Human Molecular Genetics

Figure 1. Maps of the mouse and human IMPT1 genes and PCR strategies for assessing imprinting. (A) Location of human IMPT1 relative to other chromosome 11p15.5 genes. A complete P1 clone contig spanning the illustrated region has recently been described (42) and the distances and gene order are also consistent with the estimated sizes and hybridization patterns of P1 clones in our unpublished contig spanning this region. Genes which are imprinted in either mice or humans are indicated in bold and the parental origin of the active allele is indicated above each gene. Genes which have not shown an allelic expression bias in tissues examined to date (43-45) are in plain type. The two genes in parentheses are also under study for imprinting and show allelic expression bias in some tissues (N.Qian, L.Yuan, C.Walsh, B.Tycko, unpublished observations). The 2G3-8 gene is defined by human ESTs with accession nos R07294 and N35348. m, maternal; p, paternal. (B) Size, exon/intron structure, transcriptional orientation and proximity to IPL of human IMPT1 and transcriptional orientation and proximity to Ipl of mouse Impt1. The PCR strategies for allelic expression analysis are shown under the maps; genomic PCR products (short lines immediately below the map) were contained within single exons, while the RT-PCR products (indicated as spliced segments below the genomic products) spanned exon/exon junctions. Exon/intron boundaries were sequenced in the 3 kb region of mouse Impt1 which was used for the PCR analysis and the region between the last Impt1 exon and the Ipl gene was sequenced. The exon/intron structure of the remainder of the mouse gene has not been characterized. Filled boxes are translated regions and empty boxes are 5'- and 3'-untranslated regions; ticks indicate the 3'-end of each gene. X, XbaI; N, NotI.

Figure 2. Tissue-specific expression patterns of human and mouse Impt1 genes. (A) Northern blots of poly(A)+ RNAs from human fetal and adult tissues. Probes and size markers are indicated. The blots with the IMPT1 probes were exposed for 3 days and the blots with the [beta]-actin control probes were exposed for 3 h. (B) Northern blots of total RNAs from fetal (14.5 d.p.c.) and adult mouse tissues and poly(A)+ RNAs from adult mouse tissues. Probes and size markers are indicated. Ethidium bromide (EtBr) staining is shown as a loading control for the total RNA blots. The blots hybridized with an Impt1 cDNA probe, a 1143 bp RT-PCR product from adult mouse liver RNA spanning the second to last exon, were exposed for 2 days and the blot hybridized with a [beta]-actin probe was exposed for 2 h. Ki, kidney; Li, liver; Lu, lung; Br, brain; Pa, pancreas; Skm, skeletal muscle; Pl, placenta; Ht, heart; In, intestine; Th, thymus; Te, testis; EM, extra-embryonic membranes; Lm, limb; Aki, adult kidney.

The predicted IMPT1 protein identified many similar proteins in a Blastp analysis: all were members of a well-established family of membrane proteins with multiple membrane-spanning segments and with known or suspected polyspecific transport capabilities for small organic molecules. This family includes proteins which have been identified and studied as drug efflux pumps in bacteria (19,20; and references therein), such as the multi-drug resistance protein of Bacillus subtilis (accession no. P33449), the NapC protein of Escherichia hirae (accession no. e328773) and the tetracycline transporter of Escherichia coli (accession no. Q07282), as well as certain mammalian proteins which matched with very low probability scores in the initial Blastp screen but which showed conserved regions when analyzed directly versus IMPT1 protein using the Block Maker algorithm (Internet address http://www.blocks.fhcrc.org/blockmkr; 21), such as NKT (U52842), which has been studied as a candidate organic metabolite transporter in kidney (22). The match with the strongest probability score (4.7e - 12) was to a predicted Caenorhabditis elegans gene, C53B4.3 (accession no. 1122783), which has not yet been studied functionally; other probability scores fell in what would normally be considered a low range of 1e - 7 to 0.05, but this reflects the fact that this family of proteins is recognized not only by amino acid identities and similarities but also by membrane topology and hydropathy patterns. Alignment of the human IMPT1 protein with the bacterial NapC protein (Fig. 3A) supported the proposed start codon, as the two proteins were in good register throughout their lengths, with 23.5% identical amino acids and with numerous similar amino acids, both in hydrophobic segments and in charged regions; the C.elegans protein was smaller, as were some other bacterial efflux pumps (Fig. 3B and data not shown). Analysis with the TMpred algorithm (Internet address http://ulrec3.unil.ch/software/TMPRED_form.html; 23) yielded a strongly preferred model which placed the N-terminus inside the cell and predicted 10 strong transmembrane helices. Alignment of the Kyte-Doolittle hydropathy plot of the human IMPT1 protein with selected other polyspecific transporter family members is shown in Figure 3B. Good alignment of the predicted transmembrane helices of the various proteins supports the assignment of IMPT1 to this gene family.

Figure 3. Sequence and structure of the predicted IMPT1 protein in human and mouse. (A) Alignment of the human IMPT1 predicted protein with the E.hirae NapC protein. Amino acid identities are indicated by the double dots and similarities by the single dots. (B) Kyte-Doolittle hydropathy plots of human IMPT1 protein and four other members of the polyspecific transporter/multi-drug resistance protein family. The vertical lines bracket one of the predicted transmembrane helices and are shown to bring all five proteins into register. (C) Alignment of the human and murine Impt1 predicted proteins. Amino acid identities and similarities are indicated by the double and single dots respectively.

We next identified the murine ortholog of IMPT1 by screening the mouse EST database with the human IMPT1 cDNA sequence. Complete re-sequencing of two overlapping mouse cDNAs spanning 1.46 kb (clone accession nos. W71351 and AA122896) yielded an open reading frame of 406 amino acids, which aligned well with the predicted human protein throughout its length and which showed 77% overall sequence identity to the human IMPT1 protein (Fig. 3C). Essentially, as was found for the human protein, the murine Impt1 protein was predicted to initiate from a Kozak consensus methionine codon (agcATGg), which was 9 bp downstream of an in-frame ATG embedded in a poor initiating context.

After our exon trapping and cDNA cloning was completed an extended block of as yet unannotated genomic sequence for the region of human chromosome 11p15.5 containing IMPT1 was deposited in GenBank (human 244 kb contig from chromosome 11p15.5, HTGS phase 3, complete sequence; accession no. U90582). By searching this sequence with the IMPT1 cDNA, the IMPT1 gene was found to consist of at least 11 exons distributed over 23 kb of DNA (Fig. 1B). IMPT1 and IPL are transcribed towards each other in a tail-to-tail configuration and the 3'-ends of these genes are separated by 3 kb (Fig. 1B). By mapping in an interspecific back-cross panel we previously showed that the mouse and human IPL genes are both in the region of conserved synteny spanning the known human chromosome 11p15.5/distal mouse chromosome 7 imprinted domain (17). To test possible conserved synteny of Impt1-Ipl we isolated three mouse genomic phage clones using an Ipl genomic probe and found that two of these clones hybridized with the Impt1 cDNA. By sequencing the region between Ipl and Impt1 we determined that the distance between the 3'-ends of these genes in the mouse is 2 kb and that, as is true in humans, the genes are oriented tail-to-tail (Fig. 1B). Thus, consistent with the overall conservation of synteny of the imprinted domain, these two genes have been maintained in a conserved configuration in mammalian evolution.

