www.nature.com/ejhg REVIEW ARTICLE OPEN Somatic genetic variation in healthy tissue and non-cancer diseases 2,7,8 ✉ Manuel Solís-Moruno1,2,3,4,9, Laura Batlle-Masó 5,9 , Núria Bonet 2 , Juan I. Aróstegui 3,4,6 and Ferran Casals © The Author(s) 2022 Somatic genetic variants have been studied for several years mostly concerning cancer, where they contribute to its origin and development. It is also clear that the somatic variants load is greater in aged individuals in comparison to younger ones, pointing to a cause/consequence of the senescence process. More recently, researchers have focused on the role of this type of variation in healthy tissue and its dynamics in cell lineages and different organs. In addition, somatic variants have been described to contribute to monogenic diseases, and the number of evidences of their role in complex disorders is also increasing. Thanks to recent advances in next-generation sequencing technologies, this type of genetic variation can be now more easily studied than in the past, although we still face some important limitations. Novel strategies for sampling, sequencing and filtering are being investigated to detect these variants, although validating them with an orthogonal approach will most likely still be needed. In this review, we aim to update our knowledge of somatic variation detection and its relation to healthy tissue and non-cancer diseases. 1234567890();,: European Journal of Human Genetics (2023) 31:48–54; https://doi.org/10.1038/s41431-022-01213-8 INTRODUCTION of somatic mutations. Also, certain somatic variants are the Mutation is the biological process that originates a genetic variant. causative agent of different diseases: many driver mutations have Thus, somatic variants are caused by postzygotic mutations, those been described in cancer, and some of them cause monogenic arising after the encounter of the spermatozoid and the ovule. diseases. For complex diseases, the link is harder to establish, They can occur from the beginning of the development of the although some studies have suggested that somatic variants may organisms to any point during their life. Postzygotic mutations play an important role in neurological and other disorders. Recent give rise to mosaicism, in which different cells of the same technological advances allow a better characterization of somatic individual present dissimilarities in their genomes, in contrast to variation in human tissues. However, there are still many questions germline variants, inherited by the individuals from their parents to be answered on its role and relative contribution to human and present in every cell of the body. To our knowledge, the first disorders, for which it is first crucial to understand the mention of the concept of the accumulation of somatic mutations, mechanisms that govern somatic variation origin, distribution which were referred to as “ageing hits” on somatic cells, appeared and transmission. In this review, while being virtually impossible in 1959 by Leo Szilard [1]. It represented the first remarkable effort covering the whole extensive and diverse literature on these that linked genetic damage and senescence. He developed his topics, we aim to provide a description for understanding theory from a mathematical point of view and mentioned the current state of this research field, including the evolution of mutational agents such as ionizing radiation. Also, the concept the different methodological strategies for the detection and of the somatic hypermutability in antibodies has been known analysis of somatic genetic variants, the knowledge of the since the 1980s. However, in the last 15 years, there has been an presence and dynamics of somatic genetic variants in healthy acceleration in the study of somatic variation, both in health and tissue, and their effect on non-cancer disorders. disease status, mostly because of the advances in next-generation sequencing (NGS) technologies. Somatic mutations accumulate with time, with aged individuals SOMATIC VARIANT DETECTION presenting significantly more variants than younger ones. It is also Heterozygous germline variants are present in half of the DNA well known that both extrinsic (tobacco smoke, alcohol, radiation) molecules, which is reflected in peaks of similar height in Sanger and intrinsic (cytosine deamination, oxidative damage, DNA sequencing or a presence in approximately 50 % of the reads for replication errors) mutational agents contribute to the emergence NGS. In fact, allelic imbalance, detected as a deviation from the 1 Institut de Biologia Evolutiva (CSIC-UPF), Departament de Ciències Experimentals i de la Salut, Universitat Pompeu Fabra, Doctor Aiguader 88, Barcelona, Spain. 2Genomics Core Facility, Departament de Ciències Experimentals i de la Salut, Universitat Pompeu Fabra, Parc de Recerca Biomèdica de Barcelona, 08003 Barcelona, Spain. 3Department of Immunology, Hospital Clínic, Barcelona, Spain. 4Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain. 5Pediatric Infectious Diseases and Immunodeficiencies Unit, Hospital Universitari Vall d’Hebron (HUVH), Vall d’Hebron Institut de Recerca (VHIR), Universitat Autònoma de Barcelona, Barcelona, Spain. 6Universitat de Barcelona, Barcelona, Spain. 7Departament de Genètica, Microbiologia i Estadística, Facultat de Biologia, Universitat de Barcelona, Barcelona, Spain. 8Institut de Biomedicina de la Universitat de Barcelona (IBUB), Universitat de Barcelona, Barcelona, Spain. 9These authors contributed equally: Manuel Solís-Moruno, Laura Batlle-Masó. ✉email: ferrancasals@ub.edu Received: 4 November 2021 Revised: 19 September 2022 Accepted: 3 October 2022 Published online: 27 October 2022 M. Solís-Moruno et al. 49 expected 50 % variant allele frequency (VAF) across NGS reads, is a live tissue, although the capture can include more than one cell filter commonly applied during variant calling as a strategy to and it requires advanced technical skills. Finally, microfluidic or discard sequencing and mapping artefacts from true germline droplet devices are gaining territory as emerging single-cell variants. In contrast, somatic genetic variants are usually found at isolation methods [3]. VAFs below 50 % and the proportion of carrier cells in each tissue Once cells have been isolated, one of the issues that single-cell will condition the capacity of their detection. Sanger sequencing is genomics methods encounter is the relatively large amounts of unable to detect those somatic variants with VAF below ∼10 %, genetic material required in current sequencing methods, in indistinguishable from the background signal. Also, variants with comparison to the low amount of DNA