The term semantic variant of frontotemporal dementia (svFTD) usually refers to the form of FTD characterized by an early impairment of verbal semantic knowledge, that manifests as anomia and impaired single word comprehension, linked to the predominant atrophy of the left anterior temporal lobe and the diagnosis is based on current clinical criteria (Gorno-tempini et al., 2011). The right temporal variant of FTD (rtvFTD), a rarer clinical phenotype, originally considered a right variant of svFTD, has distinct clinical features and predominantly right hemisphere involvement. Patients with rtvFTD exhibit early behavioral symptoms, particularly empathy deficits and psychiatric disorders (Ulugut Erkoyun et al., 2020), but the diagnosis is challenging, and patients are often misdiagnosed as behavioral variant of FTD (bvFTD; Younes et al., 2022). A motor neuron involvement is rare in svFTD, but it has been found in up to 28% of rtvFTD cases in a recent neuropathological cohort (Ulugut et al., 2021).

While pathology can be highly heterogeneous in bvFTD, most of patients affected by svFTD and rtvFTD has TDP-43 neuronal accumulation, particularly the Type C pathology (Rohrer et al., 2010). In a minority of patients with rtvFTD a recent study reported also tauopathy as pathological substrate of the disease (Ulugut et al., 2021).

FTD has a prominent genetic basis, with up to 40% of cases presenting a family history (Mann and Snowden, 2017). To date three major genes have been described as linked to FTD: microtubule-associated protein tau (MAPT), progranulin (GRN) and Chromosome 9 open reading frame 72 (C9ORF72; Mann and Snowden, 2017). Furthermore, other causative genes have been disclosed, underlining rarer familial cases: Sequestosome 1 (SQSTM1), Valosin-Containing Protein (VCP), Charged Multivesicular Body Protein 2B (CHMP2B), TAR DNA Binding Protein, 43-KD (TARDBP), Fused in Sarcoma (FUS), Triggering Receptor Expressed on Myeloid cells 2 (TREM2), Tank-Binding Kinase 1 (TBK1) (see Fenoglio et al., 2018 for review). Left and right variants of FTD are usually sporadic, but recent evidence suggests a higher frequency of genetic mutations for the right versus the left variant (Ulugut Erkoyun et al., 2021). However, the genetic basis of these forms is not clear and genotype–phenotype correlations need to be elucidated, to allow accurate genetic counseling.

In this study, we used a next generation sequencing (NGS) approach to perform a genetic screening of a single-center cohort of patients with svFTD and rtvFTD, aiming at evaluating the genetic contribution in those disorders, by confirming candidate or discovering new genetic variants that could be associated with these clinical phenotypes.

Demographic and clinical data of overall cohort are reported in Table 1. Patients and HC did not show significant difference in term of sex (p = 0.11) and age (p = 0.18). 39% (9/23) of patients had a family history of dementia (53% of svFTD).

Semantic impairment with naming deficits was the first symptom of svFTD patients, while rtvFTD patients showed early changes in their personality and behaviour (among them, eating behavior changes, loss of empathy, compulsive behaviors), memory loss, prosopagnosia and topographical disorientation. According with clinical diagnosis, brain magnetic resonance imaging and fluorodeoxyglucose-positron emission tomography showed predominant atrophy and hypometabolism in left anterior temporal lobe in svFTD and a predominant right temporal involvement in all rtvFTD. CSF biomarkers were in the normal range in 13/16 patients, 3 patients had isolated reduction of Abeta42. Total tau was elevated in 2 patients and one of them had also elevated p-tau, but normal Abeta42.

In this study we found several pathogenic or VUS/likely pathogenic variants, in nearly 40% of patients.

