Retroposition of autosomal mRNA yielded testis-specific gene family on
human Y chromosome
Bruce T Lahn
& David C Page
Howard Hughes Medical Institute, Whitehead Institute
and Department of Biology, Massachusetts Institute of Technology,
9 Cambridge Center, Cambridge, Massachusetts
02142, USA.
Correspondence should be addressed to David C Page dcpage@wi.mit.edu
Most genes in the human NRY (non-recombining portion of the Y chromosome)
can be assigned to one of two groups: X-homologous genes or testis-specific
gene families with no obvious X-chromosomal homologues1,
2.
The CDY genes have been localized to the human Y chromosome1,
and we report here that they are derivatives of a conventional single-copy
gene, CDYL (CDY-like), located on human chromosome 13 and mouse
chromosome 6. CDY genes retain CDYL exonic sequences but lack
its introns. In mice, whose evolutionary lineage diverged before the appearance
of the Y-linked derivatives, the autosomal Cdyl gene produces two transcripts;
one is expressed ubiquitously and the other is expressed in testes only. In
humans, autosomal CDYL produces only the ubiquitous transcript; the
testis-specific transcript is the province of the Y-borne CDY genes.
Our data indicate that CDY genes arose during primate evolution by
retroposition of a CDYL mRNA and amplification of the retroposed gene.
Retroposition contributed to the gene content of the human Y chromosome, together
with two other molecular evolutionary processes: persistence of a subset of
genes shared with the X chromosome3,
4 and transposition of
genomic DNA harbouring intact transcription units5.
We had previously identified a single full-length cDNA clone from CDY
but had mapped homologous sequences to two different locations on the
human Y chromosome1. We explored the possibility that there
might be multiple functional CDY genes by isolating 25 additional cDNA
clones from a human testis library. Sequence analysis of the 25 clones revealed
two distinct species, which we designated CDY1 (17 clones) and
CDY2 (8 clones). The sequence of CDY1 was as previously reported1. Like CDY1, CDY2 appears to encode a protein with
a combination of chromatin-binding and catalytic domains (
Fig. 1a). The predicted coding regions of CDY1 and
CDY2 were 99% identical in nucleotide sequence, and the amino acid sequences
of the predicted proteins were 98% identical. We observed an alternative 3´
region in 4 of 17 CDY1 cDNA clones. Most of the putative protein encoded
by this minor CDY1 transcript is identical to that encoded by the major
transcript; its carboxy terminus is divergent (Fig. 1a,b).
a, Schematic representation of predicted CDY1 and CDY2 proteins.
Positions of chromo and putative catalytic domains1 are indicated,
as are protein lengths in amino acid residues. CDY1 major and minor isoforms
differ only near their C termini (grey bar in C region of CDY1 minor isoform).
b, 3´ portions of CDY1 (major and minor) and CDY2
transcripts. Shown are partial cDNA and predicted amino acid sequences; coding
sequence is placed against a black background (grey in the case of differential
portion of CDY1 minor transcript); terminal residues of predicted proteins
are numbered. Part of the intron (genomic) sequence of the CDY1 minor
transcript is shown; this intron evolved from what had been exon 9 of
CDYL (Fig. 2). The fifth nucleotide of this intron
is circled; the appearance of a G at this position (where 84% of introns19 have a G, but where human CDYL and mouse Cdyl have
a T) may have contributed to the evolution of this novel intron. Alternative
poly(A) tracts observed in some CDY1 minor and CDY2 cDNA clones
are indicated. Below polyadenylation signals (underlined) are human CDYL
sequences at corresponding positions; polyadenylation sites employed
by human CDY genes apparently arose only after retroposition, through
mutations that generated appropriate signal sequences.
We had previously localized CDY-homologous sequences to Y chromosome
deletion intervals 5L and 6F (ref. 1), but it
was not apparent which of these map locations corresponded to which gene.
Exploiting sequence differences between CDY1 and CDY2, we designed
PCR assays specific to each of the two genes. Using genomic DNAs from individuals
carrying partial Y chromosomes as templates6, we mapped
CDY1 to interval 6F, and CDY2 to interval 5L (data not shown).
We conclude that at least two distinct functional CDY genes exist on
the human Y chromosome, and that they encode protein isoforms. We cannot exclude
the possibility that there are multiple CDY1 genes in interval 6F,
multiple CDY2 genes in interval 5L, or additional CDY species
that we failed to identify.
