Gonadal Mosaicism as a Rare Inheritance Pattern in Recessive Genodermatoses: Report of Two Cases with Pseudoxanthoma Elasticum and Literature Review

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This case report by Dangreau et al. covers unusual cases of pseudoxanthoma elasticum (PXE) in which the authors suspect gonadal mosaicism instead of the typical autosomal recessive inheritance. Although this is likely in both families here, the authors have not shown definitively that the deleted alleles of ABCC6 came from the suspected fathers. They need to show and describe data to confirm that these are indeed due to gonadal mosaicism in the father and not simple inheritance from a different father. There are also issues with clarity in the images shown and the description of the molecular data (including the MLPA results). The authors should also describe their literature review in more detail. These issues reduced enthusiasm for publication of the manuscript.

Major points:

1. (Methods and Results) MLPA needs to be described in sufficient detail (and with respect to images shown in Figure 2) that the readers can assess how this technique measures deletions. Some readers may not be completely familiar with this assay. There is also no description of the MLPA results presented in the text or Figure 2 legend.

2. (Results) Since the fathers of each proband apparently did not submit sperm (or other?) DNA samples, the authors must show or describe the procedure and data used to state that the father is indeed the father of the affected children. The claim of paternity in the text is not sufficient. In the second case, even though the brother has the same deletion, it remains formally possibility that a different father could have contributed this deleted allele of ABCC6 to both of these children and therefore it would not necessarily be a case of germline mosaicism. In both cases, there must be genetic evidence of paternity shown and described.

3. (Figure 2) Figure 2 needs to be enlarged and the image quality improved, particularly the size and quality of the electropherograms and MLPA results. Without this information being legible, the entire manuscript is compromised. The relevant mutation in the mother of the first family is difficult to see on the electropherogram; one can only see the wild type “T” at the splice site, Also, the relevant mutation in the proband (II-1) in the second family should be labeled “T” instead of what appears to be C (the wild type sequence). The lack of consistency in how the electropherograms are presented between the two families also takes away from the manuscript.

4. (literature review) This section is confusing as written and should be clarified and expanded. The first sentence should specify that only one patient or proband is the subject in each study (except the current one). The last sentence of the paragraph should be moved to after the first sentence, since it applies to the entire table and the 14 families. Then, clearly discuss the specific instance of PXE, stating that “among all reported (provide the number) cases of PXE, only one case (other than the current report) was suspected gonadal mosaicism.” The authors could further expand this section and their description of table 1.

Minor comments

1. (p. 2) The sentence including the primary symptoms of PXE is too long and convoluted. This should be  reworded and/or broken up into multiple sentences.

2. (Methods) Primer sequences used for PCR and sequencing of the ABCC6, GGCX, and ENPP1 genes should be included here.

3. (Methods) The authors state “Nucleotide numbers are derived from gDNA ABCC6 sequences (GenBank accession no. NM_001171).” Since this accession number provides the cDNA/mRNA sequence and excludes other genomic sequences, the gDNA term must be changed to cDNA.

4. (Discussion) I believe that increasing paternal age increases the frequency of germline mutations of all types (not only SNVs) being present in the offspring, as opposed to what the authors state. This point needs to be clarified, as it is relevant to the frequency of de novo mutations in the offspring. The cases reported here are likely due to gonadal mosaicism, but this needs to be definitively proven (as mentioned above).

5. (Discussion) The use of the term “non-homologous recombination” probably is incorrect in that all recombination that occurs during meiosis is homologous or at least homeologous. If there are homologous or homeologous sequences in the broad region surrounding ABCC6, this can promote misalignment during meiosis and deletion of the intervening region in one of the resulting germ cells.

Comments on the Quality of English Language

The authors need to use commas in certain places.

Author Response

Major points:

1. (Methods and Results) MLPA needs to be described in sufficient detail (and with respect to images shown in Figure 2) that the readers can assess how this technique measures deletions. Some readers may not be completely familiar with this assay. There is also no description of the MLPA results presented in the text or Figure 2 legend.

We understand the comment of the reviewer that not every reader may be sufficient familiar with the MLPA technology and how to interpret the data output. Therefore we have briefly summarized the MLPA technology characteristics in the methods section, including also a referral to a review paper for readers who wish to have more in-depth information about the technique (Stuppia et al.). This paragraph now reads as: 

