A new XPC gene splicing mutation has lead to the highest worldwide prevalence of xeroderma pigmentosum in black Mahori patients

DNA Repair 10 (2011) 577–585

Contents lists available at ScienceDirect

DNA Repair journal homepage: www.elsevier.com/locate/dnarepair

A new XPC gene splicing mutation has lead to the highest worldwide prevalence of xeroderma pigmentosum in black Mahori patients Franc¸ois Cartault a,∗ , Caroline Nava a , Anne-Claire Malbrunot a , Patrick Munier a , Jean-Christophe Hebert b , Patrick N’guyen a , Nadia Djeridi a , Philippe Pariaud c , Joelle Pariaud c , Aurélie Dupuy d , Frédéric Austerlitz e , Alain Sarasin d,f,∗∗ a

Department of Medical Genetics, INSERM U781 CHR Félix Guyon, La Réunion, and CHU Necker, Paris, France Department of Pediatrics, Hospital of Mamoudzou, Mayotte, France c Department of ORL, Hospital of Mamoudzou, Mayotte, France d UMR8200 CNRS, University Paris-Sud and Institut Gustave Roussy, Villejuif, France e Unité Eco-Anthropologie, UMR 7206 CNRS-Muséum d’Histoire naturelle, Université Paris Diderot, 57, rue Cuvier, 75231 Paris, France f Department of Genetics, Institut Gustave Roussy, France b

a r t i c l e

i n f o

Article history: Received 3 February 2011 Received in revised form 3 March 2011 Accepted 8 March 2011 Available online 8 April 2011 Keywords: Xeroderma pigmentosum Black skin XPC Splicing mutation Skin cancer Ocular injuries

a b s t r a c t Xeroderma pigmentosum (XP) is a rare, recessive disease characterized by sunlight hypersensitivity and early appearance of cutaneous and ocular malignancies. We report the first description of a very high incidence (around 1/5000) of black XP patients in the Mayotte population in the Indian Ocean. Among a cohort of 32 XP, we describe the clinical and genetic features of 18 living Comorian black XP patients. We discuss the remarkable clinical differences between white and black XPs. Skin and ocular abnormalities are remarkably precocious and severe XP phenotypes are recognized by the early ocular injuries. In our cohort, the first skin cancer appeared at a median age of 4.5 years with no neurological symptoms. PostUV DNA repair, cell survival and genetic complementation assigned these patients to the XP group C. All patients exhibited a new G → C homozygous substitution at 3 -end of XPC intron 12 (IVS 12-1G > C) leading to the abolition of an acceptor splicing site and the absence of the XPC protein. We found 3 different mRNA isoforms: one with retention of intron 12, one showing exon 13 skipping, and a third with a 44 bp deletion in exon 13. These 3 isoforms were differently expressed in XP-C cells compared to normal cells. This new mutation found in the Comorian islands, where consanguinity is frequent, represents a founder effect, with an estimated age of about 770 years. Due to the African origin of the black XPs from Mayotte, it would be valuable to search for this mutation in African XPs whenever possible. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Xeroderma pigmentosum (XP) is a rare, recessive disease characterized by sunlight hypersensitivity and high incidence of skin cancer, principally carcinoma and melanoma [1,2]. White XP patients have a >1000 fold increased skin cancer risk on UVexposed sites with a median age for the first cancer at 8 years. The risk factor for melanoma is also very high [3,4].

∗ Corresponding author at: Department of Medical Genetics, CHR de La Réunion, Site Félix Guyon, 97400 Saint Denis, France. Tel.: +33 2 6290 6400; fax: +33 2 6290 6405. ∗∗ Corresponding author at: UMR8200, PR2, Institut Gustave Roussy, 94805 Villejuif, France. Tel: +33 1 42 11 63 28; fax: +33 1 42 11 50 08. E-mail addresses: [email protected] (F. Cartault), [email protected] (A. Sarasin). 1568-7864/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.dnarep.2011.03.005

Classical XP patients have a defect in one of the seven XP genes, XPA to XPG. These genes belong to the nucleotide excision repair (NER) pathway, involved in the removal of bulky DNA lesions such as UV-induced damage. XP is ubiquitous in the world but its frequency and the genes involved vary between countries. The incidence is estimated at 1/1,000,000 in the United States and Europe [5], while it is much higher in Japan (1/100,000), North Africa and the Middle East (around 1/50,000) where consanguinity is common. XP-C (MIM ID #278720) patients are the most common XP worldwide (25–40% of all XP cases). XP-C patients exhibit predominantly skin damage and early malignancies without neurological deterioration. The XPC gene (MIM ID*613208) is located on chromosome 3p25, spans 33 kb and contains 16 exons (82–882 bp). The 940 amino acid XPC protein complexed with HHR23B acts during the initial step of NER in lesion recognition. The XPC–HHR23B complex and DDB1/2 complex are involved in genome overall repair. XP seems to be rare in black individuals and to our knowledge neither the frequency, nor the genes involved have been reported

578

F. Cartault et al. / DNA Repair 10 (2011) 577–585

Fig. 1. Clinical features of Mahori XP-C patients from 6 to 32 years old (at the time of the pictures) showing skin damage, depigmentation, ocular injuries, running nose and tumor sites, including the tip of the tongue. Pictures are shown with authorization.