present in a cell. Two main VAFs above ∼30 % can be misinterpreted as germline [2]. strategies have been implemented to face this limitation: single- Alternatively, Sanger sequencing of multiple clones derived from cell isolation followed by whole-genome amplification (WGA), or individual DNA molecules can be used, although this approach clonal amplification of single cells in culture for DNA extraction requires a considerable amount of lab work. Therefore the use of plus sequencing (Fig. 1). The goal of WGA is to amplify the these strategies is limited to validate or to better approach the genome of a single cell with the highest precision while frequency of candidate mutations, the accuracy of which will minimizing the amplification bias. Initially, this methodology was depend on the number of sequenced clones, rather than for new based either on the ligation of adaptors that were used as specific variant discovery. primer sites for PCR amplification, or on the use of primers NGS overcomes most of these limitations, especially when high targeting repeated sequences in the genome. Improved techni- sequencing depth is achieved. For this reason it has been the main ques were designed including degenerate oligonucleotide-primer strategy employed during the last years to discover and PCR (DOP-PCR), multiple displacement amplification (MDA), and characterize somatic variants. Different strategies have been multiple annealing and looping-based amplification cycles (MAL- developed, which mainly differ according to sample obtention BAC) [3]. DOP-PCR is an exponential PCR method that uses and type as well as the sequencing approach (Fig. 1). On the random primers and a thermostable polymerase which has a sample nature, sequencing from bulk tissue was initially the higher error rate compared to the thermolabile enzymes. As said, standard approach with the constraint of consisting in a mixture one of the improvements of MDA is the use of isothermal of cell populations with multiple clonal origins. In this sense, polymerases that have a reduced error rate at the expense of microbiopsy and microdissection resulted in higher purity of the having less uniformity. In between both approaches, MALBAC sample although it will still show cell-type heterogeneity. starts with an isothermal preamplification followed by PCR Alternatively, sequencing at the single-cell level presents the amplification. An important caveat of these techniques is the theoretical possibility of identifying any single mutation in a cell, mutation rate, up to 1–40 mutations per 100 kb [4], which needs representing one step forward towards the understanding and to be considered in the variant calling process and subsequent characterization of the somatic genetic variant load in healthy analyses. Primary template-directed amplification (PTA) is an tissues. Several technologies are available for single-cell isolation: improved alternative that takes advantage of the low error rate micromanipulation, serial dilution, Fluorescence Activated Cell of the polymerase used in MDA with some modifications in the Sorting (FACS), Magnetic-Activated Cell Sorting (MACS), Laser amplification [5]. The process is quasilinear as the amplification Capture Microdissection (LCM), or microfluidic devices. FACS and occurs mostly from the primary template. The result is that errors MACS allow the individualization of dissociated single cells that have less propagation to subsequent amplicons. Furthermore, have been previously marked, the first with fluorescence and the their developers claim a recovery of 95 % of the single-cell second with antibodies conjugated to magnetic beads, which genomes in a more uniform representation. confer to these methodologies a higher specificity. In contrast, As an alternative to WGA, in vitro clonal expansion of single LCM is a low-throughput method that can be used with fixed and cells relays in the intrinsic replication mechanisms of the cell division. Classical 2D culture was one of the earliest methods used for this purpose yet it is difficult to achieve for some cell types, particularly from a single cell. Induced pluripotent stem cell (iPSC) culture is another option. However, the culture conditions and environmental agents can be a selective bottleneck to a group of cells and can induce de novo mutations during culture [4]. In contrast, organoid culture technologies have the advantage of being compatible with multiple human cell types, and long-term genome-stable cultures can be obtained. This approach is more time and resources consuming, but still more accurate and sensitive. In addition, it produces viable cells that can be used for further experiments. As for the library preparation protocols, methods based on the use of unique molecular identifiers (UMI) also contribute to reducing error rates. Duplex sequencing was one of the first methods to include tags to independently barcode each strand of a DNA duplex so that true mutations have to be found at the same position in both strands, and presenting a theoretical background error rate claimed of less than 10−9 errors per base pair [6]. Another alternative is the Safe-Sequencing System (Safe-SeqS) which incorporated the use of unique identifiers (UID, an Fig. 1 Sample types (green), sequencing approach (blue) and equivalent to UMI) [7]. These strategies were the basis of many methodology (grey) used to study somatic variants along time. new NGS methods such as BotSeqS or its most recent derivative Gradient in each box approximately represents the intensivity of use at different periods. DOP-PCR degenerate oligonucleotide-primer NanoSeq [8] which presents error rates lower than five per billion PCR, MDA multiple displacement amplification, MALBAC multiple base pairs. Both methods take advantage of the use of Duplex-Seq annealing and looping-based amplification, PTA primary template- Adaptors (IDT) that mark individual molecules and strands. directed amplification, UID unique identifiers, UMI unique molecular NanoSeq also reduces the substitution imbalances that are still identifiers. present in standard BotSeqS protocols estimated to be caused by European Journal of Human Genetics (2023) 31:48 – 54 M. Solís-Moruno et al. 50 the end repair and nick extension processes in the library frequency, or differential mutation rates across tissues. Recently, preparation. As an alternative, NanoSeq replaced the sonication some studies performed extensive multi-tissue analysis allowing fragmentation and end repair by restriction enzyme fragmenta- to perform a direct comparison, which confirmed a non-uniform tion or sonication plus exonuclease blunting and modified the distribution