Introduction

The Notch 2 N-terminal like C gene (NOTCH2NLC), located at chromosome 1q21, differs from the other two human NOTCH2 paralogs (NOTCH2NLA and NOTCH2NLB) for the presence of a repeat sequence (GGC)9(GGA)2(GGC)2 in the 5′ untranslated region (UTR), and for its enhanced expression in brain, especially in the prefrontal cortex1,2. Starting from 2019, NOTCH2NLC GGC repeat expansions in the 5′-UTR were found in patients affected by neuronal intranuclear inclusion disease (NIID), a neurodegenerative disorder characterized by eosinophilic, p62 and ubiquitin-positive intranuclear inclusions diffuse to different tissues, including the central and peripheral nervous systems3,4,5,6,7,8. NIID is a heterogeneous disorder characterized by a variety of neurological signs and symptoms, including cognitive impairment, parkinsonism, tremor, cerebellar ataxia, epilepsy, peripheral neuropathy, and autonomic dysfunction2,5. NIID is traditionally classified in three main types based on the predominant neurological features, namely muscle weakness-dominant, parkinsonism-dominant, and dementia-dominant5. NOTCH2NLC GGC repeat expansions have been reported in all these three forms, with a higher repeat size in the muscle weakness-dominant type5. An almost pathognomonic magnetic resonance imaging (MRI) marker of NIID is represented by a curvilinear hyperintensity at the corticomedullary junction at diffusion weighted imaging (DWI) sequences. However, its sensitivity is limited2.

By employing long-read sequencing (LRS), repeat-primed polymerase chain reaction (RP-PCR) and GC-rich PCR, the screening of NOTCH2NLC GGC repeat expansions has been rapidly extended to a variety of neurological disorders, including oculopharyngodistal myopathy (OPDM)9,10, Parkinson’s disease (PD)11,12,13,14,15,16, essential tremor (ET)14,17,18,19,20,21,22, multiple system atrophy (MSA)14,23,24, spinocerebellar ataxia (SCA)5,14, dementia [i.e., Alzheimer disease (AD), frontotemporal dementia (FTD), dementia with Lewy bodies (DLB), vascular dementia (VaD)]5,25,26, hereditary spastic paraplegia (HSP)27, peripheral neuropathy5,28,29,30, adult leukoencephalopathy31,32,33,34, and specifically cerebral small vessel disease35. However, the results of these studies have been spurious, so that the pathogenic role of NOTCH2NLC in neurological disorders beyond NIID is still debated.

In the last years, the discovery of a hexanucleotide repeat expansion in chromosome 9 open reading frame 2 (C9orf72) as the main genetic cause of Amyotrophic Lateral Sclerosis (ALS) and the association between intermediate repeats in ataxin 2 (ATXN2) with this disorder have suggested that repetitive sequences in human genome play a major role in ALS pathophysiology36,37,38. In this scenario, in 2020 Yuan and colleagues estimated the number of GGC repeats in the 5′-untranslated region (UTR) of NOTCH2NLC in 545 ALS patients from mainland China39. The authors found 4 ALS subjects carrying expanded alleles: two of them in the range of intermediate repeat numbers (44 and 54 repeats) and the others with pathogenic expansions of 96 and 143 GGC repeats. None of the age-matched 1305 controls displayed expanded alleles. Based on these data, the authors suggested that GGC repeat expansions in NOTCH2NLC might be also associated with ALS3,4,5,17,31. However, different authors failed to detect similar expanded alleles in NOTCH2NLC in other Chinese and Taiwanese ALS cohorts5,40.

In this scenario, we challenged the hypothesis that NOTCH2NLC GGC repeat expansions might be associated with ALS by evaluating their number in a cohort of Italian ALS patients.

Results

We enrolled 385 ALS patients, including 357 (92.7%) sporadic cases. Mean age at onset was 60.5 years (SD 13.7), and 60.9% were males. Most patients had a spinal onset (71.8%) rather than a bulbar one. The hexanucleotide repeat expansion in C9orf72 and mutations in common ALS disease-causing genes [superoxide dismutase 1 (SOD1), TAR DNA-binding protein (TARDBP) and fused in sarcoma (FUS)] were excluded. All patients were screened, and none of them showed the typical sawtooth tail pattern on RP-PCR, thus excluding the presence of abnormal expansions in the 5′-UTR of NOTCH2NLC. The estimated repeat sizes ranged from 11 to 35 in our cohort, as shown in Fig. 1. In Supplementary Fig. 1 we provide the pattern obtained by RP-PCR analysis in a representative negative control from our cohort compared to that of a known carrier of a NOTCH2NLC GGC repeat expansion (Supplementary Fig. 1)14. The number of patients enrolled, the ethnic background, the methods employed and the estimated NOTCH2NLC GGC repeat size were compared to previous studies performed both in ALS and in other neurological conditions, specifically HSP, NIID, OPDM, PD, ET, MSA, SCA, AD, FTD, DLB, VaD, peripheral neuropathy, adult leukoencephalopathy and specifically cerebral small vessel disease (Table 1).

Figure 1
figure 1

Distribution of the GGC repeat length of NOTCH2NLC in the 385 Italian patients with amyotrophic lateral sclerosis.