for sub-Saharan African patients. Mayotte, a French community of 185,000 individuals located in the Indian Ocean, is part of the four Comorian islands with about 800,000 individuals. Thirty-two XP black Mahori patients have been referred to us since 1995. These patients are only coming from Mayotte placing this island among the regions with the highest XP incidence worldwide. Indeed, the number of patients is probably much higher than that because Mayotte is just one of the four Comorian islands where individuals have the same African origin with similar consanguinity and the same life style. The aim of this study was to accurately characterize, from 2007 to 2010, the Mahori black XP phenotype, particularly the early symptoms, for detecting patients as early as possible to allow efficient photoprotection. We have characterized the cell sensitivity to UV and identified the gene and the mutation involved in this community. We have also compared white and black XPC patients revealing major differences in the type and location of tumors and suggesting different approaches for early diagnosis and treatment. 2. Materials and methods 2.1. Patients Thirty-two patients with clinical diagnosis of XP were registered at the hospital of Mamoudzou on the island of Mayotte since 1995. We followed, from 2007 to 2010, 22 patients because 10 deceased before the end of this study. Clinical analysis (Table 1) was carried out on 18 living patients from 13 different families (2

siblings from family A, C and J, 3 from family E) with a clinical diagnosis of XP. Inclusion criteria were: Comorian origin, black skin, association of skin injuries with abnormal pigmentation before 4 years, ocular abnormalities before 2 years, skin cancer on UVexposed body sites before 15 years and/or sibling(s) suffering from XP (Fig. 1). Besides data indicated in Table 1, we have also collected familial medical history, height, weight, head circumference, audiometer and oral mucosa examination. For molecular analysis, 22 DNA samples from 17 different families were used: samples from the 18 living patients and DNA of 4 deceased XPs, whose DNA had been preserved. The informed signed consent for genetic investigation was obtained from all patients or their parents. This study was approved by the Ethics Committee from the CPP Bordeaux (France).

2.2. Cellular analysis Primary fibroblasts from unexposed skin from 3 unrelated XP patients and 2 unrelated XP heterozygous siblings were cultivated in MEM plus 10% FCS and analyzed at low passages. Normal (MRC5SV) and XP-C (XP4PA-SV) SV40-transformed fibroblasts [6] were used as controls. DNA repair was characterized by post-UV unscheduled DNA synthesis [7] and cell survival using the luminescence ATP detection assay (ATPlite-1step) as described by the manufacturer (Perkin Elmer, Waltham, USA). The XP genetic defect was characterized by complementation using recombinant retroviruses expressing wild-type XP genes, as published [8].

Table 1 Clinical data in the 18 XPC Mahori patients. A siblings

B

C siblings

D

E siblings

F

G

H

I

J siblings

K

L

M

Patients

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

Age at the examination Sex Consanguinity Age at the first symptom Nature of the first symptom Age at diagnosis Growth retardation Dysmorphy Skin abnormalities Erythema Lentigines Xerosis Skin atrophy Telangiectasia Hypopigmented patches Hyperpigmented patches Actinic keratosis Ocular abnormalities Visual acuity

18y F ? 7m Ocular

11y 2m M ? 1m Ocular

5y M + ? Cutaneous

6y M − 6m Cutaneous

2y 9m F − 1y Ocular

2y 9m M + 4m Ocular

5y 2m M + 7m Ocular

8y 5m F + 4m Ocular

6y M − 8m Cutaneous

5y F − 6m Ocular

5y 4m M + 8m Ocular

2y 4m M − 6m Ocular

9y M ? 9y Ocular

6y M ? 6y Ocular

1y 4m M ? 11m Ocular

1y 6m F ? 8m Ocular

2y 6m M ? 8m Ocular

12y − −

4y 11m − −

3y − −

10m − −

1y − −

28 y F + 2m Ocular/ cutaneous 12y − Brittle hair

4m − −

1y 6m − −

2y 6m − −

2y 7m 10 m − − Brittle hair ?

1y 8m − −

6m − −

? − −

? − −

11m − −

11m − −

14m − −

− + (1) (2) − (3) (3) −

− + (1) − − (3) (3) −

− + (1) (1) − (3) (1) +

− + − − − (1) (1) −

− + − − − (1) (2) −

− + (3) (1) − (1) (3) −

− + − − − (1) (2) −

− + − (1) − (2) (3) −

− + − (1) − (2) (3) −

− + (2) (1) − − (3) −

? ? ? ? ? ? ? ?

− + (1) − − (1) (1) −

− + − − − (1) (2) −

− + − − (2) (3) (3) −

− + − − (1) (2) (2) −

− + − − − (2) (2) −

− + − − − (2) (2) +

− + − − − (3) (3) −

Blind

<5/10

PL

PL

>5/10

>5/10

?

<5/10

PL

>5/10

>5/10

>5/10

?

<5/10

+ + + − +

+ + + − −

+ + − − −

+ + + − +

+ + − − −

+ + + − +

>5/10(R) blind (L) + + + − +

PL

+ + + − +

<5/10(R) blind (L) + + + − +

>5/10

Photophoby Conjunctivitis Loss of eyelash Keratitis Corneal opacity/calcification ORL abnormalities Tongue erythroplasia Mucous hypersecretion Neurological abnormalities Decreased deep tendon reflexes Microcephaly Mental retardation Cancers Age at the first cancer Number Type SCC BCC Melanoma Other tumor

+ + + + +

+ + ? ? ?

+ + − − −

+ + − − −

+ + − − +

+ + ? − −

+ + − − −

+ + − + −

+ + − − +

+ −

+ +

− −

+ −

− −

+ −

− −

+ −

− +

+ −

? +

+ +

− −

+ +

− +

− −

− −

− −

−

−

−

−

−

−

−

−

−

−

?

