Abstract Mosaicism in germ cells has been recognized,over the past few years, as an important and relatively fre-quent mechanism at the origin of genetic disorders. Thereare two possibilities for the existence of such a mosaicism:one is that the mutation occurs in a germ cell that contin-ues to divide. The other possibility is that the mutation oc-curs very early in a somatic cell before the separation togerminal cells and is therefore present both in somatic andgerminal cells. Depending on various factors, such as thegene involved and/or the degree of mosaicism, the carrierof a somatic and germline mosaicism may be asymptomat-ic or may present with various symptoms of the disease.There are still relatively few reports in the literature inwhich the origin of germ-line mosaicism has been ana-lyzed; nevertheless, they allow for a better insight into themechanisms involved. In some diseases, such as osteogen-esis imperfecta, new mutations are often present as asymp-tomatic somatic and germline mosaicism in one of the par-ents of the propositus. In other disorders, such as neurofi-bromatosis, somatic mosaicism is very rare in the parentsof the propositus, perhaps since such mosaicism causesclinical symptoms. These differences are particularly im-portant for genetic counseling in order to evaluate the riskfor another affected child after the birth of the propositus.

J. ZlotogoraThe Rosa and David Orzen Human Genetics Clinic, Department ofHuman Genetics, Hadassah Medical Center, Hebrew University,Jerusalem, Israel

Present address:Departement de genetique Medicale, Necker Enfants Malades,149 Rue de Sèvres, F-75743 Paris CEDEX 15, FranceTel.: +33-1-44495161, Fax: +33-1-47348514

Hum Genet (1998) 102:381–386 © Springer-Verlag 1998

REVIEW ARTICLE

Joël Zlotogora

Germ line mosaicism

Received: 15 September 1997 / Accepted: 12 January 1998

tory of the disease. In some cases, the possibility of a re-cessive variant of the disease has been postulated; howev-er, in most of the reports, the existence of mosaicism in thegerminal cells has been suspected (Hall 1988). It is only inthe last few years, with the possibility of determining themolecular basis of many diseases, that the mechanisms in-volved are beginning to be understood. The existence ofmosaicism in the germ cells of one of the asymptomaticparents of children affected with dominant disorders hasindeed been demonstrated to be the cause of many cases ofrecurrence. There are two possibilities for the existence ofsuch a mosaicism. One is that the mutation occurs in agerm cell that continues to divide. The other possibility isthat the mutation occurs very early in a somatic cell beforethe separation to germinal cells and is therefore present inboth somatic and germinal cells. Depending on variousfactors, such as the gene involved and/or the degree of mo-saicism, the carrier of a somatic and germline mosaicismmay be asymptomatic or may present with various symp-toms of the disease.

Recently, with the advances of molecular genetics, ithas been possible to try to determine when new mutationscausing mosaicism in germ cells occur. There are still rela-tively few reports in the literature in which the origin ofmosaicism in the germ cells has been analyzed; however,they allow for a better insight into the mechanisms in-volved.

Review of the literature

The literature has been reviewed for cases of autosomaldominant and X-linked disorders in which mosaicism hasbeen demonstrated by molecular analysis. These disordershave been classified into two groups according to themethod of ascertainment: one group consists of caseswhere mosaicism is suspected in one of the asymptomaticparents, because of the transmission of the abnormal alleleto more than one offspring (Tables 1, 2), and the othergroup includes individuals in whom the clinical symptoms

Introduction

There have been several clinical descriptions of families inwhich more than one child affected with a dominant disor-der has been born to unaffected parents with no family his-

are caused by a somatic mosaicism and who have had atleast one child who has received the mutation and is clini-cally affected (Table 3).

