Introduction

Inherited neurological diseases in African populations have been largely neglected. Africans will be left behind in the global quest for targeted genetic therapies without an African perspective on disease-associated mutations. While modern genomic approaches have led to new gene discoveries in complex inherited neuromuscular disorders [1], the genetic landscape of neuromuscular disorders in Africans are barely known.

Inherited neuromuscular disorders, such as hereditary spastic paraplegia (HSP) and Charcot-Marie-Tooth (CMT) disease are not rare in North America, Europe, and Asia with a global prevalence ranging between 4.3/100,000 for HSP and 82.3/100,000 for CMT [1, 2]. There are no epidemiological data for Africa. Akinyemi et al. reported that of the 58 African states, scattered reports related to the genetics of neurological disorders emanated from only 17 countries and these were heavily concentrated in four North African countries [3]. Presently in South Africa, and with relevance to this review, the National Health Laboratory Service offers one genetic screening test for CMT (the common PMP22 gene duplication/deletion) and none for HSP. Although the screening test to detect the most common cause of Spinal Muscular Atrophy (SMA) (homozygous deletion/disruption of SMN1) has been available in South Africa for more than 2 decades, only isolated cases are able to access gene therapies for SMA which are available in resource-rich countries. Therefore, there is an urgent need to address the disparate healthcare in inherited neuromuscular diseases which exist between the developed world and Africa. However, there are presently a few initiatives such as the International Centre for Genomic Medicine in Neuromuscular Diseases (ucl.ac.uk/genomic-medicine-neuromuscular- diseases/) to prioritise the advancement of genetic research in neuromuscular diseases, and the broader H3Africa Initiative to expand population reference data in sub-Saharan Africans [4], which will facilitate the analysis of pathogenic genetic variants in Africans with rare inherited diseases. This review will synthesize genetic reports from HSP, CMT and SMA in Africans, to give an overview of the genetic variants and their associated phenotypes, which have been reported and can be used as a reference resource for African researchers and clinicians. A separate review of inherited myopathies and muscle dystrophies in Africans, is in progress.

Methodology

PubMed was searched for journal articles related to the molecular genetic causes of HSP, CMT, and SMA in Africa. The following MeSH terms were used (hereditary spastic paraplegia) or (Charcot-Marie-Tooth disease) or (genetic neuropathies) or (inherited neuropathies) or (familial amyloid neuropathies) AND (Africa), and (spinal muscular atrophy) AND (Africa) for searching PubMed. We performed a google search using search terms: “genetic neuropathies Africa”, “neuromuscular inherited Africa”, “hereditary spastic paraplegia Africa”, “Charcot-Marie-Tooth disease Africa”, “spinal muscular atrophy or SMA and Africa”, “Kennedy’s syndrome Africa”. We also manually searched the reference lists of reports and review publications to look for additional references and searched for “Africa” within articles. We confined this review to studies with genetic descriptive components. Studies involving linkage analysis of a large genomic region or single genes where a genetic diagnosis was not reached were excluded as the focus of this paper was on the identified genetic causes of inherited neuromuscular disorders in Africans (European or Indian ancestries were excluded) (Fig. 1). Reports related to infectious disease-associated neuropathies were excluded. Only English articles were reviewed which resulted in the exclusion of two reports from 2002 and 2008 which were published in French.

Fig. 1
figure 1

Flow chart describing the methodology used for the literature curation

Full size image

The data collected from the reports included: the genetic results of probands with African-genetic ancestry, phenotypic features including age at onset, inheritance pattern and consanguinity, and electrophysiological features. We also noted genetic variants found in Africans but which had been previously reported in non-African families, whether there were attempts to determine segregation of the putative disease-causing variant within the family, and whether population controls were assessed for the variant. Segregation of genetic variation was scored positive if the putative disease-causing variant was (a) excluded in at least one unaffected individual of the same age or older than the affected individual for autosomal dominant inheritance, or (b) confirmed in the heterozygous state in at least one unaffected parent for autosomal recessive inheritance. Variants in which functional studies had been performed were noted. In addition, as many of these publications were published prior to the establishment of large scale public genetic databases, we also interrogated the gnomAD database (last accessed 6 Sept. 2021) to determine the frequency of putative disease-causing variants [5]. For variant nomenclature we followed the Human Genetic Variation Sequence (HGVS)(version 20.05) guidelines [6].

