Refined 2.7 Centimorgan Locus in Xp21.3-22.1 for aNonspecific X-Linked Mental RetardationGene (MRX54)

Lamia Ben Jemaa,1* Vincent des Portes,2 Ramzi Zemni,1 Ridha Mrad,1 Faouzi Maazoul,1Cherif Beldjord,2 Habiba Chaabouni,1 and Jamel Chelly2

1Service des Maladies congenitales et hereditaires, Hopital Charles Nicolle, Tunis, Tunisie2INSERM U 129-ICGM, Faculte de Medecine Cochin, Paris, France

Nonspecific X-linked mental retardation(MRX) is a heterogeneous condition inwhich mental retardation (MR) appears tobe the only consistent manifestation. A largegenetic interval of assignment obtained onindividual families by linkage analysis, ge-netic, heterogeneity, and phenotypic vari-ability usually are major obstacles to fine-map and identify the related disease genes.Here we report on a large Tunisian family(MRX54) with an MRX condition. X-linkedrecessive inheritance is strongly suggestedby the segregation of MR through seven un-affected carrier females to 14 affected malesin two generations. Two-point linkageanalysis demonstrated significant linkagebetween the disorder and several markersin Xp21.3-22.1 (maximum LOD score Zmax =3.56, recombination fraction 0 = 0 atDXS1202), which was confirmed by multi-point linkage analyses. Recombinant eventsobserved with the flanking markers DXS989and DXS1218 delineate a refined locus of ap-proximately 2.7 cM in accordance with thephysical distance between these two mark-ers. The small interval of assignment ob-served in this family overlaps not only withnine large MRX loci previously reported inXp21.3-22.1 but also with two inherited mi-crodeletions in Xp21.3-22.1 involved in non-

specific MR. Although the involvement ofseveral genes located in the Xp21.3-22.1 re-gion cannot be ruled out, data reported inthis study could be used as a starting pointfor the search of such gene(s). Am. J. Med.Genet. 85:276–282, 1999. © 1999 Wiley-Liss, Inc.

KEY WORDS: X-linked nonspecific mentalretardation (MRX); linkageanalysis; Xp21.3-22.1 region

INTRODUCTION

The frequency of mental retardation (MR) is esti-mated to be 1–2% in the general population. Most epi-demiological studies indicated a higher frequency inmales versus females with an excess between 10 and40% [Baird and Sadovnick, 1985]. This significantlyskewed sex ratio towards males probably is related to ahigh frequency of X-linked mental retardation (XLMR)that was estimated at approximately 1/600 male births[Herbst and Miller, 1980]. XLMR conditions can be cat-egorized roughly as syndromic disorders (MRXS), char-acterized by consistent, distinctive clinical findings, oras nonspecific (MRX), if MR is the only primary symp-tom among affected individuals [Neri et al., 1991]. De-spite significant advances in understanding the mo-lecular basis of some syndromic XLMR conditions, in-cluding fragile X syndrome, progress in identifyinggenes involved in the MRX group has been slow, partlybecause of the clinical variability and genetic hetero-geneity. Genetic mapping data on individual familiesshowed more than 55 MRX loci (8th International Meet-ing on Fragile X and X-linked Mental Retardation, Pic-ton, Canada, August 17–22, 1997) with broad intervalsof assignment and several overlaps between these in-tervals. However, a recent review reported by Gedeonet al. [1996] and personal data collected within theframe of an XLMR European consortium showed thatthese loci are clustered in 8 to 10 nonoverlapping re-gions, suggesting the involvement of as many genes.

In this study, we describe the clinical and linkagestudy of a large nonconsanguineous Tunisian family,

Contract grant sponsor: le comite mixte Franco-Tunisien pourla cooperation universitaire; Contract grant sponsor: le Ministerede l’Enseignement Superieur Tunisien; Contract grant sponsor:le Ministere de la Sante publique Tunisien; Contract grant spon-sor: l’Institut National de la Sante et de la Recherche Medicale;Contract grant sponsor: Assistance Publique-Hopitaux de Paris;Contract grant number: CRC950164; Contract grant sponsor: TheAssociation Francaise contre les Myopathies; Contract grantsponsor: The fondation Jerome Lejeune.