Among 38 cases of germline mosaicism that were sus-pected clinically and from whom molecular data wereavailable in 19 (50%), the mosaicism was also somatic: 14out of 23 cases of autosomal dominant disorders (Table 1)and 5 out of 15 cases of X-linked disorders (Table 2). Insome of the cases, the mosaicism was sought only bySouthern blots, a technique that is not sensitive enough todetect low levels of mutations. However, if we consideronly those cases in which the detection involved poly-merase chain reaction (PCR) techniques, somatic mosa-icism was still detected in approximately half of the cases.This is probably related to the assumption that, in order tobe present in the germ cells, a mutation in somatic cellsmust occur very early, before the commitment to germcells occurs, and must therefore be present in a relativelyhigh percentage of the somatic cells. A possible underesti-mation of the frequency of mosaicism in somatic cells mayarise because the degree of mosaicism is often different indifferent tissue, and it may not be present in the tissue ex-amined. Some stochastic events or disadvantage to thecells that carry the mutation may lead to the absence of themutation in white blood cells. Many cytogenetic exampleshave been reported in which the mosaicism was not detect-ed in white blood cells, but only in fibroblasts. Therefore,the examination of several different tissues may increasethe ability to detect mosaicism and, indeed, in two cases in

which a mutation was not found in white blood cells, themutation was revealed in other tissues, such as muscle,buccal smear, or hairs (Putman et al. 1997; Voit et al.1992). Since, in all of the cases with germ line mosaicismin which a somatic mutation was not detected, it wassought only in leukocytes (Tables 1, 2), the percentage ofsomatic mosaicism may have been underestimated.

In 14 cases, the clinical features present in the patientwere secondary to a somatic mosaicism and, since the pa-tient transmitted the mutation to his progeny, it was con-cluded that the mosaicism was also present in the germcells (Table 3). In most cases, the disorder present in themosaicism carrier was milder than that present in the off-spring. In some families, the symptoms present in the par-ent and child were so different that they were consideredas being clinically affected by different syndromes. Awoman affected with Stickler syndrome gave birth to achild with Kniest dysplasia. Molecular analysis demon-strated that she had a mosaicism for a COL2A1 mutation,whereas her child had the mutation in all its cells (Winter-pacht et al. 1993).

Commitment of the germ cells

The majority of cells in the blastocyst contributes to extra-embryonic tissue and it is estimated that only three cellswithin the inner cell mass are committed to form the em-

382

Gene Type of mutation Number, sex Somatic tissues Germline Reference

COMP Small deletion 1 M,1F b+ (h- in one) (PCR) Not done Ferguson et al. 1997

FBN2 Point mutation 1M b-, bs+, h+, (PCR) Not done Putnam et al. 1997

FGFR3 Point mutation 1F NM+ (PCR) Not done Stoilov et al. 1995

FSHD1A Rearrangements 3 F, 1 M b+ (SB) Not done Kohler et al. 1996

Rearrangements 3? b- (SB) Not done Weiffenbach et al. 1993

Griggs et al. 1993

Hb Puttelange Point mutation 1M b- (PCR) Not done Wajcman et al. 1995

NF1 Deletion 1M b- (PCR) s:10% Lazaro et al. 1994

TSC2 4-bp insertion 1F b- (PCR) Not done Yates et al. 1997

RB Mutation 3? b- (PCR) s+ in one Sippel et al. 1997

COL1A1 Point mutation 1M b+, h+, (PCR) s: 20% Cohn et al. 1990

Point mutation 1F b+, f+, (PCR) Not done Bonaventure et al. 1992

Point mutation 1M b+, f+, (PCR) s: 43% Namikawa et al. 1995

COL2A1 Point mutation 1M b+, f+, (SB) s: 40% Edwards et al, 1992

Point mutation 1M b+, f+, bs+, (PCR) Not done Raghunath et al. 1995

Point mutation 1M b+ (PCR) s: 40% Lund et al. 1996

Table 1 Autosomal dominant disorders in which the parents havemore than one affected offspring and are asymptomatic or very mild-ly affected (M male, F female, ? sex unknown, SB Southern blot,PCR polymerase chain reaction, + present, – absent b blood, f fibro-

blasts, l liver, h hair, bs buccal smear, NM not mentioned, Not doneno examination performed in the germ cells, s percentage of the mo-saicism in the sperm cells

bryo. All the available evidence places germ cell commit-ment at an early stage in development, but the exact timingis unknown, as is the number of cells ancestral to the germcells. If there is only one progenitor cell and if it carries amutation, then all the germ cells will carry it. The risk ofthe offspring inheriting the mutation will thus be 50%, andall the carriers of the same intragenic haplotype will alsobe carriers of the mutation. The data obtained from studiesin mice embryos and in humans demonstrate that there ismore than one ancestral germ cell. In mice, Soriano andJaenisch (1986) have shown, using retroviruses as probes,that the cells that give rise to the embryo intermingle ex-tensively before the final tissue allocation and that the so-matic lineages are derived from at least 8 founder cells.