Results

Most reports used the following genetic methodologies: Targeted PCR sequencing and/or Sanger sequencing; multiplex ligation-dependent probe amplification; and HSP or CMT gene panels. Some studies used appropriate microsatellite markers to construct segregating haplotypes to establish linkage in families followed by targeted Sanger sequencing of coding exons. More recent reports (from 2013) used whole exome sequencing (WES) to screen protein coding variants or  performed comprehensive whole genome sequence (WGS) analysis.

Hereditary spastic paraplegia

Hereditary spastic paraplegias (HSP) are clinically characterized by a progressive gait disturbance due to increasing spasticity of the legs. Clinicians have recognized two forms of HSP; patients who only have features of HSP (or pure HSP), or those with additional neurological system dysfunction such as ataxia, cognitive/intellectual disability, extrapyramidal signs, and features of sensory ± motor neuropathy. The latter are called complex HSP.

Although the clinical manifestations of HSP usually manifest over years rather than months, it remains important to exclude other non-degenerative conditions by performing imaging studies of the brain and spinal cord. Magnetic resonance imaging (MRI) of the brain may be normal or show atrophy, and/or may show thinning of the corpus callosum and/or increased white matter signal intensities (Table 1). In Africa, infectious causes such as HTLV1-associated tropical spastic paraparesis is a concern in adults, which can be excluded with cerebrospinal fluid examination and/or serology [7]. Lathyrism caused by excessive consumption of the chickpeas of the lathyrism family, is endemic in Ethiopia, and can result in a slowly progressive paraparesis [7].

Table 1 Genetic causal variants of HSP identified in African populations
Full size table

More than 88 genes have thus far been reported to cause HSP, which are designated as SPastic Gait/Gene or SPG genes [2, 8]. Inheritance patterns in HSP are predominant autosomal dominant (AD), except in areas with high consanguinity, such as in North Africa, where autosomal recessive inheritance (AR) patterns are prevalent [2, 8, 9] (See Table 1). X-linked and mitochondrial maternal inheritance patterns of HSP are rare [8]. World-wide SPG4 is reported to account for up to 79% of HSP cases with AD inheritance, albeit mainly in those with Caucasian ancestry [8]. Other frequent causes of AD HSP include the monoallelic pathogenic variants in KIF1A, as well as SPG3A and SPG31 [8]. Genes accounting for HSP cases with AR inheritance patterns include SPG11 and SPG7, followed by SPG15 and SPG5 in overall frequencies [8]. Interestingly, three genes (KIF1C, SPG7, KIF1A) have been reported to associate with mixed inheritance patterns related to allele-dose-dependent clinical phenotypes i.e. milder phenotypes with heterozygous variants, and more severe phenotypes with homozygous states [8].

HSP in North Africa

Most of the genetic reports on HSP in Africa are from North Africa and are based on targeted linkage analysis in families to identify a candidate gene locus that segregated with the phenotype, followed by direct gene sequencing (Fig. 2). The commonest gene harbouring a pathogenic variant identified in HSP cases, was SPG11 (KIAA1840) associated with thin corpus callosum on MRI [9] (Table 1; Additional file 1: Table A). Only four of the reports used WES for a more comprehensive gene screen in North African cases with HSP conditions, and one used WGS data [10,11,12,13]. The commonly encountered AR-HSP causal genes (SPG11, SPG15) in North African populations were also amongst the top seven genes in a large European cohort [9, 10, 14,15,16,17,18,19]. Other AR-HSP genes amongst North African families included SPG5, SPG7, SPG35, SPG46, SPG48, SPG51, and SPG57, as well as mutations in the ALS2 and SACS genes [9, 13, 14, 20, 21] However, private mutations in novel genes (RNF170, CAPN1, KLC2, B4GALNT1, DDHD1, CCT5) were also reported to be disease-causing in isolated cases or families [10,11,12, 21,22,23,24,25].