−

−

−

−

−

−

−

− − + ? 9

− − + 4y 11m 9

− − + 3y 3

− − −

− − + 8y >2

− − + 16y 6

− − −

− − −

− − −

− − + 3y 16

? ? −

− − + 3y 6m

− − −

− − +

− − −

− − −

− − −

− − + 2y 6m 2

+ + − Angiosarcoma

+ + − Pyogenic granuloma

+ − − Papilloma atypical fibroxanthoma keratoacanthoma

+ − − −

+ + − Xanthoma

+ − − Atypical fibroxanthoma

+ − + Atypical fibroxanthoma leiomyosarcoma hystiocytoma or melanoma juvenile xanthogranuloma pyogenic granuloma

? ? ? ?

+ − −

+ + − −

+ + − +

+ − + −

+ − − −

+ + + +

+ − + +

+ − + −

+

+

Localization Head and face Tongue Eye Lip

F. Cartault et al. / DNA Repair 10 (2011) 577–585

Family

Skin abnormalities (1) injuries <25% of sun exposure site (2) injuries between 25 and 50% of sun exposure site (3) injuries >50% of sun exposure site or non exposure site affected. PL: perception of light, R: right L: left. 579

580

F. Cartault et al. / DNA Repair 10 (2011) 577–585

Western blots of the XPC protein were performed as described [9]. Protein concentration of cell extracts was measured using the Quick Start Bradford Protein Assay (Bio-Rad, Philadelphia, USA). 80 ␮g were used for each diploid fibroblast line and 40 ␮g for the 3VI and 12VI amniocytes and the SV40-transformed cells. The rabbit polyclonal antibody directed against XPC: ab21078 and against vinculin: ab18058 from Abcam (Cambridge, UK) were used. 2.3. Molecular analysis 2.3.1. PCR and sequencing DNA samples were obtained from peripheral leucocytes. Mutation screening was performed by bi-directional sequencing of exons and their flanking intron–exon boundaries for the entire coding region of XPC. Direct sequencing of PCR products was performed using the Big Dye Terminator Cycle Sequencing Ready Reaction Kit (Applied Biosystems, Foster City, USA). Reaction products were run on a fluorescent sequencer (ABI 3100 Genetic Analyzer, Applied Biosystems). Sequences were aligned using Seqscape® analysis-software 2.6 (Applied Biosystems) and compared with the reference sequences for genomic DNA (GenBank accession number: XPC: NG 011763.1). Primers are shown in AppendixBSupplementary Table 1. 2.3.2. RNA and RT-PCR Total RNA was extracted from leucocytes using trisochloroform and isopropanol precipitation. Contaminating genomic DNA was removed by DNase treatment (37◦ , 30 min). A second RNA extraction was performed with phenol/chloroform. 0.5 ␮g RNA was reverse-transcribed using SuperScript II RT and cDNA was separated by electrophoresis in a 2% agarose gel. DNA bands were cut from the gels and DNA was extracted using the Nucleotrap Kit (MACHERY-NAGEL, Hoerdt, France). Following RT-PCR, amplification products were sequenced and aligned as previously described and compared with the reference sequences for the XPC mRNA (GenBank accession number: XPC mRNA: NM 004628). 2.3.3. Real-time RT-PCR The relative abundance of specific mRNA sequences was determined by real-time fluorescent Syber green PCR, using the LightCycler® 480-CYBR-Green I master Kit (Roche, Meylan, France). For each mRNA isoform, 2 ␮L of 10 times-diluted cDNA samples were amplified with 0.5 ␮M of each primers and 10 ␮L of Master Mix (2×) in a final volume of 20 ␮L. PCR amplification was performed in a LightCycler® 480 (Roche) with the LightCycler® 480 software-v1.2.9.11 (Roche) as follows: initial denaturation at 95 ◦ C for 5 min followed by 45 cycles (denaturation at 95 ◦ C for 30 s, annealing for 30 s at the optimal melting temperature (Tm ) of the primers pairs used (AppendixBSupplementary Table 2) and amplification at 72 ◦ C for 30 s). Finally, the specificity of amplified PCR product was assessed by performing a melting curve analysis. For each specific pairs of primers, the melting temperature used was optimized to obtain an optimal efficiency of amplification determined by performing a dilution curve according to the method developed by Pfaffl [10]. The fold change versus control for each gene was calculated using the GAPDH gene as control. 2.4. Estimating the age of the mutation We used the method developed by Austerlitz et al. [11] that provides a joint estimate of the age of the mutation, i.e., the time elapsed since the appearance of the common ancestor of the mutation carriers in the population and the growth rate of the number

of copies since this appearance. This method uses as inputs the current number of copy of the mutant allele in the population and the level of linkage disequilibrium of the disease locus with a surrounding haplotype constituted of several microsatellite markers linked to the disease locus. The number of copies of the mutant allele can be deduced from the prevalence of the recessive disease and the total effective size of the present population. We assumed here a prevalence for the mutated allele of 1/100 and a total population size of 850,000 (this is supposing that XP patients are, distributed randomly between the four Comorian islands, although we studied only the Mayotte population, which is historically part of this territory and is composed of about 185,000 individuals) yielding a number of copies of 8500. Regarding the haplotypes, we used eight microsatellite markers (D3S1597, D3S3611, D3S1263, D3S1259, D3S1554, D3S3613, D3S2338 and D3S2336) surrounding the XPC gene, spanning 21.21 cM. We genotyped 13 homozygous carriers and six parents. These six individuals could thus be phased without ambiguity. We phased the remaining individuals using Phase 2.1.1 [12,13]. Only two individuals could be phased at more than 95% level. This gave us a total of eight unambiguously phased carriers, hence 16 haplotypes that could be used for the analysis.