They also suggest that three or more ancestral cells formthe germ cells and are set aside prior to tissue allocation.In humans, the affected mother of a patient with neurofi-bromatosis type 2 had a somatic and germline mosaicismfor the mutation and transmitted the same haplotype to hertwo children, but only one had the mutation (Bijlsma et al.1997). Other similar examples have been reported indicat-ing that the mosaicism is present in the progenitors of thegerm cells (Sommer et al. 1995). Another indication thatthere is more than one ancestral germ cell is that, in all themales examined who had a mosaicism in their somatic andgermline cells, the mutation was found in less than 50% ofthe spermatozoids (Tables 1, 3). In these males, the muta-tion has been found in at least 1:8 of the sperm cells, sug-

383

Table 2 X-linked disorders; asymptomatic germline mosaicism with or without somatic mosaicism.(SB Southern blot, PCR polymerasechain reaction, M male, F female, b blood, m muscle, + present, – absent)

Gene Number sex Mutation Somatic tissue Germ line References

IL-2RG 1 F Point mutation b- (PCR) Not done Puck et al. 1995

Androgen receptor 1 F Point mutation b+ (PCR) Not done Boehmer et al. 1997

Hunter 1 F Point mutation b+ (PCR) Not done Froissart et al. 1997

DMD 2 F Deletion b- (SB) Not done Bakker et al, 1987

2 F Deletion b- (SB) Not done Bakker et al. 1989

1 F Deletion b+ (SB) Not done Bakker et al. 1989

2 F Deletion b- (SB) Not done Darras et al. 1988

1 F Point mutation b- (PCR) Not done Wilton et al. 1994

1 F Deletion b- (PCR) Not done Bullock et al. 1996

1 F Deletion b-, m+ Not done Voit et al. 1992

Factor VIII 1 F Deletion b- (SB) Not done Levinson et al. 1990

1 F Deletion b+ (SB) Not done Gitschier 1988

Table 3 Symptomatic somatic and germline mosaicism; autosomal dominant and X-linked disorders (M male, F female, b blood, f fibro-blasts, h hair, l liver)

Gene Mosaic carrier, sex (tissue examined) Affected child Reference

DMD Duchenne mild , 1 M (b) Duchenne muscular dystrophy Lebo et al. 1990

Factor VIII Mild hemophilia, 1 M (b, l) Hemophilia Taylor et al. 1991

K10 3 families epidermal nevus, 2 F, 1 M (f) Epidermolytic hyperkeratosis Paller et al. 1994

NF2 Neurofibromatosis type 2, 1F (b) Neurofibromatosis type 2 Bijlsma et al. 1997

TSC2 Tuberous sclerosis mild, 1 M (b) Tuberous sclerosis Verhoef et al. 1995

RB Retinoblastoma bilateral, 1 M (b) Retinoblastoma Greger et al. 1990

COL1A1 Mild ostogenesis imperfecta, 1F (b, f) Ostogenesis imperfecta type II Constantinou-Deltas et al. 1993

Mild ostogenesis imperfecta, 1F (f) Ostogenesis imperfecta type II Constantinou-Deltas et al. 1993

Mild ostogenesis imperfecta, 1F (f) Ostogenesis imperfecta type II Cohen-Solal et al. 1996

Mild ostogenesis imperfecta, 1 M (b, f) Ostogenesis imperfecta type II Wallis et al. 1990

COL2A1 Stickler, 1F (b, f, h) Kniest dysplasia Winterpacht et al. 1993

Mild spondyloepiphyseal dysplasia, 1M Kniest dysplasia Winterpacht et al. 1994

gesting that there are at least 4 ancestral germ cells in hu-mans (Cohn et al. 1990).