of somatic variation across somatic tissues [8, 14]. The A-tailing step to avoid errors from nick extension. results showed a few tens of single base substitutions (SBS) per Regarding the sequencing approach, studies to detect somatic year with up to two- to fivefold differences among tissues, which variants have used deep sequencing of target genes, whole- might be independent of the cell division rate [8]. The mutation exome sequencing (WES) and whole-genome sequencing (WGS) rates obtained by several studies exploring specific cell types and (Fig. 1). Because of the need of achieving high sequencing depths, tissues are in concordance with this observation. These rates are targeted sequencing and WES approaches have been extensively translated in a load from hundreds to a few thousands of somatic used in the analysis of tissue samples and biopsies [9, 10] as in the SNVs in a whole genome (e.g. 200–2000 in oesophageal case of cancer analysis from tumours. These studies are restricted epithelium [10] or around 1500 in neurons [18], and up to to candidate genes, which in addition to focusing on the 1500–15,000 in colonic crypts [12] which present the higher subsequent analysis and interpretation, have the advantage of substitution rates across the tissues analyzed). In contrast, providing high sequencing depths with more affordable costs. spermatogonia showed a lower rate, similar to the 1.35 SBS per Other works have used genomic approaches (WES, WGS and/or year inferred from trio analyses [8]. RNASeq) to the study of somatic genetic variation in healthy Germline cells represent a particular tissue regarding the tissues [11–14]. In most of these cases, samples are generated by analysis of genetic mosaicism. Contrary to somatic variants found microdissections or single-cell analysis, which partially compen- in non-germline tissues, genetic variants in gonadal or gonosomic sates for the smaller sequencing depths achieved in WGS mosaicisms can be transmitted to the next generations. These de compared to targeted analysis. Of importance, DNA sequencing novo variants are known to contribute importantly to the load of after clonal expansion in cell culture or developing organoids [4] genetic disorders in human populations. In that regard, there can be used as an approach to uncover all genomic regions and exists an important level of asymmetry in female and male avoid the possible biases of whole-genome amplification from a contributions with 80 % of de novo SNVs being inherited from the single cell, where each allele is represented only once. Finally, father, probably because of the elevated proliferation of WGS also confers the opportunity to expand the analysis more spermatogonia compared to oocytes with no mitotic activity. This accurately to copy number variation (CNV) in addition to SNVs and explains the increase of the genetic variants transmitted by males indels. In that line, comparative genomic hybridization has also to the offspring with age, because of replication and mitotic errors been used for structural variants detection from single cells [15]. in sperm, while other changes such as aneuploidies are more Once sequencing data is generated, several specific bioinfor- present in oocytes [19]. Sperm mosaicism has been classified into matics algorithms have been created for somatic variant calling, different types according to their origin in primordial, spermato- such as CaVEMan, EBCall, LoFreq, SomVarIUS, Strelka2, VarDict and gonia or sperm cells. While the number of mutations occurring in VarScan2. However, the comparison of the results generated by the earlier developmental steps and later at the sperm cell levels these tools has revealed, in general, poor levels of overlap [16] are constant across time, mutations in spermatogonia stem cells given the different algorithms and original purposes of such increase with paternal age. This also has important consequences callers. To resolve this issue, pipelines implementing different on the risk of recurrent transmission to offspring which is combinations of the filtering strategies have been proposed. After negligible for more recent mutations affecting the latest steps of somatic variant calling, and depending on the generated data, the sperm maturation but higher if occurred earlier, especially at the implementation of a set of filters is mandatory to exclude false level of primordial germ cells [19]. positives. Because of the modification of the frequency and The somatic mutation rate also does not remain constant over imbalance thresholds filters, the number of candidate genetic time, with higher rates in the first developmental stages and foetal variants is usually very high and true somatic variants need to be life compared to post-natal ones. Several studies have used discriminated from germline variants and sequencing, mapping or somatic variants as barcodes to define cellular phylogenies and calling errors. The commonly used filters are those based on study cellular dynamics in the embryo and development, mapping and calling quality, expected VAF, comparison to revealing an increase of the mutations per cell division in the variants called in other samples, allelic imbalance, position at first two or three cell divisions. The mutation rate is also increased the beginning/end of the reads, etc [17]. Finally, somatic genetic in several foetal tissues analyzed, up to fivefold higher than the variants should be validated by using a different sequencing same tissues in adulthood. Many factors can contribute to this approach. For this purpose, amplicon-based deep sequencing is excess of somatic mutations during early developmental stages being extensively used in current studies, also providing a more including a reduced time for DNA repair because of the shorter accurate estimation of the VAF in the sample. Of note, the duration of cell cycles or relaxation and dilution of the reparatory simultaneous analysis of different amplicons encompassing the mechanisms maternally inherited in absence of transcriptional same genetic variant is recommended to avoid inaccurate activity in the first egg divisions [20]. The mutational signatures frequency estimations due to PCR biases [2]. SBS1, a cell division clock signature and associated with the deamination of 5-methyl cytosine that happens before DNA replication [21], and SBS5, associated with the ageing process, are SOMATIC VARIANTS IN HUMAN DEVELOPMENT AND ADULT ubiquitously observed although with distinct relative contribu- TISSUES tions across tissues. SBS18, caused by oxidative damage, is also Somatic variants have been traditionally studied in the context of quite common across many