“MLPA (Multiplex ligation-dependent probe amplification) analysis. MLPA is a sensitive and specific molecular technique used for the simultaneous detection and quantification of multiple target sequences in a single reaction. MLPA employs two hybridization probes that bind adjacent to each other on the target DNA, followed by a ligation step that joins the probes only when they hybridize correctly, thus allowing the amplification of specific sequences. The amplified products are then analyzed via capillary electrophoresis to determine the presence and relative quantities of the target sequences [15]. MLPA analysis of the ABCC6 gene was performed using the SALSA MLPA kit PO92-B3 (MRC-Holland, Amsterdam, The Netherlands) according to the manufacturer’s recommendations. This kit contains 23 probes corresponding to ABCC6 exons 2, 4, 5, 7–15, 17, 18, 21–28 and 30 and control probes for quality control. The PO92-B3 kit lacks probes for ABCC6 exons 1, 3, 6, 16, 19, 20, 29 and 31. As ABCC1 is in close proximity to ABCC6 (6.5 kb telomeric), an ABCC1 probe was also included. The construction of the kit precludes generation of signals from the ABCC6 pseudogenes. MLPA fragments were detected using an ABI3130XL or ABI3730XL capillary electrophoresis system (Applied Biosystems, Foster City, USA). The genemapper software (Applied Biosystems,Foster City, USA) was used to calculate fragment size and concentration, whereas the quantification analysis was performed using Coffalyzer (MRC Holland, Amsterdam, The Netherlands). All samples were tested in duplicate.”

The MLPA results are mentioned in the text of the original manuscript but this was further emphasized in the revised version of the results section (case reports):

Family 1: (…) “ABCC6 analysis revealed compound heterozygosity for the pathogenic c.3506+2T>C splice site variant (via direct sequencing; Figure 2A,g) and a whole ABCC6 gene deletion (via MLPA; Figure 2A,h) in the proband. “

Family 2: (…) “ABCC6 analysis showed the proband to be compound heterozygous for the p.R1141* pathogenic variant (detected via direct sequencing), inherited from the mother, and a whole ABCC6 deletion (demonstrated by MLPA), which was not found in blood and buccal cells of either parent (Figure 2B,d-f).“

Furthermore, we have updated the legend of Figure 2 with more details on how to interpret the graphic MLPA results that are shown. This legend now reads as: 

“Figure 2. Clinical characteristics, pedigrees and molecular findings of the proband families. 

Panel A: proband 1. (a) Papular skin lesions in the lateral neck at the time of diagnosis; (b-e) White light funduscopy of the proband (b,c) and his mother (d,e). Peau d’orange (asterisk), comet tails (open arrows) and angioid streaks (closed arrows) are shown. (f) Pedigree of the family of proband 1 (II-2, arrowed). The ABCC6 genotype is indicated for each individual (WT = wild type). (g) Electropherograms of the ABCC6 direct sequencing results for individuals I-1, I-2, II-1 and II-2. The location of the affected nucleotide at position c.3506+2 is arrowed. Due to the presence of a heterozygous whole gene deletion, only a single peak is observed in the proband, in contrast to overlapping peaks in his mother and sister. (h) MLPA results of the proband (II-2) and his father (I-1).  Every bar is the ratio result of 1 probe pair PCR product. From left to right: ratio for the ABCC1 probe, ABCC6 exon 30, exon 28–21, exon 18–17, exon 15–7, exon 5, exon 4, exon 2, followed by 12 bars representing the control probes (contr.). The ABCC1 control probe is indicated with an asterisk. All ratios for the control samples are ∼1, indicating that there is no deletion or duplication present. In the father (I-1) all ratio’s also equal 1 indicating no deletion is present; in the proband (II-1), all ABCC6 probes and the ABCC1 probe have a ratio ∼0.5 confirming the presence of a heterozygous ABCC6 whole gene deletion. The deletion of ABCC1 is commonly associated with ABCC6 deletions and confirms that also exon 1 of ABCC6 - for which no probe is present in the MLPA kit - is deleted; the ABCC1 deletion however has no phenotypic consequences as was previously described [17]. Panel B: proband 2. (a) Papular skin lesions in the lateral neck at the time of diagnosis; (b,c) White light funduscopy of the proband. Peau d’orange (asterisk) and angioid streaks (closed arrows) are shown. (d) Pedigree of the family of proband 2 (II-1, arrowed). The ABCC6 genotype is indicated for each individual (WT = wild type). (e) Electropherograms of the ABCC6 direct sequencing results for individuals I-1, I-2, II-1 and II-2. The location of the affected nucleotide at position c.3421 is arrowed. (f) MLPA results of the proband (II-1), her father (I-1) and her brother (II-2). The ratio results of the probe pair PCR products are indicated by the dots; black dots: the ratio lies within the 95% confidence interval (CI) of the reference sample population indicating the absence of a deletion or duplication; red: the ratio lies out of the 95% CI and over the arbitrary borders (0.7 to 1.3 by default) indicating the presence of a deletion or duplication. Whiskers: 95% CI for sample value (test or reference). Boxes: 95% CI in reference sample population (by default). Blue: compared to test probes, Green: compared to reference probes. From left to right: ratio for the TSC2 probe, ABCC1 probe, ABCC6 exon 30, exon 28–21, exon 18–17, exon 15–7, exon 5, exon 4, exon 2, followed by 10 bars representing the control probes (reference.). All ratios for the negative control sample are ∼1, indicating that there is no deletion or duplication. In the father (I-1) all ratio’s also equal 1 indicating no deletion is present; in the brother (II-2) and the proband (II-1), all ABCC6 probes and the ABCC1 probe again have a ratio ∼0.5, confirming the presence of a heterozygous ABCC6 whole gene deletion. “