Germ cell mosaicism

In germ cells, if a mutation occurs before meiosis, it willalmost always be present as a mosaicism and, dependingon the stage at which it occurs, may be present in very fewto up to almost 50% of the germ cells. It is probable that,since human families are usually small, in most cases inwhich a mutation is transmitted to more than one child, themutation is present in a significant number of germ cellsand therefore must have occurred very early after a few di-visions of the committed germ cells. Similarly, since thecommitment to germ cells occurs early in development,only mutations that occurred before that period will bepresent in the germline. Therefore, in both type of mosa-icism, mutations must arise in the early embryo after asmall number of division of the respective type of cell. Ifthere is, as is thought, a relationship between the risk of amutation occurring and the number of cell divisions, thenthe frequency of the two types of events should be in asimilar range. The data from the literature (summarized inTables) are in accord with these expectations. When a mo-saicism in the germ line was diagnosed because of thebirth of more than one affected child, the mutation was al-so present in somatic cells in 50% of the cases.

In many genetic disorders caused by point mutations,such as achondroplasia or Apert syndrome, new mutationsare almost always of paternal origin (Moloney et al. 1996;Szabo et al. 1997). These observations are consistent withthe advanced paternal age observed among the parents ofsporadic cases of these disorders and are probably relatedto the increasing number of cell divisions. Since mutations

causing mosaicism occur in the early embryo, there shouldbe no differences among the sexes and, of course, no pa-ternal age effect should be detected. No data are availableon parental age but, among the carriers of somatic mosa-icism of autosomal dominant disorders, the sex ratio is in-deed not significantly different from 1:1 (15 males and13 females; Tables 1, 3). There are too few data on the car-riers of germ line mosaicism only; however, there was onefemale among the three carriers for whom the sex wasknown (Table 1).

Frequency of germ line mosaicism

Studies based on clinical and biochemical data have esti-mated that, in at least 6% of the cases of lethal osteogene-sis imperfecta type II, one of the parents of the affectedchild is a carrier of a germ cell mosaicism (Byers et al.1988). Lately, the origin of sporadic cases has been deter-mined in various X-linked and autosomal dominant disor-ders. The results are presented in Table 4, and it can beseen that two groups of diseases are distinguished: group 1in which somatic and germline mosaicism in one of theunaffected parents is relatively frequent, and group 2 inwhich mosaicism is almost never found. In group 1, so-matic mosaicism was found in 11%–20% of the cases in alarge series of sporadic patients with hemophilia A and B,Duchenne muscular dystrophy, or facioscapulohumeraldystrophy. In group 2, in large series of patients withachondroplasia, Apert syndrome or MEN2, not a singlecase of mosaicism was detected among the parents of spo-radic patients in which the mutation was known. Thesemolecular data parallel the clinical observations: ingroup 1, many families were reported in which more thanone affected child was born to unaffected parents, whereas

384

Table 4 Somatic and germline mosaicism among asymptomatic parents of isolated cases