tissues. Contrary, other signatures are cancer. More recently, enhanced by the development of NGS restricted to particular cell types pointing to a different relative technologies, the presence of somatic genetic variants in healthy contribution of mutational mechanisms and exposures. For tissues is being characterized. Many of the studies use different instance, SBS7 is associated with UV light exposure and found in combinations of the main approaches for sample obtention and the skin [9] and individual melanocytes [22]; SBS4 is associated somatic variant characterization (Fig. 1) and are performed on with tobacco smoke and found in smokers’ and ex-smokers’ lungs healthy tissues obtained in the context of cancer analysis. While [11]; and SBS16 is associated with alcohol consumption in the providing valuable information on particular tissues and organs, oesophagus. this methodological diversity makes the studies hardly compar- In addition to studies analyzing somatic variation across able among them in terms of somatic variant distribution, different tissues, many works have focused on particular cell European Journal of Human Genetics (2023) 31:48 – 54 M. Solís-Moruno et al. 51 types. Along with these studies human tissues are asymmetrically of somatic variants is a cause or consequence of the aging process represented, with a higher representation of tissues more easily is still under discussion [36]. accessible or obtained in the context of routine biopsies or An important process linked to ageing is muscle decline, surgeries, as well as because of the major interests that some especially in skeletal muscle which facilitates the mobility of tissues have generated in the research community (e.g. brain). The bodies. In that sense, somatic variants were studied in the satellite epithelium is among the most represented tissues, although with cells of muscles, where they have been observed to be a considerable variety of types analyzed including eyelid accumulated with age, also in exons and promoters. The authors epidermis, oesophageal and bronchial epithelium, liver or colonic suggest that this accumulation is a driving force behind the crypts, endometrial glands or urothelium. Some of these tissues decline of this type of muscle [37]. We are not aware of any work are particularly exposed to mutagenic agents such as sunlight, studying somatic variation in smooth muscle. Along with the tobacco and alcohol, with a confirmed role in the increase of physical decline of the ageing process comes the cognitive one. somatic mutations. Somatic variants in sun-exposed tissues We have already mentioned a few works exploring somatic present a pattern similar to that typically related to UV exposure, mutations in neurons [31], in which the accumulation of these although it is higher in intermittently sun-exposed skin (back and variants with age has been described and linked to ageing and limbs) than in sun-exposed skin (face) probably related to different neurological diseases. Actually, neurodegenerative diseases are mutation rates, DNA repair mechanisms or turnover frequency mostly diagnosed in elderly individuals, a topic developed later in [22]. The effect of tobacco on increasing the mutational load has this review. been shown in oesophageal and bronchial epitheliums. Positively selected or mutationally enriched genes have been found in some of these tissues, such as NOTCH1 (and other Notch family SOMATIC VARIANTS IN CLONAL HAEMATOPOIESIS members) in eyelid or oesophagus epithelium, or PIK3CA in Haematopoiesis is, briefly, the process by which mature blood colorectal crypts and endometrial glands. Skin fibroblasts have cells are formed in the bone marrow from haematopoietic stem also been analyzed, showing a rate of UV-induced somatic variants cells (HSCs). By studying the whole genomes of HSCs and which do not correlate with age [23]. Also, skin tone was downstream multipotent progenitor cells (MPPs) clones, from significantly impacting the load of somatic variants, with European people of different ages, an accumulation of somatic SNVs in a ancestry individuals showing a higher mutational load in skin linear fashion has been observed [38] at a ratio of 14.2 new base fibroblasts and melanocytes, probably due to the protective effect substitutions per year. Of note, indels did not seem to correlate of melanin [24]. with age. Somatic variants have been widely studied in blood due Brain and nervous system cells have also been analyzed by to the easy sampling, especially with the focus on clonal several studies, mostly in the context of analysis of neurodegen- haematopoiesis (CH). We understand CH as the phenomenon by erative disorders. Neurons are terminally differentiated post-mitotic which a single clone dominates the haematopoiesis process and it cells, thus, somatic variants in more than one cell will have occurred is overrepresented in blood. CH is related to ageing and several during development. In contrast, mutations occurred later will malignant pathologies. Among the several genes reported to be affect only the original cell. The observed enrichment in coding linked to CH [33, 39] we highlight DNMT3A, ASXL1 and TET2, three exons with a strand bias supports this, as the pattern is expected epigenetic regulators, as genes recurrently found to harbour for mutations arising in the transcription process [18]. In that line, somatic variants in individuals with CH. Recently, and taking the study of two brain areas, the prefrontal cortex and the dentate advantage of a dataset generated for cancer research with around gyrus of the hippocampus revealed interesting features such as 12,000 blood-cancer paired samples, the list of genes under differences between the regions, or a mutation pattern changing positive selection in CH has been expanded up to almost 70 [40]. with age. For example, among younger individuals, C > T mutations In the past, CH has been described as more common than (Signature B) are the most common, but their fraction decreased previously observed in people older than 65 [32]. In that study, with time. Several studies have also attempted to detect somatic genetic drift was proposed as the process behind CH, something structural variants in the human brain, as the effort to create a brain that also occurred with clonality in other tissues such as the CNV atlas for neurotypical individuals revealing important levels of stomach [41] and intestine [42]. However, in a metastudy variation across individuals and an inverse correlation with age [25] published in 2020, in which several of the aforementioned big contrary to the hypothesis of the accumulation of SV due to ageing datasets were