There are many inherited and acquired clinical disorders caused by a defect in upper or lower motor neurone or the peripheral nerve. These include amyotrophic lateral sclerosis (ALS; upper and lower motor neurone), hereditary motor and sensory neuropathies (HMSNs; motor and sensory neurone and peripheral nerves), the spinal muscular atrophies (SMAs; lower motor neurone) and inflammatory peripheral neuropathies. In addition, ageing of muscle, some metabolic conditions, alcohol abuse and some drugs (e.g. vincristine) are accompanied by a neuropathy/denervation (see Chapter 18, Chapter 19, Chapter 23 ). Several causative genes have been identified, inheritance may be dominant or recessive and there are also X-linked forms.
ALS is one of the most common neurodegenerative disorders in adults with variable clinical features ( ). The first gene to be associated with ALS was SOD1, a Cu/Zn superoxide dismutase, but the most commonly mutated gene is now considered to be C9orf72 ( ) . With the wide application of next-generation sequencing, there is a growing list of additional genes in which various types of variants have been identified. This list includes fused in sarcoma ( FUS ), alsin ( ALS2 ), senataxin ( SETX ), vesicle-associated membrane protein-associated protein B ( VAPB ), TAR DNA-binding protein (TDP-43; TARDPB ), vasolin-containing protein ( VCP ), optineurin ( OPTN ), angiogenin ( ANG ), spatasin ( SPG11 ), profilin1 ( PFN1 ) and others ( ) (see Neuromuscular Disorders Gene Table www.musclegenetable.fr/ ). Many of the defective proteins share similar biological functions and are involved in similar pathways to those in other neurogenerative disorders, including autophagy, mitophagy and oxidative stress and mitochondrial function ( ).
The hereditary peripheral neuropathies are a heterogeneous group of disorders encompassing several clinical syndromes with dominant or recessive inheritance. Many of these still carry the eponymous titles of the clinicians who described them (e.g. Charcot–Marie–Tooth), but the advent of molecular genetics has led to reclassification and the identification of several different inherited types ( ). Detailed clinical and electrophysiological studies and studies of sural nerve biopsies are required to direct molecular analysis, and muscle biopsies usually contribute very little to this. The study of sural nerves is a specialized field beyond the scope of this book, and details can be found in a variety of textbooks (e.g. ).
Changes in the muscle as a result of a denervating process are similar, irrespective of the site of the lesion, be it in the neurone or the peripheral nerve, and it is rarely possible to precisely define the disorder from a muscle biopsy, although certain patterns are suggestive. Careful clinical and electrophysiological investigations often give a clue to the defective gene, and muscle biopsies are now performed less often in neurogenic disorders. For example, molecular analysis of the survival motor neurone ( SMN ) gene identifies the majority of cases with SMA, and severity and prognosis are based on clinical features not muscle pathology.
Some neurogenic atrophies, however, may mimic some muscular dystrophies or myopathies, such as distal myopathies. Muscle pathology can then be useful in differential diagnosis, and particular patterns of change may be revealed with histological and histochemical techniques. Immunohistochemistry may contribute to the interpretation of secondary changes in the muscle, but the primary gene product is not usually studied in the muscle. As mentioned previously, changes in relation to the nerve may be informative: for example, abnormalities in the sensory sural nerve in the diagnosis of peripheral neuropathies. We describe here the general characteristics of denervated muscle. These are the changes seen in chronic disorders such as ALS and can be distinguished from the early-onset childhood disorders such as SMA. As the pathology in the different forms of SMA is variable, this is described in detail, and a muscle biopsy may well be performed in advance of a molecular diagnosis or in cases where a neurogenic problem is not initially suspected.
General Pathological Features of Denervated Muscle
Diseases which involve the motor neurone are associated with a characteristic set of pathological changes in human muscle. Denervated muscle fibres shrink in size, but there may be little change in the internal architecture of the fibres. Even the cross-striations of individual atrophic muscle fibres are preserved until late in the atrophic process. The basal lamina round individual fibres often remains as the fibre shrinks and becomes folded (see Ch. 5 on electron microscopy). In chronic conditions such as ALS the atrophic fibres have an angular shape ( Fig. 9.1 ), in contrast to the rounded shape of the atrophic fibres in SMA. All that may be visible of some atrophic fibres in chronic denervation is a clump of nuclei which may be pyknotic (see Fig. 4.21 ). Pyknotic sarcolemmal nuclei may also be seen along the course of a pre-existing fibre.
