SHORT COMMUNICATION

A second case of intrauterine growth retardation and primaryhypospadias associated with a trisomy 22 placenta but withbiparental inheritance of chromosome 22 in the fetus

Jennifer Bryan1*, Michelle Peters1, Gary Pritchard2, Sue Healey3 and Diane Payton4

1Cytogenetics Division, Mater Laboratory Services, Mater Misercordiae Adult Hospital, Brisbane, Australia2Brisbane Ultrasound for Women, Wickham Terrace, Brisbane, Australia3Molecular Genetics Laboratory, Royal Brisbane Hospital, Brisbane, Australia4Anatomical Pathology Laboratory, Mater Laboratory Services, Mater Misercordiae Adult Hospital, Brisbane, Australia

We report a case of severe intrauterine growth retardation (IUGR) and hypospadias in association withtrisomy 22 diagnosed following chorionic villus sampling (CVS). Subsequent analysis of amniotic fluidcultures showed a normal male karyotype, 46,XY. As a previous case had been reported with similarabnormalities, in association with maternal uniparental disomy (UPD) 22, molecular studies were alsoperformed. Microsatellite marker studies showed biparental inheritance. Follow-up studies after deliveryshowed a normal cell line in lymphocytes with the trisomy appearing to be confined to the placenta. The presentcase concurs with other earlier reports that maternal UPD for chromosome 22 has no impact on the phenotype.The features seen in the fetus are most likely the result of placental dysfunction due to trisomy, tissue-specificmosaicism and/or the effects of local growth restriction. Copyright # 2002 John Wiley & Sons, Ltd.

KEY WORDS: trisomy 22; CVS; IUGR; mat UPD (22); CPM

INTRODUCTION

Trisomy 22 is common in spontaneous abortions withan incidence of 2.26% (Pflueger, 1999). It is rarer inlive-borns and is associated with a severe phenotypeand a shortened lifespan with the longest reportedsurvival being 3 years (Kukolich et al., 1989). Thetrisomy appears to be compatible with a longerlifespan only in the mosaic form and a well-definedphenotype has been established including features suchas microcephaly, abnormal ears, webbed neck, cardiacabnormalities, long fingers and growth retardation(Hsu et al., 1997).

Trisomy for chromosome 22 confined to theplacenta (confined placental mosaicism, CPM) hasalso been reported in association with poor fetaloutcome (Wolstenholme, 1996). CPM, with the cyto-genetic abnormality (most often trisomy) confined tothe placenta, occurs in approximately 2% of via-ble pregnancies studied by CVS at 9 to 11 weeks(Kalousek and Vekemans, 1996). Preliminary investi-gations for other chromosomes suggest a similarcorrelation between intrauterine growth retardation(IUGR) and placental trisomy, implicating placentalinsufficiency due to trisomy for many chromosomes(Ledbetter and Engel, 1995).

Recent evaluation of the mechanisms leading to

CPM has demonstrated that most cases of mosaictrisomy 22 occur following trisomic rescue (Robinsonet al., 1997). Trisomic zygote rescue is believed tobe a predominant mechanism leading to constitu-tional uniparental disomy (UPD) (Shaffer et al., 1998;Robinson, 2000). If loss of one of the three chromo-somes from a trisomic cell occurs randomly, then UPDwould be expected in the derivative diploid lineageone-third of the time. In many cases the trisomiclineage is predominantly or exclusively confined to theplacenta and it is not uncommon to observe 100%trisomy in placental tissue with complete absence ofthe trisomy in any fetal tissue (Robinson, 2000).

While both maternal and paternal UPD forchromosome 22 have been widely documented ashaving no phenotypic effect (Palmer et al., 1980;Schinzel et al., 1994; Wang, 1999), recently a case byBalmer et al. (1999) raised an association betweenCPM for trisomy 22 and maternal UPD (22) in a childwith IUGR and hypospadias.

We report a second case of trisomy 22 confined tothe placenta and associated with a phenotype ofIUGR and primary hypospadias but with biparentalinheritance.

CASE HISTORY

A 22-year-old G1P0 Caucasian female presented forroutine high-resolution ultrasonography at 18 weeks’gestation. A single fetus with IUGR was noted. At18.3 weeks’ gestation by dates the fetus appeared the

*Correspondence to: J. Bryan, Cytogenetics Division, MaterLaboratory Services, Mater Misercordiae Adult Hospital, RaymondTerrace, South Brisbane, 4101, Queensland, Australia.E-mail: 17patah@mater.org.au

PRENATAL DIAGNOSIS

Prenat Diagn 2002; 22: 137–140.DOI: 10.1002 /pd.260

Copyright # 2002 John Wiley & Sons, Ltd. Received: 9 April 2001Revised: 27 August 2001

Accepted: 10 September 2001

size of a 16–17-week fetus. Other findings includedoligohydramnios and an enlarged echogenic placenta.Amniotic fluid and chorionic villi were collected forkaryotyping.

