The Genetic Architecture of Parkinson’s Disease in Morocco: Highlighting a Predominance of Mendelian Genes Open Access

Parkinson’s disease (PD) is a progressive neurodegenerative disorder characterized by degeneration of dopaminergic neurons in the substantia nigra, a key brain region involved in movement control. Classically defined by the symptomatic triad of resting tremor, muscle rigidity, and bradykinesia, PD is also accompanied by non-motor symptoms, including cognitive, psychiatric, and autonomic impairments [1]. These non-motor symptoms can appear well before the onset of motor disorders and contribute greatly to the deterioration of patients’ quality of life [2, 3]. Although pharmacological treatments exist, mainly based on dopaminergic substitution (such as levodopa), they remain symptomatic and not curative. In addition, their efficacy decreases over time, and patients often develop motor complications, such as levodopa-induced dyskinesias [4]. There is currently no treatment capable of stopping or slowing the progression of neuronal degeneration, making PD a major global public health problem.

The prevalence of PD increases significantly with age, estimated at 1% of the global population in people over 60 years old and 3% in people over 80 years old [5]. It is a multifactorial disease resulting from the complex interaction between environmental and genetic factors contributing to a large variability between individuals and populations in clinical manifestations. Environmental factors can include lifestyle choices and environmental pollution, whereas genetic factors can be monogenic and multigenic [6, 7]. At least twenty genes have been associated with the disease, in which rare but penetrant variants with high effect sizes are typically associated with monogenic causes for PD identified in a minority of familial cases. In contrast, the hypothesis of the multigenic cause is supported by the identification of numerous common variants with small effect sizes that, in combination, contribute to disease susceptibility, as proven by several genome-wide association studies (GWAS). However, the heritability of PD estimated by these studies only manages to explain a fraction of cases, suggesting missing heritability [8]. Furthermore, the genetic architecture of PD can vary significantly between populations due to differences in allele frequencies [9]. However, most of the PD risk variants discovered so far have been found in individuals of European ancestry, particularly in Europe and North America, and other populations, particularly those from Africa, were less well studied.

This review aimed to summarize published research data on the genetic profile of PD patients from the Moroccan population in order to describe its genetic architecture. All of these articles are from a research project conducted by the Neurology Department of the Rabat Specialized Hospital on a cohort of 211 patients, including 59 familial forms, recruited from 2016 to 2020, all of Moroccan origin. In this project, the same genetic analysis strategy was adopted, which consisted of first screening by Sanger sequencing all patients for the LRRK2 p.G2019S variant common in North Africa. The MLPA or chromosomal microarray analysis was used to detect exon rearrangements in PD genes for all negative p.G2019S patients. Then, a gene panel NGS containing 20 genes associated with PD and overlapping phenotypes was performed for the remaining patients.

The monogenic form of PD is caused by a mutation in a single gene and generally follows clear Mendelian inheritance patterns, either autosomal dominant or autosomal recessive. Overall, monogenic forms of PD account for less than 10%, including 30% for familial and 3–5% of sporadic cases [10]. Twelve established genes with pathogenic variants conferring a high risk for PD (SNCA, LRRK2, VPS35, CHCHD2 with autosomal dominant, and PRKN, PINK1, DJ-1, VPS13C, ATP13A2, PLA2G6, FBOX7, DNAJC6, SYNJ1 with autosomal recessive modes of inheritance) are responsible for either classical and more complex phenotypes [8, 11].

The LRRK2 c.6055G>A (p.Gly2019Ser) variant is the major genetic cause of PD identified to date, occurring in both familial and sporadic patients [12]. It is transmitted in an autosomal dominant mode of inheritance and is localized in the kinase domain of LRRK2, inducing a gain in the kinase activity of the protein. This variant is associated with a high risk of developing PD but with incomplete penetrance that varies according to age and ethnicity, probably due to various genetic modifiers that have not yet been elucidated. Worldwide, the prevalence of this variant is estimated at 1% in sporadic and 4% in familial PD [13] and is highly variable depending on genetic ancestry, with the highest frequencies observed in North African countries [14].

