Article Text


Genes for peripheral neuropathy and their relevance to clinical practice
  1. M Donaghy
  1. University Dept of Clinical Neurology, The Radcliffe Infirmary, Oxford, UK
  1. Correspondence to:
 Dr M Donaghy
 University Dept of Clinical Neurology, The Radcliffe Infirmary, Woodstock Road, Oxford OX2 6HE, UK;

Statistics from

What does the future hold for genetic testing of peripheral neuropathies?

Identification of neurological disease genes has expanded and transformed the neurologist’s nosology in the last decade. Yet some clinicians see the resultant overload of detail as merely making the diagnosis and management of patients more cumbersome with little tangible benefit. Using the example of inherited peripheral neuropathy it is timely for a clinician’s perspective of where neurogenetics has taken present day neurological practice and what the future might hold.

Prior to molecular genetics, most neurologists kept a simple classification of inherited peripheral neuropathy in mind: demyelinating and axonal forms of hereditary motor and sensory neuropathy (HMSN), also known as Charcot-Marie-Tooth disease (CMT);1 the hereditary sensory and autonomic neuropathies (HASN);2 hereditary liability to pressure palsies;3 and the hereditary amyloiditic polyneuropathies.4 Although workaday, this classification’s deficiencies were apparent. For example, a significant minority of patients with HMSN could not be categorised cleanly as either Type 1 (demyelinating) or Type 2 (axonal) on the basis of electrophysiology, leading to the notion of intermediate forms.5 A range of severe HMSN—recessively inherited and affecting infants and children, and including the category known as congenital hypomyelinating neuropathy—seemed to evade consistent classification.6 Our inability to make definitive diagnoses for these rare infantile disorders was particularly distressing given the family implications, profound motor disability, and sometimes death, which could result. So, when the chromosome 17 reduplication of the PMP-22 gene in CMT was described, a new dawn promised accurate diagnosis of genetic neuropathy and its implications.

Since that first flush of promise, everything has become complicated by detail. There are more genetic neuropathies than we had ever imagined, and philately seems as useful as neurology for practising this subspecialty. We should be grateful to colleagues who have taken the trouble to distil this genetic literature and provide us with contemporary classifications.7–10 A brief look at their contemporary classifications for HMSN—now named CMT again—shows 37 identified genes or loci. Of these, 22 reflect demyelinating forms, 13 axonal and 2 intermediate, with a range of autosomal dominant and recessive, and sex-linked recessive transmission. At last the causes of the severe childhood and congenital hypomyelinating forms are emerging.7–10 The deluge continues apace. New interesting forms of CMT associated with mutations of the lamin A/C gene that encodes a nuclear envelope protein have been described,11 and of the gangliocide-induced differentiation-associated protein 1 (GDAP1).12 Also, different mutations affecting the same gene can produce demyelinating, intermediate or axonal forms of CMT—examples being CMT1B (myelin protein P0 mutations) and GDAP1 mutations. Reviews get out of date quickly—the website provides a real time encyclopaedia.

A similarly burgeoning range of genetic abnormalities seems to underlie other types of genetic neuropathy; acromutilating neuropathies of the HSAN type have been associated with the genetic locus for CMT Type 2B, or with the serine palmitoyltransferase long-chain base subunit 1 (SPTLC1) gene; the Riley-Day syndrome of familial dysautonomia is associated with mutations in the IkB kinase complex-associated protein (IKAP) gene; and selective neuropathies affecting small myelinated and unmyelinated fibres are associated with mutations in the tyrosine kinase A receptor for nerve growth factor (TRKA) gene. Memorising all of this is near impossible, even for academic clinicians subspecialising in peripheral neuropathy. The numerical classification has reached almost ridiculous proportions, whilst biochemical classifications have limited appeal for the non-cognoscenti.

Identification of these myriad genetic abnormalities underlying CMT has led to one surprising, but important intellectual realisation. At the beginning of this voyage of molecular genetic discovery, many had anticipated that each new gene would add a logical piece to the jigsaw of understanding the development of peripheral nerve and the maintenance of its structure. However, in reality no all-embracing view of the biology of peripheral nerve development and structure has emerged from this wide range of naturally occurring human mutations. Indeed, few mutations seem to affect processes or proteins unique to peripheral nerves. Most seem to affect functions likely to be fundamental to the biology of the cells of many different tissues. Even in the case of mutations affecting the protein responsible for peripheral nerve myelin compaction—the myelin protein zero (MPZ) gene—we find that the 80 different point mutations cause a bewildering array of CMT Type 1B (demyelinating), CMT Type 2 (axonal), forms intermediate between CMT1 and CMT2, congenital hypomyelinating and childhood onset forms of CMT, and a late onset form of progressive demyelinating neuropathy with features resembling chronic inflammatory demyelinating polyneuropathy.13 Clinically similar phenotypes of CMT1 (demyelinating) can be associated with abnormalities of the various genes for the growth arrest protein peripheral myelin protein 22 (PMP-22), early growth response element 2 (EGR2), ganglioside-induced-differentiation associated-protein-1 (GDAP1), myotubularin-related-protein-2 (MTMR2), n-myc-downstream-regulated-gene-1 (NDRG1), epithial-growth-factor-related-protein-2 (EGR2), and periaxin (PRX).8