Full size image
Table 1 Studies describing NOTCH2NLC GGC repeat expansion screening in neurological disorders.
Full size table

Discussion

The four patients described by Yuan et al., who harbored GGC repeat expansions in NOTCH2NLC, showed similar clinical findings, including limb muscle weakness and atrophy, widespread fasciculations, dysarthria, dysphagia, dyspnea, and upper motor neuron signs39. The size of the abnormal repeat expansion was in the range of intermediate repeat numbers (between 43 and 59) in two cases, and in that of pathogenic expansions in the other two (96 and 143). Noteworthy, the two carriers of intermediate GGC repeat expansions died before reaching a definite diagnosis of ALS. High inter-individual clinical variability within families, nerve conduction abnormalities and intranuclear ubiquitin and p62-positive inclusions were identified in carriers of GGC repeat expansions in NOTCH2NLC and NIID-M patients. However, some significant differences argued against the presence of a unique clinical entity, including the significantly more severe phenotype and rapid deterioration of the four patients described by Yuan and colleagues, and the evidence of spontaneous activity on needle examination in more regions compared to NIID-M. Considering the different ranges of GGC repeat number detected in ALS (44–143) and NIID-M (118–517) patients, the authors suggested that the length of expansion might be related to the development of specific phenotypes, but further analysis are warranted to confirm this hypothesis. Alternatively, ALS with GGC repeat expansion in NOTCH2NLC might be a subtype of NIID previously undescribed.

To date, mutations in more than 30 genes have been associated with ALS. In Caucasian ALS patients the most recurring genetic defects are observed in C9orf72 (familial ALS (fALS) 33.7%, sporadic ALS (sALS) 5.1%), followed by SOD1 (fALS 14.8%, sALS 1.2%), TARDBP (fALS 4.2%, sALS 0.8%) and FUS (fALS 2.8%, sALS 0.3%)41. The proportion of mutated individuals might differ remarkably in different ethnic backgrounds, as previously observed for the mutation spectrum of ALS genes in the Chinese population42. However, a recent study failed to detect abnormal GGC repeats in NOTCH2NLC in a cohort of 304 unrelated Taiwan ALS patients, whereas an intermediate GGC repeat allele (46 repeats) was detected in 1 out of 637 control subjects40. Similarly, Tian and colleagues did not find expanded NOTCH2NLC alleles among 44 Chinese families affected by motor neuron disease, nor in 211 matched healthy controls5. While Tian et al. did not report the range of NOTCH2NLC repeats of their cohort, the one found by Jih and colleagues is in line the estimated repeat size of our ALS population (7–36 vs. 11–35)5,40.

Only six other works have performed NOTCH2NLC screening in patients of European descent, and specifically 2 in PD, 2 in ET, 1 in combined movement disorders, and 1 in adult leukoencephalopathy11,14,15,19,33. Out of a total of 38,820 European patients, only 2 were found to carry pathogenic NOTCH2NLC GGC repeat expansions (frequency 6.0 × 10–5). The first, who carried 118 NOTCH2NLC GGC repeats, was a Ukrainian woman affected by recurrent encephalopathy, whose skin biopsy revealed p62 and ubiquitin-positive inclusions in fibroblasts, endothelial cells, and serous glands. The second, instead, was an Italian man with postural tremor and a positive family history for tremor-dominant PD. The estimated number of GGC repeats in NOTCH2NLC was 90. Intriguingly, the employment of whole genome sequencing (WGS) revealed no significant differences in the repeat structure of the 5′-UTR of NOTCH2NLC, nor in its allelic frequency between individuals of European and East Asian descent14. This finding supports the hypothesis of a founder effect to explain the different distribution of NIID worldwide. Additionally, one PD patient was confirmed to carry an intermediate NOTCH2NLC GGC repeat expansions by LRS (estimated repeat size 44–48)15. Indeed, the recent development of cutting-edge techniques such as LRS has revolutionized our ability to detect long repetitive elements, copy number and structural variations, which cannot be revealed by conventional, short-read sequencing technologies. In this scenario, however, RP-PCR is still a valuable tool to confirm novel altered expansions of repeat units, or to perform screening in large cohorts.

This is the first assessment of the prevalence of GGC abnormal repeats in NOTCH2NLC in a European cohort of ALS patients. Although our study could be improved by the analysis of a control group and by the use of additional tests for a better detection of GGC repeat sizes or repeat interruptions, we think that these improvements are not expected to impact on the result.

Overall, we suggest that NOTCH2NLC expanded alleles might be absent or at least extremely rare in ALS Italian patients. Further investigations in larger cohorts with different ethnic backgrounds are required to support the involvement of NOTCH2NLC in ALS.

Methods

The patients in our cohort meet the revised El Escorial criteria for probable or definite ALS43. Patients carrying a mutation in common ALS disease-causing genes (C9orf72, SOD1, TARDBP and FUS) were excluded. A screening analysis of GGC repeats in NOTCH2NLC was performed at the Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico by RP-PCR, as previously described4. Specifically, a slow-down PCR protocol was employed. After denaturation at 95 °C for 5 min, the cycling conditions were followed by: 50 cycles of 95 °C for 30 s, 98 °C for 10 s, 62 °C for 30 s and 72 °C for 2 min. We set the ramp rate to 95 °C and 72 °C to 2.5 °C s−1, and the one to 62 °C to 1.5 °C s−1. Electrophoresis was performed on a 3130 Genetic analyzer (Thermo Fisher Scientific, Waltham, MA) and the data were analyzed using GeneMapper software (Thermo Fisher Scientific). We performed RP-PCR also on a known positive control (patient B)14, as quality control assessment (Supplementary Fig. 1). All the probands provided written informed consent. The “Comitato Etico Milano Area 2 Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico” (Milan, Italy) approved the study. The study is in accordance with relevant guidelines and regulations.