To determine the intron/exon structures of CDY1 and CDY2,
we partially sequenced genomic BAC clones corresponding to each of the genes.
CDY1 and CDY2 genomic sequences were found to be collinear with
the major CDY1 and CDY2 transcripts as captured in cDNA clones,
with no evidence that any intronic sequences had been excised. In the case
of the minor CDY1 transcript, it appeared that a single intron had
been excised; the splice donor site was located within, and near the 3´
end of, the single exon of the major transcript (Fig. 1
b). Further evidence of the atypical nature of CDY genes
arose from examining their polyadenylation sites. For 13 of 17 CDY1
and 5 of 8 CDY2 cDNA clones, the poly(A) tract was located immediately 3´
of the predicted stop codon (Fig. 1b). CDY1
and CDY2 may be the only mammalian nuclear genes in which polyadenylation
is known to occur so close to the stop codon.
To address the evolutionary origin of CDY on the human Y chromosome,
we sought homologues that might exist elsewhere in the genomes of humans or
mice. We identified an autosomal homologue in both humans and mice by incubating
a CDY-derived probe, at low stringency, with testis cDNA libraries
from the two species. We will refer to the human and mouse autosomal homologues
as CDYL and Cdyl, respectively. We determined the sequence of
the human CDYL transcript by merging the sequences of four incomplete
but substantially overlapping cDNA clones; we found no sequence differences
among the four cDNA clones in regions of overlap. Similarly, we derived the
sequence of mouse Cdyl by merging the sequences of ten incomplete but
overlapping cDNA clones. In both species, the autosomal transcript, similar
to human CDY, appears to encode a protein with an amino-terminal chromo
domain and a C-terminal catalytic domain (Fig. 2). The
predicted mouse and human CDYL proteins have 93% amino acid identity overall,
with even greater similarity evident in the chromo and catalytic domains (Fig. 2). In contrast, the predicted human CDYL and CDY proteins
have only 63% amino acid identity; again, greater similarity is evident in
the chromo and catalytic domains (Fig. 2). These results
suggest that human CDYL and mouse Cdyl are orthologues, and
human CDYL and CDY are paralogues. They also suggest that evolutionary
constraints operating on the chromo and catalytic domains of the CDYL and
CDY proteins account for the relatively high amino acid identities observed
there.
Figure 2. Comparison of transcripts and encoded proteins from mouse Cdyl,
human CDYL and human CDY1 genes.
Exons are numbered; positions of introns are indicated. Black and grey
bars depict translated regions. Percentage identities are shown in several
regions for both DNA and predicted amino acid sequences; na, not applicable.
Note: 64% nucleotide identity was observed between the 3´ portion of
human CDYL transcript and the intron/second exon of human CDY1
minor transcript. Sequence similarity between CDY1 genomic locus and
CDYL falls abruptly (to less than 50% nucleotide identity) at both the 5´
and 3´ ends: immediately 5´ of CDYL exon 4, and 3´
of the polyadenylation site of CDY1 minor transcript. Poly(A) tracts,
including alternative locations observed in some cDNA clones, are indicated.
Chromosomal mapping of the human CDYL and mouse Cdyl genes
indicate that they are indeed orthologues. In humans, we localized CDYL
to the distal short arm of chromosome 6 (distal to marker NIB1876 and
proximal to WI-4489) by radiation hybrid mapping. In mice, we localized
Cdyl to the proximal portion of chromosome 13 (distal to Fim1 and
proximal to D13Mit18) by meiotic linkage mapping. These portions of
human chromosome 6 and mouse chromosome 13 were previously shown to be syntenic7. We conclude that a gene much like human CDYL and mouse
Cdyl existed in the common ancestors of mice and humans.
Our experiments to this point left another evolutionary question unanswered:
which came first, CDY or CDYL? The first indication came from
studies in mice, in which we were unable to identify a Y-linked homologue
of human CDY by either Southern-blot analysis (Fig.
3a) or screening of testis cDNA libraries. These findings are
consistent with CDYL having given rise to CDY sometime after
divergence of the mouse and human lineages.