3. Results 3.1. Clinical description Among a series of 32 XPs, this clinical analysis included 18 Comorian XP patients, 6 girls and 12 boys, from 13 different families. The median age during examination was 7.2 years (between 1.4 and 28 years). The principal clinical features are summarized in Table 1. The median age at first symptom was 6.5 months while the median age of diagnosis was 3.4 years. In 83% of cases (15/18), the first symptoms were ocular (photophobia or conjunctivitis). Ocular lesions were found in all patients and occurred in the first year of life. All had conjunctivitis, 9/17 had corneal opacity, 8/16 had eyelid abnormalities (hyperpigmentation or eyelash loss) and 9/16 had bad visual acuity (less than 5/10), 4 of them having only light perception or were blind. ORL examination (Ear, Nose and Throat) showed tongue erythroplakia (8/17) or runny nose (6/18) (Fig. 1). All our XP patients had classic and severe skin abnormalities including xerosis, skin atrophy and abnormal pigmentation (Fig. 1) that covered more than 50% of sun-exposed areas in 53% of cases (9/17). None presented dysmorphic features or growth retardation. Neurological status and psychomotor development were normal for all patients (Table 1). Nine patients had skin cancer at the first examination. The age for the first cancer varied from 2.5 to 16 years (median = 4.5 years). The first cancer observed was squamous cell carcinoma (SCC) in 5 cases and unknown for the others. These 9 patients (aged between 2.5 and 16 years old) exhibited numerous cancers including Basal Cell Carcinoma (BCC) (3/9) and SCC (8/9). Lip and tongue cancers and ocular malignancies, uncommon in the general black population, are frequent in XP Mahori patients. Atypical neoplasia including third phalanx angiosarcoma was diagnosed at 15 years in a patient and leiomyosarcoma in another. Atypical fibroxanthoma was found in 3 patients and was multiple for one of them. This lesion is a pleiomorphic tumor that most commonly arises on sun-exposed skin of the white elderly and has been rarely reported in XP [14]. Cutaneous benign tumors were frequent and included pyogenic granuloma, kerathoacanthoma, xanthoma and juvenile xantogranuloma. Only one patient had melanoma (no. 12; Table 1), which is less frequent than in white XPs [4]. All deceased patients had developed cutaneous cancers between the age of 4 and 15 years (median = 7 years). The median age at death was 13.5 years.

F. Cartault et al. / DNA Repair 10 (2011) 577–585

581

Fig. 3. Absence of the XPC protein in Mahori XP diploid fibroblasts. Western blot of proteins have been extracted from normal (911VI, 405VI, 3VI and 12VI), XP heterozygote (XPHZ918VI and XPHZ929VI) and XP (XP844VI, XP845VI, XP936VI) diploid fibroblasts or amniocytes from Mahori patients (A) and from wild-type and XP-C SV40-transformed cells as control (B), as described in Section 2. The rabbit polyclonal ab21078 antibody raised against the N-terminus of the human XPC protein was used. “XPC” indicates the unique full length 120 kDa XPC protein and vinculin was used as a loading marker. 3VI and 12VI are normal amniocytes, not related to the Mahori population as control.

Fig. 2. XP cell responses to UV irradiation: (A) cell proliferation following UV irradiation. Wild type cells (405VI) and XP cells (XP844VI and XP936VI, patients 6 and 14 in Table 1) were irradiated with the indicated UVC doses. Cell proliferation was quantified 5–7 days following irradiation. Results are given in % of luminescence in arbitrary units (% RLU) and the SD is less than 10%. (B) DNA repair deficiency and complementation group in our XP patients UV-induced DNA repair synthesis of primary fibroblasts (UDS) is expressed as mean number of autoradiographic grains per nucleus as a function of UVC doses as already described by Sarasin et al. [7]. XP-C fibroblasts have been complemented by recombinant retroviruses expressing the wild type XPC gene (XP844VI + XPC) as already described [8]. The SD is around +/− 10%.

3.2. Cellular analysis and genetic screening of the XPC gene The cells from 2 XP patients (XP844VI and XP936VI: patients 6 and 14 in Table 1) exhibited hypersensitivity to the lethal effects of UVC compared to cells from normal individuals (Fig. 2A). The cells from 3 XP patients XP844VI, XP845VI and XP936VI (patients 6, 2 and 14 in Table 1) showed very little repair synthesis (about 15% of controls) as measured by unscheduled DNA synthesis (UDS) following UV irradiation while 2 heterozygous siblings (XPHZ918VI, brother of patient 12 and XPHZ929VI sister of patient 4) had normal UDS levels (Fig. 2B). Using the previously described assay with recombinant retroviruses expressing XP genes [8], we found that only retroviruses expressing the wild-type XPC gene were able to complement the low UDS levels of XP cells (Fig. 2B). Western blots were performed on total cell extracts using XPC specific antibodies. No XPC protein was detectable in fibroblasts from the three tested XP patients, while control cells exhibited a unique XPC band at 120 kDa as also seen in cells from heterozygous siblings (Fig. 3). A large variability in the level of XPC protein in normal cell lines is observed may be in relation with the speed of cell division. These patients belong to the XP-C group. This result is in agreement with the clinical examination: severe and early ocular and skin damage, early tumor development and normal neurology.