Since one motor nerve supplies many muscle fibres, denervation will result in atrophic fibres scattered at random in a biopsy. These atrophic fibres are often clustered into groups, and the number of fibres within these groups increases with increasing severity of denervation, until whole fascicles may be rendered atrophic. The presence of this ‘small-group’ or ‘large-group’ atrophy is pathognomonic of denervation (see Figs. 9.1 and 9.3 and see section on SMA).
Two populations of muscle fibres are seen in denervated muscle: atrophic ones that are denervated and those that are relatively normal in size or hypertrophied. Histographic representation of this frequently shows a bimodal or twin-peaked configuration (see Ch. 4 ). This occurs because not all motor neurones are involved simultaneously. Fibres supplied by an intact motor nerve will obviously not atrophy and compensatory hypertrophy may occur as they take over the function of the atrophic muscle fibres, which have lost their motor supply. In addition, surviving nerves may sprout and reinnervate clusters of denervated fibres, causing them to enlarge again.
All the changes described previously can be seen with routine stains such as Gomori trichrome and haematoxylin and eosin (H&E). In addition, histological examination of nerves present in a sample may reveal changes such as loss of myelin or fibrosis ( Fig. 9.2 ), but these should be interpreted with caution as variability in these features can be seen in the absence of a neural defect. In chronic denervation, histochemical reactions are particularly useful in demonstrating the angular atrophic fibres ( Fig. 9.3 ). Atrophic angulated fibres and non-angulated fibre stain intensely with techniques for non-specific esterase, and oxidative enzymes, suggesting they are type 1 fibres (see Figs. 9.3 and 9.5 ), but with adenosine triphosphatase (ATPase) staining or immunolabelling of myosin isoforms they may be of either type ( Figs. 9.4 and 9.5 ). This is an important distinction, and a diagnosis of a denervating disease should not be made unless the atrophy involves both fibre types. In addition, these denervated fibres tend to have a positive esterase but negative acid phosphatase reaction, in contrast to regenerating or necrotic fibres which are strongly positive for both enzyme reactions. Denervated fibres may in some incidences lose their glycogen content and appear unstained and white with periodic acid–Schiff (PAS). With immunohistochemistry, some, but not all, angulated fibres may show fetal myosin and some may co-express fast myosin (see Fig. 9.4 ). As discussed in Chapter 6 , it is possible this reflects re-expression of fetal myosin rather than immaturity resulting from regeneration, but the possibility of regeneration stimulated by denervation cannot be excluded. Immature and non-innervated fibres, such as regenerating fibres, show neural cell adhesion molecule (NCAM; CD56) on all the sarcolemma, whereas on normal mature fibres it is confined to the neuromuscular junction ( ). Immunolabelling of NCAM may also highlight satellite cells, but this relates to their state of activation and thus they are not always apparent (see Ch. 6 ). Denervated fibres also show extrajunctional N no hyphen ie NCAM please ensure this is consistent through out the book CAM but, in contrast, neuronal nitric oxide synthase (nNOS) is not detected and reappears after reinnervation (see Fig. 9.4 ; ). Assessment of nNOS must always take account of immaturity, as nNOS is absent from immature fibres (see Ch. 6 ), but the combination of nNOS, NCAM and fetal myosin can help to distinguish denervated fibres from those that may be immature. There is no single marker for distinguishing an atrophic from a denervated fibre in human muscle, and the presence of some proteins that are associated with regeneration/development makes distinguishing atrophy from regeneration difficult. A number of studies of denervation and atrophy in animal species have identified quantitative alterations in several proteins that may prove in the future to be of pathological use in human muscle: in particular, components of the ubiquitin-proteasome, the autophagy-lysosome and AKT signalling pathways ( ). For example, Ring finger 1 (MuRF-1), antrogen-1 and the sodium channel Nav 1.5 are higher in atrophic fibres, while myosin-binding protein C is reduced ( ).