The following references were added to the reference list of the revised manuscript: 

15. Stuppia L, Antonucci I, Palka G, Gatta V. Use of the MLPA assay in the molecular diagnosis of gene copy number alterations in human genetic diseases. Int J Mol Sci. 2012, 13, 3245-3276. doi: 10.3390/ijms13033245. 

17. Costrop LM, Vanakker O, Van Laer L, Le Saux O, Martin L, Chassaing N, et al. Novel deletions causing pseudoxanthoma elasticum underscore the genomic instability of the ABCC6 region. J Hum Genet. 2010, 55, 112-117. doi: 10.1038/jhg.2009.132. 

 

2. (Results) Since the fathers of each proband apparently did not submit sperm (or other?) DNA samples, the authors must show or describe the procedure and data used to state that the father is indeed the father of the affected children. The claim of paternity in the text is not sufficient. In the second case, even though the brother has the same deletion, it remains formally possibility that a different father could have contributed this deleted allele of ABCC6 to both of these children and therefore it would not necessarily be a case of germline mosaicism. In both cases, there must be genetic evidence of paternity shown and described.

The author is correct that sperm samples of the respective fathers could not be obtained, one because of a vasectomy which would require a testicular punction to possibly obtain sperm which was considered as unethical because of the potential complications of the procedure; the other father did not give consent for a sperm sample. Obviously, both fathers did provide other DNA samples, c.q. an EDTA blood sample for genomic DNA extraction and a buccal swab as otherwise the DNA sequencing and MLPA results shown for both fathers in Figure 2 could not have been obtained nor would paternity testing have been possible. This is also mentioned in the results section of the manuscript: 

Family 1: (…) “Segregation analysis confirmed the mother to carry the c.3506+2T>C variant, while the second variant was not identified in a blood sample nor in buccal cells of the father (Figure 2A,f).” 

Family 2: (…) “ABCC6 analysis showed the proband to be compound heterozygous for the p.R1141* pathogenic variant (detected via direct sequencing), inherited from the mother, and a whole ABCC6 deletion (demonstrated by MLPA), which was not found in blood and buccal cells of either parent (Figure 2B,d-f).“

Paternity was confirmed in both families, as stated in the methods and results of the original manuscript. In the methodology section of the revised version, we have added further details on the method of paternity testing which is also used in diagnostic and legal paternity testing in Belgium. This paragraph now reads as: 

“Paternity testing. Paternity testing was performed using the PowerPlex 16 system (Promega, Madison, USA) kit according the manufacturer’s guidelines, based on the analysis of Short Tandem Repeats (STR), short tenderly repeated DAN sequences that involve a repetitive unit of 2 to 7 basepairs. The PowerPlex 16 System is a multiplex STR system that  allows co-amplification and three-color detection of sixteen loci (fifteen STR loci and Amelogenin to determine the sex): Penta E, D18S51, D21S11, TH01, D3S1358, FGA, TPOX, D8S1179, vWA, Amelogenin, Penta D, CSF1PO, D16S539, D7S820, D13S317 and D5S818. These STR are very polymorphic making them ideal markers for paternity testing, allowing to achieve a probability of (non-)paternity > 99,99%. One primer for each of the Penta E, D18S51, D21S11, TH01 and D3S1358 loci is labeled with fluorescein (FL); one primer for each of the FGA, TPOX, D8S1179, vWA and Amelogenin loci is labeled with carboxy-tetramethylrhodamine (TMR); and one primer for each of the Penta D, CSF1PO, D16S539, D7S820, D13S317 and D5S818 loci is labeled with 6-carboxy-4´,5´-dichloro-2´,7´-dimethoxy-fluorescein (JOE). All sixteen loci are amplified simultaneously in a single tube and analyzed in a single injection or gel lane. The 15 autosomal STR loci were amplified in a GeneAmp PCR System 9700 thermocycler (Applied Biosystems, Foster City, CA, USA). Separation and detection of PCR products were carried out with an an Applied Biosystems 3730xl Sequencer (Applied Biosystems, Foster City, CA, USA) and genotyping was performed by comparison with the allelic ladder included in the kit, using GeneScan 2.1 software (Applied Biosystems, Foster City, CA, USA). Statistical analysis, including the calculation of the paternity index and determination of the probability of paternity was done using the analytics described by Gjertson and Brenner (most recent version on www.dna-view.com) [16]. 

