysosomal storage disorders (Miranda et al. 2013; Patil and Maegawa1 2013; Pierret et al. 2008). Enzyme replacement therapy (ERT) is undergoing clinical trials. Small molecules capable of crossing the blood–brain or blood–nerve barriers can be used as enzyme enhancers of diverse ASA mutants, either as pharmacological chaperones or as proteostasis regulators, to reduce the biosynthesis of sulfatides or target different affected pathways downstream of the primary ARSA deficiency. Hematopoietic stem cell transplantation using bone marrow-derived or umbilical cord blood-derived sources represents another therapeutic possibility, especially using cells which overexpress the ARSA gene. Bone marrow-derived mesenchymal stromal cells (BM-MSCs) have also been investigated as therapeutic agents (Miranda et al. 2013; reviewed in Patil and Maegawal 2013). BM-MSCs can provide the defective protein, secrete a number of neurotrophic factors and chemokines, have low immunogenicity, lack tumorigenic potential, and have a strong ability to home to injured tissues after being systemically injected. Some small molecule pharmacological chaperones may assist misfolded-prone mutant enzymes by assisting their ER folding and others may work on the lysosomal environment normalizing calcium levels (Patil and Maegawal 2013).

Globoid cell leukodystrophy (GLD, MIM#245200) is an autosomal recessively inherited disease caused by mutation in the galactosylceramidase gene (GALC) resulting in defective galactosylceramide-β-galactosidase activity causing accumulation of galactosylceramide in neural and nonneural tissues. Similar to MLD, an atypical form of GLD (MIM#611722) is produced by saposin A activator deficiency. Onset is usually in the first year of life with retarded development and irritability, rapid progression to spasticity, blindness, deafness, and decerebration, and death following within a few years of onset (Hagberg 1984). Some patients, usually those with onset later in childhood, have a more indolent course characterized by visual failure, ataxia, spasticity, and dementia and may survive into the second or third decade (Kolodny et al. 1991). Onset in adulthood is very rare (Hagberg 1984; Hedley-Whyte et al. 1988). Peripheral neuropathy is a usually a minor part of the infantile disease, but may be evidenced by diminished distal reflexes or hypotonia, and becomes more prominent with increasing age (Hogan et al. 1969; Kolodny et al. 1991; Dunn et al. 1969). Cases with neuropathy as the dominant manifestation have been described (Hedley-Whyte et al. 1988; Lyon et al. 1991). More often there is only electrophysiological evidence of neuropathy, marked conduction slowing being typical (Dunn et al. 1969; Suzuki and Grover 1970; Kolodny et al. 1991; Lyon et al. 1991). Diagnosis may be made by assay of β-galactosidase activity in cultured fibroblasts, leukocytes, or serum (Kolodny et al. 1991) or, more recently, from gene analysis.

Globoid cell leukodystrophy derives its name in part from the prominent demyelination that afflicts the CNS and in part from the presence of “globoid” cells, multinucleated giant cells up to 120 μm in diameter containing PAS-positive material (Allen and de Veyra 1967; Suzuki and Grover 1970). Globoid cells are seen predominantly in the brain, often around vessels, and may be seen in nonneural tissues, but are not present in peripheral nerves. However, ultrastructural examination reveals that the material stored in globoid cells is the same as the storage product detected in the peripheral nervous system.