Northern blots showed that human IMPT1 mRNA is expressed at high levels in many but not all fetal and adult tissues. Tissues with relatively low expression included brain and lung, while tissues with high expression included kidney and liver (Fig. 2A). This distribution of expression appears consistent with the predicted role of the IMPT1 protein as a metabolite transporter, since kidney tubular epithelium and liver parenchyma are responsible for active transport of multiple metabolites. As noted above, in addition to the major 1.5 kb mRNA the complex northern band pattern suggested the presence of alternative splicing, alternative transcriptional initiation sites or other forms of alternative transcripts. The simplified band pattern observed with the isolated exon 8 probe supports alternative exon usage in the various transcripts from this locus (Fig. 2A). A parallel analysis using the murine Impt1 cDNA probe (Fig. 2B) gave similar results and also provided additional information. First, mouse Impt1 mRNA was detected at high levels in tissues with metabolite transport functions, such as the extra-embryonic membranes (yolk sac), which have an important role in metabolite exchange in the early embryo, as well as the liver, kidney and intestine. Tissues which lack general transport functions, such as heart, skeletal muscle, brain, thymus and spleen, did not show signals with the mouse probe, although Impt1 mRNA could be amplified from these sources by reverse PCR (below). Second, the northern blot band pattern observed in the murine tissues was considerably simpler than the corresponding human pattern and showed primarily the 1.5 kb band; only the placenta showed an alternative transcript which was slightly smaller than that seen in the other tissues (Fig. 2B). This supports the likelihood that the major 1.5 kb mRNA is the most physiologically important spliced form of IMPT1 mRNA. Alternative transcripts could, however, be important in particular tissues and could in principal give rise to sequence diversity in the protein, but this remains to be tested.

Interspecific mouse crosses have been widely used to study parental imprinting, since these crosses reliably generate heterozygosity for exonic polymorphisms in the F1 progeny. Most recently we used an extensive series of staged interspecific crosses to characterize the imprinting of murine Ipl in fetal and adult tissues (17). We used cDNA preparations from these same types of crosses to investigate possible imprinting of Impt1. The maternal and paternal alleles in cDNA and genomic PCR products were distinguished by three methods: direct sequencing, RFLP analysis and SSCP analysis. We previously showed that all three methods yield reliable qualitative information concerning allele-specific expression levels and that RFLP and SSCP produce accurate quantitative data (17). We used precise reciprocal crosses as well as two different types of crosses, involving BL/6 and two different divergent subspecies, Mus mus molossinus (MOLD) and M.m.castaneus (CAST), in order to test the generality of any functional imprinting which might be observed, i.e. to exclude possible strain-specific effects. The results are shown in Figure 4 and Table 1. A clear and consistent allelic expression bias diagnostic of parental imprinting was detected in fetal tissues of all crosses. Tissues showing very strong functional imprinting (virtually monoallelic expression) included placenta, extra-embryonic membranes (yolk sac) and fetal liver. In these tissues >80% of total expression derived from the maternal allele (Fig. 4A and Table 1). In every cross there was preferential expression of the maternal Impt1 allele (relative paternal repression or `paternal imprinting') and there was no evidence of somatic allele switching when the direction of bias was compared across multiple organs. In adult tissues the allelic expression bias became less pronounced or disappeared entirely. We conclude that Impt1 is imprinted in the same parental `direction' as Ipl, Mash2, p57^Kip2, KvLQT1 and H19 and in the opposite direction to Igf2 and Ins2.

Figure 4. Parental imprinting of murine Impt1. (A) Allelic expression of Impt1 in tissues of F1 progeny of interspecific crosses between BL/6 and MOLD or CAST assessed by RFLP analysis with HpaII (crosses with MOLD) or KspI (crosses with CAST). The mixing experiment was with known amounts of gel-isolated PCR templates from the two parental strains; densitometry shows an apparent 5% `bias' in favor of amplification of the MOLD (upper) allele, which was also observed in F1 genomic DNA controls of this cross and the crosses with CAST (see lane 9 of bottom panel). This is due to the fact that the MOLD and CAST alleles are the larger bands and therefore contain several additional radiolabeled cytosines. The asterisks indicate that the values have been adjusted accordingly. (Right) Lane 1, BL/6 liver; lane 2, MOLD liver; lane 3, 14.5 d.p.c. MOLD × BL/6 brain; lane 4, 14.5 d.p.c. BL/6 × MOLD brain; lane 5, adult BL/6 × MOLD spleen; lane 6, adult BL/6 × MOLD brain; lane 7, 12.5 d.p.c. BL/6 × MOLD brain. (Bottom) Lane 8, CAST genomic DNA; lane 9, CAST × BL/6 F1 genomic DNA. Note that most of the fetal tissues show strong functional imprinting, with virtually monoallelic expression, but that this attenuates in many of the adult tissues. In the fetal tissues the darker exposures show very low but detectable paternal expression in most organs. Additional data are shown in Table 1. (B) Allelic expression of Impt1 assessed by direct sequencing of RT-PCR products. Polymorphic residues are highlighted by the boxes and are indicated by the asterisks. M, maternal allele; P, paternal allele; Co, colon; Ri, ribs; other abbreviations as in Figure 2.

Table 1. Organs and tissues analyzed for imprinting of Impt1 in interspecific mouse crosses

Cross	Stage	Organ/tissue	mRNA (% from BL/6)
BL/6 × CAST	12.5 d.p.c.	Placenta	91
		EM	97
		Liver	89
		Fetus (-liver)	79
	14.5 d.p.c.	Placenta	81
		EM	86
		Liver	85
		Kidney	74
		Lung	78
		Skeletal muscle	80
		Brain	87
	Adult	Liver	60
		Kidney	75
		Brain	63
		Spleen	50
		Lung	70
		Colon	62
CAST × BL/6	12.5 d.p.c.	Placenta	0
		EM	0
		Fetus	10
		Brain	9
	14.5 d.p.c.	Placenta	0
		EM	0
		Heart	0
		Lung	5
		Liver	8
		Kidney	22
		Limbs	8
		Brain	17
		Ribs	14
	Adult	Brain	17
BL/6 × MOLD	12.5 d.p.c.	Placenta	95
		EM	94
		Liver	83
		Fetus (-liver)	79
		Brain	86
	14.5 d.p.c.	placenta	86
		EM	93
		Liver	81
		Kidney	55
		Lung	79
		Brain	76
	Adult	Brain	78
		Spleen	56
MOLD × BL/6	14.5 d.p.c.	Placenta	5
		EM	6
		Liver	7
		Kidney	8
		Lung	6
		Brain	15
	Adult	Lung	47
		Liver	41
		Kidney	54
		Intestine	36

The allelic bias was measured by densitometry of RFLP films. Crosses are indicated by the usual convention of (female × male). There is strongly selective expression of the maternal allele in fetal tissues and attenuation of the functional imprint in adult tissues. Values are from single determinations; similar results were obtained in two experiments.