Fig. 2
figure 2

Bubble map depicting the number of genetic reports in HSP- and CMT-related disorders in African countries

Full size image

The most frequent gene variants accounting for autosomal dominant inheritance patterns, were found in SPG4 (SPAST) [26,27,28].

HSP in sub-Saharan Africa

Six reports were found from sub-Saharan Africa of which two screened a targeted panel of 58 HSP genes [29, 30] and two used WES [17, 31] (Table 1; Additional file 1: Table A; Fig. 2). The cases from consanguineous families from Kenya and Mali with homozygous pathogenic alleles were most frequent with SPG11 variants, followed by SPG7, SPG35 and SPG43 [32].

There were two reports on autosomal dominant HSP; one black South African family with a novel SPG3A variant [33] and a family from Mali with SPG10 [30]. Therefore, the common SPG genes present in Europeans [19], viz. SPG3, SPG4 and SPG10, have been found in isolated African cases.

Genetic neuropathies

The largest group of genetic neuropathies are referred to as the Hereditary Sensory Motor Neuropathies or Charcot Marie Tooth (CMT) disease. CMT affects predominantly the motor and sensory nerves, although the CMT-spectrum includes rare forms with autonomic and motor only involvement [34]. The clinical features of CMT disease are progressive and symmetrical weakness and wasting of distal muscles of the foot and ankle which may result in clumsy feet, foot deformities such as pes cavus, and loss of deep tendon jerks. Later, there may be involvement of the distal arms with wasting and weakness although clawing of the hands is less common. Some genetic neuropathies may have early and predominant upper limb involvement. Sensory involvement ranges from mild distal numbness to severe loss of sensation with ulcers, and/or sensory ataxia. The insidious clinical progression of CMT distinguishes it from subacute acquired inflammatory neuropathies in most cases, although rare forms of CMT can give a patchy electrophysiological picture with conduction blocks that may resemble treatment-resistant chronic inflammatory demyelinating polyradiculoneuropathy [34]. In Southern African populations, where HIV-infection is prevalent, small fibre painful neuropathies may be considered in cases with more advanced HIV-infection, and/or with concomitant tuberculosis and isoniazid therapies, but weakness is extremely rare [35]. This contrasts with CMT where the absence of motor involvement is unlikely [34].

In the pre-molecular era, CMT was categorized by the electrophysiological involvement of the sensory and motor nerves, whereas the CMT neuropathies are further categorized according to their electrophysiological findings into three types; the demyelinating forms (nerve conduction velocities (NCVs) < 38 m/s in the upper limbs), axonal forms (NCV > 45 m/s), or the intermediate types of CMT (NCV in the upper limbs between 25 and 45 m/s) [1]. All neuropathies categorized as HSMN or CMT, would show evidence of motor and sensory nerve abnormalities on electrophysiological testing, whereas hereditary motor neuropathy (HMN) by definition would have normal sensory nerve action potential responses. However, there appears to be genetic overlap between CMT2 and HMN subtypes [36].

In North America and European populations, most CMT neuropathies show AD inheritance compared to AR inheritance which comprises < 10% of cases. In contrast, in North Africa, where consanguinity is high [37], most of the cases published showed AR inheritance (Table 2). Similar to what is observed in HSP, CMT shows substantial genetic heterogeneity with > 100 genes identified which can cause genetic neuropathies [1]. The most common autosomal dominantly inherited CMT in North America and Europe, the demyelinating CMT1A caused by a duplication in the PMP22 gene, accounts for ~ 40% of genetic neuropathies [38], yet remains unreported in those with African genetic ancestry.