Figure 3. Homologues of CDYL and CDY in diverse mammalian species.
a, Southern blots of male and female genomic DNAs from 13 mammalian
species hybridized with cDNA fragments corresponding to the entire coding
sequence of either human CDYL (top, TaqI digest) or human
CDY1 (bottom, BglII digest). At low stringency, CDYL probe
hybridized with male-female common (presumably autosomal) homologues in all
13 species. It also cross-hybridized with Y-encoded homologues in some primates,
especially orangutan and squirrel monkey. CDY probe identified Y-encoded
homologues in all primates except lemurs. It also cross-hybridized weakly
to the autosomal homologue in some species. Positions of size markers are
shown at right. The apparent smears in the squirrel monkey lanes are composed
of intense, discrete fragments, as revealed by shorter exposures. b,
Summary and interpretation of Southern-blot data. The 13 species tested are
arranged phylogenetically20,
21,
22.
Additional evidence that CDYL gave rise to CDY came from
analysis of the human CDYL and mouse Cdyl gene structures. We
found that each of these autosomal genes contains eight introns (
Fig. 2). As one would expect for orthologues, most introns were positioned
at corresponding sites in human CDYL and mouse Cdyl. The presence
of numerous introns in the CDYL/Cdyl orthologues, contrasted
with the paucity of introns in the human CDY genes, led us to consider
whether CDY had arisen by reverse transcription of a spliced CDYL
mRNA, with subsequent integration into the Y chromosome.
Closer examination of the sequence data corroborated this retroposition
model. Excluding a few hundred nucleotides at the 5´ terminus, the nucleotide
sequence of the human CDY1 (or CDY2) genomic locus is essentially
collinear with that of the mature human CDYL (or mouse Cdyl)
transcripts. Even the single intron of the minor CDY1 transcript is
collinear with, and homologous to, CDYL exon 9 (Fig.
2). Thus, with respect to evolutionary origins, all but the most 5´
portions of human CDY1 and CDY2 genomic loci can be accounted
for by mature CDYL/Cdyl transcripts. These findings are as one
would predict if CDYL/Cdyl intronic sequences had been excised
before retroposition.
How long ago did the retroposition event occur? To address this question,
we incubated probes derived from human CDYL and CDY with Southern
blots of male and female genomic DNAs from various mammalian species. Two
conclusions emerged from this analysis (Fig. 3a):
(i) the autosomal (male-female common) CDYL gene appears to be widely
conserved among the mammals tested, including opossum, a marsupial; and (ii),
the Y-chromosomal (male-specific) CDY family appears to be restricted
to simian primates, including apes, Old World monkeys (for example macaques)
and New World monkeys (for example squirrel monkeys, whose sequences appear
to have undergone extraordinary amplification on the Y). We observed no male-specific
hybridization fragments in prosimians (lemurs) or any of the non-primate mammals
tested. The simplest interpretation of these findings is that the retroposition
event occurred in the simian lineage after its divergence from prosimians
but before the split between Old and New World monkeys, perhaps 40-50 million
years ago (Fig. 3b). Given the age of the retroposed
CDY genes, it is not unexpected that certain hallmarks of retroposition
are absent: we found no evidence of target site duplication or of a poly(A)
tract in CDY genomic sequences. The observed 99% identity between the
coding sequences of human CDY1 and CDY2 (Fig.
1a; as contrasted with 73% identity between human CDY
and human CDYL) suggests that their coexistence is the result of subsequent
amplification on the Y chromosome.
Do all members of the CDY/CDYL gene family serve male-specific
functions? We had previously demonstrated by northern-blot analysis that human
CDY genes are abundantly transcribed in adult testis but are not detectable
in other adult tissues1 (Fig. 4). In contrast,
human CDYL appears to be expressed at a modest and comparable level
in all tissues examined (Fig. 4), indicating that it
may perform a housekeeping function. In mice, Cdyl is characterized
by two transcripts: (i) a ubiquitous one that is similar in length to the
ubiquitous transcript of human CDYL; and (ii) a testis-specific one
that is similar in length to the testis-specific transcript of human CDY
(Fig. 4). These expression studies in humans and
mice suggest that a genomic partitioning of housekeeping and testis-specific
functions evolved in primates after retroposition. We interpret the mouse,
which possesses no CDY homologue on its Y chromosome, as representing
the evolutionarily ancestral condition. In humans, the autosomal gene
CDYL appears to retain the housekeeping function but relinquished the
testis-specific function, the latter of which seems to have transferred to
the Y-linked CDY genes.
Figure 4. Tissue distributions of mouse Cdyl, human CDYL and human
CDY transcripts.