Serial ultrasound monitoring throughout the restof the pregnancy showed continued IUGR and at 31weeks’ gestation just prior to delivery the estimatedfetal weight was <3rd percentile (Figure 1). In parti-cular, the femoral length parameters were the equi-valent of a 24.5-week fetus. At this point amnioticfluid volumes were normal and the only other ultra-sound finding was the presence of hypospadias.

Following delivery at 32 weeks’ gestation, due tosome evidence of vascular redistribution to thecerebral circulation, IUGR and primary hypospadiaswere confirmed. All other parameters were normal.Continued follow-up has shown the baby to bemeeting targets and at 9 months of age appears tobe the correct weight and size for age with acompletely normal phenotype.

CYTOGENETIC ANALYSIS

Due to the necessity for a rapid result, and receipt ofonly a small amount of chorionic villi, only a directand a short-term (overnight) culture were initiated,according to standard protocols. GTG-banded analy-sis of 14 cells from these cultures showed a male fetuswith trisomy 22 in all cells. The amniotic fluid wascultured and harvested by standard protocols. GTG-banded analysis of 21 colonies from the two primary insitu cultures of amniotic fluid showed a normal malekaryotype with no evidence of trisomy 22.

Follow-up studies after delivery were performed onchorionic villi, cord, and peripheral blood accordingto standard protocols. Long-term flask cultures of

chorionic villi showed 100% of cells with trisomy 22 intwo of the three sites sampled, with one cultureunsuccessful. The cord sample also failed to grow.Peripheral blood cultures showed a normal karyotypein all of the 100 cells analyzed (Table 1).

MOLECULAR ANALYSIS

Genotypes for eight microsatellite markers (D22S420,D22S539, D22S315, D22S275, D22S280, D22S283,D22S43, and D22S274) mapping to chromosome 22were determined using radioactively labeled PCRanalysis according to standard protocols. The amnio-cytes demonstrated biparental inheritance for threemarkers: D22S315, D22S283, and D22S274 (Table 2and Figure 2).

Following delivery, DNA from the cultured chor-ionic villus sample (CVS) was tested in tandem withthe amniocytes. Comparison of band intensitiesbetween the amniocytes and the CVS in markerD22S283 suggested that the additional chromosome22 in the CVS was paternal in origin and isodisomic innature (Figure 2). Although this methodology is notquantitative, the same dosage effect was apparentupon repeating the analysis and did not appear to beeither a PCR or gel electrophoresis artefact. MarkersD22S539, D22S315, and D22S274 also showed pat-terns consistent with paternal isodisomy although thedosage effect was not as obvious.

PATHOLOGY FINDINGS

Analysis of the placenta post delivery showed accel-erated villus maturation as well as prominent areas ofintermediate trophoblasts throughout the placental

Figure 1 — Fetal weight parameters from 15 weeks’ gestation to delivery (proband HB)

J. BRYAN ET AL.138

Copyright # 2002 John Wiley & Sons, Ltd. Prenat Diagn 2002; 22: 137–140.

cake. The significance of this, apart from possiblyrelating to the abnormal formation of the placentaassociated with a chromosome abnormality, was notclear.

DISCUSSION

The finding of trisomic cells in both cytotropho-blast and mesenchymal cells of the placenta butnot in amniocytes is consistent with CPM Type III(Robinson et al., 1997) and is associated with anincreased risk of spontaneous abortion or IUGR(Ledbetter and Engel, 1995).

CPM may arise from a diploid zygote by a mitoticerror confined to placental progenitors resulting in anabnormal line in the placenta. The fetus arising from adiploid zygote will almost always inherit a homologuefrom each parent (biparental disomy). AlternativelyCPM can result from a trisomic zygote with rescue ofthe embryonic progenitors and some of the placentaby mitotic loss. Theoretically one-third of embryoswould inherit both homologues from a single parent(UPD)(Hansen et al., 1997). Trisomic zygote rescueresulting in fetal UPD is associated with CPM TypeIII involving non-mosaic trisomies (Los et al., 1998)and recent evaluation of the mechanisms leading toCPM has demonstrated that most cases of mosaictrisomy 22 occur following trisomic rescue (Berghellaet al., 1998). The number of abnormal cells and theirtissue distribution are dependent on the timing and thetissue compartment in which rescue occurs.

The main effect of UPD on the prenatal or post-natal development is dependent on the presence ofimprinted genes carried by the involved chromosomalpair. Imprints are defined as the epigenetic modifica-tion that distinguishes the two parental alleles (Sleutelset al., 2000). As a result the same gene may functiondifferently depending on whether it is maternal or

paternal in origin (Kalousek and Barrett, 1994; Wang,1999).

