Mammalian Genome 3: 625-632, 1992

~enome 9 Springer-VerlagNew YorkInc. 1992

Genetic mapping of the murine gene and 14 related sequences encoding chromosomal protein HMG-14 Kenneth R. Johnson, 1 Sue A. Cook, 1 Michael Bustin, z and Muriel T. Davisson 1 ~The Jackson Laboratory, 600 Main Street, Bar Harbor, Maine 04609, USA; ZLaboratory of Molecular Carcinogenesis, National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892, USA Received May 26, 1992; accepted July2, 1992

Abstract. The high-mobility-group chromosomal pro-

tein HMG-14 preferentially binds to nucleosomal core particles of mammalian chromatin and may modulate the chromatin configuration of transcriptionally active genes. The human gene for HMG-14 has been localized to the Down syndrome region of Chromosome (Chr) 21 and may be involved in the etiology of this syndrome. Here we show, by means of genetic linkage analysis of interspecific and intersubspecific backcross mice, that the murine functional gene, Hmg14, is located on the distal end of mouse Chr 16, a region known to have conserved synteny with human Chr 21. In addition to the functional gene for HMG-14, both human and mouse genomes contain many related sequences that are probably processed pseudogenes. Here we map the locations of 14 Hmgl4-related sequences in two mouse genomes. The 14 mapped loci are widely dispersed on ten chromosomes (Chrs 3, 5, 7, 9, 11, 12, 16, 17, 19, and X) and can be detected efficiently with a single c D N A probe. Thus, the Hmg14 multigene family is well suited to serve as genetic markers for other linkage studies in mice.

Introduction

The high-mobility-group (HMG) proteins are among the most abundant class of nonhistone proteins associated with mammalian chromosomal DNA. HMG-14, and the closely related protein HMG-17, bind specifically to the nucleosomal core particle. It has been suggested that these two proteins may modulate the chromatin structure of transcriptionally active genes (Bustin et al. 1990). The gene for HMG-14 maps to the Down syndrome region of human Chr 21, suggesting that HMG-14 may be a contributing factor in the etiology of this syndrome (Pash et al. 1990). Offprint requests to: K.R. Johnson

Although HMG-14 message and protein have been shown to be overexpressed in mouse embryos that are trisomic for Chr 16 (Pash et al. 1990), the murine gene for HMG-14, Hmgl4, has not been genetically mapped. A confounding factor in mapping this gene in both human and mouse genomes has been the presence of multiple sequences that cross-hybridize with cDNA probes in Southern blot analysis (Landsman et al. 1986, 1989). These related sequences are probably processed pseudogenes, on the basis of the following evidence: (1) in mammals, there is only a single species of HMG-14 protein and a single-sized mRNA, suggesting that there is only a single functional gene for HMG-14 (Landsman et al. 1986) (2) the chicken genome contains only a single gene copy for the homolog of the HMG-14 protein found in mammals (Srikantha et al. 1990); (3) the multiple sequences that cross-hybridize with the cDNA for the closely related chromosomal protein HMG-17 have been shown by sequence analysis to be retropseudogenes (Srikantha et al. 1987); and (4) in humans, only a single gene hybridizes with a DNA probe derived from an intron sequence; the remaining related sequences apparently lack introns (Landsman et al. 1989; Pash et al. 1990), a hallmark of processed pseudogenes (Vanin 1985). We report here the genetic map locations in the mouse of the functional Hmg14 gene and 14 Hmg14related sequences (probably processed pseudogenes). The map locations were determined by genetic linkage analysis of 144 interspecific and 144 intersubspecific backcross mice. To identify the functional gene from among the many related sequences that hybridized with the murine Hmg14 cDNA probe, we used a probe derived from the first intron of the murine gene. We show that the functional gene is located on the distal end of mouse Chr 16, in a region known to have homologies with the Down syndrome region of human Chr 21. DNA probes that detect multiple loci can be used to quickly and efficiently scan the mouse genome.