*Correspondence to: Dr. Lamia Ben Jemaa, Service des Mala-dies congenitales et hereditaires, Hopital Charles Nicolle, 1006Tunis Bab Souika, Tunisie.

Received 22 August 1997; Accepted 21 July 1998

American Journal of Medical Genetics 85:276–282 (1999)

© 1999 Wiley-Liss, Inc.

Fig

.1.

Ped

igre

eof

fam

ily

MR

X54

.Th

eob

serv

edh

aplo

type

sfo

rn

ine

mar

kers

ofth

eX

p21.

3-22

.1re

gion

are

show

nfo

rin

divi

dual

incl

ude

din

the

lin

kage

anal

ysis

,an

dh

aplo

type

sin

bra

cket

sar

ede

duce

dh

aplo

type

s.T

he

com

plet

esh

aded

squ

ares

den

ote

indi

vidu

als

wit

hm

enta

lre

tard

atio

n.

Th

ed

oub

le-h

ead

edar

row

sde

not

ein

form

ativ

ecr

ossi

ng

over

.T

he

alle

les

shar

edby

affe

cted

and

tran

smit

tin

gin

divi

dual

sar

esh

own

asb

oxed

.*D

NA

avai

labl

efo

rli

nka

gean

alys

is.

MRX54, affected with a nonspecific X-MR. Most of thechromosome X was excluded and significant linkagewith markers in Xp21.3-22.1 was found with a refined2.7 cM interval of assignment. Our data showed thatthis interval overlaps with a microdeletion in Xp21.3-22.1 present in a nonspecific XLMR family and alloweda better definition of a critical region potentially in-volved in Xp21.3-22.1-linked MR.

PATIENTS AND METHODSClinical Report

The pedigree of the affected family (Fig. 1), shows thesegregation of moderate to severe MR through sevenunaffected carrier females to 14 affected males in twogenerations. X-linked recessive inheritance is stronglysuggested by the occurrence of MR exclusively in malesand lack of male-to-male transmission.

A thorough family history was collected from physi-cians, social workers, and relatives. Seven of the af-fected individuals underwent clinical examination.Three of them are shown in Figure 2.

Cytogenetic Analysis and Fragile Sites Study

Chromosomes were prepared from peripheral lym-phocytes by standard techniques and analyzed after Rbanding. Low folate cultures were also performed forfragile sites screening. DNA samples of one affectedboy and an obligate carrier were tested by Southernblot analysis with the FraxA (StB12.3) and FraxE(OxE20) probes on EcoRI and HindIII digested genomicDNA, respectively [Rousseau et al., 1991; Knight et al.,1993].

Genotyping

Blood was collected from each relative included inthe genetic study after informed consent was obtained.Linkage analysis for a primary mapping was carriedout in 23 individuals using 26 highly polymorphic

Genethon microsatellite markers evenly distributed onthe X chromosome (DXS1073, DXS1193, DXS8028,DXS1227, DXS1205, DXS1192, DXS1047, DXS1001,DXS1059, DXS8020, DXS9990, DXS1196, DXS986,DXS1275, DXS991, DXS1003, DXS993, DXS1068,DXS1214, DXS1202, DXS1226, DXS987, DXS1223,and DXS7107) [Gyapay et al., 1994; Dib et al., 1996],DXS424 [Huang et al., 1992], and DXS1108 (primerssequences in Genome Data Base). Then six furthermarkers located in Xp21.3-22.1 were tested in refinethe locus interval: DXS989, DXS1061, DXS8010,DXS1218, DXS8039, and DXS985 [Gyapay et al., 1994;Dib et al., 1996]. For (CA)n markers, polymerase chainreaction (PCR) was performed as described by Gyapayet al. [1994]; PCR amplification products were electro-phoresed on a 6%-denaturing polyacrylamide gel,transferred to a nylon membrane, and then hybridizedovernight at 42°C with [32P]dCTP-labeled poly-ACprobe.