Northern blots incubated cDNA fragments corresponding to the entire coding
sequence of either mouse Cdyl (top), human CDYL (middle), or
human CDY1 (bottom) reveal tissue expression patterns; the phylogenic
relationship of these genes (illustrated on left) is based on data presented
in Fig. 3.
We propose, on the basis of these and other findings, that the gene content
of the human NRY was forged during evolution through the interplay of three
molecular processes: retroposition, transposition and persistence (
Fig. 5). In certain respects, the process of CDY retroposition
is reminiscent of the transposition process by which the DAZ gene family
arose on the Y chromosome5. The net result of both evolutionary
processes was the duplicative transfer of an autosomal gene to the Y chromosome,
with retention of the progenitor autosomal gene. Subsequent to retroposition
(as exemplified by CDY) and transposition (as exemplified by DAZ
) the genes transferred to the Y chromosome were therein amplified. The
expression pattern of the Y-linked CDY genes, however, is very different
to that of their autosomal progenitor, Cdyl/CDYL (Fig.
4), whereas expression of the Y-linked DAZ genes is quite similar
to that of their autosomal progenitor, DAZL (also known as DAZLA
, DAZH or SPGYLA; Refs 5,8-11). In the case of CDY, the nascent Y-linked
gene was fashioned from a fully processed, reverse-transcribed cDNA which,
we speculate, fortuitously integrated near an existing promoter or into an
otherwise transcriptionally permissive locale on the Y chromosome. In contrast,
in the case of DAZ, a segment of genomic DNA containing an entire transcription
unit was duplicated from an autosome onto the Y chromosome5.
The transposed genomic DNA of DAZ probably included not only introns
and exons but also the promoter of the ancestral gene.
Figure 5. Schematic representation of three molecular evolutionary processes
that contributed genes to human NRY.
Persistence (top): an autosomal pair gave rise to the neo-Y and neo-X (subsequently
enlarged by fusion with other autosomes or autosomal segments; data not shown;
refs 23,24).
SRY, RPS4Y and several other genes derived from these ancestral
autosomes persist as X-homologous genes in the human NRY (refs 1,2,3,4,25). Transposition: the
DAZ genes arose by transposition (and subsequent amplification; data not
shown) of autosomal genomic DNA containing the entire DAZL transcription
unit5. Retroposition: the CDY genes arose by integration
(and subsequent amplification; data not shown) of a reverse-transcribed copy
of a processed mRNA derived from the autosomal CDYL gene. Gene sizes
are not to scale.
Retroposition has long been suspected to figure prominently in the evolution
of animal Y chromosomes—but solely as a means of marking the decay of
genes previously shared with the X chromosome. In Drosophila melanogaster
, molecular studies have suggested that insertional mutagenesis via retroposition
was a major driving force in Y gene decay12,
13,
14. An accumulation
of retroposed elements on the mouse Y chromosome15 suggested
that this paradigm might extend to mammals16. Thus, it is reasonable
to speculate that retroposition—especially of highly transposable elements—may
have had an important role in Y gene decay in many animal lineages. As illustrated
here by CDY, however, retroposition has also provided a mechanism for
gene building during the evolution of the human Y chromosome.
Methods Identification of cDNA clones. The cDNA insert of plasmid
pDP1660 contains the entire ORF of human CDY1, as previously reported1. To identify additional CDY clones, we labelled the plasmid
insert with 32P-dCTP by random priming and used this probe
to screen a human adult testis cDNA library (Clontech). We incubated blots
overnight at 65 °C in NaiPO4 (0.5 M), 7% SDS, followed
by three washes of 15 min each at 65 °C in 0.1SSC, 0.1% SDS. We
identified human CDYL clones by rescreening the same library with the
same probe at lower stringency (incubation at 60 °C, as above; washing
at 55 °C in 0.1SSC, 0.1% SDS). We also used this probe and conditions
to screen a mouse adult testis cDNA library (Clontech), resulting in the identification
of mouse Cdyl clones.
PCR analysis of human genomic DNAs. We tested human
genomic DNAs for the presence or absence of CDY1 and CDY2 using
PCR assays specific to each of the genes. PCR primers for CDY1 were
5'-GGCGAAAGCTGACAGCAA-3' and 5'-GGGTGAAAGTTCCAGTCAA-3´. PCR primers
for CDY2 were 5'-GACCACAAGAAAACTGTGAG-3' and 5'-GATCTGCTGCAATAGGGTC-3´.
Thermocycling conditions were 30 cycles of 1 min at 94 °C, 45 s at 60
°C and 45 s at 72 °C.