Table 2 Genetic causal variants of Charcot-Marie-Tooth (CMT)-related disorders reported in African populations
Full size table

CMT in North Africa

Due to high levels of consanguinity in Algeria, Morocco, and Tunisia, AR-CMTB1 (LMNA) was by far the commonest, followed by CMT4A (GDAP1), CMT4C (SH3TC2), and CMT4B2 (MTMR13) [24, 39,40,41,42,43,44,45,46,47,48,49,50](Table 2). These are present in < 1% of AR-CMT cases in non-Africans [38]. Two Algerian families had compound heterozygous pathogenic variants with the common GDAP1 S194* variant [51], which has a population frequency of 2.3 × 10–5 (Additional file 1: Table B). Isolated cases were reported with CMT4B1 and CMT4F [37].

Four Algerian families with distal HMN (dHMN5A) and AD inheritance patterns were reported with the rare [38] GARS pathogenic variants characterised by predominant upper limb weakness and hand wasting [52, 53].

CMT in Sub-Saharan Africa

Three reports were found (Fig. 2). One CMT1B (MPZ) Nigerian AD pedigree with late-onset demyelinating neuropathy [54]; and an intermediate CMT phenotype with conduction blocks and a novel PLEKHG5 variant which segregated in the family [55]. A consanguineous pedigree from Mali was reported with a heterozygous GARS variant, but without evidence of segregation or population screening [56]. Caution must be used in interpreting variants with “incomplete penetrance” to explain incomplete segregation of variants particularly in Africans where the population data are sparse and genetic variation is increased [57].

Familial amyloid neuropathies

There are three types of familial amyloid neuropathies (FAP) which are categorised according to the abnormal precursor protein which will result in downstream deposition of amyloid fibrils viz. transthyretin (TTR), apolipoprotein A-1 and gelsolin [58]. Although some TTR mutations can cause FAP, which characteristically manifests with sensory and autonomic nerve dysfunction alone, a rare manifestation is oculoleptomeningeal amyloidosis (OLMA) which may present with additional features such as subarachnoid haemorrhage, epilepsy, hearing and visual loss, and headaches [59]. OLMA was described in a Nigerian adult heterozygous for TTR L21P, a variant which was previously reported in several European-ancestry cases [59]. Another common variant, at least among African-Americans (and found amongst West Africans), is the TTR V122I variant which was detected in the heterozygous state in 4% of African-Americans [60] and is associated with hypertrophic restrictive cardiomyopathy in older individuals, but without neuropathy. A man from Benin was reported with cognitive changes, a sensori-motor neuropathy with autonomic involvement and sensory ataxia, as well as hypertrophic cardiomyopathy, and a TTR I107V variant, which has been found in several Europeans with inherited amyloidosis [61].

Spinal muscular atrophies

Classical Spinal Muscular Atrophy (SMA) due to the homozygous loss of exon 7 (± exon 8) of SMN1 results in a critical loss of protein production and progressive degeneration of the lower motor neurons of the spinal cord [62]. We will refer to this as SMN1-SMA. Clinically, SMN1-SMA is characterized by proximal muscle atrophy and weakness, and eventually distal paresis as well. The clinical subtypes of SMN1-SMA (types I–IV) were categorized based on the disease severity and age at onset, which also informed the prognosis and survival; Type I is most severe and manifests in early infancy, SMA II manifests in late infancy to early childhood (< 18 months), SMA III in childhood (> 18 months)[62] and SMA IV has adult-onset [63].

There is increasing recognition of SMN1-negative SMA, although this groups accounts for < 5% of SMA and is often associated with overlapping central nervous system/brainstem signs, and even cardiomyopathy [63]. However, in reports from Africa there are between 25 and 65% of the clinical cohorts categorised as either congenital hypotonia or SMA phenotypes, which can be categorized as SMN1-negative SMA (absence of homozygous exon 7 deletion). In addition, there are several types of distal SMA (DSMA) which overlap with classifications of distal HMN/dHMN [63] (see Table 2).