The mechanism of myelin damage and cell death in GLD remains uncertain (Suzuki and Suzuki 1989). Defective function of the enzyme galactosylceramide-β-galactosidase (also called galactosylceramidase) is always associated with GLD. Dogs and rats with the same defect have an identical disease (Suzuki and Suzuki 1989), and GLD affects only those tissues where galactosylceramide accumulates. Injection of cerebroside, but not other sphingolipids, into animal tissue produces typical globoid cells (Olsson et al. 1966). The “indigestibility” of the storage substance of GLD (as described above) may relate to the formation of multinucleate giant cells. A related sphingolipid, psychosine, also broken down by galactosylceramide-β-galactosidase, is consistently found to be increased in GLD tissue and is highly cytotoxic (Suzuki and Suzuki 1989). Various neuropathic mechanisms have been proposed in studies of the twitcher mouse model of GLD including activation of the pro-apoptotic protease caspase 3 in sciatic nerves (which occurs before demyelination), axonal instability, defective axonal transport, and deregulated ion channels including loss of Na⁺ channel concentration at the nodes of Ranvier (Siddiqi et al. 2006; Sakai 2009; Kagitani-Shimono et al. 2008; Smith et al. 2011). Psychosine rapidly accumulates in lipid rafts (White et al. 2009) and may interfere in multiple signaling pathways (Suzuki 1998). Thus, it is likely that the accumulation of galactosylceramide and its toxic metabolite psychosine causes the histologic features and tissue injury seen in GLD (Suzuki and Suzuki 1989).

As in other lysosomal disorders, experimental treatment methods have been investigated in human or animal models including hematopoietic stem cell transplantation, chaperone therapy, enzyme replacement therapy, deprivation of the substrate, gene therapy, and cytokine therapy (Sakai 2009). Siddiqi et al. (2006) found that nerve conduction studies improved in 7 (60 %) of 12 patients after hematopoietic stem cell transplantation followed for an average of 18 months.

Fabry disease (MIM#301500), also called angiokeratoma corporis diffusum in reference to the frequent presence of cutaneous angiokeratomas, is an X-linked disease resulting from a mutation in alpha–galactosidase A (GLA), has been localized to the long arm of the X chromosome, and codes for α-galactosidase A which in its full blown form affects only males, although heterozygote females may be mildly symptomatic (Morgan et al. 1990; Brady 1993; Biegstraaten et al. 2012). Age of symptom onset is usually childhood or adolescence, with diagnosis often delayed into adult life, the significance of skin lesions and corneal “whorl” dystrophy (Fig. 20.9) going unappreciated for many years (Morgan et al. 1990). Presentation with renal dysfunction or episodic painful paresthesia is typical, with cardiac and cerebral vascular disease representing important causes of morbidity and renal failure in mid-adult life being the usual cause of death prior to the availability of dialysis and renal transplantation. Autonomic features include hypo- or anhidrosis, reduced salivation and lacrimation, and gastrointestinal tract disordered motility; however, decreased sweating has been attributed both to autonomic neuropathy and to direct sweat gland involvement (Bersano et al. 2012). Routine clinical and electrophysiological testing for evidence of a peripheral neuropathy is often unrewarding, but specific assessment of small fiber function (Ohnishi and Dyck 1974; Morgan et al. 1990) and quantitative determination of numbers of axons within the epidermis (IENFD, Biegstraaten et al. 2012) will reveal impairment. Accumulation of glycolipid in DRG neuron cell bodies and their eventual loss may also play a role (Schiffmann 2006). Diagnosis is often suspected on the basis of the clinical picture and is confirmed by assay of the activity of the affected enzyme, α-galactosidase A (lysosomal hydrolase a-galactosidase A), in plasma or leukocytes. This is typically 10 % or less of the normal value in affected individuals and 50 % of normal value in heterozygote females. Molecular techniques may be used to demonstrate GLA, alpha–galactosidase A, gene defects.

Nerve biopsy has minimal diagnostic value in Fabry disease because of the characteristic clinical manifestations and the presence of reliable noninvasive tests. If tissue is taken, the storage material of Fabry disease will be found predominantly in vascular cells: endothelial cells, smooth muscle cells, and pericytes; skin biopsy will readily demonstrate characteristic inclusions. However, when the diagnosis is unsuspected, nerve biopsy reveals pathognomonic histology, as has occurred once in our experience. Other detailed reports of nerve histology are available (Ohnishi and Dyck 1974; Kocen and Thomas 1970; Bischoff et al. 1968; Gemignami et al. 1984; Sima and Robertson 1978; Tabira et al. 1974; Fukuhara et al. 1975; Vital et al. 1984; Tome et al. 1977; Pellissier et al. 1981; Schiffmann 2006).