Characterization of intron/exon structures. We identified
BAC clones corresponding to the human CDY1, human CDY2, human
CDYL and mouse Cdyl genomic loci by hybridization screening on
high-density filters (Research Genetics) of CITB BAC libraries prepared from
human or mouse genomic DNA. We then used these BAC clones to characterize
the intron/exon structures of the genes. We designed a series of PCR assays,
based on the cDNA sequence of each gene, each yielding a cDNA product of approximately
50 bp and collectively encompassing the entire cDNA sequence. We then repeated
each of these PCR assays using the corresponding BAC as template. In some
cases, we obtained PCR products of identical length using cDNA or genomic
BAC as template, indicating that no introns were located between the two primers.
In other cases, a larger PCR product was obtained using the BAC as template
(in some cases long-range PCR employing TaqExtender enzyme (Stratagene) was
required), indicating the presence of an intron. We sequenced all intron-containing
PCR products to identify intron/exon boundaries precisely.
Radiation hybrid mapping of human CDYL. Using
PCR, we tested DNAs from the 93 hybrid cell lines of the GeneBridge 4 panel17 (Research Genetics) for the presence of CDYL. PCR primers
were 5'-CTGAGCAGGAGAACATCACC-3' and 5'-GCTACGGGTGAGCTTGTTTC-3'. Thermocycling
conditions were as listed above. Analysis of the results positioned CDYL
with respect to the radiation hybrid map constructed at the Whitehead/MIT
Center for Genome Research18
(http://www-genome.wi.mit.edu/cgi-bin/contig/phys_map).
Genetic mapping of mouse Cdyl. We typed genomic
DNA from the BSS ([C57BL/6JEiSPRET/Ei]F1 femaleSPRET/Ei male)
backcross panel (Jackson Laboratory) for a Cdyl polymorphism by PCR.
PCR primers were 5'-AAAAATGACCCTCACTAAGTTAC-3' and 5'-AGGCTCCCTGCAGTAAGTA-3'.
Thermocycling conditions were as listed above. This PCR assay yielded a 53-bp
product when C57BL/6 genomic DNA was used as template, but no product with
Mus spretus DNA (because of sequence polymorphism at priming sites). Analysis
of the results positioned Cdyl with respect to the genetic linkage
map maintained at the Jackson Laboratory (http://www.informatics.jax.org/map.html).
Southern- and northern-blot analyses. Each Southern-blot
lane contained genomic DNA (7 g). We incubated blots overnight at 65 °C
in NaiPO4 (0.5 M), 7% SDS, followed by three washes
of 15 min each at 60 °C in 1SSC, 0.1% SDS. For northern-blot analysis,
commercial filters (Clontech) doped with poly(A)+ RNA (2 g/lane) were
incubated under the same conditions, then washed at 65 °C in 0.1SSC,
0.1% SDS.
Received 9 November 1998; Accepted 25 February 1999
REFERENCES
Lahn, B.T. & Page, D.C. Functional coherence of the human Y chromosome. Science278, 675−680 (1997). | Article | PubMed | ISI | ChemPort |
Vogt, P.H. et al. Report of the Third International Workshop on Y chromosome mapping 1997. Cytogenet. Cell Genet.79, 1−20 (1997). | PubMed | ChemPort |
Graves, J.A.M. The origin and function of the mammalian Y chromosome and Y-borne genes—an evolving understanding. BioEssays17, 311−321 (1995). | PubMed | ISI | ChemPort |
Jegalian, K. & Page, D.C. A proposed path by which genes common to mammalian X and Y chromosomes evolve to become X inactivated. Nature394, 776−780 (1998). | Article | PubMed | ISI | ChemPort |
Saxena, R. et al. The DAZ gene cluster on the human Y chromosome arose from an autosomal gene that was transposed, repeatedly amplified and pruned. Nature Genet.14, 292−299 (1996). | Article | PubMed | ISI | ChemPort |
Vollrath, D. et al. The human Y chromosome: a 43-interval map based on naturally occurring deletions. Science258, 52−59 (1992). | PubMed | ISI | ChemPort |