DNA of Centre d’Etudes du Polymorphisme Humain(CEPH) individual number 1347 was included in eachexperiment as a control for allele size estimation.

Linkage AnalysisTwo-point LOD scores, maximum LOD scores, and

multipoint location score analyses were conducted withthe MLINK, ILINK, and LINKMAP programs [Lath-rop and Lalouel, 1984] respectively, using the softwarepackage FASTLINK (version 2.3P) [Cottingham, et al.,1993]. The mutation rate, phenocopy rate, penetrance,and gene frequency were set at 0.001, 0.0, 1, and0.0005, respectively. Equal allele frequencies were as-sumed for the highly polymorphic markers. Because noclinical information about individual I-1 was available,his phenotype was set as “unknown.”

RESULTSClinical Findings

All affected males were born at home at term afternormal pregnancy and have moderate to severe MR; no

Fig. 2. Nonspecific facial appearance in affected males: individuals III-16 (left), IV-9 (middle), and IV-10 (right).

278 Ben Jemaa et al.

statural growth deficiency was observed. Although in-dividuals IV-9, IV-10, IV-11, and IV-20 have a long facewith thin lips and large ears, no characteristic patternof anomalies was shared by all affected individuals.Seven affected males were examined thoroughly, andclinical findings are summarized in Table I. Three ofthem are shown in Figure 2.

Individual IV-9 was first examined at age 14 years.His height was 150 cm, weight 43 kg, and occipito-frontal circumference (OFC) 54 cm. No visceromegalywas observed; neurological findings were normal. Hehad convulsions in the first year of life and is severelymentally retarded (IQ 4 40). Brain computed tomog-raphy scan, magnetic resonance imaging biochemicalblood, and urine screening (amino and organic-acidchromatography) were normal.

Individual IV-10 was first seen at age 26 years. Hewas 169 cm tall, weighed 59 kg, and his OFC was 53.5cm. He has macroorchidism and mild to moderate men-tal retardation; he never attended normal school and isnot able to read and write.

Individual IV-11 had severe epileptic seizures andbehavior problems. He was 172 cm tall, weighed 70 kg,and his OFC was 55 cm. He has macroorchidism andsevere mental retardation.

Individual IV-12 had moderate mental retardationand attended a school for the developmentally dis-abled.

Individual IV-20 also had a moderate mental retar-dation and does not work.

Individual III-5 had severe mental retardation with-out specific findings. He was unable to attend schooland does not work.

Individual III-6 is 60 years old. Despite mild to mod-erate mental retardation, he is married and has fourhealthy children. He has a long face with large ears.

Obligate carrier females do not exhibit any clinicalmanifestations.

Cytogenetic Analysis and Fragile X Screening

No chromosomal abnormalities were observed in thethree brothers (IV-9, IV-10, IV-11) and their mother.Repeat expansions at FRAXA and FRAXE site wereruled out by Southern blot analysis.

Linkage Analysis

Pairwise linkage analysis with 32 informative mark-ers spanning the X chromosome excluded almost thewhole chromosome and indicated a significant linkagebetween the disease and several markers in Xp21.3-22.1 (Table II) with maximum two-point LOD score of Z4 3.56 at u 4 0 for DXS 1202 and Z 4 3.60 at u 4 0 forDXS 8039. However, haplotype observation and multi-point linkage analysis (Fig. 3) exclude a significantlinkage between the disease and DXS 8039 with scoreof 0.42 at u 4 0 for this marker. In contrast, the mul-tipoint linkage analysis confirms the significant link-age with DXS 1061 and DXS1202 (Z 4 3.38 at u 4 0)(Fig. 3). However, taking into account haplotype analy-sis (Fig. 1), it is worth noting that the mutated X chro-mosome was inherited from the grandfather. As thismale is dead and no information is available, his phe-notypic status was set as unknown for linkage studies.Recombination events observed for individuals IV-9and III-16 at the flanking markers DXS989 andDXS1218, respectively, define a 2.7-cM region, accord-ing to genetic [Dib et al., 1996] and physical [Ferrero etal., 1995] maps of the Xp21.3-22.1 region.