reanalyzed, positive selection was seen as the main [26] which detected variants with higher VAFs (up to fivefold) in driving force of CH, the authors argue, among other reasons, that older people. Finally, a yet underexplored branch of somatic VAFs observed were too high to be caused by drift [43]. The structural variation is the role of transposable elements. Using relationship between CH and disease is explored in the previous single-cell studies, although Evrony et al. initially showed that LINE1 works. Not surprisingly, different haematological cancers are the insertions were not a source of mosaicism in neurons, the same most linked diseases to this process. Within CH we find a linked authors later reported clonal LINE1 retrotransposition events in the term: clonal haematopoiesis of indeterminate potential (CHIP). It is human brain [27]. Upton et al 2015 [28] described similar results in defined by the existence of cancer-associated clonal mutations, hippocampal neurons. This shows the rapid evolution of the field with VAF > 0.02, in blood cells or bone marrow, but without and encourages the debate about the comparison of somatic having an actual malignancy [44]. The probability to develop events using different tissues and technologies. cancer in individuals with CHIP is low, and it is calculated to be increased by this condition by around 0.5 %-1 % per year. Also, the presence of CHIP almost doubles the risk of having coronary heart SOMATIC VARIANTS IN AGING disease and makes the risk of myocardial infarction four times We accumulate somatic mutations throughout life, a fact higher [45]. supported by a great number of studies, as these in the epithelium of oesophagus [10, 29]; endometrium [14] lung [11]; colon [12] and bladder [13]. Also in bulk skin [9], individual melanocytes [22]; SOMATIC VARIANTS IN MONOGENIC DISEASES liver [30]; neurons [31] and blood [32, 33]. This happens in fully In the same way as germline variants, somatic variants can be the differentiated cells, but it has also been observed in adult stem cause of monogenic diseases if they reach a sufficient VAF in the cells of the small intestine, colon and liver [34]. In fact, even the affected organ or tissue. This is especially relevant for diseases speed of accumulation of these mutations is accelerated when the with an autosomal dominant inheritance pattern, or those aging process is boosted in mice [35]. Whether the accumulation X-linked in men, where only one altered copy of the causal allele European Journal of Human Genetics (2023) 31:48 – 54 M. Solís-Moruno et al. 52 is needed to develop the disease. The first study proving the direct identified by linkage and association studies. Later, NGS allowed impact of somatic variants in disease is more than 25 years old exploring the role of rare genetic variation thanks to the [46] which described mutations in the PIGA gene as causative of development of new statistical approaches mostly optimized to paroxysmal nocturnal haemoglobinuria. identify the enrichment of this type of variants in patients A very remarkable example of somatic variants in monogenic compared to controls, or even to explore the role of de novo disorders are inborn errors of immunity (IEI), with growing genetic variants in particular individuals that would therefore evidence of the role of genetic mosaicism in recent years [47]. better fit to a Mendelian inheritance. The de novo scenario has In addition to particular mechanisms such as clonal hematopoiesis been particularly explored for neurodevelopmental disorders, that can increase the frequency of a certain genetic variant, a where these mutations would be unlikely to be observed as methodological bias because of the accessibility and universal use germinal because of the low reproductive rates of affected of the affected tissue, blood, could also contribute to an individuals [61]. Similarly, new genetic variants arising during overrepresentation of IEI among the reported cases. The pioneer development or even in adult neurons could also contribute to studies reporting the first somatic variants associated with IEI are the development of these disorders. from 2004 [48] and 2005 [49], in which the NLRP3 (former CIAS1) Several works propose a possible relation of somatic variants to and FAS genes were identified as harbouring causative somatic neurological disorders such as Alzheimer’s disease [62] or variants. Interestingly, NLRP3 is nowadays the gene with the Parkinson’s disease [63]. Also, individuals with a neurodegenera- greatest number of reported somatic variants linked to autoin- tive condition showed an increase in the number of somatic SNVs flammatory syndromes [50], a family of disorders classified within in comparison with matched healthy individuals of similar age. IEI. On the other hand, the FAS gene, underlying autoimmune A somatic copy number alterations was found to be responsible lymphoproliferative syndrome (ALPS), is an interesting example of for the neuropsychiatric disease [64] at VAFs below 20 %. Also, in various genetic mechanisms converging in similar phenotypes. autism spectrum disorder, structural variants affecting 3–70 % of FAS causes ALPS in patients who harbour pathogenic germline the cells were detected by microarrays [65]. Interestingly, in this variants, somatic variants, and a combination of them [51]. Up to case, the event size positively correlated with the severity of the 20 % of the cases are caused exclusively by somatic variation, patients’ symptoms [65]. Recently, following the idea that LINE1 which is restricted to a specific cell population and can be does play a role in the diversity of the neuron’s genome, Zhu et al. detected by NGS [51]. Other outstanding cases are somatic corroborated the possible pathogenic effect of retrotransposon variants in NOD2 causing Blau syndrome [52]; NLRC4 causing insertion on neurologic disease [66]. NOMID (neonatal-onset multisystem inflammatory disease) [53]; As mentioned before, CH is related to a group of complex and UBA1 associated with the late-onset VEXAS (vacuole, E1 diseases, also beyond cancer. For instance, there is a relationship enzyme, X-linked, autoinflammatory, somatic) syndrome, a severe between chronic ischaemic heart failure and the mutations X-linked disorder affecting males and restricted to cells of the causing the CH [67]. The somatic variant landscape has been also myeloid line [54]. We could also extend the list of genes studied in endometriosis [68], in which the authors performed harbouring somatic variants being causative of an IEI to BTK, WES on 13 endometriotic and 11 normal endometrial epithelium ELANE, IL2RG, NRAS, PIK3CD, STAT3 and WAS [2]. Interestingly, we samples. They also performed targeted sequencing on a larger have recently reported that those causal genetic variants are validation dataset of 94 endometriotic epitheliums from 45 detectable from moderately high sequencing coverages in WES individuals and 71 normal epitheliums from 29 individuals. They data from blood, by adapting the variant calling bioinformatics discovered two oncogenes, KRAS and PIK3CA, to be the most pipelines and establishing a set of stringent filters [55]. Thus, it frequently mutated in both scenarios, but with VAFs significantly could be recommendable considering the detection of somatic higher in endometriosis. Thus, the authors suggested a potential variants in non-diagnosed cases after NGS, which if restricted to a role in the development of endometriosis of the generated clonal set of candidate genes can be a feasible and affordable approach. expansion due to the presence of these cancer-associated The list of reported cases of diseases includes other organs and mutations. tissues beyond blood, such as Alport syndrome, in which somatic Following a previous publication on colonic crypts [12], and variants in COL4A3 can impair the function of kidneys [56], or using the generated samples in that study as healthy controls, Huntington’s disease with a somatic expansion of the CAG triplet inflammatory bowel disease (IBD) has been explored. In this new in the HTT gene in the brain [57]. We also want to highlight the study, WGS was performed on 446 crypts from 46 IBD patients existence of several diseases in which only somatic variants, and with a median coverage of 18.2× [69]. They sequenced 28 not germline, have been described. In these cases, the most ulcerative colitis patients and 18 Crohn’s disease patients and feasible hypothesis is that germline variants would be highly described an average of ~2.4-fold increase in the mutation rates lethal. This is supported by the severe effects of the somatic of IBD patients compared to controls, and reported no alterations, such as somatic-derived hemimegalencephaly [58]. significant differences between the patients of both diseases. Of interest, a somatic mutation could also reverse a pathogenic They observed that around 80 % of the increase in mutation variant acting as a spontaneous reversion of germline inherited burden is explained by mutational signatures also present in the disease [59]. These new variants may confer an advantage to the healthy colon, pointing to an acceleration of the normal age- cell and would be targeted by natural selection and increase in related mutational processes. Besides this, the clonal expansion frequency. Therefore, the presence of the new variant may have a found in patients with IBD is much larger than in healthy positive impact on the phenotype of the affected individuals, who individuals, with an important number of clones bigger than can experience significant clinical improvements. This is called 2 mm. They also found signatures of positive selection in genes somatic genetic rescue and has been described in various ARID1A, FBXW7, PIGR and ZC3H12A, as well as in some of the IL- diseases, among which many IEI [60]. The presence of such 17 and Toll-like receptor pathways, suggesting differences in events has important diagnostic implications, as it could explain selection mechanisms in IBD colons. Finally, the authors suggest unexpected clinical improvements, and also therapeutic implica- a potential causal role of somatic mutations in IBD. The case of tions, as an in vivo proof of the potential effect of gene therapy. ulcerative colitis has also been studied in another work, in which WES was performed on 76 clonal human colon organoids [70]. Similarly, the authors observed an accumulation of somatic SOMATIC VARIANTS IN COMPLEX DISEASES mutations in genes of the IL-17 signalling pathway such as The study of the genetic aetiology of complex disorders was NFKBIZ, ZC3H12A and PIGR. These genes are not particularly traditionally focused on the role of common genetic variants related to colorectal cancer. European Journal of Human Genetics (2023) 31:48 – 54 M. Solís-Moruno et al. 53 CONCLUSIONS 18. Ju YS, Martincorena I, Gerstung M, Petljak M, Alexandrov LB, Rahbari R, et al. The number of evidences of the role of postzygotic mutations in Somatic mutations reveal asymmetric cellular dynamics in the early human non-cancer diseases is growing in parallel with the development embryo. Nature. 2017;543:714–8. of experimental and analytical methodologies to detect and 19. Breuss MW, Yang X, Gleeson JG. Sperm mosaicism: implications for genomic diversity and disease. Trends Genet. 2021;37:890–902. characterize somatic genetic variants. Sampling methods are 20. Manders F, van Boxtel R, Middelkamp S. The dynamics of somatic mutagenesis being refined allowing the characterization of genetic variants at during life in humans. Front Aging. 2021;2:802407. the single-cell level in an increasing number of cell types. 21. Nik-Zainal S, Alexandrov LB, Wedge DC, Van Loo P, Greenman CD, Raine K, et al. Simultaneously, bioinformatics pipelines are being adapted to Mutational processes molding the genomes of 21 breast cancers. Cell. different experimental designs to identify real somatic variants 2012;149:979–93. while discarding false positives generated by sequencing, map- 22. Tang J, Fewings E, Chang D, Zeng H, Liu S, Jorapur A, et al. The genomic land- ping and mostly calling artefacts because of the relaxation of scapes of individual melanocytes from human skin. Nature. 2020;586:600–5. allelic imbalance thresholds. Of importance, a proper biological 23. Abyzov A, Tomasini L, Zhou B, Vasmatzis N, Coppola G, Amenduni M, et al. One interpretation of the results can only be done based on the thousand somatic SNVs per skin fibroblast cell set baseline of mosaic mutational load with patterns that suggest proliferative origin. Genome Res. 2017;27:512–23. understanding of the presence and distribution of somatic genetic 24. Saini N, Giacobone CK, Klimczak LJ, Papas BN, Burkholder AB, Li JL, et al. UV- variants in healthy tissue, which provides a baseline for the exposure, endogenous DNA damage, and DNA replication errors shape the analysis of somatic variation in disease. While there exist some spectra of genome changes in human skin. PLoS Genet. 2021;17:1–23. commonalities across tissues as the accumulation of somatic 25. Proukakis C. Somatic mutations in neurodegeneration: an update. Neurobiol Dis. variants with age and exposure to external agents such as 2020;144:105021. tobacco, alcohol or sun radiation, we should ideally consider the 26. Laurie CC, Laurie CA, Rice K, Doheny KF, Zelnick LR, McHugh CP, et al. Detectable specificities and different mutagenic patterns shown by each clonal mosaicism from birth to old age and its relationship to cancer. Nat Genet. cell type. 2012;44:642–50. 