We sequenced the XPC gene in 22 DNA samples (18 living XPs and 4 deceased XPs with conserved DNA). We found a unique G > C homozygous substitution in the 3 end of the XPC intron 12 (IVS 12-1G > C) in all patients (Fig. 4). This mutation was searched in both parents for 4 cases and in 5 healthy siblings: all parents were heterozygous G/C, indicating the absence of a de novo mutation. Healthy siblings were homozygous G/G (3/5) or heterozygous G/C (2/5). Moreover, this substitution was not found in a series of 55 normal subjects with similar ethnic background. No SNP was found for this sequence in databases. This is in favor of a pathogenic mutation. We analyzed the known single nucleotide polymorphism (rs2279017) A > C at −6 bp at the 3 end of intron 12 inside the splice acceptor. Twenty-two patients DNA were homozygous for C while parents and siblings were homozygous C/C (5 parents, 2 siblings) or heterozygous A/C (3 parents, 3 siblings). Among 55 normal subjects from Mayotte, the C allele was found in 67% of cases and the A allele in 33%. A p-value <4 × 10−5 (Fisher two-sided) shows a very significant association between the C allele and the XPC mutation. 3.3. XPC mRNA in XPC cells Using primers located between exons 3 and 7 (AppendixBSupplementary Table 2 and Fig. 5D), we found that the amount of total XPC mRNA is 10 times higher in wild-type cells as compared to XP-C cells (p = 0.0001, Student’ test). Heterozygous cells expressed only 75% of the wild type level of mRNA (p = 0.001) (Fig. 5A). This strongly suggests that in XP-C cells the mutated mRNAs, containing premature termination codons, were subject to the nonsense-mediated decay pathway. RT-PCR products using primer pairs, exon 10 (5 ) – exon 13 (3 ) (AppendixBSupplementary Table 2 and Fig. 5D) resulted in two bands (bands a and b) on a 2% agarose gel for the XPs while control subjects and heterozygous showed only one band (Fig. 5B, band c). Sequence analysis revealed 2 XPC mRNA isoforms in patients: retention of intron 12 and deletion of 44 bases at the 5 of exon 13. These two isoforms are present in very small quantities in heterozygous and wild-type cells and are not visible on gels. Analysis of RT-PCR products using primer pairs, 5 -exon 12 and 3 -exon 16 (AppendixBSupplementary Table 2) showed a band in XP cells cor-

582

F. Cartault et al. / DNA Repair 10 (2011) 577–585

Fig. 4. Sequences of the XPC gene in black Mahori patients. Sequences of the end of intron 12 and beginning of exon 13 revealed a C homozygous mutation in all XP-C patients at the −1 bp of splice acceptor site, the normal G sequence in wild type individual (control) and the C/G heterozygous sequence in an XP-C parent (boxed sequences). The rs2279017 SNP (C/A) at −6 bp of the splice acceptor site is indicated by the arrow.

responding to a third isoform with a skipped exon 13 as shown by sequencing (Fig. 5C, band d). Samples from heterozygous parents showed two bands using these primers: one with normal sequence and one with exon 13 skipping while control subjects showed only the normal band (Fig. 5C, band c). We compared by real-time RT-PCR the relative abundance of the three most expressed corresponding mRNAs in two patients versus two healthy controls. The total mRNA was 10 times more expressed in the healthy control than in the XP cells (Fig. 5A), whereas the two other mRNAs quantified (the 44 bp deleted exon 13 and the intron 12 conserved isoforms) were, respectively 9 and 8.8 fold more highly expressed in the XP as compared to control cells. 3.4. Age and growth rate of the mutation We estimated a per generation growth rate of 1.24 (confidence interval = [1.15–1.41]) and an age of 30.9 generations (confidence interval = [24.5–39.7]), which translates into 773 years [614–945], assuming a 25-year generation time. This estimate would be 618 years [490–794] assuming a 20-year generation time or 927 years [734–1191] for a 30-year generation time [15]. 4. Discussion 4.1. XP-C in the African black population of Mayotte This study is the first clinical and molecular report of a large group of black skinned XP patients. Few African patients with XP have been reported in the literature up till now but only as small groups of patients and none of them were screened for XP

mutations, except one black XP woman (XP1MI, GM2096 cells from Coriell Institute) and the African American patient (XP25BE) already described [16], who will be discussed below. The available clinical data report that black XPs have very early clinical symptoms, 100% of black XPs exhibit ocular abnormalities and cutaneous lesions are very typical [17–22]. Similarly to these previously reported black XP patients, these Mahori patients suffered from a severe and precocious form of XP. We found severe ocular damage as being most often the first symptom of the disease as well as cancers occurring earlier than classically reported in white XP patients. The largest XP patient report concerns a group of 830 XP patients published by Kraemer et al. [4], where at least 90% of these patients were white skinned. In comparing the data from the two populations, black patients are found to have more ocular damage (100% versus 40%; p < 0.001) including ocular tumors and less cutaneous cancers (50% versus 78%; p < 0.02). This large group of 830 white XPs was, however, not genotyped and corresponded to a mixture of XP-C and other XP complementation groups. To determine the difference between white and black XP patients, we have compared the XP-C population from Mayotte with a more homogenous population of XP-C patients, originating from North Africa [23]. The comparison with the Mahori XP-C population is particularly valid because XP-C patients from North Africa are considered as Caucasians with a phototype of 3–4 and live also in high sun-exposure countries. In both groups, a very high consanguinity rate is observed and a founder effect is also linked to their unique mutation type, leading to the same absence of XPC protein. If we compare clinical data between the two populations: ocular abnormalities are very high in both populations but occur earlier in the black XPC