In Morocco, located in the northwest of the African continent, in a consecutive monocentric series of 100 PD patients, we showed that the overall frequency of the G2019S mutation was 41%, with 76% among dominantly transmitted familial forms and 28% among sporadic cases. The mutation was also found in 5% out of 100 healthy control individuals tested [15]. The PD series was subsequently expanded to 211 patients, and the overall prevalence of the mutation remains high at 38% (Fig. 1), with 63% among dominant familial forms and 32.9% among sporadic cases. Apart from the G2019S mutation, no other pathological mutations were found in the LRRK2 gene or in the VPS35 and SNCA genes responsible for autosomal dominant PD.

The prevalence rates of the G2019S variant in Morocco are the highest in the world reported to date, decreasing toward the Middle East and the North of Europe, while it is absent in sub-Saharan Africa and rare in the remaining continents. The worldwide distribution of the G2019S mutation follows a geographical gradient with northwest Africa as the center of dispersal [15, 16]. This variant has been reported to be also prevalent among Ashkenazi Jews, reaching up to 18% [17‒19], suggesting the possibility of a shared common ancestor and founder effect with the North African population. Indeed, the vast majority of G2019S gene carriers share the same haplotype 1, which is thought to have originated from a Near Eastern founder at least 4,000 years ago. [14]. By studying this haplotype in a large sample of Moroccan carriers and their uniparental markers of Y chromosome and mitochondrial DNA, we showed that the LRRK2 G2019S variant originated in an autochthonous founder who lived at least 5,000 years ago and who belonged to the Berber ethnicity [20].

Clinically, the age of onset of the disease of G2019S carriers (N = 41) in our Moroccan cohort showed a strong variability ranging from 28 to 78 years of age, with a mean of 52.15 (±11.28) years. In comparison to non-carriers (N = 59) with idiopathic PD and who have a similar age at disease onset, G2019S carriers exhibit a less tremor-dominant phenotype with more dystonia and a slow disease progression. Regarding non-motor symptoms, sleep symptoms and cognitive impairment tend to be more common in G2019S carriers [15]. Our results align with a recent study on a larger multi-ethnic series, which showed that G2019S carriers had a similar age of onset, slow disease progression, and a similar prevalence of motor symptoms as non-carriers [21]. However, this study reports a lower prevalence of non-motor features, particularly sleep disorder and cognitive impairment, in contrast to previous studies that showed more sleep disturbances with a higher prevalence of insomnia and less REM sleep behavior disorders, excessive daytime sleepiness [22, 23], or nocturia [22, 24] in G2019S carriers.

In recessive genes, 17 different genotypes, either homozygous or compound heterozygous, were found in 21 patients representing 10% of the cohort (Table 1), and which were reported in [25‒28]. Twelve of the 21 cases were familial forms, and nine were sporadic. Pathogenic variants of the PRKN gene were most frequently found in 8 cases with mainly exon deletions involving exons 3–9, a region that appears to be a hotspot of deletion events. The second prevalent gene found was PINK1 with 3 familial and 3 sporadic cases, of which 4 were carriers of the p.Gln456X mutation specific to the North African population [27, 29], supporting a founder origin. Patients carrying pathogenic variants in PRKN and PINK1 developed the disease at ages between 30 and 55 years and presented with a typical clinical phenotype. Two familial and sporadic patients with two novel mutations in DJ-1 (Tyr141Ter and Thr110Pro) started the disease at a very young age of 20 and 27 years and developed relatively more severe parkinsonism. Two other patients, with ages at onset of 12 and 13, carry the same pathogenic mutation p.Trp258X in ATP13A2, also specific to the North African population, and presented with even more severe parkinsonism, which started in adolescence [27]. Finally, patients with genotypes 6 and 15–17 (Table 1), all of them familial forms and carrying a pathogenic allele and a rare likely pathogenic allele, developed a moderate phenotype that started at a very late age between 60 and 80 years. It should be noted that the genes involved, including SYNJ1, PRKN, and VPS13C, are generally associated with early forms of the disease and rapid progression [11].