Further, we have added the results of the paternity test in both families in a Supplementary Data file, as Table S2 and Table S3.   

The following references were added to the reference list of the revised manuscript: 

16. Gjertson DW, Brenner CH, Baur MP, Carracedo A, Guidet F, Luque JA, et al. Recommendations on biostatistics in paternity testing. Forensic Sci Int Genet. 2007, 1, 223-231. doi: 10.1016/j.fsigen.2007.06.006. 

 

3. (Figure 2) Figure 2 needs to be enlarged and the image quality improved, particularly the size and quality of the electropherograms and MLPA results. Without this information being legible, the entire manuscript is compromised. The relevant mutation in the mother of the first family is difficult to see on the electropherogram; one can only see the wild type “T” at the splice site, Also, the relevant mutation in the proband (II-1) in the second family should be labeled “T” instead of what appears to be C (the wild type sequence). The lack of consistency in how the electropherograms are presented between the two families also takes away from the manuscript.

We understand the comment of the reviewer and have re-organized the original Figure 2 so that the genetic results - electropherograms and MLPA results - are enlarged and better visible.

Regarding the electropherogram of proband II-1, this electropherogram does show a T in the proband (both in letter and in color of the peak). The wild type sequence is noted above the electropherogram, while the patient DNA sequence is noted below. We have added this to Figure 2 for clarity. 

The author is correct that there is a graphic difference in the electropherograms in both families; this is because they are from different time periods and hence different software; the electropherograms of family 1 also needed to be scanned because they were only available on paper. We find it however important to show the original data and not to redesign the electropherograms with graphical editing programs to make them look more alike, as this does not add to the findings. We do not agree with the reviewer that this takes away from the manuscript or its key messages. On the contrary, it confirms the authenticity of the data.

 

4. (literature review) This section is confusing as written and should be clarified and expanded. The first sentence should specify that only one patient or proband is the subject in each study (except the current one). The last sentence of the paragraph should be moved to after the first sentence, since it applies to the entire table and the 14 families. Then, clearly discuss the specific instance of PXE, stating that “among all reported (provide the number) cases of PXE, only one case (other than the current report) was suspected gonadal mosaicism.” The authors could further expand this section and their description of table 1.

This section has been revised and now reads as: 

“Using a systematic literature review following the PRISMA-P guidelines, we identified 10 autosomal recessive genodermatoses where a suspected gonadal mosaicism was reported (Table 1). These include disorders with a predominant skin phenotype such as cutis laxa or epidermolysis bullosa as well as multi systemic diseases such as Cockayne syndrome, Vici syndrome and PXE.

Almost all reports describe a single patient with suspected germline mosaicism, except for a report on epidermolysis bullosa with late-onset muscular dystrophy due to PLEC1 pathogenic variants that reports 2 patients and the current report on PXE [23]. The phenotypes that were reported in the probands did  not differ in symptoms or severity from other patients with the same disease. The parents with the suspected germline mosaicism did not present any symptoms as expected.

At the molecular level, 9 of the pathogenic variants suspected to be only present in the germline of a parent were single nucleotide variants, mostly nonsense variants. Though most of these variants were unique, a recurrent COL7A1 variant p.R1933* was noted in dystrophic EB patients, which was suggested to be a hotspot variant as it occurred in a CpG dinucleotide [20,21]. The other variants were CNVs, mostly whole gene deletions of ERCC6, GORAB and ABCC6 respectively, and one single exon duplication (in the EPG5 gene) [18,19,29,30].  

All but three cases were found to be due to paternal germline mosaicism. This assumption was usually based on the fact that one of the pathogenic variants of the proband was not found in the father and after paternity was confirmed. In only one of the reported families, genetic testing was performed on actual gonadal cells [20].

Specifically for PXE, after reviewing all previously reported PXE patients and families, only one family was found with similar segregation data. The proband was an 11 year old boy who presented with typical papular skin lesions with middermal elastic fiber calcification and fragmentation on skin histology; ophthalmological symptoms were not yet present, which is not unexpected at this young age. He was compound heterozygote for p.R1164Q, which was inherited from the father and the p.R518* nonsense variant which was not found in both parents and suggested to be either a de novo mutation or reflecting germline mosaicism in the clinically unaffected mother. Similar to the two families in the current report, the clinical presentation or natural history was no different from other patients and families with PXE. There was no indication of a specific genotype-phenotype correlation [29].”

Minor comments

1. (p. 2) The sentence including the primary symptoms of PXE is too long and convoluted. This should be  reworded and/or broken up into multiple sentences.