F. Cartault et al. / DNA Repair 10 (2011) 577–585

583

Fig. 5. Analysis of the XPC mRNA isoforms in XP-C, heterozygous and wild type cells. (A) The level of total XPC mRNA is measured by RT-PCR using the RT2 primers (Fig. 5D). Quantification shows that the total mRNA is statistically much lower in XP-C cells compared to wild type and heterozygous parent cells (**p = 0.0001 between normal versus XP-C cells and *p = 0.001 between normal and XP heterozygous parent cells (Student’s test)). (B) Using the RT6 primers, 2 abnormal XPC mRNA isoforms (retention of intron 12 and the 44 bp deletion of exon 13) are found in the XP-C cells (bands a and b, respectively) while the normal mRNA is observed in wild type and heterozygous parent cells (band c). (C) Using primers between exon 12 and exon 16, a unique band corresponding to the exon 13 skipping is found in XP-C cells (band d) while a unique normal band is found in wild type cells (band c) and both bands are observed in the heterozygous parental cells. (D) Schematic representation of the XPC gene with the location of the primers and of the XPC Mahori mutation (IVS12-1G > C). Coding exons are represented in black boxes. Experiments have been carried out twice with two cell lines of each XP-C, heterozygous parent and control cells.

patients. More cutaneous tumors are present in the white population and the types and localization of cutaneous cancers seem also to be different. The tip of the tongue and eyes seem to be more frequent cancer sites in black patients. Also melanomas are more frequent in white than in black XPC patients [4,23,24]. Interestingly, we showed that young white XP patients develop essentially the lentigo maligna melanoma (LMM) type as found essentially in the elderly [24]. This melanoma class could be due to the accumulation with time of unrepaired UV-induced lesions in young XPs. The black skinned XP patients do not develop melanoma on pigmented areas and are, therefore, protected from high level of UV-induced lesions by their melanin, confirming clearly the essential role of melanin in protecting individuals from UV-induced melanomas. We have found a new mutation in the XPC gene from the Mahori population. This mutation was found in all patients as well as in their parents in heterozygous state. This corresponds to a founder effect associated with consanguinity, which is frequent in Mayotte. We calculated that the common ancestor mutation occurred approximately 770 years ago. This founder effect may be common to other black patients because the Mahori population migrated from Africa several hundreds of years ago. Mutation

screening should search in priority for this mutation in African XP patients. 4.2. Splicing mutations at intron 12 According to the latest published compilation of pathogenic XPC mutations [23], the most common mutations are deletions (17/46), substitutions (16/46), splicing (9/46) and insertions (4/46). Splicing mutations have been described for the XPC gene, such as for introns 3, 5.1, 8, 9 and 11 [25–27]. Concerning the splicing of intron 12, an African American patient (XP25BE) has been reported to exhibit a homozygous deletion of AG (−1 and −2 bp) and an insertion CC (−6 and −7 bp) at the 3 end of intron 12 [16]. This mutation, distinct from the unique Comorian mutation, we describe here, produces the same final DNA sequence by different mechanisms and the same three isoforms that we found in our XP patients. The G > C substitution at the 3 splice acceptor site of intron 12 in the Mahori XP-C patients alters the obligatory AG acceptor dinucleotide to AC and reduces the information content of the native splice acceptor in intron 12 from 5.1 bits to −2.2 bits, leading to

584

F. Cartault et al. / DNA Repair 10 (2011) 577–585

the abolition of the splice site [28]. Sequencing of the mutant XPC cDNA showed that 3 abnormally spliced mRNAs were generated (exon 13 skipping, intron 12 retention, and 44 bp exon 13 deletion). Real-time quantitative RT-PCR in normal cells showed a high excess of the full-length XPC mRNA while the two abnormal isoforms were found at very low levels (Fig. 5). The relative abundance of alternatively spliced transcripts could, however, lead to a reduced nucleotide excision repair capacity and to a higher risk of cancer according to data previously reported. Khan et al. [27] showed that an increasing proportion of alternatively spliced XPC mRNA with exon 13 skipping was associated with reduced DNA repair. 4.3. SNP-6bp intron 11 A common single nucleotide polymorphism at −6 bp in the end of intron 12 (rs2279017) corresponding to allele frequencies of 67% for C and 33% for A has been found in the control Mahori population. These frequencies match those previously reported for black populations of Sub-Saharan Africa (for 120 individuals, the C allele was 0.71%) and of African Americans (for 46 individuals, the C allele was 0.65%) (http://www.ncbi.nlm.nih.gov/projects/SNP/snp ref.cgi?rs=2279017). These C values are slightly higher than in white populations: in an American group of 97 controls, allele C represented 58% of cases [27], in a German cohort of 669 subjects the allele frequency was 63% for C [29] and in 2 control groups of French population the allele frequency for C was 56% among 382 individuals and 59% among 344 individuals [30,31]. The presence of the A/A genotype leads to an increased proportion of mRNA isoform with exon 13 skipping, associated with reduced DNA repair activity in normal cells [27]. This polymorphism is in a linkage-disequilibrium with 2 other polymorphisms: an intron 9 poly-AT insertion/deletion polymorphism (PAT) and an exon 15 A/C polymorphism. Haplotype intron 9 PAT+, intron 12 −6A, exon 15 2920C may be associated with differences in cancer susceptibility, notably melanoma [29]. In the Mahori XP patients, this polymorphism is statistically associated with the XPC mutation with a p-value < 4 × 10−5 , indicating that the C allele is part of the mutated and rare haplotype. This SNP could modify the splicing efficiency and may be important for the level of repair in heterozygous individuals exhibiting the A sequence on the wild type allele. 3