In addition to patients carrying high penetrant variants responsible for the monogenic form of PD, 59 patients representing 28% of the Moroccan cohort were carriers of rare variants under 1%, classified as likely pathogenic (LP) or variant of unknown significance (VUS), and which were reported in [28, 30]. Twenty-seven of them, representing 12.8% of the Moroccan cohort, were carriers of variants in a single gene, of which 8 were familial and 19 were sporadic cases (Table 2). Eight rare LP/VUS variants were identified in 11 patients in LRRK2, either monoallelic or biallelic, more frequently in sporadic cases. Three of these variants were novel, including the p.Lys739Arg, p.Gln1379Arg, and p.Glu1492Lys variants. Finally, 12 patients carried LP/VUS variants in MAPT and POLG, the most frequent of which was the p.Arg222Ser variant in MAPT found in 2 familial and 4 sporadic cases. Variants in VPS35, GBA, and UCHL1 were found in rare patients. This group of patients presented clinically with a late disease onset with a tremor-dominant or mixed phenotype and a classic PD phenomenology in terms of non-motor symptoms and motor complications [28, 30] as reported in other populations [31, 32].

The co-occurrence of rare high-risk variants involving more than one gene was observed in 32 PD patients (15.2%), all of them were sporadic cases (Fig. 2). Eight of them carried 2 variants in Mendelian genes, exclusively in genes with autosomal recessive inheritance, and the remaining 24 patients carried 2–5 rare variants in risk factor and Mendelian genes that may contribute collectively to the risk of PD. In this latter group of patients, LP/VUS variants were mainly found in MAPT and POLG, and additionally, one or more variants in Mendelian genes, particularly in LRRK2 and VPS13C. Therefore, the co-occurrence in patients of rare variants in these four genes appears to be most common in the Moroccan population, affecting PD risk. These genes, in addition to GBA, have been reported to be among the most decisive genetic risk factors for sporadic PD [33‒35]. However, the frequency of high-risk variants in GBA was very low in the Moroccan population and was found only in 3 PD patients among the 211 studied.

Clinically, in this group of sporadic PD patients, data showed globally a late disease onset (above 50), with resting tremor as a revealing symptom evolving into a mixed phenotype with motor fluctuations and variable non-motor symptoms, of which cognitive impairment was highly represented (41.7%) [28]. This could be related to the involvement of MAPT and VPS13C genes, both reported as genetic risk factors for cognitive impairment [32, 36, 37].

Finally, fifty-two patients (24%) do not carry any rare variants under 1% in the 20 PD genes tested, suggesting that there may be novel high-risk genes that have not yet been identified and, thus, further genetic heterogeneity for PD in the Moroccan population. Furthermore, according to the common variant hypothesis in multifactorial diseases, the disease in these patients could be due to numerous common variants that have not been taken into account when studying the Moroccan cohort.

Our report provides a first overview of the genetic architecture of PD in Morocco (Fig. 3). In this cohort, familial forms represent 28%, which is similar in neighboring North African or Middle Eastern populations but high compared to populations of European ancestry, which account only for 5–15% of patients [5, 38].

In the Moroccan population, the LRRK2 p.G2019S mutation was the most prevalent associated variant causing monogenic PD in both familial and sporadic cases, accounting for 38%, and no other pathological variant was identified in known dominantly inherited genes, including LRRK2, SNCA, and VPS35. Biallelic pathogenic mutations were identified in five of the nine genes known to cause autosomal recessive PD, including PRKN, PINK1, DJ-1, ATP13A2, SYNJ1, and VPS13C, representing 10% of the cohort, where PRKN and PINK1 were the most frequently causative genes. Overall, a monogenic cause of PD in the Moroccan population involves 7 genes, accounting for almost 50%. This high rate of Mendelian heritability of PD in the Moroccan population is due to the high prevalence of the North African founder variant and the high rate of consanguinity. Additionally, rare high-risk variants for PD were found in 28% of the Moroccan cohort, involving four genes, including LRRK2, VPS13C, MAPT, and POLG, alone or in co-occurrence. Finally, 24% of patients did not carry such variants in the twenty PD genes analyzed.