As has been pointed out earlier (see Ch. 3 ), the motor nerve has an important role in determining the particular type of muscle fibre. Thus, motor neurones may be thought of as supplying either type 1 or type 2 (slow or fast) motor units. Although it should be possible to have a denervating disease selectively involving a single type of motor neurone, this is not found in human disorders and both fibre types undergo denervation. If this criterion is used in making the diagnosis of denervation, it safeguards against diagnosing denervation in a patient showing, for example, selective fibre-type atrophy from other causes.
Grouping of fibres of the same type is pathognomonic of denervated muscle and results from collateral sprouting of the surviving nerves which reinnervate the denervated fibres. The number of fibres in a group can vary, but if a fibre is completely surrounded by fibres of the same type this is usually considered as a group. Some idea of prognosis may be obtained from the presence of type grouping, since extensive fibre-type grouping, in association with very little fibre atrophy, should imply good compensation of the denervating process and thus a milder or more chronic process. It is important to distinguish fibre type predominance from fibre type grouping, especially in a small sample, and atrophy of both types should be present to make a diagnosis of denervation; however, group atrophy may not always be apparent (see Fig. 9.5 ). Grouping may sometimes be difficult to assess and it may not be very obvious in all biopsies from neurogenic conditions ( Fig. 9.6 ). In addition, areas of a biopsy may vary (see Fig. 9.6b ). Co-expression of myosin isoforms can be a feature of denervated fibres and may also make fibre type grouping difficult to define, especially in sections stained for ATPase (see Figs. 9.4 and 9.12 ).
Another change seen in denervating diseases is the presence of target fibres ( ). In some biopsies the dark rim round the pale zone devoid of oxidative enzyme activity may not be prominent and the fibres then have an appearance similar to a fibre with a central core (see Figs. 9.3, 9.11 and 9.13 ). These core-like areas can occur in fibres of varying size (see section on SMA), and, although they can be quite long, they rarely extend down the whole length of a fibre, in contrast to central cores ( ). The possibility has been raised, based on experimental studies in animals, that target fibres may represent reinnervation rather than denervation of the fibre ( ). In addition, their presence in acute recovering neuropathies and chronic peripheral neuropathies in which reinnervation is a prominent feature may also support this view.
Other architectural changes, such as moth-eaten fibres and whorled fibres, may occasionally be seen with oxidative enzyme stains. Excess endomysial connective tissue is not usually a feature of denervated muscle but can occur, particularly in chronic conditions. Internal nuclei may also be seen in some hypertrophic fibres and, occasionally, vacuoles may occur. These features, in the absence of fibre type grouping, may make it difficult to distinguish from a ‘myopathic’ appearance (see section on SMA).
Spinal Muscular Atrophy
There are various forms of SMA ( ). The form linked to chromosome 5q is not only one of the most common forms but also is one of the most common neuromuscular disorders, with incidence being quoted as about 1 in 6000 to 1in11,000 births ( ). It is an autosomal recessive condition in which the degenerative process results in loss of anterior horn cells of the spinal cord. The genetic defect is in the SMN gene on chromosome 5q, and over 95% of SMA patients have a deletion affecting exons 7 and 8 of the telomeric copy of the SMN gene ( SMN1 ). Due to an evolutionary genetic duplication, humans (but not many of the animal species used for research) have a centromeric gene ( SMN2 ) that is almost identical to SMN1 , but the SMN2 transcript lacks exon 7. Severity of the disorder correlates with the amount of full-length SMN protein produced by the variable number of copies of the SMN2 gene ( ) . Some current therapeutic strategies, such as antisense oligonucleotide therapy, are aimed at the up-regulation of centromeric SMN2 with some encouraging results ( ) . The SMN gene product has an important role in several cell functions, including assembly of a spliceosomal complex, in pre-RNA splicing, axonal mRNA transport and actin cytoskeletal dynamics ( ). The affected region of chromosome 5 is complex, and other genes that are also duplicated and have telomeric and centromeric copies may also be deleted, in particular, the neuronal apoptosis inhibitory protein gene ( NAIP ) and p44 . The high proportion of patients with a deletion involving exon 7 and 8 of SMN1 provides a reliable molecular screen that identifies the majority of SMA cases and also heterozygous carriers. The presence of exons 7 and 8, however, does not totally exclude a diagnosis of SMA, as a few atypical cases with mutations outside this region or with point mutations have been documented ( ). The reliability of molecular analysis has revolutionized diagnosis of SMA, and the role of muscle biopsy has diminished. Muscle biopsies, however, may be taken in cases where the diagnosis is less obvious clinically. No deletions of SMN1 have been found in patients with adult-onset motor neurone disease, but it has been suggested that there is a significantly higher incidence of deletions of exon 7 of centromeric SMN2 in these patients than in the normal population ( ).