5. Conclusion For the first time, we have identified a new mutation responsible for XP-C in a homogenous group of black patients. We found major differences between black and white XP patients. The diagnosis of XP in almost all black patients was associated with severe ocular abnormalities, including malignant lesions. The patients with skin tumors always had SCC, sometimes at unusual sites (lips and tip of the tongue) and melanoma was less frequent than in white patients. This new mutation abolishes a splice acceptor site leading to an aberrant splicing of XPC mRNA. Three abnormal isoforms are predominant in patients, while the normal isoform represents the vast majority in healthy subjects. The clinical description found in the group of XP-C Mahori patients is amazingly similar to most of the previously reported forms of XP in black patients. It is possible that this common mutation on the XPC gene, which we calculated to be as old as 770 years, is present as a founder effect in Africa as well. African XP patients with a severe XP phenotype with predominant ocular abnormality and precocious skin cancers, but without neurological abnormalities, should be screened for this mutation. The common polymorphism A > C at −6 bp at the 3 end of intron 12 seems to influence splicing and be correlated with cancer risk

in the general population. This could modify the severity of the XPC mutation and, therefore, be important for evaluating the cancer risk in XP-C heterozygotes. Further data and the pursuit of our multidisciplinary study are essential to analyze the evolution of XP in black skinned patients from Mayotte and the other neighboring Comorian islands. Early detection or prenatal diagnosis of XP in black populations allows us to propose stringent UV protection and, therefore, to help young patients to live a better and healthier life. Conflict of interest We declare no conflicts of interest. Acknowledgments We thank Mrs. D. Pham for cell culture and UDS analysis, Mrs. O. Lagente-Chevallier for complementation study with retroviruses and Dr. S. Benhamou (Institut Gustave Roussy) for help in statistical analysis and M.J. Martinez for technical assistance (CHR-Réunion). We are thankful to Dr. L. Daya-Grosjean for her help in reading the manuscript. This research was supported by Ministère délégué à l’enseignement supérieur et à la recherche (PHRC) to F.C. and by CNRS, ANR (Paris, France) and the French “Association des Enfants de la Lune” to A. S. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.dnarep.2011.03.005. References [1] J.E. Cleaver, Defective repair replication of DNA in xeroderma pigmentosum, Nature 218 (1968) 652–656. [2] J.H. Hoeijmakers, Genome maintenance mechanisms for preventing cancer, Nature 411 (2001) 366–374. [3] A. Stary, A. Sarasin, The genetics of the hereditary xeroderma pigmentosum syndrome, Biochimie 84 (2002) 49–60. [4] K.H. Kraemer, M.M. Lee, J. Scotto, Xeroderma pigmentosum. Cutaneous, ocular, and neurological abnormalities in 830 published cases, Arch. Dermatol. 123 (1987) 241–250. [5] W.J. Kleijer, V. Laugel, M. Berneburg, T. Nardo, H. Fawcett, A. Gratchev, N.G. Jaspers, A. Sarasin, M. Stefanini, A.R. Lehmann, Incidence of DNA repair deficiency disorders in western Europe: xeroderma pigmentosum Cockayne syndrome and trichothiodystrophy, DNA Repair 7 (2008) 744–750. [6] L. Daya-Grosjean, M.R. James, C. Drougard, A. Sarasin, An immortalized xeroderma pigmentosum, group C, cell line which replicates SV40 shuttle vectors, Mutat. Res. 183 (1987) 185–196. [7] A. Sarasin, C. Blanchet-Bardon, G. Renault, A.R. Lehmann, C. Arlett, Y. Dumez, Prenatal diagnosis in a subset of trichothiodystrophy patients defective in DNA repair, Br. J. Dermatol. 127 (1992) 485–491. [8] C. Arnaudeau-Bégard, F. Brellier, O. Chevallier-Lagente, J. Hoeijmakers, F. Bernerd, A. Sarasin, T. Magnaldo, Genetic correction of DNA repairdeficient/cancer-prone xeroderma pigmentosum group C keratinocytes, Hum. Gene Ther. 14 (2003) 983–996. [9] L. Zeng, X. Quilliet, O. Chevallier-Lagente, E. Eveno, A. Sarasin, M. Mezzina, Retrovirus-mediated gene transfer corrects DNA repair defect of xeroderma pigmentosum cells of complementation groups A, B and C, Gene Ther. 4 (1997) 1077–1084. [10] M.W. Pfaffl, A new mathematical model for relative quantification in real-time RT-PCR, Nucleic Acids Res. 29 (2001) e45. [11] F. Austerlitz, L. Kalaydjieva, E. Heyer, Detecting population growth, selection and inherited fertility from haplotypic data in humans, Genetics 165 (2003) 1579–1586. [12] M. Stephens, N.J. Smith, P. Donnelly, A new statistical method for haplotype reconstruction from population data, Am. J. Hum. Genet. 68 (2001) 978–989. [13] M. Stephens, P. Donnelly, A comparison of Bayesian methods for haplotype reconstruction from population genotype data, Am. J. Hum. Genet. 73 (2003) 1162–1169. [14] N. Youssef, P. Vabres, T. Buisson, N. Brousse, S. Fraitag, Two unusual tumors in a patient with xeroderma pigmentosum: atypical fibroxanthoma and basosquamous carcinoma, J. Cutan. Pathol. 26 (1999) 430–435. [15] M. Tremblay, H. Vézina, New estimates of intergenerational time intervals for the calculation of age and origins of mutations, Am. J. Hum. Genet. 66 (2000) 651–658.