P301L, carried by P05, is one of the most known and studied MAPT mutations, whose pathogenicity has been demonstrated in several genetic, functional and neuropathological studies, as well as in animal models (Hutton et al., 1998; Barghorn et al., 2000; Lewis et al., 2000; Rossi et al., 2008). Different FTD clinical phenotypes have been associated with the mutation, including svFTD (Ishizuka et al., 2011). Q336H MAPT mutation, carried by patient P12, was recently published as associated with svFTD (Villa et al., 2022). This mutation had been previously functionally and neuropathologically characterized as pathogenic (Tacik et al., 2015). Patient P01 carried the N286N MAPT and M93V PSEN1 variants. As for the N286N MAPT variant, only once cited without additional details (Rohrer et al., 2009), the classification as likely benign does not take into account the pathogenetic mechanism typical of some silent MAPT mutations localized in exon 10, that is the alteration of an exon 10 splicing enhancer or silencer (D’Souza and Schellenberg, 2002; Qian and Liu, 2014). In fact, an in silico functional analysis demonstrated the effect on exon 10 splicing of this variant (Tubeuf et al., 2020). The M93V variant in PSEN1, so far never described, is very interesting as localized in a mutational hot spot in a critical region for the protein function, the transmembrane domain 1: in fact, the adjacent mutations C92S (Zhang et al., 2000; Tedde et al., 2003) and V94M (Arango et al., 2001; Somavarapu and Kepp, 2016) are associated with Alzheimer’s disease (AD). In addition, several other pathogenic AD mutations have been described in the same protein region. It is well known that PSEN1 mutations give rise to Alzheimer’s disease by affecting the beta amyloid levels; however, a few mutations have been reported as associated with a FTD phenotype, even pathologically confirmed as a tauopathy (Dermaut et al., 2004), although how loss of functional presenilin 1 could predispose to tauopathy has not been clarified. Our patient had a clinical phenotype of rtvFTD with early memory deficits and behavioural disturbances including lack of empathy, hyperphagia and loss of inhibition, the clinical phenotype suggesting the prevalence of FTD neurodegeneration pathway, while the AD pathway may cooperate in the pathology. Moreover, amyloid-PET was negative in our patient, making AD diagnosis unlikely.

MAPT mutations have been previously identified in 4 rtvFTD cases (Ulugut Erkoyun et al., 2021). However, in our cohort, both svFTD (P05 and P12) and rtvFTD (P01) patients carried MAPT mutations, not confirming a prevalent association of MAPT mutations with rtvFTD.

Patient P06 had Q314L OPTN and R795H DCTN1 variants. OPTN, earlier reported as causative of primary open-angle glaucoma, was afterwards linked to ALS (Maruyama et al., 2010), or to FTD (Dominguez et al., 2021), with dominant or recessive transmission. Heterozygous Q314L in OPTN was described in some sporadic ALS cases (Del Bo et al., 2011; Pensato et al., 2020). DCTN1, firstly linked to ALS and Perry syndrome, was then associated with progressive supranuclear palsy and FTD phenotypes (Caroppo et al., 2014). R795H variant was only described in a case of Charcot–Marie-Tooth. These variants are classified as VUS-near likely pathogenic and may contribute to the FTD pathology in our patient, although with very different mechanisms, as OPTN is involved in autophagy whereas DCTN1, binding to microtubules and to dynein, has a role in retrograde axonal transport (LaMonte et al., 2002). Patient P21 carried the G376E VCP and I207T TBK1 variants. Mutations in VCP were identified as the cause of inclusion body myopathy with Paget disease of bone and frontotemporal dementia (IBMPF); (Watts et al., 2004). Afterwards, VCP mutations were also found in cases of pure FTD (Saracino et al., 2018). The G376E variant has never been described and is not present in population databases, constituting a possible new mutation. In fact, it is adjacent to the ATPase domain 1, where other mutations have been described (Mol et al., 2021), and it is classified as VUS-near likely pathogenic, having at least 19 pathogenicity scores. I207T in TBK1 was described in a patient affected by sporadic ALS (Tohnai et al., 2018). The mutation is within the serine/threonine kinase domain and has 10 pathogenicity scores. TBK1 is now recognized also as a FTD gene (van Mossevelde et al., 2018). Since VCP has several functions, including the maintenance of lysosomal homeostasis (Arhzaouy et al., 2019), it may cooperate with TBK1, whose principal role is in the autophagy. The coexistence of variants in genes involved in the same pathway reinforces their pathogenic role and may explain the pathology. Both P06 and P21 patients were svFTD, without any sign of ALS.