This sentence has been revised; it now reads as: 

(…) “Among autosomal recessive genodermatoses, pseudoxanthoma elasticum (PXE; OMIM# 264800) is considered a paradigm disorder, in which fragmentation of mineralized elastic fibers results in dermal, ocular and cardiovascular symptoms. Cutaneous lesions are typically seen in the flexural areas and include (plaques of)  papular lesions and increased skin laxity with excessive skin folds. Ocular symptoms entail in fundo abnormalities such as peau d’orange, comet tails and angioid streaks. The latter may lead to subretinal neovascularization and hemorrhage with subsequent vision loss. Cardiovascular symptoms arise due to media calcifications in midsized arteries and include peripheral artery disease and stroke [5-12]. ” 

 

2. (Methods) Primer sequences used for PCR and sequencing of the ABCC6, GGCX, and ENPP1 genes should be included here.

The original manuscript included the statement that ‘primer sequences are available upon request’. As requested by the reviewer we have added the primer sequences in the Supplementary Data in Table S1, to which we refer in the revised methods section; the former sentence was omitted from the methods section, which now reads as: 

“Molecular analysis of the ABCC6, ENPP1 and GGCX gene. Genomic DNA was isolated from whole blood (QIAamp blood kit, Qiagen®, Hilden, Germany) and the coding regions of the ABCC6, GGCX and ENPP1 gene were amplified using an established protocol. Primer sequences are listed in Supplementary Table S1. Direct sequencing was performed using an Applied Biosystems 3730xl Sequencer®, with ABI PRISM BigDye Terminator Cycle Sequencing Kit (Applied Biosystems®, Foster City, USA). Nucleotide numbers are derived from cDNA ABCC6 sequences (GenBank accession no. NM_001171).” 

 

3. (Methods) The authors state “Nucleotide numbers are derived from gDNA ABCC6 sequences (GenBank accession no. NM_001171).” Since this accession number provides the cDNA/mRNA sequence and excludes other genomic sequences, the gDNA term must be changed to cDNA.

This typo has been corrected in the revised version of the manuscript.

 

4. (Discussion) I believe that increasing paternal age increases the frequency of germline mutations of all types (not only SNVs) being present in the offspring, as opposed to what the authors state. This point needs to be clarified, as it is relevant to the frequency of de novo mutations in the offspring. 

We do not completely agree with the reviewer. The declining integrity of germ cells with increased age is suggested as a potential source of CNVs in the offspring, but the association between advancing paternal age and disease risk is not always consistent or reproducible. A 2020 study (Wadhawan et al.) investigated the association of CNVs and paternal age and could not find solid evidence for an association. We acknowledge that, even though the analysis was done on a large sample size, this study still has limitations. In this respect we have modified this paragraph in the discussion, adding also additional references. This paragraph now reads as: 

(…) “Interestingly, in eleven families the variants were due to paternal germline mosaicism. Indeed it is known that most new pathogenic variants are observed in fathers and increasing paternal age positively correlates with the risk of new single nucleotide variants [31]. In both our families the variant was however a copy number variant (CNV). Though many factors can contribute to the formation of CNVs, they usually occur due to non-allelic homologous recombination between identical sequences of repeated DNA, which can occur during meiosis [32]. Indeed, several types of repeats (such as long and short interspersed nuclear elements and Alu repeats) are abundantly present in the intra- and extragenic region of ABCC6, making the gene more prone for intragenic and whole gene deletions [33,34]. It has been shown that these structural variations are more present on paternal chromosomes, emphasizing the contribution of the paternal germline to structural variation, though a link with paternal age has not always been consistent [35-38]. Indeed, in our report the first proband’s father was 42 years old at conception while the second was only 30.” 

The following references were added to the reference list of the revised manuscript:

32. Gu W, Zhang F, Lupski JR. Mechanisms for human genomic rearrangements. Pathogenetics. 2008, 1, 4. doi: 10.1186/1755-8417-1-4.

33. Chassaing N, Martin L, Bourthoumieu S, Calvas P, Hovnanian A. Contribution of ABCC6 genomic rearrangements to the diagnosis of pseudoxanthoma elasticum in French patients. Hum Mutat. 2007, 28, 1046. doi: 10.1002/humu.9509.

34. Le Saux O, Beck K, Sachsinger C, Silvestri C, Treiber C, Göring HH, et al. A spectrum of ABCC6 mutations is responsible for pseudoxanthoma elasticum. Am J Hum Genet. 2001, 69, 749-764. doi: 10.1086/323704.

37. Wadhawan I, Hai Y, Foyouzi Yousefi N, Guo X, Graham JM Jr, Rosenfeld JA. De novo copy number variants and parental age: Is there an association? Eur J Med Genet. 2020, 63, 103829. doi: 10.1016/j.ejmg.2019.103829.