Other genes are responsible for clinical syndromes related to SMA. Some predominantly affect the lower limbs, in particular the distal muscles, rather than proximal muscle involvement (see below later), and some are asymmetrical or more focal in presentation. Some of the adult forms have a dominant inheritance, some recessive, and there are also X-linked forms. An X-linked bulbospinal form with a benign course and associated facial fasciculation, severe muscle cramps and gynaecomastia (Kennedy disease) is caused by a CAG nucleotide expansion of the first exon of the androgen receptor ( AR ) on chromosome Xq12 and mutations influence the proteasomal and autophagic pathways ( ). A severe X-linked form with associated arthrogryposis is caused by mutations in the UBA1 gene mapped to Xp11, the protein product of which has a role in ubiquitin homeostasis and neurogeneration ( ). A distinct autosomal recessive form with severe diaphragmatic and respiratory involvement (diaphragmatic spinal muscular atrophy with respiratory distress; SMARD1) is caused by mutations in the gene encoding immunoglobulin microbinding protein 2 ( IGHMBP2 ) on chromosome 11q13-q21 ( ). The severe diaphragmatic involvement is a useful distinguishing clinical feature from classical chromosome 5 SMA, in which the diaphragm is selectively spared, in spite of severe involvement of the intercostal muscles. Cases with pontocerebellar hypoplasia may also resemble SMA clinically and pathologically ( ), but have associated features of central nervous system involvement. They were shown not to be associated with the SMN1 gene. The function of another gene ( MEGF10 ) defective in a disorder often with early diaphragmatic involvement with some similarities to SMARD1 is not yet clear but the features relate to a myopathy rather than a neuropathy and it may have a role in myogenesis ( ). The typical features of denervation in muscle biopsies are not always obvious in these more rare forms of SMA-like syndromes (see below), and in MEGF10 cases the features are myopathic (see Ch. 15 ).
In this section we cover the muscle pathology associated with the three main clinical groups (SMA I, II and III) of autosomal chromosome 5q SMA, both the cases manifesting the classical features and those where these may be less obvious.
Three main clinical groups, types I, II and III, have been arbitrarily defined on relative severity and whether the ability to sit or stand unaided and walk is achieved (see below). There is clinical variability within each group, and some cases are not easily categorized into these rigid groups ( ). Some clinicians distinguish cases with a mild adult onset as a fourth group (SMA IV), while others include these within the spectrum of SMA III. Similarly, severe cases with onset in utero are sometimes referred to as SMA0 rather than being grouped with type I patients ( ). Type I SMA (Werdnig–Hoffmann disease) is the severest form, with onset in utero or early infancy and patients are never able to sit; they usually die in the first year and rarely survive beyond 2 years of age. In type II (intermediate SMA), onset is usually between 6 and 12 months and the ability to sit unsupported is achieved, but they fail to stand. The clinical spectrum of SMA III (Kugelberg–Welander syndrome), in which patients achieve ambulation, is broad, ranging from onset in the second year of life, with little or occasionally with marked progression, to adult onset with mild progression. Differential diagnosis from a myopathic condition is important in these cases. The typical features of each group are summarized in Tables 9.1, 9.2 and 9.3 , and clinical features are constantly being updated and reassessed as therapies are developed ( ).