The defective gene (GLA, alpha–galactosidase A, MIM#300644) has been localized to the long arm of the X chromosome and codes for α-galactosidase A. A number of different deletions and point mutations causing an abnormally formed or an abnormally processed protein have been found in patients with Fabry disease (Bernstein et al. 1989; Lemansky et al. 1987). Precisely how the genetic defect relates to disease manifestations is unclear. The peripheral nerve disease is thought to be a neuronopathy preferentially involving small myelinated and unmyelinated axon cell bodies (Ohnishi and Dyck 1974; Gemignami et al. 1984; Morgan et al. 1990). Alternatively, the prominent vascular involvement seen in Fabry disease has been suggested as a cause of peripheral nerve dysfunction (Fukuhara et al. 1975).

Enzyme replacement therapy and small molecule chaperones have increased residual enzyme activity in animal models and cultured cells from Fabry disease patients (Bersano et al. 2012; Toyooka 2011, 2013). Nonetheless, the IENFD was unchanged after 12–18 months of enzyme replacement treatment (Schiffmann et al. 2006), although there was evidence of improved small nerve fiber function (Hilz et al. 2004).

The Niemann–Pick (NP) group of diseases is characterized by tissue storage of sphingomyelin and/or cholesterol (Spence and Callahan 1989). All subtypes are inherited as autosomal recessive traits. Niemann–Pick disease is currently subdivided on the basis of biochemical and molecular criteria into several separate classes: types A (MIM#257200) and B (MIM#607616) (both type I in older classification) are deficient in acid sphingomyelinase and are sphingomyelin storage diseases (Takahashi et al. 1992), and types C and D (MIM#257220) (both previously type II) are deficient in the NPC1 protein and represent lysosomal disease of intracellular cholesterol trafficking. Foam cell infiltration and visceromegaly are common to all, but severe neurological involvement occurs only in types A (a rapidly progressive neurodegenerative course leading to death by 2–3 years of age) and C and not in type B (McGovern and Desnick 2011).

Hepatosplenomegaly is a central element of the NP clinical picture, and biopsy reveals “foam cells,” macrophages distended with storage material ultrastructurally similar to that described below. Peripheral neuropathy is usually not a clinically significant part of the disease (Gumbinas et al. 1975). Nearly all the NP patients with neuropathy discussed below had conduction slowing suggestive of a demyelinating neuropathy.

Diagnosis traditionally has been made through clinical suspicion and the histologic demonstration of “foam cells” on bone marrow or liver biopsy. More recently assays for sphingomyelinase in tissue culture or fluids have been replaced by genetic studies. Niemann–Pick type A is caused by mutation in the sphingomyelin phosphodiesterase-1 gene (SMPD1), which encodes acid sphingomyelinase (ASM; Schuchman et al. 1992) and Niemann–Pick disease types C and D are caused by mutation in the NPC1 gene. Recently a number of small molecule agents including cyclodextrin have been used to improve visceral and neurodegeneration (Perez-Poyato and Pineda 2011; Liu et al. 2010).

Clinicopathological studies of peripheral nerve in NP are rare (Landrieu and Said 1984; Da Silva et al. 1975; Gumbinas et al. 1975; Hahn et al. 1994; Elleder 1989; Elleder et al. 1986; Babel et al. 1970 p 130; Bischoff 1979; Dubois et al. 1990). Most reports have considered NP type A except those of Elleder (1989) and Hahn et al. (1994). Our experience consists of three nerve biopsies in a brother and sister with NP type A who survived well into adult life (Case 20.2). Two biopsies in the sister were taken 15 years apart.