38. Hehir-Kwa JY, Rodríguez-Santiago B, Vissers LE, de Leeuw N, Pfundt R, Buitelaar JK, et al. De novo copy number variants associated with intellectual disability have a paternal origin and age bias. J Med Genet. 2011, 48, 776-778. doi: 10.1136/jmedgenet-2011-100147. 

 

5. (Discussion) The use of the term “non-homologous recombination” probably is incorrect in that all recombination that occurs during meiosis is homologous or at least homeologous. If there are homologous or homeologous sequences in the broad region surrounding ABCC6, this can promote misalignment during meiosis and deletion of the intervening region in one of the resulting germ cells.

We agree with the reviewer and have rephrased the sentence using the term ‘non-allelic homologous recombination’. 

Reviewer 2 Report

Comments and Suggestions for Authors

The article entitled “Gonadal mosaicism as a rare inheritance pattern in recessive genodermatoses: report of two cases with pseudoxanthoma elasticum and literaure review” focuses on two cases characaterized by germline mosaicism in PXE. This information is associated with literature review. In the scientific literature is reported that PXE can be compliated by vascular calcification that causes  arteriosclerosis. I think that the authors should say a few words on this topic and add updated citations regarding this important clinical topic (for example Gilles Kauffenstein et al, Biology 2024; Kristina Pfau et al, Progress in Retinal and Eye Research 2024; etc)) that may be of interest to the reader. Therefore, I think that this article is not suitable for publication in its current version.

Author Response

1. The article entitled “Gonadal mosaicism as a rare inheritance pattern in recessive genodermatoses: report of two cases with pseudoxanthoma elasticum and literaure review” focuses on two cases characaterized by germline mosaicism in PXE. This information is associated with literature review. In the scientific literature is reported that PXE can be compliated by vascular calcification that causes  arteriosclerosis. I think that the authors should say a few words on this topic and add updated citations regarding this important clinical topic (for example Gilles Kauffenstein et al, Biology 2024; Kristina Pfau et al, Progress in Retinal and Eye Research 2024; etc)) that may be of interest to the reader. Therefore, I think that this article is not suitable for publication in its current version.

We thank the reviewer for his comment; however, we fear that the reviewer may have misinterpreted the title and the scope of the paper. The term ‘literature review’ in the titel refers to 'gonadal mosaicism as a rare inheritance pattern in recessive genodermatoses’ and not to PXE itself. As the reviewer can appreciate in the revised results paragraphs of this literature review, we have overviewed all reported cases of gonadal mosaicism in autosomal recessive genodermatoses. 

As suggested by the reviewer, we have clarified the occurrence of the cardiovascular symptoms due to vascular calcification in PXE in the revised introduction. However, the scope of  this paper is neither to provide a review of PXE in general nor of its cardiovascular phenotype specifically. To guide the readers to more extensive information about PXE, its complex phenotype, pathophysiology, (differential) diagnosis and management, we have added some additional references of excellent reviews on this topic:

“Among autosomal recessive genodermatoses, pseudoxanthoma elasticum (PXE; OMIM# 264800) is considered a paradigm disorder, in which fragmentation of mineralized elastic fibers results in dermal, ocular and cardiovascular symptoms. Cutaneous lesions are typically seen in the flexural areas and include (plaques of)  papular lesions and increased skin laxity with excessive skin folds. Ocular symptoms entail in fundo abnormalities such as peau d’orange, comet tails and angioid streaks. The latter may lead to subretinal neovascularization and hemorrhage with subsequent vision loss. Cardiovascular symptoms arise due to media calcifications in midsized arteries and include peripheral artery disease and stroke [5-12].”

 

The following references were added to the reference list of the revised manuscript: 

8. Pfau K, Lengyel I, Ossewaarde-van Norel J, van Leeuwen R, Risseeuw S, Leftheriotis G, et al. Pseudoxanthoma elasticum - Genetics, pathophysiology, and clinical presentation. Prog Retin Eye Res. 2024, 102, 101274. doi: 10.1016/j.preteyeres.2024.101274.

9. Ghaoui N, Abou-Rahal J, Nasser N, Kurban M, Abbas O. Pseudoxanthoma Elasticum-Like Changes:Associations- and Underlying Mechanisms. Skinmed. 2024, 22, 172-177.

10. Stumpf MJ, Schahab N, Nickenig G, Skowasch D, Schaefer CA. Therapy of Pseudoxanthoma Elasticum: Current Knowledge and Future Perspectives. Biomedicines. 2021, 9, 1895. doi: 10.3390/biomedicines9121895.

11. Verschuere S, Van Gils M, Nollet L, Vanakker OM. From membrane to mineralization: the curious case of the ABCC6 transporter. FEBS Lett. 2020, 594, 4109-4133. doi: 10.1002/1873-3468.13981.