DISCUSSION

A linkage study carried out in the large family de-scribed herein excluded most of the X chromosome andshowed the existence of an MRX gene in Xp21.3-22.1within 2.7-cM interval between DXS989 and DXS1218.Two-point LOD scores obtained with markers atXp21.3-22.1 are summarized in Table II. SignificantLOD scores (>3 at u 4 0) were obtained with threeclustered markers in Xp22.1 (DXS1202, DXS1061, andDXS8010) and with a more proximal marker DXS8039.All other markers showed either nonsignificant LODscore or at least one recombination, indicated by anLOD score of “− inf” at u 4 0, including markerDXS1218, which maps between DXS8010 andDXS8039 (Table II). Therefore, two-point linkageanalysis suggested the existence of two potential loci inXp21.3-22.1. However, consecutive multipoint linkageanalyses using informative markers in the Xp21.3-22.1

TABLE I. Summary of the Clinical Findings in MRX Patients*

IndividualIV-9

IndividualIV-10

IndividualIV-11

IndividualIV-12

IndividualIV-20

IndividualIII-5

IndividualIII-16

Age (years) 14 26 22 16 24 30 60

Physical measurementsHeight: cm (centile) 150 (10) 169 (25) 172 (25–50) 170 (50) 177 (50–75) 165 (10) 165 (10)Weight: Kg (centile) 43 (25) 59 (25) 70 (75–90) 58 (50) 70 (75–90) 57 (10–25) 60 (25)OFC: cm (centile) 54 (50) 54 (2–50) 55 (50) 53 (2) 56 (50) 56 (50) 55 (50)Testicular volume (ml) 12 22 28 14 15 13.5 15

Clinical observationsMental status Severe MR Mild MR Severe MR Moderate MR Moderate MR Severe MR Mild MRIQ 40 NE NE NE NE NE NEMinor anomalies Long face, thin

lips, largeears

Long face, thinlips, largeears

Long face, thinlips, largeears

No Long face, thinlips, largeears

No Long face, largeears

Malformations No No No No No No NoNeurological signs

Seizures + − + − − − −Behavior Normal Normal Agressive Normal Normal Normal Normal

*NE, observation not made.

Linkage Analysis in Nonspecific XLMR 279

Fig. 3. Multipoint LOD scores between MRX54 phenotype and 10 different markers from Xp21.3-22.1. Three overlapping five-point runs have beenperformed and compiled. This multipoint LOD score analysis allowed exclusion of linkage of the disease with DXS8039 that showed a two-point LOD scoreof 3.60.

TABLE II. Pairwise LOD Scores Between MRX54 Locus and 18 PolymorphicMarkers Spanning the X Chromosome, and Seven Further Markers Clustered