K.R. J o h n s o n et al.: Mapping of the HMG-14 multigene family

626

Once mapped, members of multigene families can be used as genetic reference markers for mapping other genes and mutations. Many endogenous murine retroviruses (Kozak et al. 1987; Taylor and Rowe 1989; Frankel et al. 1990, 1992) and members of genepseudogene families (Siracusa et al. 1991; RichardsSmith and Elliott 1992) have been previously characterized and are currently being used as markers for mouse gene mapping. Here, we report the map locations of 14 members of a new multigene family--the Hmgl4-related sequences--on ten different mouse chromosomes (Chrs 3, 5, 7, 9, 11, 12, 16, 17, 19, and X). Because they are widely dispersed and can be efficiently detected with a single c D N A probe, the mapped Hmg14-related loci are well suited to serve as reference markers for future genetic linkage studies in mice.

Materials and methods

Mice

with C57BL/6J; F1 females were then b a c k c r o s s e d to C57BL/6J males. All mice were maintained u n d e r standard Jackson Laboratory conditions.

Identification of Hmg14 loci The m o u s e c D N A for HMG-14 described by L a n d s m a n and Bustin (1990) was u s e d as a probe for detecting the multiple Hmg14-related s e q u e n c e s in m o u s e genomic D N A . The single functional Hmg14 locus was identified by hybridization with the cloned first intron of the m o u s e Hmg14 gene. This 181-bp intron s e q u e n c e was obtained by PCR amplification of a genomic clone ( L e h n and Bustin, unpublished data). The primers used for amplification c o r r e s p o n d e d to the murine sequences flanking intron I as predicted from the intronexon boundaries of the h u m a n HMG-14 gene. Intron I consists of nucleotides 220-540 of the complete 6804-bp-long h u m a n HMG-14 gene, as described by L a n d s m a n and co-workers (1989). T h e corresponding position of this intron is b e t w e e n nucleotides 87 and 88 of the m o u s e Hmg14 c D N A sequence, as reported by L a n d s m a n and Bustin (1990). The amplified s e q u e n c e was cloned into Bluescript II (Stratagene), and the insert from this clone (pM14glntronl) w a s used as a probe in Southern blot analysis. Probe labeling, S o u t h e r n blotting, and hybridization procedures were as previously described (Johnson et al. 1992).

Reference loci

All m o u s e stocks u s e d in this s t u d y were obtained from the M o u s e M u t a n t R e s o u r c e colony of T h e J a c k s o n Laboratory (Bar Harbor, Me.). The interspecific and intersubspecific b a c k c r o s s e s have been previously described (Johnson et al. 1992). For the interspecific backcross, mice from an inbred strain of Mus spretus (SPRET/Ei) were m a t e d with C57BL/6J; F~ females were then b a c k c r o s s e d to SPRET/Ei males. F o r the intersubspecific backcross, mice from an inbred strain o f Mus musculus castaneus (CAST/Ei) were m a t e d

Table 1 lists the prepositioned reference loci that were u s e d in the linkage analysis of Hmgl4 and Hmg14-related sequences. T w e n t y one of the reference loci were typed as D N A restriction fragment length p o l y m o r p h i s m s (RFLPs), and s e v e n were typed as protein p o l y m o r p h i s m s (Table 1). For R F L P analysis, only the c D N A inserts from clones were used as probes. Oligonucleotide probes were hybridized as previously described (Johnson 199I).

Table 1. Reference loci used in the linkage analysis of Hmgl4 and Hmg14-related sequences. Gene name Friend MuLV integration site-3 amylase-l, salivary P glycoprotein-1 phosphogiucomutase-1 polytropic routine leukemia virus-4 ornithine decarboxylase-related sequence-7 tyrosinase glucosephosphate isomerase-1 hemoglobin 13-chain complex triosephosphate isomerase related sequence-4 mannose phosphate isomerase-1 interleukin-4 xenotropic murine leukemia virus-42 esterase-3 ct-l-antitrypsin modified polytropic murine leukemia virus-11 protamine-1 pituitary-specific transcription factor-1 E26 avian leukemia oncogene-2 myxovirus resistance-1 proviral integration site-1 t-complex protein-10b complement component-3 cytochrome P450, 2c adrenergie receptor, c~-2 hypoxanthine-guanine phosphoribosyl transferase phosphoglycerate kinase-1 ornithine decarboxylase