Patients P20 and P23, both svFTD, carried a SQSTM1 mutation. SQSTM1 mutations were firstly identified as causative of Paget Disease of Bone (PDB; Laurin et al., 2002), and afterwards this gene was disclosed to be also causative of ALS (Fecto et al., 2011) and FTD (Le Ber et al., 2013). The P387L mutation, carried by patient P23, was previously described in patients affected by PDB (Johnson-Pais et al., 2003), and also in familiar cases of FTD, where a segregation with the disease was demonstrated (Le Ber et al., 2013). The amino acid deletion E280, carried by patient P20, was described as a susceptibility factor for FTD (Van Der Zee et al., 2014).

Patient P10, rtvFTD, had the P80L CHCHD10 mutation, previously reported in patients affected by familial or sporadic ALS (Ronchi et al., 2015; Zhang et al., 2015). CHCHD10 codes for a mitochondrial protein associated not only with ALS but also with FTD (Chaussenot et al., 2014).

While homozygous or heterozygous compound mutations in PRKN lead to early-onset Parkinson’s disease (PD; Lücking et al., 2000), the role of heterozygous mutations is still controversial, ranging from benign condition to susceptibility factor (Marder et al., 2010; Moura et al., 2013). PRKN T240M mutation, carried by the svFTD patient P11, is predicted to eliminate a phosphorylation site for casein kinase II and occurs in the same codon as other mutations (T240R and T240K), indicating that this is an important functional residue. Furthermore, functional studies in neuronal cell cultures demonstrated that this mutation impairs glutamatergic signaling (Zhu et al., 2018). Biallelic mutations in PKRN can lead to FTD (Zimmermann et al., 2018), whereas we suppose that heterozygous mutations may only constitute a susceptibility factor.

In summary, as for single variant findings, we found 12 mutations in 9 patients out of 23, evidencing a strong genetic component in FTD phenotypes usually regarded as sporadic. We also showed that the mutations appeared to be more associated with svFTD (7/15, 47%) than with rtvFTD (2/8, 25%) at variance with what suggested by others (Ulugut Erkoyun et al., 2021). Our data show that a gene strongly involved in our cohort of patients is MAPT, one of the three major genes causative of the FTD spectrum. Along with DCTN1, the involvement of this gene clearly demonstrates the relevance of the correct microtubule dynamics and transport in the nervous system. The mutations found in our patients outline cell pathways whose alterations are common to different neurodegenerative diseases such as FTD, ALS and Parkinson’s disease (PD). Lysosomal and autophagic functions are sustained by VCP, which plays a role in lysosome homeostasis, and by SQSTM1, TBK1 and OPTN, which interact in the autophagy (Abramzon et al., 2020; Fleming et al., 2022), all mutated in ALS and FTD. In addition, mitophagy, a specialized form of autophagy concerning mitocondria, is carried out by the interaction of TBK1, PRKN and OPTN (Harding et al., 2021). Mitochondrial homeostasis is the function of CHCHD10, linked to FTD, ALS and PD (Jiang et al., 2022).

We also performed a gene-level association analysis (burden analysis) in order to disclose genes which may contain a significantly higher number of variants in patients than in controls, suggesting the gene to be relevant to the pathology. We are aware of the small size of our cohorts, partly justified by the rarity of the disease, and in fact we did not achieved the Bonferroni significant threshold, but only suggestive p values highlighting 4 genes: CD33, PSEN1 and ABCA1, showing a simultaneous presence of causal and non-causal variants with opposing directions, and OPTN, showing unidirectional risk association with rare variants exclusively present or more frequent in patients than in HC.

