Allelic IGF2R Repression Does Not Correlate with Expression of Antisense RNA in Human Extraembryonic Tissues Cees B. M. Oudejans, 1 Bart Westerman, Diana Wouters, Sascha Gooyer, Peter A. J. Leegwater, Inge J. van Wijk, and Frank Sleutels 2 Molecular Biology Laboratory, Department of Clinical Chemistry, University Hospital ‘Vrije Universiteit’, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands Received November 27, 2000; accepted February 12, 2001; published online April 10, 2001 In the mouse, expression of an antisense Igf2r RNA (Air) is correlated with Igf2r repression on the pater- nal allele. One of the possible models for Igf2r repres- sion could be through promoter competition or through the action of the Air RNA, in, e.g., transcrip- tional interference or repressor binding. These models predict the conservation of AIR RNA in human sam- ples with monoallelic IGF2R expression and the pro- duction of AIR RNA in first-trimester human tissues. However, by strand-specific RT-PCR and by ribonucle- ase protection assay we have not detected any AIR RNA in first-trimester placental tissue samples, not even in samples that downregulate IGF2R expression in an allele-specific manner. This indicates that in con- trast to the mouse, allelic IGF2R repression in the developing human placenta does not correlate with AIR expression. © 2001 Academic Press INTRODUCTION The human insulin-like growth factor type 2 receptor gene, also known as the cation independent mannose 6-phosphate receptor gene (IGF2R/CIMPR), contrasts with the mouse Igf2r/Cimpr gene (Barlow et al., 1991) and the majority of other imprinted genes in that it shows a generalized loss of imprinted expression in adult tissues (Ogawa et al., 1993; Kalscheuer et al., 1993). Although originally described as “not imprinted” (Ogawa et al., 1993; Kalscheuer et al., 1993), analyses of earlier developmental stages showed that preferen- tial expression of the maternal allele of the IGF2R gene exists up to the second-trimester fetal stage in both fetal and placental tissues, but only in a minority of individuals (Xu et al., 1993, 1997; Riesewijk et al., 1998). Out of a total of 20 heterozygous fetuses, only 2 expressed the IGF2R gene monoallelically (Ogawa et al., 1993; Kalscheuer et al., 1993; Xu et al., 1993, 1997; Riesewijk et al., 1998). In the mouse, maternal-specific expression of Igf2r is dependent on a CpG island, named region 2, situated within intron 2. Deletion of region 2 results in a failure to imprint Igf2r correctly (Wutz et al., 1997). Recently, a novel antisense RNA (termed Air) consisting of a 108-kb noncoding intronless transcript was shown to be expressed from the unmethylated region 2 on the paternal chromosome (Lyle et al., 2000). One of the mechanisms of Igf2r repression by region 2 could be through the action of antisense RNA in, e.g., transcrip- tional interference or repressor binding (Sleutels et al., 2000). The human IGF2R gene also contains a CpG island in intron 2 that is maternally methylated, but retains this methylation imprint even in tissues that lack imprinted IGF2R expression (Smrzka et al., 1995). Thus, although the human IGF2R gene has features commonly shared by imprinted genes (CpG island pro- moters, gene clusters, allelic DNA methylation, mono- allelic expression, replication asynchrony) (Kalscheuer et al., 1993; Smrzka et al., 1995), evidence for a corre- lation of AIR expression with IGF2R repression on the same allele is lacking. In this study, human first-trimester placental cells were screened for monoallelic IGF2R expression and tested for the presence of antisense IGF2R transcripts by strand-specific RT-PCR as well as RNase protection. MATERIALS AND METHODS Tissues. Normal first-trimester placental tissue fragments (n = 9) (gestational age 6 –10 weeks) and cells (n = 25) (gestational age 9 –10 weeks) were obtained from pregnancy terminations and by explant culture of chorionic villus samplings, respectively. The latter cells were obtained by growth of intact chorionic villus fragments— removed for diagnostic reasons by transabdominal sampling— on Matrigel or laminin in complete Amniomax medium (Life Technolo- gies). Following explantation, cells were maintained in complete Amniomax medium. By immunocytochemistry, karyotyping, and flu- orescence in situ hybridization with XY probes (Vysis), these cells were found to consist of villus fibroblast cells (95%) of normal fetal karyotype with no contamination of maternal cells. In addition, the fetal sex of all samples was confirmed by multiplex XY PCR (van 1 To whom correspondence should be addressed. Telephone: (00 31) 20 444 3872. Fax: (00 31) 20 444 3895. E-mail: cbm.oudejans@azvu.nl. 2 Present address: The Netherlands Cancer Institute (H5), Ples- manlaan 121, 1066 CX Amsterdam, The Netherlands. Genomics 73, 331–337 (2001) doi:10.1006/geno.2001.6522, available online at http://www.idealibrary.com on 331 0888-7543/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.