Disorder (gene) Type of mutation Proportion of Referencesmosaicism

Facioscapulohumeral dystrophy (FSHD1A) Rearrangements 11/57 Kohler et al. 1996

Hemophilia (Factor VIII) All types 3/16 Becker et al. 1996

1/4 Levinson et al. 1990

Christmas disease (Factor IX) All types 5/45 Sommer and Ketterling 1996

Duchenne muscular dystrophy (DMD) Deletions 15%a Passos Bueno et al. 1992

Duchenne muscular dystrophy (DMD) Deletions 4/9 Robinson et al. 1995

Apert (FGRF2) Point mutations 0/55 Moloney et al. 1996

Achondroplasia (FGRF3) Point mutations 0/91 Bellus et al. 1995

MEM 2 B (RET) Point mutations 0/14 Carlson et al. 1994

MEN 2A (RET) Point mutations 0/8 Schuffenecker et al. 1997

Osteogenesis imperfecta 6%a Byers et al. 1988

a Germinal mutations, the exact type of which was not determined

in group 2, such reports were rare. One possibility that hasbeen proposed to explain these differences is a differencein the type of mutations causing these diseases. Deletionsare present in most cases of Duchenne muscular dystro-phy, with somatic and germline mosaicism and chromo-some rearrangements in those with facioscapulohumeraldystrophy. In Duchenne muscular dystrophy, a higher fre-quency of proximal gene deletions has been found amonggerm cells mosaic cases, and it has been proposed that dif-ferent types of deletions occur at different periods of de-velopment (Passos-Bueno et al. 1992). Proximal deletionsmost probably occur early in embryonic development andhave a higher chance of being present than somatic andgermline mosaicism. Distal deletions occur later and areoften found in nonfamilial isolated cases of Duchennemuscular dystrophy. Therefore, according to the site of thedeletion, the recurrence risk may be high, i.e., approxi-mately 30% in proximal deletions, or relatively low, i.e.,4% in distal deletions. However, the differences in the typeof mutations cannot be the only explanation for the differ-ences in the occurrence of somatic and germline mosa-icism, since single base substitutions have also been asso-ciated with somatic and germline mosaicism. Not enoughdata are available to draw conclusions, even though possi-ble mechanisms have been proposed that may be related tothe different types of mutations (Sommer and Ketterling1996). Another possibility that may explain some of thedifferences in the frequency of the somatic and germlinemosaicism between the two groups of disorders is that thedifferences reflect differences in the clinical expression ofmosaicism for the different genes (Zlotogora 1993). Thosecases in which the carriers of the mosaicism are asymp-tomatic are discovered only because they are the parents ofsporadic patients, whereas those in whom the somatic mo-saicism is symptomatic will be included among patientswho appear to be sporadic. Therefore, although some ofthe parents of sporadic patients with some diseases willhave a somatic and germline mosaicism, somatic mosa-icism will appears to be very rare in others. Recently, be-cause of the improved ability to detect mosaicism, it hasindeed been demonstrated that some of the patients affect-ed with a genetic disease apparently because of a ”newmutation” in one of the parents are carriers of a somaticmosaicism.

Conclusions

New mutations may be present as mosaicisms and, with abetter ability to detect them, it seems that their frequencyis relatively high. Depending on various factors, such asthe gene involved and/or the degree of mosaicism, theclinical effect of the mosaicism may be different in differ-ent disorders. In some diseases, such as hemophilia or os-teogenesis imperfecta, new mutations are often present asasymptomatic somatic and germline mosaicism in one ofthe parents of the propositus. In other disorders, such asneurofibromatosis, somatic mosaicism is very rare in the

parents of the propositus, perhaps since such a mosaicismcauses clinical symptom. These differences are particular-ly important for genetic counseling, since after the birth ofthe propositus, the risk of having another affected childmay be low, as in Apert syndrome (less than 1%), moder-ate as in osteogenesis imperfecta (6%), or high as in proxi-mal deletions of the Duchenne muscular dystrophy gene(30%).

References

Bakker E, Broeckhoven C van, Bonten EJ, Vooren MJ van de, Veen-ema H, Hul W van, Den Dunnen JT, Ommen GJB van, Vanden-berghe A, Pearson PL (1987) Germ cell mosaicism and Duchen-ne muscular dystrophy mutations. Nature 329:554–558

Bakker E, Veenema H, Den Dunnen JT, Broeckhoven C van,Grootscholten PM, Bonten EJ, Ommen GJB van, Pearson PL(1989) Germinal mosaicism increases the recurrence risk for”new” Duchenne muscular dystrophy mutations. J Med Genet26:553–559

Becker J, Schwaab R, Moller-Taube A, Schwaab U, Schmidt W,Brackmann HH, Grimm T, Olek K, Oldenburg J (1996) Charac-terization of the factor VIII defect in 147 patients with sporadichemophilia A: family studies indicate a mutation type dependentsex ratio of mutation frequencies. Am J Hum Genet 58:657–670

Bellus GA, Hefferon TW, Ortiz de Luna RI, Hecht JT, Horton WA,Machado M, Kaitila I, McIntosh I, Francomano CA (1995):Achondroplasia is defined by recurrent G380R mutations inFGFR3. Am J Hum Genet 56:368–373