12. Terry SF, Uitto J. Pseudoxanthoma Elasticum. In: Adam MP, Feldman J, Mirzaa GM, Pagon RA, Wallace SE, Bean LJH, Gripp KW, Amemiya A, editors. GeneReviews®. Seattle (WA): University of Washington, Seattle. 2020 Jun 4.

Reviewer 3 Report

Comments and Suggestions for Authors

The report discussed cases of germline mosaicism in autosomal recessive disorders, specifically focusing on two families with pseudoxanthoma elasticum (PXE) where paternal germline mosaicism for an ABCC6 whole gene deletion was identified. The authors highlight the clinical challenges associated with PXE, particularly in cases where a heterozygous carrier parent exhibits typical PXE retinopathy. The article also concludes that while germline mosaicism is rare, its implications for diagnosis and patient counseling are significant, necessitating further research and awareness in clinical practice. Here are a few points that need to be further addressed for the findings.

1.       Given that the frequency of the germline mosaicism, what factors did the authors thin that can contribute the rarity in patients with PXE compared with other autosomal recessive disorders?  What are the proposed molecular mechanisms underlying germline mosaicism in PXE?

2.       How does the presence of germline mosaicism affect the phenotypic variability observed in first-degree relatives of PXE patients? Are there specific features that are more commonly associated with mosaicism? Did any segregation analysis perform in these reported cases?

3.       The authors should include a session discussing on the recommendations provided for the genetic counseling in families with identified germline mosaicism. What strategies do the authors think that can effectively communicate the risks to families?

4.       How do the findings related to germline mosaicism in PXE compare with those in other autosomal recessive genodermatoses, e.g. dystrophic epidermolysis bullosa or neurofibromatosis type 1?

5.       What are the long-term outcomes for individuals with PXE who have parents with germline mosaicism? Is there evidence to suggest differences in disease progression or severity?

Author Response

1.       Given that the frequency of the germline mosaicism, what factors did the authors thin that can contribute the rarity in patients with PXE compared with other autosomal recessive disorders?  What are the proposed molecular mechanisms underlying germline mosaicism in PXE?

There is currently no evidence that germline mosaicism in PXE is more frequent or more rare compared to other autosomal recessive genodermatoses; indeed, our literature review provides a variety of autosomal recessive genodermatoses for whom germline mosaicism has been reported, but for each disease the number of reported cases is limited. If anything, we suggest that germline mosaicism may be underreported because in many patients - both those that are reported in literature as those who are seen in the clinics on a daily basis - segregation analyses of the identified pathogenic variants is not done. This is also stated in the discussion of the manuscript. 

Regarding the mechanism of germline mosaicism, it seems reasonable to assume that this will be identical to what is most commonly shown to lead to mutations in the germline, namely non-allelic homologous recombination. This has been detailed in the discussion which now reads as: 

“Though many factors can contribute to the formation of CNVs, they usually occur due to non-allelic homologous recombination between identical sequences of repeated DNA, which can occur during meiosis [32]. Indeed, several types of repeats (such as long and short interspersed nuclear elements and Alu repeats) are abundantly present in the intra- and extragenic region of ABCC6, making the gene more prone for intragenic and whole gene deletions [33,34]. It has been shown that these structural variations are more present on paternal chromosomes, emphasizing the contribution of the paternal germline to structural variation, though a link with paternal age has not always been consistent [35-38]. Indeed, in our report the first proband’s father was 42 years old at conception while the second was only 30. “ 

The following references were added to the reference list of the revised manuscript: 

33. Chassaing N, Martin L, Bourthoumieu S, Calvas P, Hovnanian A. Contribution of ABCC6 genomic rearrangements to the diagnosis of pseudoxanthoma elasticum in French patients. Hum Mutat. 2007, 28, 1046. doi: 10.1002/humu.9509.

34. Le Saux O, Beck K, Sachsinger C, Silvestri C, Treiber C, Göring HH, et al. A spectrum of ABCC6 mutations is responsible for pseudoxanthoma elasticum. Am J Hum Genet. 2001, 69, 749-764. doi: 10.1086/323704.

 

2.       How does the presence of germline mosaicism affect the phenotypic variability observed in first-degree relatives of PXE patients? Are there specific features that are more commonly associated with mosaicism? Did any segregation analysis perform in these reported cases?

When a pathogenic variant is only present in the germline (i.c. sperm cells or egg cells), this usually does not cause any symptoms as the variant is not present in any other body cells (contrary to somatic mosaicism, where a variant is present in a percentage of the cells of the whole body). As stated in the presented cases, the fathers who had a germline mosaicism for the ABCC6 variant did not present any symptoms. As a result, germline mosaics cannot account for any degree of variability seen in PXE.

Regarding segregation analysis, it should be noticed that a germline mosaicism always occurs de novo in the individual where it is demonstrated or suspected and that no other family members can inherit the same germline mosaicism. Indeed, when a variant that is present only in the germline of an individual is passed on to the next generation, it will by definition be present in all somatic cells. 