27. Evrony GD, Lee E, Mehta BK, Benjamini Y, Johnson RM, Cai X, et al. Cell lineage analysis in human brain using endogenous retroelements. Neuron. REFERENCES 2015;85:49–59. 1. Szilard L. On the nature of the aging process. Proc Natl Acad Sci USA. 28. Upton KR, Gerhardt DJ, Jesuadian JS, Richardson SR, Sánchez-Luque FJ, Bodea 1959;45:30–45. GO, et al. Ubiquitous L1 mosaicism in hippocampal neurons. Cell. 2. Mensa-Vilaró A, García-Morato MB, de la Calle-Martin O, Franco-Jarava C, 2015;161:228–39. Martínez-Saavedra MT, González-Granado LI, et al. Unexpected relevant role of 29. Yokoyama A, Kakiuchi N, Yoshizato T, Nannya Y, Suzuki H, Takeuchi Y, et al. Age- gene mosaicism in primary immunodeficiency diseases. J Allergy Clin Immunol. related remodelling of oesophageal epithelia by mutated cancer drivers. Nature. 2019;143:359–68 2019;565:312–7. 3. Gawad C, Koh W, Quake SR. Single-cell genome sequencing: current state of the 30. Brunner SF, Roberts ND, Wylie LA, Moore L, Aitken SJ, Davies SE, et al. Somatic science. Nat Rev Genet. 2016;17:175–88. mutations and clonal dynamics in healthy and cirrhotic human liver. Nature. 4. Youk J, Kwon HW, Kim R, Ju YS. Dissecting single-cell genomes through the clonal 2019;574:538–42. organoid technique. Exp Mol Med. 2021;53:1503–11. 31. Lodato MA, Rodin RE, Bohrson CL, Coulter ME, Barton AR, Kwon M, et al. Aging 5. Gonzalez-Pena V, Natarajan S, Xia Y, Klein D, Carter R, Pang Y, et al. Accurate and neurodegeneration are associated with increased mutations in single human genomic variant detection in single cells with primary template-directed ampli- neurons. Science. 2018;359:555–9. fication. Proc Natl Acad Sci USA. 2021;118:e2024176118. 32. Zink F, Stacey SN, Norddahl GL, Frigge ML, Magnusson OT, Jonsdottir I, et al. 6. Schmitt MW, Kennedy SR, Salk JJ, Fox EJ, Hiatt JB, Loeb LA. Detection of ultra-rare Clonal hematopoiesis, with and without candidate driver mutations, is common mutations by next-generation sequencing. Proc Natl Acad Sci USA. in the elderly. Blood. 2017;130:742–52. 2012;109:14508–13. 33. Jaiswal S, Fontanillas P, Flannick J, Manning A, Grauman PV, Mar BG, et al. Age- 7. Kinde I, Wu J, Papadopoulos N, Kinzler KW, Vogelstein B. Detection and quanti- related clonal hematopoiesis associated with adverse outcomes. N Engl J Med. fication of rare mutations with massively parallel sequencing. Proc Natl Acad Sci 2014;371:2488–98. USA. 2011;108:9530–5. 34. Blokzijl F, de Ligt J, Jager M, Sasselli V, Roerink S, Sasaki N, et al. Tissue-specific 8. Abascal F, Harvey LMR, Mitchell E, Lawson ARJ, Lensing SV, Ellis P, et al. Somatic mutation accumulation in human adult stem cells during life. Nature. mutation landscapes at single-molecule resolution. Nature. 2021;593:405–10. 2016;538:260–4. 9. Martincorena I, Roshan A, Gerstung M, Ellis P, Van Loo P, McLaren S, et al. High 35. Odagiri Y, Uchida H, Hosokawa M, Takemoto K, Morley AA, Takeda T. Accelerated burden and pervasive positive selection of somatic mutations in normal human accumulation of somatic mutations in the senescence-accelerated mouse. Nat skin. Science. 2015;348:880–6. Genet. 1998;19:116–7. 10. Martincorena I, Fowler JC, Wabik A, Lawson ARJ, Abascal F, Hall MWJ, et al. 36. Vijg J, Montagna C. Genome instability and aging: cause or effect? Transl Med Somatic mutant clones colonize the human esophagus with age. Science. Aging. 2017;1:5–11. 2018;362:911–7. 37. Franco I, Johansson A, Olsson K, Vrtačnik P, Lundin P, Helgadottir HT, et al. 11. Yoshida K, Gowers KHC, Lee-Six H, Chandrasekharan DP, Coorens T, Maughan EF, Somatic mutagenesis in satellite cells associates with human skeletal muscle et al. Tobacco smoking and somatic mutations in human bronchial epithelium. aging. Nat Commun. 2018;9:800. Nature. 2020;578:266–72. 38. Osorio FG, Rosendahl Huber A, Oka R, Verheul M, Patel SH, Hasaart K, et al. 12. Lee-Six H, Olafsson S, Ellis P, Osborne RJ, Sanders MA, Moore L, et al. The land- Somatic mutations reveal lineage relationships and age-related mutagenesis in scape of somatic mutation in normal colorectal epithelial cells. Nature. human hematopoiesis. Cell Rep. 2018;25:2308–2316.e4. 2019;574:532–7. 39. Xie M, Lu C, Wang J, McLellan MD, Johnson KJ, Wendl MC, et al. Age-related 13. Lawson ARJ, Abascal F, Coorens THH, Hooks Y, O’Neill L, Latimer C, et al. Extensive mutations associated with clonal hematopoietic expansion and malignancies. heterogeneity in somatic mutation and selection in the human bladder. Science. Nat Med. 2014;20:1472–8. 2020;370:75–82. 40. Pich O, Reyes-Salazar I, Gonzalez-Perez A, Lopez-Bigas N. Discovering the drivers 14. Moore L, Cagan A, Coorens THH, Neville MDC, Sanghvi R, Sanders MA, et al. The of clonal hematopoiesis. Nat Commun. 2020;13:4267. mutational landscape of human somatic and germline cells. Nature. 41. Barker N, Huch M, Kujala P, van de Wetering M, Snippert HJ, van Es JH, et al. 2021;597:381–6. Lgr5+ve stem cells drive self-renewal in the stomach and build long-lived gastric 15. Jacobs K, Mertzanidou A, Geens M, Thi Nguyen H, Staessen C, Spits C. Low-grade units in vitro. Cell Stem Cell. 2010;6:25–36. chromosomal mosaicism in human somatic and embryonic stem cell popula- 42. Lopez-Garcia C, Klein AM, Simons BD, Winton DJ. Intestinal stem cell replacement tions. Nat Commun. 2014;5:1–10. follows a pattern of neutral drift. Science. 2010;330:822–5. 16. Cai L, Yuan W, Zhang Z, He L, Chou KC. In-depth comparison of somatic point 43. Watson CJ, Papula AL, Poon GYPP, Wong WH, Young AL, Druley TE, et al. The mutation callers based on different tumor next-generation sequencing depth evolutionary dynamics and fitness landscape of clonal hematopoiesis. Science. data. Sci Rep. 2016;6:1–9. 2020;367:1449–54. 17. Lobon I, Solís-Moruno M, Juan D, Muhaisen A, Abascal F, Esteller-Cucala P, et al. 44. Steensma DP, Bejar R, Jaiswal S, Lindsley RC, Sekeres MA, Hasserjian RP, et al. Somatic mutations detected in Parkinson disease could affect genes with a role Clonal hematopoiesis of indeterminate potential and its distinction from mye- in synaptic and neuronal processes. Front Aging. 2022;3:851039. lodysplastic syndromes. Blood. 2015;126:9–16. European Journal of Human Genetics (2023) 31:48 – 54 M. Solís-Moruno et al. 54 45. Jaiswal S, Natarajan P, Silver AJ, Gibson CJ, Bick AG, Shvartz E, et al. Clonal 66. Zhu X, Zhou B, Pattni R, Gleason K, Tan C, Kalinowski A, et al. Machine learning hematopoiesis and risk of atherosclerotic cardiovascular disease. N Engl J Med. reveals bilateral distribution of somatic L1 insertions in human neurons and glia. 2017;377:111–21. Nat Neurosci. 