Bijlsma EK, Wallace AJ, Evans DGR (1997) Misleading linkage re-sults in an NF2 presymptomatic test owing to mosaicism. J MedGenet 34:934–936

Boehmer ALM, Brinkmann AO, Niermeijer MF, Bakker L, HalleyDJJ, Drop SLS (1997) Germ-line and somatic mosaicism in theandrogen insensitivity syndrome: implications for genetic coun-seling. Am J Hum Genet 60:1003–1006

Bonaventure J, Cohen-Solal L, Lasselin C, Maroteaux P (1992) Adominant mutation in the COL1A1 gene that substitutes glycinefor valine causes recurrent lethal osteogenesis imperfecta. HumGenet 89:640–646

Bullock S, Felix C, Iskander-Gabra S, Davison V (1996) Detectionof germinal mosaicism in a DMD family. Biochem Soc Trans24:273S

Byers PH, Tsipouras P, Bonadio F, Starman BJ, Schwartz RC (1988)Perinatal lethal osteogenesis imperfecta (OI type II): a biochemi-cally heterogeneous disorder usually due to new mutations in thegenes for type I collagen. Am J Hum Genet 42:237–248

Carlson KM, Dou S, Chi D, Scavarda N, Toshima K, Jakson CE,Wells SA, Goodfellow PJ, Donis-Keller H (1994) Single missen-ce mutation in the kinase catalytic domain of RET protooncoge-ne is associated with multiple endocrine neoplasia type 2B. ProcNatl Acad Sci USA 91:1579–1583

Cohen-Solal L, Bonaventure J, Maroteaux P (1991) Dominant muta-tions in familial lethal and severe osteogenesis imperfecta. HumGenet 87:297–301

Cohn DH, Starman BJ, Blumberg B, Byers PH (1990) Recurrence oflethal osteogenesis imperfecta due to parental mosaicism for adominant mutation in a human type I collagen gene (COL1A1).Am J Hum Genet 46:591–601

Colman SD, Rasmussen SA, Ho VT, Abernathy CR, Wallace MR(1996) Somatic mosaicism in a patient with neurofibromatosistype I. Am J Hum Genet 58:484–490.

Constantinou-Deltas CD, Ladda RL, Prockop DJ (1993) Somatic cellmosaicism: another source of phenotypic heterogeneity in nucle-ar families with osteogenesis imperfecta. Am J Med Genet45:246–251

385

Darras BT, Blattner P, Harper JF, Spiro AJ, Alter S, Franke U (1988)Intragenic deletions in 21 Duchenne muscular dystrophy (DMD)Becker muscular dystrophy (BMD) families studied with the dys-trophin cDNA: location of breakpoints on HindIII and BglII ex-on-containing fragment maps, meiotic and mitotic origin of themutations. Am J Hum Genet 43:620–629

Edwards MJ, Wenstrup RJ, Byers PH, Cohn DH (1992) Recurrenceof lethal osteogenesis imperfecta due to parental mosaicism for amutation in the COL1A2 gene of type I collagen. The mosaicparent exhibits phenotypic features of a mild form of the disease.Hum Mutation 1:47–54

Ferguson HL, Deere M, Evans R, Rotta J, Hall JG, Hecht JT (1997)Mosaicism in pseudoachondroplasia. Am J Med Genet70:287–291

Froissart R, Maire I, Bonnet V, Levade T, Bozon D (1997) Germ-line and somatic mosaicism in a female carrier of Hunter disease.J Med Genet 34:137–140

Gitschier J (1988) Maternal duplication associated with gene dele-tion in sporadic hemophilia. Am J Hum Genet 43:274–279

Greger V, Passarge E, Horsthemke B (1990) Somatic mosaicism in apatient with bilateral retinoblastoma. Am J Hum Genet46:1187–1193

Griggs RC, Tawil R, Storvick D, Mendell JR, Altherr MR (1993)Genetics of facioscapulohumeral muscular dystrophy: new muta-tions in sporadic cases. Neurology 43:2369–2372