Farber disease is a rare autosomal recessive infantile disease (MIM#228000) caused by mutation in the ASAH1 gene encoding acid ceramidase resulting in the deficiency of acid ceramidase, necessary for metabolism of the sphingolipid ceramide (Moser et al. 1989; Zappatini-Tommasi et al. 1992; Alayoubi et al. 2013). The characteristic clinical features are disseminated nodular subcutaneous, periarticular swellings and vocal cord thickening. The central nervous system is affected in about half of the patients. Peripheral nerve involvement is common, manifesting with hypotonia and atrophy, and nerve conduction studies have shown significant slowing (Fujiwaki et al. 1992; Pellissier et al. 1986). The usual outcome is rapid progression to death in the first few years of life, although some patients survive into the third decade (Moser et al. 1989).

Histologic examination of the nodules of Farber disease reveals a granulomatous infiltrate, sometimes with mature granuloma formation. Storage material may take on a variety of ultrastructural appearances including Farber bodies, “banana” bodies, and “zebra” bodies (Zappatini-Tommasi et al. 1992; Burck et al. 1985). Farber bodies are curved tubular structures 14–17 nm in diameter found in great numbers within membrane-bound vacuoles or free within the cytoplasm and believed to correspond to stored ceramide (Rauch and Aubock 1983; Takahashi and Naito 1985; Pellissier et al. 1986). They bear a slight resemblance to the curvilinear profiles of Batten–Kufs disease, but are smaller and tubular, not lamellated. Farber bodies are seen in fibroblasts, histiocytes, and endothelial cells in extraneural tissues. Zebra bodies have been described in endothelial cells (Rauch and Aubock 1983).

A few nerve biopsy specimens have been closely studied (Vital and Vallat 1987; Vital et al. 1976; Pellissier et al. 1986; Rauch and Aubock 1983; Burck et al. 1985). Light microscopic examination of nerves has shown widespread hypomyelination and may reveal mild loss of large axons not sufficient to reduce the total MF density below normal (Fig. 20.21a) (Pellissier et al. 1986). Vital et al. (1976) indicated that unmyelinated fibers were spared in their specimen, but this was not quantitated. High-power LM can reveal pathognomonic vacuoles in myelinating Schwann cells (Fig. 20.21b). Frozen sections did not reveal any sudanophilic or metachromatic material in the case of Vital et al. (1976).

Ultrastructural examination is diagnostic (Fig. 20.22). Myelinating Schwann cells contain electron-lucent membrane-bound clefts, which take a variety of shapes ranging from needlelike to elongated rectangular, angulated, polygonal, and even annular profiles (Fig. 20.23a, b). These may be up to 5 μm in diameter, may indent the axon, and are almost exclusively seen in myelinating Schwann cells (Rauch and Aubock 1983; Schmoeckel and Hohlfed 1979). Presumably the content of these structures is lipid, washed out in preparation. At times there is a residual content which Pellissier et al. (1986) described as “… either homogenous and osmiophilic, or less dark with an electron-dense lamellar component.” We have also observed, in nonmyelinating Schwann cells only, novel membrane-bound elongated bodies containing a central longitudinal striated region surrounded by amorphous electron-dense material (Fig. 20.24).

A mouse model has been developed (Alayoubi et al. 2013) via “knock-in” of a known human ASAH1 mutation into a conserved murine Asah1 gene locus. In that model it has been determined that the accumulation of ceramide prompts the release of MCP-1 which recruits circulating monocytes to tissues where they accumulate and fail to degrade the sphingolipids and ceramide they engulf, amplifying the foamy appearance which is characteristic of FD. Neonatal gene therapy using recombinant lentivectors encoding the human acid ceramidase gene has shown a salutary effect in this experimental model. In patients allogeneic bone marrow transplantation has shown success in reducing the peripheral symptoms in Farber disease but failed to improve neurological symptoms and patient lifespan (Ehlert et al. 2007).