Table 2. Organs and tissues analyzed for allelic expression of human IMPT1

Stage	ID no.	Organ	Polymorphism	Allelic bias (active allele; % RNA expression)>
Term	1835	Placenta	HinfI; 5'SSCP	Yes (a 92 ± 2)
		Chorioamnion		Yes (a 84 ± 9)
Term	1845	Placenta	HinfI; 5'SSCP	Yes (a 76 ±11)
		Chorioamnion		Yes (a 62 ± 3)
Term	1947	Placenta	HinfI; 5'SSCP	Yes (g 86 ± 1)
		Chorioamnion		Yes (g 72 ± 1)
Term	1849	Placenta	HinfI; 5'SSCP	Yes (a 74 ± 3)
		Chorioamnion		Yes (a 95 ± 2)
Term	1879	Placenta	HinfI; 5'SSCP	Yes (g 64)
		Chorioamnion		Yes (a 61)
Term	1851	Placenta	3'SSCP (g -> a 1448)	Yes (g 96)
		Chorioamnion		Yes (g 89)
Term	1883	Placenta	3'SSCP (g -> a 1448)	Yes (a 78)
		Chorioamnion		Yes (a 76)
Fetus	1063	Placenta	HinfI; 5'SSCP	yes (a 82 ± 5)
		Lung		Yes (a 61 ± 1)
		Brain		Yes (a 62)
		Liver		Yes (a 69 ± 1)
		Kidney		No
		Thymus		No
		Spleen		Yes (a 61)
		Adrenal		Yes (a 81 ± 2)
		Ovary		Yes (a 62)
Fetus	994	Kidney	HinfI; 5'SSCP	Yes (a 65)
		Adrenal		Yes (g 61)
		Spleen		No
		Lung		Yes (a 65)
		Liver		Yes (a 77)
		Muscle		Yes (a 75)
Fetus	1932	Brain	HinfI; 5'SSCP	No
		Liver		No
		Kidney		Yes (g 60)
		Adrenal		No
		Lung		Yes (g 61)
		Intestine		Yes (g 70)
		Heart		Yes (g 76)
		Skeletal muscle		No
		Meninges		No
Fetus	1863	Kidney	HinfI; 5'SSCP	No
		Adrenal		Yes (g 85)
		Lung		Yes (g 71)
		Liver		No
		Spleen		No
		Placenta		Yes (g 100)
		Intestine		No
Child	803	Kidney	HinfI; 5'SSCP	No
Child	1779	Kidney	3'SSCP (g -> a 1448)	No
Adult	1720	Liver	HinfI; 5'SSCP	Yes (g 74)
		Kidney		No
		Lung		Yes (g 91)
		Spleen		Yes (g 90)
		Adrenal		Yes (g 74)
Adult	861	Heart	3'SSCP (c -> t 1382)	Yes (t 64)
		Lung		No
		Pancreas		No
		Thyroid		Yes (c 63)
		Liver		Yes (c 60)
Adult	1606	Kidney	3'SSCP (g -> a 1448)	No
Adult	1761	Lung	3'SSCP (g -> a 1448)	No
		Spleen	HinfI	Yes (a 72)
		Skeletal muscle		No
		Adrenal		No
		Heart		Yes (a 69)
		Kidney		No
Newborn	1040	Liver	HinfI; 5'SSCP	Yes (a 68)
		Lung		Yes (a 80)
		Kidney		Yes (a 61)
		Thymus		Yes (a 66)
		Adrenal		Yes (a 74)
		Spleen		Yes (a 66)
Newborn	1583	Liver	HpaII/AvaII	No
		Kidney		No
		Adrenal		No
		Muscle		No
		Thymus		No
		Spleen		No

For all cases the allelic bias was scored qualitatively by inspection of SSCP and RFLP films and quantitatively by densitometry of the SSCP films, except for case 1863, which was scored by densitometry of RFLP films; where shown, the standard errors are based on three or four determinations with independent cDNA aliquots, other values are from single determinations, but these data were also reproduced in independent cDNA aliquots. A designation of `no allelic bias' indicates <10% bias on densitometry. The more active allele is indicated for each sample. There is marked interindividual variation in strength of functional imprinting even when the same organs are compared at similar developmental stages.

We screened the human IMPT1 gene for exonic polymorphisms which would allow us to distinguish the two alleles. By SSCP and direct sequencing of genomic and cDNA PCR products we identified four useful polymorphisms, two in exon 2 and two in exon 11, as well as several less common polymorphisms (Materials and Methods and Table 2). We then analyzed relative expression levels from the two alleles in autopsy or surgical tissues from four fetuses (16-23 weeks gestation), two newborns, two children and four adults, as well as seven term placentas and corresponding chorioamnionic membranes (Table 2 and Fig. 5A-D). We used reverse transcription-PCR strategies, which are shown in Figure 1Band which were analogous to those used for analysis of the mouse gene. Genomic PCR reactions were run in each experiment to establish the relative band intensities on the SSCP and RFLP analyses corresponding to equal allelic representation. An allelic expression bias was observed in multiple tissues of several individuals and the experimental data were highly reproducible when multiple aliquots of cDNA were examined (Fig. 5B and Table 2). Some tissues showed a strong allelic bias, with densitometry indicating that >80% of the total cDNA derived from one allele, but other tissues showed little or no allelic expression bias. As shown in Table 2 and Figure 5A-D, certain organs, such as fetal and adult kidney, consistently showed a weak allelic expression bias, while others, such as placenta, chorioamnion, liver and adrenal gland, tended to show a stronger bias, but there were major interindividual variations in the strength or even the presence or absence of an allelic bias, even when identical organs were examined at comparable developmental stages (compare different placentas, chorioamnions, adrenal glands and lungs in Table 2). Consistent with genomic imprinting and with our findings in the interspecific mouse crosses, in almost all cases within a single individual all or most of the organs examined showed the same `direction' of allelic expression bias (Table 2 and Fig. 5A-D). Moreover, in one placenta/amnion pair for which parental DNAs were available and informative we found that the expressed allele was maternal, consistent with the direction of imprinting observed in the mouse crosses (Fig. 5A).

Figure 5. Allelic expression analysis of human IMPT1. (A) Fetal, newborn and adult organs assessed for allelic expression bias by HinfI RFLP and parallel SSCP analysis of RT-PCR products from IMPT1 exons 2 and 3. (Upper left) Maternal origin of the preferentially expressed allele; the experiments in the left panels used cDNA primers IMPT1-5'-3/IMPT1-3'-5, while those in the right panels used alternative cDNA primers IMPT1-5'-2/IMPT1-3'-2. Allele-specific bands are indicated and only the informative regions of the gels are shown; other bands were present at other positions. Note that the SSCP band patterns differ among the cases, reflecting additional sequence polymorphisms outside the HinfI site. The preferentially expressed allele is indicated and is highlighted in bold for samples with the strongest bias. Similar results were obtained for all samples when the analysis was repeated with independent aliquots of cDNA and, in some cases, with independent cDNA preparations. The slight splitting of the upper SSCP band in the lung of adult 1720 is a gel loading artifact which was not seen in a replication. G, genomic DNA; M, mother; F, father; Pl, placenta; Am, chorioamnion. (B) Reproducibility and quantitation of the SSCP and RFLP results for fetus 1063. Lane 1, homozygote control; 2, genomic DNA from fetus 1063. (C) Adult organs assessed by SSCP of RT-PCR products from exons 10 and 11. Expression is essentially biallelic, but case 861 shows a suggestion of somatic allele switching on comparison of heart, liver and thyroid. Polymorphic bands are indicated by asterisks. The first genomic lane is from a homozygote. (D) Somatic allele switching in one placenta/chorioamnion pair. Note the characteristic SSCP fingerprint of this case. The results were reproduced in six cDNA aliquots from three independent cDNA preparations. For assignment of the allelic bias compare the cDNA band intensities with the corresponding bands in the genomic control lanes. (Right) Genotyping for the H19 RsaI polymorphism (7) as an additional control for sample identification; all genomic and cDNA samples from case 1879 show exclusively the RsaI `+' allele.