Hall JG (1988) Somatic mosaicism: observations related to clinicalgenetics. Am J Hum Genet 43:355–363

Knoll A, Ketterling RP, Sommer SS (1996) Absence of somatic mo-saicism in 17 families with hemophilia B: an analysis with a sen-sitivity 10- to 1000-fold greater than that of sequencing gels.Hum Genet 98:539–545

Kohler J, Rupilius B, Otto M, Bathke K, Koch MC (1996) Germ-line mosaicism in 4q35 facioscapulohumeral muscular dystrophy(FSHD1A) occurring predominantly in oogenesis. Hum Genet98:485–490

Lazaro C, Gaona A, Lynch M, Kruyer H, Ravella A, Estevill X(1995) Molecular characterization of the breackpoints of a 12-kbdeletion in the NF1 gene in a family showing germ-line mosa-icism. Am J Hum Genet 57:1044–1049

Lebo RV, Olney RK, Golbus MS (1990) Somatic mosaicism at theDuchenne locus. Am J Med Genet 37:187–190

Levinson B, Lehesjoki A-E, Chapelle A de la, Gitschier J (1990)Molecular analysis of hemophilia A mutations in the Finnishpopulation. Am J Hum Genet 46:53–62

Lund AM, Schwartz M, Raghunath M, Steinmann B, Skovby F(1996) Gly802Asp substitution in the pro α (1) collagen chain ina family with recurrent osteogenesis imperfecta due to paternalmosaicism. Eur J Hum Genet 4:39–45

Moloney DM, Slaney SF, Oldridge M, Wall SA, Sahlin P, StenmanG, Wilkie OM (1996) Exclusive paternal origin of new mutationsin Apert syndrome. Nat Genet 13:48–53

Namikawa C, Suzumori K, Fukushima Y, Sasaki M, Hata A (1995)Recurrence of osteogenesis imperfecta because of paternal mosa-icism: Gly862Ser substitution in a type I collagen gene(COL1A1). Hum Genet 95:666–670

Paller AS, Syder AJ, Chan Y-M, Yu Q-C, Hutton E, Tadini G, FuchsE (1994) Genetic and clinical mosaicism in a type of epidermalnevus. N Engl J Med 331:1408–1415

Passos-Bueno MR, Bakker E,Kneppers ALJ, Takata RI, Rapaport D,Den Dunnen JT, Zatz M, Ommen GJB van (1992) Differentmosiacism frequencies for proximal and distal Duchenne muscu-lar dystrophy (DMD) mutations indicate differences in etiologyand recurrence risk. Am J Hum Genet 51:1150–1155

Puck JM, Pepper AE, Bedard P-M, Laframboise R (1995) Femalegerm line mosaicism as the origin of a unique IL-2 receptor γ-chain mutation causing X-linked severe combined immunodefi-ciency. J Clin Invest 95:895–899

Putnam EA, Park E-S, Aalfs CM, Hennekam RCM, Milewicz DM(1997) Parental somatic and germ-line mosaicism for a FBN2mutation and analysis of FBN2 transcript levels in dermal fibro-blasts. Am J Hum Genet 60:818–827

Raghunath M, Mackay K, Dalgleish R, Steinnmann B (1995) Genet-ic counselling on brittle grounds: recurring osteogeneis im-perfecta due to parental mosaicism for a dominant mutation. EurJ Pediatr 154:123–129

Robinson DO, Bunyan DJ, Connarty M, Crolla JA (1995) Somaticmosaicism in female carriers of Duchenne muscular dystrophy.Am J Hum Genet 57:A157

Schuffenecker I, Ginet N, Goldgar D, Eng C, Chambe B, Boneu A,Houdent C, Pallo D, Schlumberger M, Thivolet C, Lenoir GM(1997) Prevalence and parental origin of de novo RET mutationsin multiple endocrine neoplasia type 2A and medullary thyroidcarcinoma. Am J Hum Genet 60:233–237

Sippel KC, Fraioli RE, Smith GD, Schalkoff ME, Dryja TP (1997)Frequent somatic and germline mosaicism in retinoblastoma: rel-evance to genetic counseling. Am J Hum Genet 61:A16