The genetic architecture of PD in patients of Moroccan ancestry, and generally in North African ancestry, is substantially different compared to patients of European and non-European ancestry, in whom approximately 5–10% are diagnosed with monogenic forms [6]. In these populations, the p.G2019S is the most common variant, but it remains low and represents only 4% of familial cases and 1% of sporadic cases, while it is scarce in populations of non-European ancestry, including Asians, Latin-Americans, Indians, Asians, and sub-Saharan Africans, in whom additional LRRK2 pathogenic variants have been reported [13, 39]. Besides that, the majority of PD heritability comes from multiple low-risk alleles in several genes, as demonstrated by the largest GWAS that identified 90 common independent variants. A large number of them are in close proximity to the Mendelian genes, particularly SNCA, LRRK2, GBA, and VPS13C [8, 9, 40].

The present genetic architecture of PD has been deduced from a monocentric study. However, it should be noted that Rabat Hospital is the oldest in Morocco and receives patients from all regions of the country except the southern Saharan region, which is underrepresented. From an epidemiological perspective, a weak selection bias is certainly present, but it keeps the results close to reality. Studies with larger samples covering all regions of the country are needed to more accurately estimate the frequency of monogenic forms of PD in Moroccans known to be a heterogeneous population that has received influences from different ethnicities over time but remains predominantly indigenous Berber.

One of the immediate consequences of the knowledge of this genetic architecture of PD is the establishment of a diagnostic strategy of genetic testing specific to the Moroccan population. This must first go through the search for the LRRK2 G2019S variant, which will be able to provide a positive genetic diagnosis in almost 30% of isolated cases and 60% of familial cases. As a second consequence, this allows us to subdivide the clinical variability of the disease into groups of common genetic and biological pathway causes, essential for a better therapeutic management strategy. This could also include deep brain stimulation therapy, which is effective in some patients with monogenic PD, such as those carrying LRRK2 p.G2019S or PRKN mutations [41]. Furthermore, the high frequency of the p.G2019S mutation makes Morocco, and thus the other countries of North Africa, a suitable region for testing the molecules of the ongoing clinical trials, including antisense oligonucleotides and LRRK2 kinase inhibitor molecules leading to the reduction of LRRK2 kinase activity [42, 43].

The methodological approach used for the genetic characterization of Moroccan patients with PD, limited to the combination of MLPA, Sanger, and gene panel sequencing techniques, allowed the discovery of the majority of genetic causes responsible for PD in this North African population. However, the use of other methods such as GWAS, whole genome, and long-read sequencing could increase this heritability by the discovery of new genes, new risk variants, or new types of mutations not detectable by the methodology used. The GP2 consortium using these methods will help complete the PD architecture, not only in North African countries but also worldwide.

This review brings together all the results of genetic studies on PD in the Moroccan population and contributes to the understanding of its genetic architecture. It shows that LRRK2 p.G2019S as well as recessive genes play an important role as a monogenic cause and that rare risk variants in LRRK2, VPS13C, MAPT, and POLG may contribute to increasing the genetic risk of the disease in this North African population.

The findings published in scientific journals used in this review article were carried out as part of the “Project on the genetic bases of Parkinson’s disease” supported by the “Ministère de l’Enseignement Supérieur, de la Recherche Scientifique et de la Formation des Cadres” (MESRSFC) of Morocco and the Mohammed V University in Rabat, Morocco.

The authors declare that they have no competing interests.

The authors did not receive external funding for this manuscript.

A.B., H.T., and W.R. drafted and critically revised the manuscript. A.B. designed the project. All authors read and approved the final manuscript.