CPM has been reported in association with IUGRand confined chorionic trisomy 22 although mosaicismfor trisomy 22 in fibroblast cells or a very weakmosaicism in lymphocyte cells cannot be excluded ascontributing to the phenotype. Distinguishing theclinical effects due to trisomy mosaicism from thosedue to UPD can be difficult as trisomy that is confinedto the placenta can cause IUGR and/or fetal demise. Itis also impossible to exclude the presence of low levelsof trisomy in the fetus itself as trisomic cells may beconfined to one tissue type only (Robinson, 2000) orbe below levels detectable in the laboratory.

It is possible that the IUGR in these pregnanciesrelates to the level of mosaicism in the placenta and isnot due to UPD in the fetus. The clinical implicationsof UPD on fetal growth and development remainunclear (Berghella et al., 1998). The phenotypic effectof mosaic or partial trisomy for chromosome 22 wouldmost likely overshadow any potential phenotypiceffect due to UPD (22) (Balmer et al., 1999).

Previous reported cases of maternal UPD (22) havesuggested that there is no impact on the phenotype(Palmer et al., 1980; Schinzel et al., 1994). Schinzelet al. (1994) reported a case of a normal probandidentified due to multiple miscarriage in his wife, with

Table 1 — Summary of cytogenetic results

Tissue type Karyotype Cells analyzed (n)

Chorionic villi direct/short-term cultures 47,XY,+22 14Chorionic villi long-term cultures 47,XY,+22 30Amniotic fluid in situ cultures 46,XY 21Cord NA Culture failedPeripheral blood 46,XY 100

NA, Not applicable.

Table 2 — Results of polymerase chain reaction (PCR)analysis of the amniocytes, chorionic villi and parentsshowing informative markers on chromosome 22a

Marker Maternal Amniocytes CVS Paternal

D22S539 2 2 2 2 2 2 1 2D22S315 2 4 3 4 3 4 1 3D22S283 1 3 1 2 1 2 2 3D22S274 3 4 1 3 1 3 1 2

aMarkers suggestive of isodisomy are in bold for CVS.

Figure 2 — Gel electrophoresis for microsatellite marker D22S283demonstrating biparental inheritance in the amniocytes (AFS).Comparison of band intensities suggested paternal isodisomy inthe chorionic villi (CVS)

TRISOMY 22 CONFINED TO PLACENTA 139

Copyright # 2002 John Wiley & Sons, Ltd. Prenat Diagn 2002; 22: 137–140.

an apparent isochromosome 22, showing completematernal homozygosity. It was concluded that mater-nal UPD (22) therefore had no effect and that nomaternally imprinted genes with major effect werepresent on chromosome 22. In 1999, Balmer et al.reported a case of maternal UPD (22) in associationwith IUGR and hypospadias. Non-mosaic trisomy 22was identified in the placenta and all lymphocytesshowed a normal male karyotype, 46,XY. Molecularstudies showed maternal UPD (22) with heterodisomyat three of six informative markers.

The present case helps to clarify the findings ofBalmer et al. (1999). Identical clinical findings ofIUGR and hypospadias, yet with biparental inherit-ance in the present case instead of maternal UPD,indicate that these clinical findings were unlikely to bedue to maternal UPD (22) and resultant imprintingeffects. This is in keeping with reports to date thatchromosome 22 is not a maternally imprinted chro-mosome. It is still unclear to what extent these featuresmay have been caused by tissue-specific mosaicism orthe effects of local growth retardation as suggested byBalmer et al. (1999).

In many cases direct fetal sampling to evaluate thedistribution of the aneuploid cell line in the fetus maybe more appropriate. While fetal blood sampling maybe helpful, this approach is limited by the potentialabsence of trisomic cells in fetal blood, despite aphenotypically abnormal infant, with the mosaictrisomy 22 cell line present in other tissues. To date,seven cases of phenotypically abnormal infants havingonly euploid lymphocytes but mosaic trisomy 22 skincells have been reported (Berghella et al., 1998).

Given the potential absence of trisomic cells inlymphocytes, a skin biopsy may be more beneficialthan fetal blood sampling in phenotypically abnormalfetuses with an abnormality detected at amniocentesisor CVS (Papenhausen et al., 1991; Yokoyama et al.,1992). The technical aspects of fetal skin samplinghave continued to improve and in the largest seriesreported no complications were observed in 54 cases(Berghella et al., 1998). In expert hands it appears thatfetal skin sampling compares well with the 1–2%incidence of pregnancy loss following fetal bloodsampling and may have a far higher diagnostic valuethan fetal blood in certain cases of mosaicism detectedin amniocytes (Berghella et al., 1998).