in the Xp21.3–Xq22.1 Region

Locus

Z (lod score) at recombination frequencies u

umax Zmax0.0 0.01 0.05 0.1 0.2 0.3 0.4

DXS7107 -inf −2.0 −0.7 −0.3 0.0 0.0 0.0 0.28 0.06DXS1223 -inf −2.0 −0.8 −0.3 0.0 0.0 0.0 0.28 0.05DXS987 -inf −0.1 0.5 0.6 0.6 0.4 0.1 0.12 0.66DXS1226 -inf 1.2 1.7 1.7 1.5 1 1.6 0.08 1.75DXS989 -inf 1.4 1.9 1.9 1.6 1.2 0.6 0.07 1.98DXS1202 3.56 3.5 3.2 2.9 2.3 1.5 0.8 0.00 3.56DXS1061 3.26 3.2 2.9 2.7 2.1 1.4 0.7 0.00 3.26DXS8010 3.14 3.1 2.9 2.6 2.0 1.4 0.7 0.00 3.14DXS1218 -inf 1.4 1.9 2.0 1.7 1.3 0.7 0.08 2.02DXS8039 3.60 3.5 3.3 2.9 2.3 1.6 0.8 0.00 3.60DXS985 -inf 1.3 1.8 1.8 1.5 0.9 0.3 0.07 1.80DXS1214 -inf 1.7 2.2 2.2 1.9 1.4 0.8 0.07 2.28DXS1068 -inf −0.5 0.7 1.1 1.2 1.0 0.6 0.17 1.17DXS1003 -inf −0.1 0.5 0.6 0.6 0.4 0.1 0.12 0.66DXS986 -inf 0.0 0.5 0.7 0.6 0.4 0.1 0.12 0.68DXS990 -inf 0.0 0.5 0.6 0.6 0.4 0.1 0.12 0.66DXS424 -inf −0.1 0.5 0.6 0.6 0.4 0.1 0.12 0.63DXS1001 -inf 0.0 0.5 0.6 0.6 0.4 0.1 0.12 0.66DXS1047 -inf 0.0 0.5 0.6 0.6 0.4 0.1 0.12 0.66DXS1192 -inf −0.1 0.5 0.6 0.6 0.4 0.1 0.13 0.63DXS1205 -inf 0.0 0.5 0.6 0.6 0.4 0.1 0.12 0.66DXS1227 -inf 0.0 0.5 0.6 0.6 0.4 0.1 0.12 0.66DXS8028 -inf −0.3 0.2 0.4 0.4 0.3 0.1 0.14 0.42DXS1193 -inf −4.0 −2.0 −1.2 −0.5 −0.2 0.0 0.50 0.00DXS1108 -inf −4.1 −2.1 −1.2 −0.5 −0.2 0.0 0.50 0.00

280 Ben Jemaa et al.

region showed a significant decrease of the maximumLOD score at DXS8039 (Z 4 0.4 at u 4 0) and alloweda powerful exclusion of any linkage around this marker(Fig. 3). As indicated in Table II and Figure 3, perform-ing multipoint linkage analysis did not raise the maxi-mum LOD score because the markers around DXS1061were completely informative in all available meiosesand the status of the grandparents was set up as un-known because of the absence of reliable clinical infor-mations. Haplotype analysis shown in Fig. 1 confirmedthat the disease gene is located between DXS989 andDXS1218 and suggests an inheritance of the grandfa-ther’s haplotype by affected children. A recombinationfrom carrier III-12 to patient IV-9 occured betweenmarkers DXS989 and DXS1202 locating the geneproximal to DXS989. Furthermore a recombination oc-cured between marker DXS8010 and DXS1218 in pa-tient III-16 mapping the gene distal to DXS1218. Ge-netic and physical distances between DXS989 andDXS1218 were estimated at approximately 2.7 cM [Dibet al., 1996] and 2.5 Mb [Ferrero et al., 1995], respec-tively.

It is interesting to point out that this small intervalis encompassed so far by at least nine other MRX locicorresponding to nine unrelated MRX families, i.e.,MRX2 [Hu et al., 1994], MRX13 [Kerr et al., 1992],MRX21 [Kozak et al., 1993], MRX29 [Hane et al.,1996], MRX32 [Lubs et al., 1996], MRX33 [Holinski-Feder et al., 1996], MRX36 [Claes et al., 1996], MRX38

[Schutz et al., 1996], and MRX43 [Hamel et al., 1997](Fig. 4). In addition to overlapping with these geneticintervals, the locus herein described overlaps with twoinherited microdeletions in Xp21.3-22.1 associatedwith nonspecific MR. The first deletion, reported at the7th International Workshop “Fragile X and X-linkedMental Retardation,” concerns the marker DXS1218and extends on both sides of this marker [Raeymaekerset al., 1996]. The second deletion, reported by Billuartet al. [1996] extends from DXS1202 to DXS1065(Fig. 4).