CD33 is a transmembrane protein and a sialic acid-binding immunoglobulin-like lectin that regulates innate immunity. In the CNS, it is expressed by the microglia. CD33 has an immunoreceptor tyrosine-based inhibitory motif (ITIM) and one ITIM-like domain that facilitates an inhibitory signal. When CD33 is activated by sialic-acid-containing glycoproteins and glycolipids, it results in inhibition of microglia phagocytosis. Genome-wide association studies (GWAS) identified several genetic loci associated with increased susceptibility to late onset Alzheimer’s disease, including CD33 (Naj et al., 2011). Some polymorphisms increase CD33 expression, giving rise to reduced microglial phagocytosis and amyloid clearance. Although most polymorphisms were related to AD dementia, others were described as associated with a FTD phenotype (Rendina et al., 2020). In our cohorts we found 4 CD33 rare or low frequency variants. In particular, S305P was found in 9.1% of patients and 0.7% of control subjects, while F243L and V267I were only found in one patient each and no controls. G156fs was found approximately with the same frequency in patients and controls. All these variants are classified as benign. Besides the involvement in amyloid clearance, a pathological feature of AD, CD33 may be regarded as a gene more generally involved in immune and inflammatory pathways, which are also altered in FTD (Ferrari et al., 2014, 2015).

We have already discussed the contribution of a PSEN1 likely pathogenic variant, M93V, to FTD phenotype (see above); in addition, the gene-based analysis revealed the presence of the E318G variant, previously described as a risk factor for AD but afterwards producing conflicting results (Dermaut et al., 1999). The D333V variant, present in a control subject, was never reported before and it is classified as likely pathogenic for cardiovascular disease (Li et al., 2006). As for OPTN, we have already described the Q341L variant (see above), while the M98K is reported in databases as likely benign.

ABCA1 is an ATP-binding cassette transporter that controls whole brain cholesterol homeostasis. Polymorphisms in ABCA1 have been demonstrated to influence susceptibility to dementia, in particular to AD (Lupton et al., 2014; Chen et al., 2016). In our cohorts we found 9 ABCA1 rare or low frequency variants. Four of them were found in patients and the others in controls, showing opposite directions in terms of risk association, that is not giving a true causal significance to this gene as regards to FTD phenotypes. Literature data about ABCA1 concern primarily cardiovascular disease and variants classified as benign due to their relatively high frequency in the population are often associated with a certain level of risk for these diseases. As for dementias, in experimental models, ABCA1 activities appear to influence neuroinflammation and neurodegeneration (Karasinska et al., 2013), thus possibly affecting different forms of dementia, including FTD.

In conclusion, our study revealed the presence of variants in genes involved in pathways relevant for the pathology, especially autophagy and inflammation, common to different neurodegenerative diseases such as FTD, ALS, AD and PD. The findings support the role of these pathways in semantic phenotypes of FTD, as recently suggested (Pascual et al., 2021). Molecular analysis of dementia-related genes should be performed in all patients with svFTD and rtvFTD, to better understand the genotype–phenotype correlations and to study the pathogenetic mechanisms that could drive the clinical phenotypes in FTD.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

The studies involving human subjects were reviewed and approved by Ethics Committee of Fondazione IRCCS Istituto Neurologico Carlo Besta (Prot. n.39–05/04/2017). All the subjects provided their written informed consent to participate in this study.

GR and PC: conceptualization. ES, EM, and MR: formal analysis. CV, SP, FM, GF, PT, VR, CC, GK, EC, MF, and FA: investigation. FA and GG: funding acquisition. GR, ES, and EM: methodology. GR, PC, and ES: writing – original draft. GR, PC, and FA: writing – review and editing. All authors contributed to the article and approved the submitted version.

This research was funded by the European Research Council (StG-2016_714388_NeuroTRACK) and the Italian Ministry of Health, Ricerca Corrente RRC.

EC has received research supports from the Italian Ministry of Health. MF is Editor-in-Chief of the Journal of Neurology; received compensation for consulting services and/or speaking activities from Bayer, Biogen Idec, Merck-Serono, Novartis, Roche, Sanofi Genzyme, Takeda, and Teva Pharmaceutical Industries; and receives research support from Biogen Idec, Merck-Serono, Novartis, Roche, Teva Pharmaceutical Industries, Italian Ministry of Health, Fondazione Italiana Sclerosi Multipla, and ARiSLA (Fondazione Italiana di Ricerca per la SLA). FA is Section Editor of NeuroImage: Clinical; has received speaker honoraria from Biogen Idec, Roche and Zambon; and receives or has received research supports from the Italian Ministry of Health, AriSLA (Fondazione Italiana di Ricerca per la SLA), the European Research Council and the Foundation Research on Alzheimer Disease.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fnagi.2022.1085406/full#supplementary-material