However, while 12 cases (including the placenta/amnion pairs) in which two or more tissues showed an allelic expression bias showed a consistent `direction' of the bias, in one placenta/chorioamnion pair there was an opposite allelic expression bias in placenta versus chorioamnion (Fig. 5D and Table 2). Since the PCR products spanned a region of the IMPT1 gene which contains multiple sequence polymorphisms (D.Dao and B.Tycko, unpublished observations), this case showed a characteristic SSCP band pattern which was shared by the two tissues (albeit with variations in relative band intensities due to the allelic bias) and which therefore provided an internal control to exclude sample identification error as an artifactual explanation for the allele switching. As a further control for sample identification in this case we ascertained the representation of H19 exon 3-5 polymorphisms (7,18,24) in the same preparations and the results were consistent in both tissues (homozygosity for the RsaI `+' allele in both genomic samples and exclusive representation of this allele in the cDNAs; Fig. 5D). In two other cases (adult 861 and fetus 994) there were weaker suggestions of allele switching in certain organs (Fig. 5A and C and Table 2). Somatic allele switching is not a common feature of imprinted genes, but we previously observed this unusual phenomenon for the imprinted human H19 gene in the cerebellum, a site of low H19 RNA expression and unusual H19 CpG methylation, in one adult (24). Allele switching has recently been suggested for the human WT1 gene, which is functionally imprinted in some tissues (though not in kidney) and which can show either a maternal or a paternal expression bias depending on the tissue source (25). To our knowledge allele switching has not been reported for any imprinted genes in mice and, consistent with this, we did not observe allele switching at the murine Impt1 locus in any fetal or post-natal tissues (Table 1).

DISCUSSION

Genomic imprinting has persisted in mammalian evolution and so it is suspected to confer some reproductive advantage. According to the influential maternal/paternal competition model (1) imprinting regulates the effective dosage of genes which control fetal and placental growth. Based on the strong functional imprinting (allelic expression bias) of mouse Impt1 in extra-embryonic and some fetal tissues, it is likely that imprinting does regulate the effective dosage of this gene in these tissues. This needs to be tested directly by germline deletion, but it is possible that a metabolite transporter expressed in yolk sac and placenta might indeed play an important role in regulating growth of these tissues and of the embryo proper. At this point it is more difficult to understand the implications of our data in terms of the second prediction of the maternal/paternal competition model, that maternally expressed/paternally repressed (`paternally imprinted') genes should be growth inhibitory and that oppositely imprinted genes should be growth promoting. Why would a metabolite transport protein be expected to function as a growth inhibitor? Germline deletion experiments in mice, genetic characterization of the C53B4.3 C.elegans homolog and biochemical studies may shed light on this. Human IMPT1 maps in the Wilms' tumor-2 (WT2)/Beckwith-Wiedemann syndrome (BWS) region of chromosome 11p15.5. IMPT1 is strongly expressed but not strongly imprinted in kidney. Expression of IMPT1 mRNA does seem to track with the tissues which are most affected by fetal overgrowth in BWS and studies of genetic and epigenetic lesions in this gene in BWS might be of interest. The impetus for these studies would be stronger if genetic and biochemical experiments eventually point to a growth inhibition pathway related to metabolite transport, as would be predicted from our data if the paternal/maternal competition theory is valid. It is interesting that another gene belonging to the polyspecific transporter family and distantly related to Impt1 by sequence similarity has recently been mapped very close to the imprinted Igf2r gene on mouse chromosome 17 (26). This gene, Lx1, was shown to be biallelically expressed in at least some adult mouse tissues, but functional imprinting of this locus in other tissues has not been excluded.

The host defense model for imprinting (2-4) posits that imprinted genes resemble integrated retroviruses or retrotransposons and are therefore specifically recognized by DNA methyltransferase enzyme(s) and targeted for inactivation in one of the parental germlines. This model has been supported by a statistical argument showing that among the known imprinted genes there is on average a more compact exon/intron structure than is observed in the genome as a whole (3) and by descriptions of repetitive sequences near imprinted genes (2). Our recent finding that the very compact IPL gene is strongly imprinted in several mouse and human tissues adds incremental support to this theory (17). The human IMPT1 gene does not appear to be particularly compact, since it includes at least 11 exons spaced over 23 kb of DNA. Additional examples of imprinted genes are needed for a definitive statistical test of this theory. Data in this report show that imprinting of human IMPT1 is variable; additional quantitative studies may show that imprinting of large genes is more variable and `leakier' than that of small ones.

Interindividual variations in the strength of imprinting (27,28) might cause phenotypic variations. Such variability within populations could be accounted for either by trans-acting modifiers or by cis-acting variations in gene structure. In view of the major interindividual variations which we have observed in the strength of IMPT1 allelic expression bias, in the future it may be interesting to correlate the strength of imprinting with polymorphisms in IMPT1 gene structure. We have found at least one frequent insertion/deletion polymorphism of an Alu retroelement in IMPT1 intron 8 (D.Dao and B.Tycko, unpublished data) and this type of polymorphism could be informative if analyzed in a large group of cases. In addition, the extended genomic sequence spanning IPL, IMPT1 and other genes (human 244 kb contig from chromosome 11p15.5, HTGS phase 3, complete sequence; GenBank accession no. U90582) indicates that there are several blocks of simple direct repeat sequences both near and within the IMPT1 gene (see Methods section of 17). It is also possible that the more complex pattern of IMPT1 mRNA transcripts in human compared with mouse might account for the greater variability in functional imprinting of the human gene: we do not have data on transcript-specific imprinting of IMPT1, but a precedent for transcript-specific imprinting exists in the human IGF2 locus (29) and, if different individuals express different ratios of imprinted versus non-imprinted transcripts or even oppositely imprinted transcripts, then this could explain the interindividual variability in the observed functional imprinting. The murine Impt1 gene produces a simpler set of transcripts and, while it is difficult to precisely match the mouse and human tissues for developmental stage, the imprinting in mice does appear to be more robust and consistent. It is also possible that neighboring or overlapping transcripts might influence the apparent strength and direction of IMPT1 imprinting and further analysis of the 2G3-8 gene, which produces some transcripts which are partially antisense to IMPT1 (Fig. 1A and unpublished observations) may be relevant. In terms of the most interesting possibility, the existence of modifier genes which affect the strength of imprinting and which act in trans at multiple imprinted loci, the fact that both IMPT1 and IPL (17), as well as p57^KIP2 (14) and KvLQT1 (15), show leaky imprinting with interindividual variability suggests that in the future it may be possible to correlate data on the strength of imprinting of genes throughout the domain to ask whether imprinting modifier genes act in humans to produce weak and strong phenotypes for domain-wide imprinting.