2021;24:186–96. 46. Takeda J, Miyata T, Kawagoe K, Iida Y, Endo Y, Fujita T, et al. Deficiency of the GPI 67. Dorsheimer L, Assmus B, Rasper T, Ortmann CA, Ecke A, Abou-El-Ardat K, et al. anchor caused by a somatic mutation of the PIG-A gene in paroxysmal nocturnal Association of mutations contributing to clonal hematopoiesis with prognosis in hemoglobinuria. Cell. 1993;73:703–11. chronic ischemic heart failure. JAMA Cardiol. 2019;4:25–33. 47. Van Horebeek L, Dubois B, Goris A. Somatic variants: new kids on the block in 68. Suda K, Nakaoka H, Yoshihara K, Ishiguro T, Tamura R, Mori Y, et al. Clonal human immunogenetics. Trends Genet. 2019;35:935–47. expansion and diversification of cancer-associated mutations in endometriosis 48. Holzelova E, Vonarbourg C, Stolzenberg MC, Arkwright PD, Selz F, Prieur AM, et al. and normal endometrium. Cell Rep. 2018;24:1777–89. Autoimmune lymphoproliferative syndrome with somatic Fas mutations. N Engl J 69. Olafsson S, McIntyre RE, Coorens T, Butler T, Jung H, Robinson PS, et al. Somatic Med. 2004;351:1409–18. evolution in non-neoplastic IBD-affected colon. Cell. 2020;182:672–684.e11. 49. Saito M, Fujisawa A, Nishikomori R, Kambe N, Nakata-Hizume M, Yoshimoto M, 70. Nanki K, Fujii M, Shimokawa M, Matano M, Nishikori S, Date S, et al. Somatic et al. Somatic mosaicism of CIAS1 in a patient with chronic infantile neurologic, inflammatory gene mutations in human ulcerative colitis epithelium. Nature. cutaneous, articular syndrome. Arthritis Rheum. 2005;52:3579–85. 2020;577:254–9. 50. Tanaka N, Izawa K, Saito MK, Sakuma M, Oshima K, Ohara O, et al. High incidence of NLRP3 somatic mosaicism in patients with chronic infantile neurologic, cuta- neous, articular syndrome: results of an international multicenter collaborative AUTHOR CONTRIBUTIONS study. Arthritis Rheum. 2011;63:3625–32. All authors have contributed to the study design, literature revision, discussion and 51. López-Nevado M, Docampo-Cordeiro J, Ramos JT, Rodríguez-Pena R, Gil-López C, writing of the study. Sánchez-Ramón S, et al. Next generation sequencing for detecting somatic FAS mutations in patients with autoimmune lymphoproliferative syndrome. Front Immunol. 2021;12:656356. 52. Mensa-Vilaro A, Tarng Cham W, Ping Tang S, Chin Lim S, González-Roca E, Ruiz- FUNDING This study was funded by grants PID2021-125106OB-C32 (FC), RTI2018-096824-B-C22 Ortiz E, et al. First identification of intrafamilial recurrence of Blau syndrome due (FC), RTI2018-096824-B-C21 (JIA) and PID2021-125106OB-C31 (JIA) funded by MCIN/ to gonosomal NOD2 mosaicism. Arthritis Rheumatol. 2016;68:1039–44. AEI /10.13039/501100011033/ and FEDER Una manera de hacer Europa; Direcció 53. Kawasaki Y, Oda H, Ito J, Niwa A, Tanaka T, Hijikata A, et al. Identification of a high- General de Recerca- Generalitat de Catalunya (2017SGR-702) (FC); and CERCA frequency somatic NLRC4 mutation as a cause of autoinflammation by pluripotent Programme/Generalitat de Catalunya (JIA). MS-M was supported by the Ministerio de cell-based phenotype dissection. Arthritis Rheumatol. 2017;69:447–59. Economía y Competitividad, Spain (Maria de Maetzu grant MDM-2014-0370-16-3). 54. Beck DB, Ferrada MA, Sikora KA, Ombrello AK, Collins JC, Pei W, et al. Somatic mutations in UBA1 and severe adult-onset autoinflammatory disease. N Engl J Med. 2020;383:2628–38. 55. Solís-Moruno M, Mensa-Vilaró A, Batlle-Masó L, Lobón I, Bonet N, Marquès-Bonet COMPETING INTERESTS T, et al. Assessment of the gene mosaicism burden in blood and its implications The authors declare no competing interests. for immune disorders. Sci Rep. 2021;11:12940. 56. Bruttini M, Vitelli F, Meloni I, Rizzari G, Della Volpe M, Mazzucco G, et al. Mosai- cism in Alport syndrome and genetic counseling. J Med Genet. 2000;37:717–9. ADDITIONAL INFORMATION 57. Swami M, Hendricks AE, Gillis T, Massood T, Mysore J, Myers RH, et al. Somatic Correspondence and requests for materials should be addressed to Ferran Casals. expansion of the Huntington’s disease CAG repeat in the brain is associated with an earlier age of disease onset. Hum Mol Genet. 2009;18:3039–47. Reprints and permission information is available at http://www.nature.com/ 58. Lee JH, Huynh M, Silhavy JL, Kim S, Dixon-Salazar T, Heiberg A, et al. De novo reprints somatic mutations in components of the PI3K-AKT3-mTOR pathway cause hemimegalencephaly. Nat Genet. 2012;44:941–5. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims 59. Revy P, Kannengiesser C, Fischer A. Somatic genetic rescue in Mendelian hae- in published maps and institutional affiliations. matopoietic diseases. Nat Rev Genet. 2019;20:582–98. 60. Miyazawa H, Wada T. Reversion mosaicism in primary immunodeficiency dis- eases. Front Immunol. 2021;12:783022. 61. D’Gama AM, Pochareddy S, Li M, Jamuar SS, Reiff RE, Lam ATN, et al. Targeted DNA sequencing from autism spectrum disorder brains implicates multiple Open Access This article is licensed under a Creative Commons genetic mechanisms. Neuron. 2015;88:910–7. Attribution 4.0 International License, which permits use, sharing, 62. Park JS, Lee J, Jung ES, Kim M-H, Kim IB, Son H, et al. Brain somatic mutations adaptation, distribution and reproduction in any medium or format, as long as you give observed in Alzheimer’s disease associated with aging and dysregulation of tau appropriate credit to the original author(s) and the source, provide a link to the Creative phosphorylation. Nat Commun. 2019;10:3090. Commons license, and indicate if changes were made. The images or other third party 63. Coxhead J, Kurzawa-Akanbi M, Hussain R, Pyle A, Chinnery P, Hudson G. Somatic material in this article are included in the article’s Creative Commons license, unless mtDNA variation is an important component of Parkinson’s disease. Neurobiol indicated otherwise in a credit line to the material. If material is not included in the Aging. 2015;38:217.e1–217.e6. article’s Creative Commons license and your intended use is not permitted by statutory 64. Cai X, Evrony GD, Lehmann HS, Elhosary PC, Mehta BK, Poduri A, et al. Single-cell, regulation or exceeds the permitted use, you will need to obtain permission directly genome-wide sequencing identifies clonal somatic copy-number variation in the from the copyright holder. To view a copy of this license, visit http:// human brain. Cell Rep. 2014;8:1280–9. creativecommons.org/licenses/by/4.0/. 65. Sherman MA, Rodin RE, Genovese G, Dias C, Barton AR, Mukamel RE, et al. Large mosaic copy number variations confer autism risk. Nat Neurosci. 2021;24:197–203. © The Author(s) 2022 European Journal of Human Genetics (2023) 31:48 – 54