Thankfully the inability to develop or maintain an axon in CMT2 is founded in straightforward neurobiological logic when due to mutations in the genes for kinesin-motor-protein-1-B (KIFIBβ) and neurofilament light chain (NFL), which involve axonal transport motor proteins and axonal structural intermediate filament proteins, respectively. But why do you get autosomal recessive CMT2 with one mutation of the laminin A/C nuclear envelope protein whilst other mutations produce the Emery-Dreyfus muscular dystrophies, limb girdle muscular dystrophies, cardiomyopathy, and partial lipodystrophy8 Molecular genetic analysis of diseases has taught us something of the specific biology maintaining the structure of the peripheral nervous system, but it has told us rather more about the vulnerability of neurones to seemingly more fundamental aberrations of cell biology.

Present day clinical practice has not been revolutionised by molecular genetic tests for peripheral neuropathy. Of course, such tests can provide patients with diagnosis couched in fact rather than diagnostic opinion; not unimportant in an era when scientific explanations dominate medical thinking. More confident prediction of prognosis and risk to offspring can follow accurate genetic diagnosis, but these were reasonably accurate previously given the defined clinical nature of most neuropathies and their clear inheritance patterns. In an era of ever smaller sibships, molecular genetics can allow proof of diagnosis when there is no opportunity to look for another affected family member. One’s experience in a peripheral neuropathy clinic shows that diagnostic genetic testing for HMSN does have its drawbacks. Routine molecular genetic testing is available for the 70% of HMSN due to reduplication of the PMP-22 gene, and another 20% due to mutations of myelin protein P0 and Connexin 32 genes. However, standard HMSN1A, due to a PMP-22 reduplication, is pretty easily diagnosable on grounds of clinical, electrophysiological, and pedigree features. A positive genetic test often merely adds icing to the cake. To a lesser extent the same can be said of the sex-linked neuropathy due to Connexin 32 mutations, and some of the more common myelin protein P0 mutations. Molecular genetic testing is not routinely available for the panoply of rarer genetic forms of HMSN, at least not in the United Kingdom. What this leaves is a sense of frustration for the clinician because, although routinely available molecular genetic testing is good for proving a diagnosis of which you are already fairly sure on clinical grounds, it offers nothing in that 10% of patients where you are sure that they have HMSN, but the features are atypical for the main syndromes. Screening for PMP-22 and P0 in infants and children can help enormously in differentiating Dejerine-Sottas disease and congenital hypomyelinating neuropathy from potentially treatable chronic inflammatory demyelinating polyneuropathy.

What does the future hold for genetic testing of peripheral neuropathies? Firstly, the difficulty of genetic diagnosis for the rarer and clinically more difficult syndromes would be revolutionised by the design of DNA chip, allowing simultaneous screening of a patient’s DNA for all known CMT gene mutations.14 Second, considerable ethical questions will face the use of molecular genetics in prenatal diagnosis of patients and families at risk of neuropathy. For a few with the potentially lethal or severely disabling infantile and childhood forms, such diagnosis might allow parents to make a decision about selective abortion. However, for the vast majority of patients with genetic peripheral neuropathy, the disorder is only slowly progressive or relatively stable throughout life, without causing overwhelming disability, and generally without significantly affecting life expectancy. When the subject is discussed in a hypothetical manner with some of my HMSN1A patients, none of them have felt that their foot deformity and reduced athletic abilities have made them wish they had never been born. Third, the holy grail of molecular genetic testing, as outlined by its pioneers15 was to allow new genes to be flown into the body’s cells to reverse the abnormality as early as possible to prevent permanent neural damage. The attraction of this remains as strong as ever, though the scientific methodology and commercial realisation remain far off. In the meantime, we should take heart from the reports of improvements in the CMT phenotype of animal models where overexpression of PMP-22 is modulated by administering ascorbic acid16 or progesterone receptor antagonists.17 One suspects that lower priority will apply to genetic modification of peripheral neuropathy than to some of the lethal or severely disabling genetic disorders of childhood. Nonetheless, the opening of such opportunities for all seems to have been signalled by the somewhat Orwellian UK White Paper,18 which has asked the Human Genetics Commission to report by the end of 2004 on “the case for screening babies at birth and storing information about their genetic profile for future use in tailoring healthcare according to their needs and their genetic make up.”

What does the future hold for genetic testing of peripheral neuropathies?


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.