The fact that of the eight genes known to be imprinted in this region at least six are paternally repressed (Fig. 1A) suggests that the primary effect of domain-wide human chromosome 11p15.5/mouse distal chromosome 7 imprinting is repression of the paternal allele relative to the maternal allele. This is consistent with the chromatin accessibility model for imprinting, in which the domain is postulated to correspond to a large region of `open' and therefore epigenetically modifiable chromatin in pre-meiotic and meiotic sperm progenitors (30,31). Recombination frequencies are in fact higher for chromosome 11p15.5 markers in paternal as compared with maternal meioses (31). If one assumes the opposing model of random imprinting through the domain, then by at least one test the probability of observing six of eight genes paternally repressed is 0.3 (not significant; two-tailed statistical test on proportions). However, since the relative maternal repression of Igf2 and Ins2 may reflect a secondary phenomenon resulting from enhancer competition with the paternally repressed H19 gene, as suggested by findings in human Wilms' tumors (32-34) and more definitively by mouse knockout data (35), the simplest explanation of the data is that all of the genes in this region which are the targets of primary imprinting are paternally repressed. Given a model of random imprinting, the probability of observing six of six genes imprinted in the same direction is 0.016, so there is strong statistical support for non-random domain-wide relative paternal repression. Moreover, our preliminary data indicate that a seventh gene in the region, Tapa1, is subject to leaky imprinting with relative repression of the paternal allele in the extra-embryonic tissues (L.Yuan, C.Walsh and B.Tycko, in preparation).

Given the very dense clustering of imprinted genes which is highlighted by this and previous reports, a direct analysis of chromatin structure throughout the known extent of the domain in spermatocytes, together with additional annotation of imprinting in the region, should be able to test differential chromatin accessibility as the mechanism dictating the locations and extents of imprinted domains. This model does not make a prediction as to the specific type of epigenetic modification which confers and maintains the regional imprint. DNA methylation has been a strong candidate (reviewed in 36) and this type of local modification plays an essential role in maintaining and possibly initiating imprinting of some strongly repressed loci, but extensive CpG methylation is not present in all imprinted genes (14,37) and the strongly imprinted human and mouse H19 genes, which have been used as models systems for imprinting, are unusually extensively methylated (24,32-34,38) and may be unusually sensitive to reactivation by demethylation (14,39). For the extended imprinted domain there are as yet few data implicating DNA methylation as a maintenance mechanism. Other factors which can regulate chromatin structure and also propagate epigenetic states, such as templating by histone acetylation patterns and repressive chromatin proteins (40,41), deserve more attention.

MATERIALS AND METHODS

Genomic clones and exon trapping

The human P1 clones shown in Figure 1A were described in our previous report (17). Exon trapping of clone 12766 used the pSPL3 vector (Exon Trapping System, Life Technologies, Gaithersburg, MD). Mouse genomic clones, including [lambda]Ipl-1, containing Ipl and the 3'-portion of Impt1, were isolated from a strain 129/Sv [lambda] library as described previously (17).

cDNAs, sequences and accession numbers

To obtain the human IMPT1 cDNA sequence we used a combined strategy of EST database searches, complete sequencing of a 1.5 kb cDNA isolated from a placental cDNA [lambda]gt11 library (1G2-5 clone 8) and partial sequencing of a second 1.4 kb cDNA (1G2-5 clone 11) from this same library. The murine Impt1 cDNA sequence was obtained by re-sequencing two EST-derived overlapping cDNA clones (clone accession nos. W71351 and AA122896). The results were confirmed by generating a 1143 bp RT-PCR product spanning the second to the last exon from mouse liver RNA. The genomic sequence between the 3'-ends of the mouse Impt1 and Ipl genes was obtained from a plasmid subclone of [lambda]Ipl-1. The genomic structure of human IMPT1 was deduced by screening the IMPT1 cDNA, sequencing against the extended genomic sequence in the nr database (human 244 kb contig from chromosome 11p15.5, HTGS phase 3, complete sequence; accession no. U90582). The human and mouse Impt1 cDNA sequences have accession nos. AF028738 and AF028739.

Northern blotting

RNA was prepared from mouse tissues using Trizol reagent (Life Technologies). Northern blots containing poly(A)⁺ RNA from adult mouse and fetal and adult human tissues were from a commercial source (Clontech, Palo Alto, CA). Blots containing total RNA from mouse tissues were from formaldehyde-containing 1% agarose gels which contained 8 µg RNA/lane.

PCR, SSCP and direct sequencing

Reverse transcription of total RNA was done with the Superscript Preamplification System (Life Technologies), direct sequencing of PCR products and cDNA and genomic clones was done with the fmol Cycle Sequencing System (Promega, Madison, WI) and SSCP analyses were carried out as previously described (46), but the gels were run at room temperature (500 V, 10-16 h) and the gels for analysis of mouse Impt1 included 6% glycerol. In some cases the radiolabeled PCR products were digested with restriction enzymes prior to SSCP analysis to generate comparably sized genomic and cDNA fragments (see below). PCR amplification of genomic DNA and cDNAs was with standard Taq polymerase (Boehringer-Mannheim) using standard buffer supplied by the manufacturer, with addition of DMSO to 5%. Primers for human IMPT1 had the sequences: IMPT1-5'-2 (exon 1), TCACCCCACCGGCACCCGT; IMPT1-5'-3 (exon 2), AGGCGGGTGCTGCCTGGGA; IMPT1-3'-2 (exon 4), CCCCGCGCTGGTCTGCGAAC; IMPT1- 3'-5 (exon 3), GGCCCGCCCAGCAGCTGCA; IMPT1-3'-11 (exon 2), CACGATGGAGAACTGCAT; IMPT1-5'-7 (exon 10), GGCTGTCTCCACCTCGGAC; IMPT1-3'-8 (exon 11), TCTTTATTGCCAGTCTGTG; IMPT1-1G2-5gen (exon 11), GGCCTCTGCGCCTCTGTA

Analogous to strategies which we used previously for analysis of other imprinted genes (7,14,17), the PCR schemes were such that the cDNA products, made with IMPT1-5'-2/IMPT1-3'2 or alternatively with IMPT1-5'-3/IMPT1-3'-5 or IMPT1-5'-7/IMPT1-3'-8, each spanned an exon/exon junction, while the genomic products, made with IMPT1-5'-3/IMPT1-3'-11 or IMPT1-1G2-5gen/IMPT1-3'-11, were within single exons. Primers for murine Impt1 cDNAs had the following sequences. For crosses between BL/6 and CAST: Impt1-5'-1 (exon 2), ATCAACAGGACTTTTGCCCC, and Impt1-3'-6 (exon 3), ACAGAATCTAGGCCCAGTG, or alternatively Impt1-3'-1 (exon 4), GCCCGCCAGGAAGGAGAG. For crosses between BL/6 and MOLD: Impt1-5'-6 (exon 2), TGTCTGCCTGGGATGTCTG, and Impt1-3'-1. For genomic PCR in crosses between BL/6 and either CAST or MOLD: Impt1-5'-1 (exon 2) and Impt1-3'-5 (exon 2), CATGAAGAGACACGTTAGC.

PCR conditions for human IMPT1 cDNAs and genomic DNA were an initial denaturation at 94°C for 5 min,followed by 35 cycles of denaturation at 94°C for 45 s, annealing at 62°C for 30 s and extension at 72°C for 60 s, with a final extension at 72°C for 7 min. PCR conditions for murine Impt1 cDNAs and genomic DNA were an initial denaturation at 95°C for 3 min, followed by 95°C for 30 s, 60°C for 30 s and 68°C for 1 min for 35 cycles, with a final extension at 68°C for 7 min. Radiolabeling was by PCR for an additional eight cycles after dilution of the gel-isolated primary product 1:10 into complete PCR mixture with addition of [³²P]dCTP. For RFLP analysis of human IMPT1 allelic expression PCR products were digested with HinfI, AvaII or HpaII. For RFLP analysis of mouse Impt1 allelic expression PCR products were digested with either HpaII (crosses with MOLD) or KspI (crosses with CAST). Digested products were electrophoresed on 12% (w/w) non-denaturing acrylamide gels. For SSCP analysis of human IMPT1 the labeled PCR products were digested with HpaII (exon 2-3 products) or CfoI (exon 10-11 products) to generate fragments of equal lengths in the cDNA and genomic lanes.