F. Cartault et al. / DNA Repair 10 (2011) 577–585 [16] S.G. Khan, K.S. Oh, T. Shahlavi, T. Ueda, D.B. Busch, H. Inui, S. Emmert, K. Imoto, V. Muniz-Medina, C.C. Baker, J.J. DiGiovanna, D. Schmidt, A. Khadavi, A. Metin, E. Gozukara, H. Slor, A. Sarasin, K.H. Kraemer, Reduced XPC DNA repair gene mRNA levels in clinically normal parents of xeroderma pigmentosum patients, Carcinogenesis 27 (2006) 84–94. [17] W.K. Jacyk, Xeroderma pigmentosum in black South Africans, Int. J. Dermatol. 38 (1999) 511–514. [18] H. Ahmed, R.Y. Hassan, U.H. Pindiga, Xeroderma pigmentosum in three consecutive siblings of a Nigerian family: observations on oculocutaneous manifestations in black African children, Br. J. Ophthalmol. 85 (2001) 110–111. [19] M.T. Dieng, S.O. Niang, J.M. Dangou, B. Ndiaye, Xeroderma pigmentosum: report of 6 cases in Dakar, Bull. Cancer 88 (2001) 199–202. [20] J.C. Hébert, J.F. Lefait, O. Hébert, Xeroderma pigmentosum in negroid children. 5 cases in Mahori children, Ann. Dermatol. Venereol. 121 (1994) 382–386. [21] M.M. Chidzonga, Lip cancer in Zimbabwe. Report of 14 cases, Int. J. Oral Maxillofac. Surg. 34 (2005) 149–151. [22] Y.M. Olumide, Photodermatoses in Lagos, Int. J. Dermatol. 26 (1987) 234–236. [23] N. Soufir, C. Ged, A. Bourillon, F. Austerlitz, C. Chemin, A. Stary, J. Armier, D. Pham, K. Khadir, J. Roume, S. Hadj-Rabia, B. Bouadjar, A. Taieb, H. De Verneuil, H. Benchiki, B. Grandchamp, A. Sarasin, A prevalent mutation with founder effect in xeroderma pigmentosum group C from North Africa, J. Invest. Dermatol. 130 (2010) 1537–1542. [24] A. Spatz, G. Giglia-Mari, S. Benhamou, A. Sarasin, Association between DNA repair deficiency and high level of p53 mutations in melanoma of xeroderma pigmentosum, Cancer Res. 61 (2001) 2480–2486. [25] F. Chavanne, B.C. Broughton, D. Pietra, T. Nardo, A. Browitt, A.R. Lehmann, M. Stefanini, Mutations in the XPC gene in families with xeroderma pigmento-

[26]

[27]

[28] [29]

[30]

[31]

585

sum and consequences at the cell, protein, and transcript levels, Cancer Res. 60 (2000) 1974–1982. S.G. Khan, A. Metin, E. Gozukara, H. Inui, T. Shahlavi, V. Muniz-Medina, C.C. Baker, T. Ueda, J.R. Aiken, T.D. Schneider, K.H. Kraemer, Two essential splice lariat branchpoint sequences in one intron in a xeroderma pigmentosum DNA repair gene: mutations result in reduced XPC mRNA levels that correlate with cancer risk, Hum. Mol. Genet. 13 (2004) 343–352. S.G. Khan, V. Muniz-Medina, T. Shahlavi, C.C. Baker, H. Inui, T. Ueda, S. Emmert, T.D. Schneider, K.H. Kraemer, The human XPC DNA repair gene: arrangement, splice site information content and influence of a single nucleotide polymorphism in a splice acceptor site on alternative splicing and function, Nucleic Acids Res. 30 (2002) 3624–3631. P.K. Rogan, B.M. Faux, T.D. Schneider, Information analysis of human splice site mutations, Hum. Mutat. 12 (1998) 153–171. S. Blankenburg, I.R. König, R. Moessner, P. Laspe, K.M. Thoms, U. Krueger, S.G. Khan, G. Westphal, C. Berking, M. Volkenandt, K. Reich, C. Neumann, A. Ziegler, K.H. Kraemer, S. Emmert, Assessement of 3 xeroderma pigmentosum group C polymorphisms and risk of cutaneous melanoma: a case-control study, Carcinogenesis 26 (2005) 1085–1090. S. Michiels, P. Danoy, P. Dessen, A. Bera, T. Boulet, C. Bouchardy, M. Lathrop, A. Sarasin, S. Benhamou, Polymorphism discovery in 62 DNA repair genes and haplotype associations with risks for lung and head and neck cancers, Carcinogenesis 28 (2007) 1731–1739. S. Michiels, A. Laplanche, T. Boulet, P. Dessen, B. Guillonneau, A. Méjean, F. Desgrandchamps, M. Lathrop, A. Sarasin, S. Benhamou, Genetic polymorphisms in 85 DNA repair genes and bladder cancer risk, Carcinogenesis 30 (2009) 763–768.