Yen, P.H., Chai, N.N. & Salido, E.C. The human autosomal gene DAZLA: testis specificity and a candidate for male infertility. Hum. Mol. Genet.5, 2013−2017 (1996). | Article | PubMed | ISI | ChemPort |
Shan, Z. et al. A SPGY copy homologous to the mouse gene Dazla and the Drosophila gene boule is autosomal and expressed only in the human male gonad. Hum. Mol. Genet.5, 2005−2011 (1996). | Article | PubMed | ISI | ChemPort |
Cooke, H.J., Lee, M., Kerr, S. & Ruggiu, M. A murine homologue of the human DAZ gene is autosomal and expressed only in male and female gonads. Hum. Mol. Genet.5, 513−516 (1996). | Article | PubMed | ISI | ChemPort |
Reijo, R. et al. Mouse autosomal homolog of DAZ, a candidate male sterility gene in humans, is expressed in male germ cells before and after puberty. Genomics35, 346−352 (1996). | Article | PubMed | ISI | ChemPort |
Ganguly, R., Swanson, K.D., Ray, K. & Krishnan, R. A BamHI repeat element is predominantly associated with the degenerating neo-Y chromosome of Drosophila miranda but absent in the Drosophila melanogaster genome. Proc. Natl Acad. Sci. USA89, 1340−1344 (1992). | PubMed | ChemPort |
Steinemann, M. & Steinemann, S. Degenerating Y chromosome of Drosophila miranda: a trap for retroposons. Proc. Natl Acad. Sci. USA89, 7591−7595 (1992). | PubMed | ChemPort |
Steinemann, M. & Steinemann, S. The enigma of Y chromosome degeneration: TRAM, a novel retrotransposon is preferentially located on the Neo-Y chromosome of Drosophila miranda. Genetics145, 261−266 (1997). | PubMed | ISI | ChemPort |
Eicher, E.M., Hutchison, K.W., Phillips, S.J., Tucker, P.K. & Lee, B.K. A repeated segment on the mouse Y chromosome is composed of retroviral-related, Y-enriched and Y-specific sequences. Genetics122, 181−192 (1989). | PubMed | ISI | ChemPort |
Charlesworth, B., Sniegowski, P. & Stephan, W. The evolutionary dynamics of repetitive DNA in eukaryotes. Nature371, 215−220 (1994). | Article | PubMed | ISI | ChemPort |
Gyapay, G. et al. A radiation hybrid map of the human genome. Hum. Mol. Genet.5, 339−346 (1996). | Article | PubMed | ISI | ChemPort |
Hudson, T.J. et al. An STS-based map of the human genome. Science270, 1945−1954 (1995). | PubMed | ISI | ChemPort |
Burge, C. & Karlin, S. Prediction of complete gene structures in human genomic DNA. J. Mol. Biol.268, 78−94 (1997). | Article | PubMed | ISI | ChemPort |
Kumar, S. & Hedges, S.B. A molecular timescale for vertebrate evolution. Nature392, 917−920 (1998). | Article | PubMed | ISI | ChemPort |
Pilbeam, D. The descent of hominoids and hominids. Sci. Am.250, 84−96 (1984). | PubMed | ISI | ChemPort |
Ohno, S. Sex Chromosomes and Sex-linked Genes (Springer-Verlag, Berlin, 1967).
Watson, J.M., Spencer, J.A., Riggs, A.D. & Graves, J.A. Sex chromosome evolution: platypus gene mapping suggests that part of the human X chromosome was originally autosomal. Proc. Natl Acad. Sci. USA88, 11256−11260 (1991). | PubMed | ChemPort |
Foster, J.W. & Graves, J.A. An SRY-related sequence on the marsupial X chromosome: implications for the evolution of the mammalian testis-determining gene. Proc. Natl Acad. Sci. USA91, 1927−1931 (1994). | PubMed | ChemPort |
Acknowledgments We thank H. Skaletsky and F. Lewitter for assistance with sequence analysis;
the San Diego Zoo and the Duke Primate Center for animal specimens; and P.
Bain, A. Bortvin, L. Brown, C. Burge, B. Charlesworth, A. Chess, S. Gilbert,
R. Jaenisch, T. Kawaguchi, K. Kleene, D. Menke, R. Saxena, C. Sun, C. Tilford
and J. Wang for helpful discussions and comments on the manuscript.
SSC, 0.1% SDS. We
identified human CDYL clones by rescreening the same library with the
same probe at lower stringency (incubation at 60 °C, as above; washing
at 55 °C in 0.1
g). We incubated blots overnight at 65 °C
in NaiPO4 (0.5 M), 7% SDS, followed by three washes
of 15 min each at 60 °C in 1