Exonic polymorphisms and assessment of allelic expression

Polymorphisms in mouse Impt1 were (relative to the domesticus strain cDNA sequence): CAST, c108 -> t (deleting a KspI site); MOLD, c194 -> a (deleting an HpaII site). Polymorphisms in human IMPT1 were: g210 -> a (creating an HinfI site); g259 -> a (deleting an HpaII/AvaII site); c1382 -> t (creating a variant SSCP pattern); g1448 -> a (creating a variant SSCP pattern). Several other human polymorphisms were found, but these were not used in the current study. Where indicated, allelic expression bias was measured by densitometry of the allele-specific bands in lightly exposed RFLP or SSCP films with a flatbed scanner (SuperVISTA S-12, UMAX) using NIH Image v.1.49 software. Mixing experiments used defined amounts of gel-isolated cDNA PCR products as templates. As a second control, genomic DNA PCR products were included and the ratios of the allele-specific band intensities were taken as 50:50 allelic representation. Where possible, SSCP results were validated by parallel RFLP analysis of the same radiolabeled products. For mouse Impt1 the RFLP results were validated qualitatively by direct sequencing of the same templates.

ACKNOWLEDGEMENTS

We thank Tim Bestor for interspecific F1 mice and for comments on the manuscript, Ruth Ottman for advice on statistics and Lin Feng and Luwa Yuan for technical assistance. This work was supported by grant RO1CA60765 from the NIH to B.T., by a Cancerfonden post-doctoral fellowship to C.P.W. and by a Tony Bennet Scholar Award to R.J.V.

REFERENCES

1. Moore, T. and Haig, D. (1991) Genomic imprinting in mammalian development: a parental tug of war. Trends Genet., 7, 45-49. MEDLINE Abstract

2. Neumann, B., Kubicka, P. and Barlow, D.P. (1995) Characteristics of imprinted genes. Nature Genet., 9, 12-13. MEDLINE Abstract

3. Hurst, L.D., McVean, G. and Moore, T. (1996) Imprinted genes have few and small introns. Nature Genet., 12, 234-237. MEDLINE Abstract

4. Bestor, T.H. and Tycko, B.(1996) Creation of genomic methylation patterns. Nature Genet., 12, 363-367. MEDLINE Abstract

5. DeChiara, T.M., Robertson, E.J. and Efstratiadis, A. (1991) Paternal imprinting of the mouse insulin-like growth factor II gene. Cell, 64, 849-859. MEDLINE Abstract

6. Bartolomei, M.S., Zemel, S. and Tilghman, S.M. (1991) Parental imprinting of the mouse H19 gene. Nature, 351, 153-155. MEDLINE Abstract

7. Zhang, Y. and Tycko, B. (1992) Monoallelic expression of the human H19 gene. Nature Genet., 1, 40-44. MEDLINE Abstract

8. Rainier, S., Johnson, L., Dobry, C.J., Ping, A.J., Grundy, P.E. and Feinberg, A.P. (1993) Relaxation of imprinted genes in human cancer. Nature, 362, 747-749. MEDLINE Abstract

9. Ohlsson, R., Nystrom, A., Pfeifer-Ohlsson, S., Tohonen, V., Hedbourg, F., Schofield, P., Flam, F. and Ekstrom, T.J. (1993) IGF2 is parentally imprinted during human embryogenesis and in the Beckwith-Wiedemann syndrome. Nature Genet., 4, 94-97. MEDLINE Abstract

10. Giannoukakis, N., Deal, C., Paquette, J., Goodyer, C.G. and Polychronakos, C. (1993) Parental genomic imprinting of the human IGF2 gene. Nature Genet., 4, 98-101. MEDLINE Abstract

11. Giddings, S.J., King, C.D., Harman, K.W., Flood, J.F. and Carnaghi, L.R. (1994) Allele-specific inactivation of insulin 1 and 2, in the mouse yolk sac, indicates imprinting. Nature Genet., 6, 310-313. MEDLINE Abstract

12. Guillemot, F., Caspary, T., Tilghman, S.M., Copeland, N.G., Gilbert, D.J., Jenkins, N.A., Anderson, D.J.., Joyner, A.L., Rossant, J. and Nagy, A. (1995) Genomic imprinting of Mash2, a mouse gene required for trophoblast development. Nature Genet., 9, 235-241. MEDLINE Abstract

13. Hatada, I. and Mukai, T. (1995) Genomic imprinting of p57Kip2, a cyclin-dependent kinase inhibitor, in mouse. Nature Genet., 11, 204-206. MEDLINE Abstract

14. Chung, W.-Y., Yuan, L., Feng, L., Hensle, T. and Tycko, B. (1996) Chromosome 11p15.5 regional imprinting: comparative analysis of KIP2 and H19 in human tissues and Wilms' tumors. Hum. Mol. Genet., 5, 1101-1108. MEDLINE Abstract

15. Lee, M.P., Hu, R.-J., Johnson, L.A. and Feinberg, A.P. (1997) Human KVLQT1 gene shows tissue specific imprinting and encompasses Beckwith-Wiedemann syndrome chromosomal rearrangements. Nature Genet., 15, 181-185. MEDLINE Abstract

16. Alders, M., Hodges, M., Hadjantonakis, A., Postmus, J., van Wijk, I., Bliek, J., de Meulemeester, M., Westerveld, A., Guillemot, F., Oudejans, C., Little, P. and Mannens, M. (1997) The human Achaete-Scute homologue 2 (ASCL2, HASH2 ) maps to chromosome 11p15.5, close to IGF2 and is expressed in extravillus trophoblasts. Hum. Mol. Genet., 6, 859-867. MEDLINE Abstract

17. Qian, N., Frank, D., O'Keefe, D., Dao, D., Zhao, L., Yuan, L., Wang, Q., Keating, M., Walsh, C. and Tycko, B. (1997) The IPL gene on chromosome 11p15.5 is imprinted in humans and mice and is similar to TDAG51, implicated in Fas expression and apoptosis. Hum. Mol. Genet., 6, 2021-2029. MEDLINE Abstract

18. Moulton, T., Chung, W.-Y., Yuan, L., Hensle, T., Waber, P., Nisen, P. and Tycko, B. (1995) Genomic imprinting and Wilms' tumor. Med. Pediat. Oncol., 27, 476-483.

19. Neyfakh, A.A., Bidenko, V.E. and Chen, L.B. (1991) Efflux-mediated multidrug resistance in Bacillus subtillus: similarities and dissimilarities with the mammalian system. Proc. Natl Acad. Sci. USA, 88, 4781-4785. MEDLINE Abstract

20. Allard, J.D. and Bertrand, K.P. (1993) Sequence of a class E tetracycline resistance gene from Escherichia coli and comparison of related tetracycline efflux proteins. J. Bacteriol., 175, 4554-4560. MEDLINE Abstract

21. Henikoff, S., Henikoff, J.G., Alford, W.J. and Pietrokovski, S. (1995) Automated construction and graphical presentation of protein blocks from unaligned sequences, Gene-COMBIS. Gene, 163, 17-26.