In view of the nonspecific clinical data and the broadintervals of assignment reported in the previously de-scribed nine families, it is difficult to assess whethermore than one gene is involved in these MRXs localizedin Xp22. However, the overlaps between the deletionsand the 2.7 cM interval of assignment reported in thisstudy could be used as a basis to better define the criti-cal genomic region involved in MR linked to Xp21.3-22.1. Although the size of deletion reported by Raey-maekers et al. [1996] has not been reported yet, theoverlap between our deletion [Billuart et al., 1996] andthe 2.7 genetic interval suggests that the genomic re-gion of approximately 1 Mb [Ferrero et al., 1995], be-tween DXS1202 (distal end of the deletion) andDXS1218 (proximal recombinant marker), could beconsidered as critical for further investigations aimingto identify the disease-related gene.

Fig. 4. Five of nine genetic intervals and deletions in nonspecific XLMR localized in Xp21.3-22.1. (Genetic intervals localized in this region larger than30 cM are not represented on this figure.) The position and order of markers within Xp21.3-22.1 were deduced from available genetic and physical maps.Dotted lines correspond to genetic intervals and bold lines to deletions associated only with nonspecific mental retardation. MRX34 is a deletionreported by Reaymackers et al. [1996], and his size has not yet been reported but involves DXS1218 and not the flanking markers DXS1065 and 1061.The second deletion was described by Billuart et al. [1996] and it extends from XDXS1065 to DXS1202.

Linkage Analysis in Nonspecific XLMR 281

ACKNOWLEDGMENTS

We thank members of the family for their participa-tion in this study. This work was supported in part bygrants from le comite mixte Franco-Tunisien pour lacoop e rat ion univers i ta i re , l e Minis t e re del’Enseignement Superieur Tunisien, le Ministere de laSante publique Tunisien, l’Institut National de laSante et de la Recherche Medicale (INSERM), Assis-tance Publique-Hopitaux de Paris (CRC950164), theAssociation Francaise contre les Myopathies (AFM),and the fondation Jerome Lejeune.

REFERENCESBaird PA, Sadovnick AD. 1985. Mental retardation in over half-a-million

consecutive livebirths: An epidemiological study. Am J Ment Defic 98:323–330.

Billuart P, Vinet MC, Des Portes V, Llense S, Richard L, Moutard ML,Recan D, Bruls T, Bienvenu T, Kahn A, Beldjord C, Chelly J. 1996.Identification by STS PCR screening of a microdeletion in Xp21.3-22.1associated with non-specific mental retardation. Hum Mol Genet 7:977–979.

Claes S, Gu X, Legius E, Lorenzetti E, Marynen P, Fryns JP, Cassiman JJ,Raeymaekers P. 1996. Linkage analysis in three families with nonspecific X-linked mental retardation. Am J Med Genet 64:137–146.

Cottingham RW, Idury RM, Schaffer A. 1993. Faster sequential geneticlinkage computations. Am J Hum Genet 53:252–263.

Dib C, Faure S, Fizames C, Samson D, Drouot N, Vignal A, Millasseau P,Marc S, Hazan J, Seboun E, Lathrop M, Gyapay G, Morissette J, Weis-senbach J. 1996. A comprehensive genetic map of the human genomebased on 50264 microsatellites. Nature 380:152–154.

Ferrero GB, Franco B, Roth EJ, Firulli BA, Borsani G, Delmas-Mata J,Weissenbach J, Halley G, Schlessinger D, Chinault AC, Zoghbi HY,Nelson DL, Ballabio A. 1995. An integrated physical and genetic mapof a 35 Mb region on chromosome Xp22.3-21.3. Hum Mol Genet 4:1821–1827.

Gedeon AK, Donnelly AJ, Mulley JC, Kerr B, Turner G. 1996. How manyX-linked genes for non-specific mental retardation (MRX) are there?Am J Med Genet 64:158–162.

Gyapay G, Morissette J, Vignal A, Dib C, Fizames C, Millasseau P, Marc S,Bernardi G, Lathrop M, Weinssenbach J. 1994. The 1993–94 Genethonhuman genetic linkage map. Nat Genet 246:246–339.