In conclusion, the present case reinforces theaccumulating evidence that chromosome 22 is notmaternally imprinted and that the previously reportedclinical findings of IUGR and hypospadias more likelyreflect the presence of a trisomic placenta, tissue-specific mosaicism or the effects of local growthrestriction.

ACKNOWLEDGEMENTS

The authors wish to thank Dr Neill Astill and Dr RonJames for their clinical insights and Aurelia Citraro,

Cecelia Draheim, Darryl Irwin, and Brent Wilson fortechnical assistance.

REFERENCES

Balmer D, Baumer A, Rothlisberger B, Schinzel A. 1999. Severeintra-uterine growth retardation in a patient with maternaluniparental disomy 22 and a 22-trisomic placenta. Prenat Diagn19: 1061–1064.

Berghella V, Wapner RJ, Yang-Feng T, Mahoney M. 1998. Prenatalconfirmation of true fetal trisomy 22 mosaicism by fetal skinbiopsy following normal fetal blood sampling. Prenat Diagn18: 384–389.

Hansen WF, Bernard LE, Langlois S, et al. 1997. Maternaluniparental disomy of chromosome 2 and confined placentalmosaicism for trisomy 2 in a fetus with intrauterine growthrestriction, hypospadias, and oligohydramnios. Prenat Diagn17: 443–450.

Hsu LYF, Yu MT, Ney RL, et al. 1997. Rare trisomy mosaicismdiagnosed in amniocytes, involving an autosome other thanchromosomes 13, 18, 20, and 21 : karyotype/phenotype correla-tions. Prenat Diagn 17: 201–242.

Kalousek DK, Barrett I. 1994. Genomic imprinting related toprenatal diagnosis. Prenat Diagn 14: 1191–1201.

Kalousek DK, Vekemans M. 1996. Confined placental mosaicism.J Med Genet 33: 529–533.

Kukolich MK, Kulharya A, Jalal SM, Drummond-Borg M. 1989.Trisomy 22: no longer an enigma. Am J Med Genet 34: 541–544.

Ledbetter DH, Engel E. 1995. Uniparental disomy in humans:development of an imprinting map and its implications forprenatal diagnosis. Hum Mol Genet 4: 1757–1764.

Los FJ, Van Opstal D, Van den Berg C, et al. 1998. Uniparentaldisomy with and without confined placental mosaicism: a modelfor trisomic zygote rescue. Prenat Diagn 18: 659–668.

Palmer CG, Schwartz S, Hodes ME. 1980. Transmission of abalanced homologous t(22q;22q) translocation from mother tonormal daughter. Clin Genet 17: 418–422.

Papenhausen P, Mueller OT, Bercu B, Salazar J, Tedesco TA. 1991.Cell line segregation in a 45,X/46,XY mosaic child with asym-metric leg growth. Clin Genet 40: 237–241.

Pflueger SMV. 1999. Cytogenetics of spontaneous abortion. In ThePrinciples of Clinical Cytogenetics, Gersen SL, Keagle MB (eds).Humana Press: New Jersey, NY.

Robinson WP. 2000. Mechanisms leading to uniparental disomy andtheir clinical consequences. Bioessays 22: 452–459.

Robinson WP, Barrett IJ, Bernard L, et al. 1997. Meiotic origin oftrisomy in confined placental mosaicism is correlated withpresence of fetal uniparental disomy, high levels of trisomy introphoblast, and increased risk of fetal intrauterine growthrestriction. Am J Hum Genet 60: 917–927.

Schinzel AA, Basaran S, Bernasconi F, Karaman B, Yuksel-ApakM, Robinson WP. 1994. Maternal uniparental disomy 22 has noimpact on the phenotype. Am J Hum Genet 54: 21–24.

Shaffer LG, McCaskill C, Adkins K, Hassold TJ. 1998. Systematicsearch for uniparental disomy in early fetal losses: the results anda review of the literature. Am J Med Genet 79: 366–372.

Sleutels F, Barlow DP, Lyle R. 2000. The uniqueness of theimprinting mechanism. Curr Opin Genet Dev 10: 229–233.

Wang JC. 1999. Genomic imprinting and uniparental disomy. In ThePrinciples of Clinical Cytogenetics, Gersen SL, Keagle MB (eds).Humana Press: New Jersey, NY.

Wolstenholme J. 1996. Confined placental mosaicism for trisomies 2,3, 7, 8, 9, 16, and 22: their incidence, likely origins, and mech-anisms for cell lineage compartmentalization. Prenat Diagn16: 511–524.

Yokoyama Y, Narahara K, Kamada M, Tsuji K, Seino Y. 1992.Tissue-specific mosaicism for trisomy 21 and congenital heartdisease. J Pediatr 121: 80–82.

J. BRYAN ET AL.140

Copyright # 2002 John Wiley & Sons, Ltd. Prenat Diagn 2002; 22: 137–140.