22. Lopez-Nieto, C.E., You, G., Bush, K.T., Barros, E.J.G, Beier, D.R. and Nigam, S.K. (1997) Molecular cloning and characterization of NKT, a gene product related to the organic cation transporter family that is almost exclusively expressed in kidney. J. Biol. Chem., 272, 6471-6478. MEDLINE Abstract

23. Hofmann, K. and Stoffel, W. (1993) TMbase-a database of membrane spanning proteins segments. Biol. Chem. Hoppe-Seyler, 347, 166.

24. Zhang, Y., Shields, T., Crenshaw, T., Hao, Y., Moulton, T. and Tycko, B.(1993) Imprinting of human H19: allele-specific CpG methylation, loss of the active allele in Wilms' tumor, and potential for somatic allele switching. Am. J. Hum. Genet., 53, 113-124. MEDLINE Abstract

25. Mitsuya, K., Sui, H., Meguro, M., Kugoh, H., Jinno, Y., Niikawa, N. and Oshimura, M. (1997) Paternal expression of WT1 in human fibroblasts and lymphocytes. Hum. Mol. Genet., 6, 2243-2246. MEDLINE Abstract

26. Schweifer, N., Valk, P.J., Delwel, R., Cox, R., Francis, F., Meier-Ewert, S., Lehrach, H., and Barlow, D.P. (1997) Characterization of the C3 YAC contig from proximal mouse chromosome 17 and analysis of allelic expression of genes flanking the imprinted Igf2r gene. Genomics, 43, 285-297. MEDLINE Abstract

27. Xu, Y., Goodyer, C.G., Deal, C. and Polychronakos, C. (1993) Functional polymorphism in the parental imprinting of the human IGF2R gene. Biochem. Biophys. Res. Commun., 747-754. MEDLINE Abstract

28. Jinno, Y., Yun, K., Nishiwaki, K., Kubota, T., Ogawa, O., Reeve, A.E. and Niikawa, N. (1994) Mosaic and polymorphic imprinting of the WT1 gene in humans. Nature Genet., 6, 305-309. MEDLINE Abstract

29. Vu, T.H. and Hoffman, A.R. (1994) Promoter-specific imprinting of the human insulin-like growth factor-II gene. Nature, 371, 714-717. MEDLINE Abstract

30. Thomas, B.J. and Rothstein, R. (1991) Sex, maps, and imprinting. Cell, 64, 1-3. MEDLINE Abstract

31. Paldi, A., Gyapay, G. and Jami, J. (1995) Imprinted chromosomal regions of the human genome display sex-specific meiotic recombination frequencies. Curr. Biol., 5, 1030-1035. MEDLINE Abstract

32. Steenman, M.J.C., Rainier, S., Dobry, C.J., Grundy, P., Horon, I.L. and Feinberg, A.P. (1994) Loss of imprinting of IGF2 is linked to reduced expression and abnormal methylation of H19 in Wilms' tumor. Nature Genet., 7, 433-439.

33. Moulton, T., Crenshaw, T., Hao, Y., Moosikasuwan, J., Lin, N., Dembitzer, F., Hensle, T., Weiss, L., McMorrow, L., Loew, T., Kraus, W., Gerald, W. and Tycko, B. (1994) Epigenetic lesions at the H19 locus in Wilms' tumor patients. Nature Genet., 7, 440-447. MEDLINE Abstract

34. Taniguchi, T., Sullivan, M.J., Osamu, O. and Reeve, A. (1995) Epigenetic changes encompassing the IGF2/H19 locus associated with relaxation of IGF2 imprinting and silencing of H19 in Wilms tumor. Proc. Natl Acad. Sci. USA, 92, 2159-2163. MEDLINE Abstract

35. Leighton, P.A., Ingram, R.S., Eggenschwiler, J., Efstratiatis, A. and Tilghman, S.M. (1995) Disruption of imprinting caused by deletion of the H19 gene region in mice. Nature, 375, 34-39. MEDLINE Abstract

36. Tycko, B. (1997) DNA methylation in genomic imprinting. Mutat. Res., 386, 131-140. MEDLINE Abstract

37. Feil, R., Walter, J., Allen, N.D. and Reik, W. (1994) Developmental control of allelic methylation in the mouse Igf2 and H19 genes. Development, 120, 2933-2943. MEDLINE Abstract

38. Tremblay, K.D., Saam, J.R., Ingram, R.S., Tilghman, S.M. and Bartolomei, M.S. (1995) A paternal-specific methylation imprint marks the alleles of the mouse H19 gene. Nature Genet., 9, 407-413. MEDLINE Abstract

39. Li., E., Beard, C. and Jaenisch, R. (1993) Role for DNA methylation in genomic imprinting. Nature, 366, 362-365. MEDLINE Abstract

40. Wade, P.A., Pruss, D. and Wolffe, A.P. (1997) Histone acetylation: chromatin in action. Trends Biochem. Sci., 22, 128-132. MEDLINE Abstract

41. Grunstein, M. (1997) Histone acetylation in chromatin structure and transcription. Nature, 389, 349-352. MEDLINE Abstract

42. Reid, L.H., Davies, C., Cooper, P.R., Crider-Miller, S.J., Sait, S.N.J., Nowak, N.J., Evans, G., Stanbridge, E.J., deJong, P., Shows, T.B., Weissman, B.E. and Higgins, M.J. (1997) A 1-Mb physical map and PAC contig of the imprinted domain in 11p15.5 that contains TAPA1 and the BWSCR1/WT2 region. Genomics, 43, 366-375. MEDLINE Abstract

43. Tsang, P., Gilles, F., Yuan, L., Kuo, Y.-H., Lupu, F., Samara, G., Moosikasuwan, J., Goy, A., Zelenetz, A.D., Selleri, L. and Tycko, B. (1995)A novel L23-related gene 40 Kb downstream of the imprinted H19 gene is biallelically expressed in mid-fetal and adult human tissues. Hum. Mol. Genet., 4, 1499-1507. MEDLINE Abstract

44. Yuan, L., Qian, N., Feng, L. and Tycko, B. (1996) An extended region of biallelic gene expression and rodent-human synteny downstream of the imprinted H19 gene on chromosome 11p15.5. Hum. Mol. Genet., 5, 1931-1937. MEDLINE Abstract

45. Hu, R.-J., Lee, M.P., Johnson, L.A. and Feinberg, A.P. (1996) A novel human homologue of yeast nucleosome assembly protein, 65 kb centromeric to the p57 KIP2 gene, is biallelically expressed in fetal and adult tissues. Hum. Mol. Genet., 5, 1743-1749. MEDLINE Abstract

46. Tycko, B., Feng, L., Nguyen, L., Francis, A., Hays, A., Chung, W.-Y., Tang, M.-X., Stern, Y., Sahota, A., Hendrie, H. and Mayeux, R. (1996) Polymorphisms in the human apolipoprotein-J/clusterin gene: ethnic variation and distribution in Alzheimer's disease. Hum. Genet., 98, 430-436. MEDLINE Abstract

---

*To whom correspondence should be addressed. Tel: +1 212 305 1287; Fax: +1 212 305 5498; Email: bt12@columbia.edu
⁺Present address: Department of Surgery, Division of Urology, Memorial Sloan-Kettering Cancer Center, New York, NY 10021, USA
^§The first two authors contributed equally to this work

---

This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 14 Mar 1998
Copyright© Oxford University Press, 1998.