The SMN2 gene is a highly homologous centromeric copy of SMN1 in which a C > T variant in exon 7 splicing enhancer distinguishes SMN2 from SMN1 [64]. Although genetic variation in SMN2 does not cause disease, SMN2 copy numbers may modify disease severity and age at onset [65].

SMN1-SMA in North Africa

SMN1-SMA in North African populations have been reported in families with and without high consanguinity rates [66,67,68,69,70,71,72,73,74,75] (Additional file 2: Table C). Similar to European cohorts, 57/60 (95%) Tunisian cases with presumed SMN1-SMA showed homozygous deletion of SMN1 exon 7 [70], although the other samples showed lower proportions of SMN1-SMA particularly in older individuals [69].

SMN1-SMA in sub-Saharan Africa

Five reports on SMA in sub-Saharan Africans were found, mostly involving South Africans and one each from Congo and Mali (Additional file 2: Table C) [73,74,75]. Several cases from two regions in South Africa reported SMN1-SMA with homozygous loss of exon 7 (± exons 8) ranging between 35 and 100% of their clinical samples, indicating a substantial number of cases with an alternative molecular diagnosis [76,77,78]. An SMN1 gene dosage assay in 300 random black SA samples showed the heterozygote exon 7 deletion in 6 individuals (1/50 population controls; 2%) which was similar to the frequency of SMN1 copy numbers in Kenyans and Nigerians [74], but roughly half of the heterozygote frequency found in European ancestry controls (3–4%)[77]. In comparison, the heterozygote frequency amongst 628 Malians was found to be 0.5% [74].

Humans have variable copies of an SMN2 gene, between 0 and 8 copies, and transcripts of this gene can modify the expression of SMN1-SMA [63]. Interestingly, the architecture of the SMN region differs substantially between Europeans and Africans, although African-Americans roughly followed the same trends in terms of SMN2 copy numbers as Europeans and Asians [79]. Amongst 75 black South African SMN1-SMA patients, 11% had > 2 SMN2 copies compared with 37% (of 30) SMN1-SMA patients with European ancestry [78]. Taken together, these results underscore the fact that the genetic architecture and disease pathogenic mechanisms in African ancestry individuals may vary from Europeans, and requires further study.

Complex inherited conditions with neuromuscular features

Although there are numerous complex multi-system conditions in which the presence of neuropathy may be present but not prominent [34], we mention two reports in Africans in which the recognition and initiation of appropriate treatment underscores their importance. Two families/probands with Allgrove or Triple A syndrome was described from North Africa/Algeria with the homozygous pathogenic variant in the AAAS gene (IVS14 + 1G > A); 1 family was consanguineous [80] and in the other both parents were heterozygous for the variant [81]. The main features were ACTH-resistant adrenal deficiency, achalasia and dry eyes, as well as features of distal motor neuropathy with/without spasticity, with the clinical onset during childhood. The importance is to recognise the treatable metabolic disturbances. A case of acute intermittent porphyria in a black South African man due to the HMBS R149* variant, was reported to mimic severe subacute motor neuropathy [82].

Conclusion

Although the high rate of consanguinity and occurrence of large families from North Africa have resulted in several molecularly confirmed cases of HSP and CMT, the genetic studies related to identifying the pathogenic variants in these conditions in sub-Saharan Africans, are sparse (Fig. 2). The high proportion of SMN1-negative SMA cases in particularly sub-Saharan Africa, identifies another group of patients with an as yet molecularly undiagnosed condition. Although the low rates of genetic reports in these complex disorders are likely due to the lack of resources and limited access to genetic screening, the clinical and genetic characteristics of these disorders need to be described and identified so that the burden of genetic variants and disorders are curated as the first steps to address accessibility to potential therapeutic trials. Collaborations among African researchers are slowly gaining momentum and will strengthen future funding applications to extend specialist clinical training of clinicians and genetic councellors, as well as increasing the number of cases and genomics capabilities in Africa. Increasing neurogenomics capacity and the development of appropriate genetic screening panels for Africans with inherited neuromuscular diseases, would help improve diagnostic capabilities.