Hamel BCJ, Smits APT, Van Roosmalen T, Van Bokhoven H, Van denHelm B, Kremer H, Knoers N, Yntema H, Assman-Hulsman CF,Smeets DF, Mariman EC, Ropers H. 1997. Non specific X-linked men-

tal retardation in 5 families (MRX43, 44, 45, 46, and 52): clinical, psy-chometric and molecular data. 8th International Workshop on Fragile Xsyndrome and X-linked Mental Retardation. Picton, Canada. (Ab-stract.)

Hane B, Schroer RJ, Arena JF, Lubs HA, Schwartz CE, Stevenson CE.1996. Nonsyndromic X-linked mental retardation: review and mappingMRX29 to Xp21. Clin Genet 50:176–183.

Herbst DS, Miller JR. 1980. Non-specific X-linked mental retardation II:the frequency in British Columbia. Am J Med Genet 7:461–469.

Holinski-Feder E, Golla A, Rost I, Seidel H, Rittinger O, Meindl A. 1996.Regional localization of two MRX genes to Xq28 (MRX28) and twoXp11.4-Xp22.12 (MRX33). Am J Med Genet 64:125–130.

Hu LJ, Blumenfeld-Heyberger S, Hanauer A, Weissenbach J, Mandel JL.1994. Non-specific X-linked mental retardation: linkage analysis inMRX2 and MRX4 families revisited. Am J Med Genet 51:569–574.

Huang TH-M, Cottingham R, Ledbetter D, Zoghbi H. 1992. Genetic map-ping of four dinucleotide repeat loci, DXS453, DXS458, DXS454, andDXS424, on the X chromosome using multiplex polymerase chain re-action. Genomics 13:375–380.

Kerr B, Gedeon A, Mulley J, Turner G. 1992. Localization of non-specificX-linked mental retardation genes. Am J Med Genet 43:392–401.

Knight SLJ, Flannery AV, Hirst MC, Campbell L, Christodoulou Z, PhelpsSR, Pointon J, Middleton-Price HR, Barnicoat A, Pembrey ME, OostraBA, Bobrow M, Davies KE. 1993. Trinucleotide repeat amplificationand hypermethylation of a CpG island in FRAXE mental retardation.Cell 74:127–134.

Kozak L, Chiurazzi P, Genuardi M, Pomponi MG, Zollini M, Neri G. 1993.Mapping of a gene for non-specific X-linked mental retardation: evi-dence for linkage to chromosomal region Xp21.1-Xp22.3. J Med Genet30:866–869.

Lathrop GM, Lalouel JM. 1984. Easy calculations of LOD scores and ge-netic risks on small computers? Am J Hum Genet 37:482–498.

Lubs HA, Schwartz CE, Stevenson R, Arena J. 1996. Study of X-linkedmental retardation (XLMR): summary of 61 families in the Miami/Greenwood study. Am J Med Genet 64:169–175.

Neri G, Gurrieri F, Gull A, Lubs HA. 1991. XLMR Genes: update 1990. AmJ Med Genet 38:186–189.

Raeymaekers P, Lin J, Gu X, Soekarman D, Cassiman JJ, Fryns JP,Marynen P. 1996. A form of non-specific mental retardation is probablycaused by a microdeletion in a belgian family. Am J Med Genet 64:16.

Rousseau F, Heitz D, Biancalana V, Blumenfeld S, Kretz C, Boue J, Tom-merup N, Van Der Hagen C, De Lozier-Blanchet C, Croquette MF,Gilgenkrantz S, Jalbert P, Voelckel MA, Obele I, Mandel JL. 1991.Direct diagnosis of DNA analysis of the fragile X syndrome of mentalretardation. N Engl J Med 325:1675–1681.

Schutz C, Ives EJ, Chalifoux M, MacLaren L, Farrell S, Robinson P, WhiteB, Holden J. 1996. Regional localization of an X-linked mental retar-dation gene to Xp21.1-p22.1 (MRX38). Am J Med Genet 64:89–96.

282 Ben Jemaa et al.