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				P. J. Rugg-Gunn, A. C. Ferguson-Smith, and R. A. Pedersen Status of genomic imprinting in human embryonic stem cells as revealed by a large cohort of independently derived and maintained lines Hum. Mol. Genet., October 15, 2007; 16(R2): R243 - R251. [Abstract] [Full Text] [PDF]

				L. Myatt Placental adaptive responses and fetal programming J. Physiol., April 1, 2006; 572(1): 25 - 30. [Abstract] [Full Text] [PDF]

				P. de Lonlay, A. Simon-Carre, M.-J. Ribeiro, N. Boddaert, I. Giurgea, K. Laborde, C. Bellanne-Chantelot, V. Verkarre, M. Polak, J. Rahier, et al. Congenital Hyperinsulinism: Pancreatic [18F]Fluoro-L-Dihydroxyphenylalanine (DOPA) Positron Emission Tomography and Immunohistochemistry Study of DOPA Decarboxylase and Insulin Secretion J. Clin. Endocrinol. Metab., March 1, 2006; 91(3): 933 - 940. [Abstract] [Full Text] [PDF]

				T. Li, T. H. Vu, G. A. Ulaner, E. Littman, J.-Q. Ling, H.-L. Chen, J.-F. Hu, B. Behr, L. Giudice, and A. R. Hoffman IVF results in de novo DNA methylation and histone methylation at an Igf2-H19 imprinting epigenetic switch Mol. Hum. Reprod., September 1, 2005; 11(9): 631 - 640. [Abstract] [Full Text] [PDF]

				T. Hirota, I. Ieiri, H. Takane, S. Maegawa, M. Hosokawa, K. Kobayashi, K. Chiba, E. Nanba, M. Oshimura, T. Sato, et al. Allelic expression imbalance of the human CYP3A4 gene and individual phenotypic status Hum. Mol. Genet., December 1, 2004; 13(23): 2959 - 2969. [Abstract] [Full Text] [PDF]

				T. L. Thomae, E. Glover, and C. A. Bradfield A Maternal Ahr Null Genotype Sensitizes Embryos to Chemical Teratogenesis J. Biol. Chem., July 16, 2004; 279(29): 30189 - 30194. [Abstract] [Full Text] [PDF]

				Y. Wang, K. Joh, S. Masuko, H. Yatsuki, H. Soejima, A. Nabetani, C. V. Beechey, S. Okinami, and T. Mukai The Mouse Murr1 Gene Is Imprinted in the Adult Brain, Presumably Due to Transcriptional Interference by the Antisense-Oriented U2af1-rs1 Gene Mol. Cell. Biol., January 1, 2004; 24(1): 270 - 279. [Abstract] [Full Text] [PDF]

				R. Weksberg, A. C. Smith, J. Squire, and P. Sadowski Beckwith-Wiedemann syndrome demonstrates a role for epigenetic control of normal development Hum. Mol. Genet., April 2, 2003; 12(90001): R61 - 68. [Abstract] [Full Text] [PDF]

				W. Reik, M. Constancia, A. Fowden, N. Anderson, W. Dean, A. Ferguson-Smith, B. Tycko, and C. Sibley Regulation of supply and demand for maternal nutrients in mammals by imprinted genes J. Physiol., February 15, 2003; 547(1): 35 - 44. [Abstract] [Full Text] [PDF]

				H. Yatsuki, K. Joh, K. Higashimoto, H. Soejima, Y. Arai, Y. Wang, I. Hatada, Y. Obata, H. Morisaki, Z. Zhang, et al. Domain Regulation of Imprinting Cluster in Kip2/Lit1 Subdomain on Mouse Chromosome 7F4/F5: Large-Scale DNA Methylation Analysis Reveals That DMR-Lit1 Is a Putative Imprinting Control Region Genome Res., December 1, 2002; 12(12): 1860 - 1870. [Abstract] [Full Text] [PDF]

				S. Engemann, M. Strodicke, M. Paulsen, O. Franck, R. Reinhardt, N. Lane, W. Reik, and J. Walter Sequence and functional comparison in the Beckwith-Wiedemann region: implications for a novel imprinting centre and extended imprinting Hum. Mol. Genet., November 1, 2000; 9(18): 2691 - 2706. [Abstract] [Full Text] [PDF]

				M. Paulsen, O. El-Maarri, S. Engemann, M. Strodicke, O. Franck, K. Davies, R. Reinhardt, W. Reik, and J. Walter Sequence conservation and variability of imprinting in the Beckwith-Wiedemann syndrome gene cluster in human and mouse Hum. Mol. Genet., July 22, 2000; 9(12): 1829 - 1841. [Abstract] [Full Text] [PDF]

				N. Blagitko, S. Mergenthaler, U. Schulz, H. A. Wollmann, W. Craigen, T. Eggermann, H.-H. Ropers, and V. M. Kalscheuer Human GRB10 is imprinted and expressed from the paternal and maternal allele in a highly tissue- and isoform-specific fashion Hum. Mol. Genet., July 1, 2000; 9(11): 1587 - 1595. [Abstract] [Full Text] [PDF]

				C. Schwienbacher, A. Angioni, R. Scelfo, A. Veronese, G. A. Calin, G. Massazza, I. Hatada, G. Barbanti-Brodano, and M. Negrini Abnormal RNA Expression of 11p15 Imprinted Genes and Kidney Developmental Genes in Wilms' Tumor Cancer Res., March 1, 2000; 60(6): 1521 - 1525. [Abstract] [Full Text]

				L. Carrel and H. F. Willard Heterogeneous gene expression from the inactive X chromosome: An X-linked gene that escapes X inactivation in some human cell lines but is inactivated in others PNAS, June 22, 1999; 96(13): 7364 - 7369. [Abstract] [Full Text] [PDF]

				M. P. Lee, M. R. DeBaun, K. Mitsuya, H. L. Galonek, S. Brandenburg, M. Oshimura, and A. P. Feinberg Loss of imprinting of a paternally expressed transcript, with antisense orientation to KVLQT1, occurs frequently in Beckwith-Wiedemann syndrome and is independent of insulin-like growth factor II imprinting PNAS, April 27, 1999; 96(9): 5203 - 5208. [Abstract] [Full Text] [PDF]

				J. G. Falls, D. J. Pulford, A. A. Wylie, and R. L. Jirtle Genomic Imprinting: Implications for Human Disease Am. J. Pathol., March 1, 1999; 154(3): 635 - 647. [Abstract] [Full Text] [PDF]

				J. M. Gabriel, M. J. Higgins, T. C. Gebuhr, T. B. Shows, S. Saitoh, and R. D. Nicholls A model system to study genomic imprinting of human genes PNAS, December 8, 1998; 95(25): 14857 - 14862. [Abstract] [Full Text] [PDF]

				C. Schwienbacher, L. Gramantieri, R. Scelfo, A. Veronese, G. A. Calin, L. Bolondi, C. M. Croce, G. Barbanti-Brodano, and M. Negrini Gain of imprinting at chromosome 11p15: A pathogenetic mechanism identified in human hepatocarcinomas PNAS, May 9, 2000; 97(10): 5445 - 5449. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (62) Request Permissions Google Scholar Articles by Dao, D. Articles by Tycko, B. Search for Related Content PubMed PubMed Citation Articles by Dao, D. Articles by Tycko, B. Social Bookmarking          What's this?