 

3.       The authors should include a session discussing on the recommendations provided for the genetic counseling in families with identified germline mosaicism. What strategies do the authors think that can effectively communicate the risks to families?

We have added on the importance of genetic counseling, the associated risks and the impact for a couple in the decision regarding whether or not to choose for prenatal testing in the discussion section of the revised manuscript. This paragraph now reads as: 

“ The reported cases of suspected germline mosaicism in recessive genodermatoses emphasize the importance of variant verification in the parents and sibs of a newly diagnosed patient to enable accurate preconceptional counseling. Even though difficult to estimate, in cases of germline mosaicism recurrence risks would be much lower than the traditional 25% recurrence risk for autosomal recessive disorders. Indeed, usually the recurrence risk for any apparent de novo pathogenic variant is considered 1%, though this may be an overestimation. Irrespective of the exact percentage it implies that for a couple that has a first child with an autosomal recessive disease due to bi-allelic pathogenic variants, in which one of the partners is a heterozygous carrier and the other has (suspected) germline mosaicism, the recurrence risk of the same disease in future children would be low. This information may influence the choices couples make regarding whether or not to perform prenatal testing or pre-implantation genetic testing. In this regard, it should be noted that for a considerable number of probands, no segregation data were mentioned in the respective reports or case series. This may suggest that such events could occur more often than currently recognized. It also highlights the importance of profession genetic counseling to assure proper assessment of all aspects of a genetic result.”

4.       How do the findings related to germline mosaicism in PXE compare with those in other autosomal recessive genodermatoses, e.g. dystrophic epidermolysis bullosa or neurofibromatosis type 1?

As detailed in table 1, most of the cases with parental germline mosaicism are paternal in origin; however, a previously described family with PXE and suspected germline mosaicism is an exception to this rule. Apart from this, the molecular and clinical characteristics are very alike. We have described this in more detail in the revised version of the results section (Literature review) of the manuscript, which now reads as: 

“Using a systematic literature review following the PRISMA-P guidelines, we identified 10 autosomal recessive genodermatoses where a suspected gonadal mosaicism was reported (Table 1). These include disorders with a predominant skin phenotype such as cutis laxa or epidermolysis bullosa as well as multi systemic diseases such as Cockayne syndrome, Vici syndrome and PXE.

Almost all reports describe a single patient with suspected germline mosaicism, except for a report on epidermolysis bullosa with late-onset muscular dystrophy due to PLEC1 pathogenic variants that reports 2 patients and the current report on PXE [23]. The phenotypes that were reported in the probands did  not differ in symptoms or severity from other patients with the same disease. The parents with the suspected germline mosaicism did not present any symptoms as expected.

At the molecular level, 9 of the pathogenic variants suspected to be only present in the germline of a parent were single nucleotide variants, mostly nonsense variants. Though most of these variants were unique, a recurrent COL7A1 variant p.R1933* was noted in dystrophic EB patients, which was suggested to be a hotspot variant as it occurred in a CpG dinucleotide [20,21]. The other variants were CNVs, mostly whole gene deletions of ERCC6, GORAB and ABCC6 respectively, and one single exon duplication (in the EPG5 gene) [18,19,29,30].  

All but three cases were found to be due to paternal germline mosaicism. This assumption was usually based on the fact that one of the pathogenic variants of the proband was not found in the father and after paternity was confirmed. In only one of the reported families, genetic testing was performed on actual gonadal cells [20].

Specifically for PXE, after reviewing all previously reported PXE patients and families, only one family was found with similar segregation data. The proband was an 11 year old boy who presented with typical papular skin lesions with middermal elastic fiber calcification and fragmentation on skin histology; ophthalmological symptoms were not yet present, which is not unexpected at this young age. He was compound heterozygote for p.R1164Q, which was inherited from the father and the p.R518* nonsense variant which was not found in both parents and suggested to be either a de novo mutation or reflecting germline mosaicism in the clinically unaffected mother. Similar to the two families in the current report, the clinical presentation or natural history was no different from other patients and families with PXE. There was no indication of a specific genotype-phenotype correlation [29].”

5.       What are the long-term outcomes for individuals with PXE who have parents with germline mosaicism? Is there evidence to suggest differences in disease progression or severity?

In the probands with PXE, who have a parent with a germline mosaic pathogenic variant, the two pathogenic ABCC6 variants will be present somatically - as in all PXE patients and similar to the probands whose parents are traditional heterozygous carriers. Hence for these probands with PXE there is no reason why there would be any difference in the disease characteristics, progression or severity. In them, PXE is equally variable and unpredictable as in any other PXE patient. There is also no mechanistic or molecular reason why the long-term outcome of these patients would be any different.