Sommer SS, Ketterling RP (1996) The factor IX gene as a model foranalysis of human germline mutations: an update. Hum MolGenet 5:1505–1514

Sommer SS, Knoll A, Greenberg CR, Ketterling RP (1995) Germ-line mosaicism in a female who seemed to be a carrier by se-quence analysis. Hum Mol Genet 4:2181–2182

Soriano P, Jaenisch R (1986) Retroviruses as probes for mammaliandevelopment: allocation of cells to the somatic and germ cell lin-eages. Cell 46:19–29

Stoilov I, Kilpatrick MW, Tsipouras P (1995) A common FGFR3gene mutation is present in achondroplasia but not in hypo-chondroplasia. Am J Med Genet 55:127–133

Szabo JK, Wilkin DJ, Cameron R, Henderson S, Bellus G, Kaitila I,Loughlin J, Munnich A, Sykes B, Bonaventure J, FrancomanoCA (1997) The achondroplasia mutation occurs exclusively onthe paternally derived fibroblast growth factor receptor 3(FGFR3) allele. Am J Hum Genet 61:A348

Taylor SAM, Deugau KV, Lillicrap DP (1991) Somatic mosaicismand female to female transmition in a kindred with hemophilia B(factor IX deficiency). Proc Natl Acad Sci USA 88:39–42

Verhoef S, Vrtel R, Essen T van, Bakker L, Sikkens E, Halley D,Lindhout D, Ouweland A van den (1995) Somatic mosaicism andclinical variation in tuberous sclerosis complex. Lancet 345:202

Voit T, Neuen-Jacob E, Mahler V, Jauch A, Cremer M (1992) Somat-ic mosaicism for a deletion of the dystrophin gene in a carrier ofBecker muscular dystrophy. Eur J Pediatr 151:112–116

Wajcman H, Girodon E, Prome D, North ML, Plassa F, Duwig I, Ki-ster J, Bergerat JP, Oberling F, Lampert E, Lonsdorfer J, Goos-sens M, Galacteros F (1995) Germ cells mosaicism for an alanineto valine substitution at residue 140 in hemoglobin Puttelange, anew variant with high oxygen affinity. Hum Genet 96:711–716

Wallis GA, Starman B, ZinnAB, Byers PH (1990) Variable expres-sion of osteogenesis imperfecta in a nuclear family is explainedby somatic mosaicism for a lethal point mutation in the 1α (1)gene (COL1A1) of type I collagen in a parent. Am J Hum Genet46:1034–1040

Weiffenbach B, Dubois J, Storvick D, Tawil R, Jacobsen SJ, GilbertJ, Wijmenga C, Mendell JR, Winokur S, Altherr MR, Schultz P,Olandt S, Frants RR, Periak-Vance M, Griggs RC (1993) Map-ping the facioscapulohumeral muscular dystrophy gene is com-plicated by chromosome 4q35 recombination events. Nat Genet4:165–169

Wilton SD, Chandler DC, Kakulas BA, Laing NG (1994) Identifica-tion of a point mutation and germinal mosaicism in a Duchennemuscular dystrophy family. Hum Mutat 3:133–140

Winterpacht A, Hilbert M, Schwarze U, Mundlos S, Spranger J,Zabel B (1993) Kniest and Stickler dysplasia phenotypes causedby collagen II gene (CLO2A1) defect. Nat Genet 3:323–326

Winterpacht A, Hilbert M, Schwarze U, Mundlos S, Spranger J,Zabel B (1994) Autosomal dominant spondylarthropathy due to atype II procollagen gene (COL2A1) point mutation. Hum Mutat4: 257–262

Yates JRW, Bakel I van, Sepp T, Payne SJ, Webb DW, Nevin NC,Green AJ (1997) Female germline mosaicism in tuberous sclero-sis confirmed by molecular genetic analysis. Hum Mol Genet6:2265–2269

Zlotogora J (1993) Mutations in von Recklinghausen neurofibroma-tosis. Am J Med Genet 46:182–184

386