Protein a or DNA probe b

Locus

Chr

Referencec

mouse DNA protein hamster cDNA protein JS-5* oligonucleotide mouse cDNA* human cDNA protein protein mouse cDNA* protein mouse cDNA JS-6* oligonucleotide protein mouse cDNA JS-4* oligonucleotide mouse cDNA mouse cDNA mouse DNA mouse cDNA mouse cDNA mouse cDNA protein human cDNA human cDNA mouse cDNA

Fim-3 Amy-I Pgy-1 Pgm-1 Pmv-4 Odc-rs7 Tyr Gpi-1 Hbb Tpi-4 Mpi II-4 Xmv-42 Es-3 Aat Mpmv-ll Prm-1 Pit-1 Ets-2 Mx-1 Pim-1 Tcp-lOb C3 Cyp2c Adra-2 Hprt

3 3 5 5 7 7 7 7 7 9 9 11 11 11 12 12 16 16 16 16 17 17 17 19 19 X

Sola et aI. 1988 Kaplan et al. 1973 ATCC # 39839 Shows et al 1969 Frankel et al. 1989b Richards-Smith and Elliott 1992 ATCC # 59510 DeLorenzo and Ruddle 1969 Whitney 1978 ATCC # 63090 Nichols et al. 1973 ATCC # 37561 Frankel et al. 1989a Martin and Petras 1971 D'Eustachio 1984 Frankel et al. 1990 Reeves et al. 1987 ATCC # 63102 Reeves et al. 1987 ATCC # 63004 Cuypers et al. 1984 ATCC # 63003 Natsuume-Sakai et al. 1978 ATCC # 61248 ATCC # 59302 ATCC # 37424

human cDNA* mouse cDNA*

Pgk-1 Odc-rsl3

X X

ATCC # 57222 Richards-Smith and Elliott 1992

a The word "protein" in this column indicates that a protein polymorphism rather than a DNA RFLP was used to type the locus. b An asterisk indicates that the DNA probe hybridizes wth multiple dispersed loci. Full descriptions of ATCC clones are given in ATCC/NIH Repository Catalogue of Human and Mouse DNA Probes and Libraries, 5th edition, 1991, American Type Culture Collection, Rockville, Md. 20852.

K.R. Johnson et at.: Mapping of the HMG-14 multigene family

Statistical analysis Genetic linkage was determined by segregation analysis of the interspecific and intersubspecific backcross progeny. Map locations, gene order, and recombination frequencies were calculated by the method of maximum likelihood by use of the computer program MAPMAKER (Lander et at. 1987).

Results

We used DNA RFLPs to identify and map Hmg14 and individual Hmgl4-related loci. Southern blot autoradiograms of genomic DNA from 12 inbred mouse strains hybridized with the murine Hmgl4 cDNA probe are shown in Fig. 1. Those DNA fragments that were mapped by means of the interspecific and intersubspecific backcrosses are indicated by numbers in Fig. 1 corresponding to their locus designations. The letter " f " is used to indicate DNA fragments containing intron I of the functional Hmg14 gene; the identification of the functional gene is described below. Based on our Southern blot analysis (Figs. 1 and 2), we estimate that there are 15-22 Hmg14-related sequences in the genomes of the mouse strains we examined. All of the mouse strains contained approximately the same number of Hmg14-related DNA fragments. There is much less D N A fragment size polymorphism among the standard inbred strains of mice (lanes a, b, c, d, f, g, i, j, 1; Fig. 1) than there is between these strains and CAST-Ei (lane e, Fig. 1),

Fig. 1. Southern blot autoradiograms illustrating RFLPs of Hrng14related sequences among 12 inbred strains of mice. Genomic DNA from each strain was digested with BamHl (panel A), PvulI (panel B), and TaqI (panel C) and hybridized with the mouse Hmgl4 cDNA probe. The strains are as follows: female BALB/cBy (lane a), female C3H/HeJ (lane b), male C57L/J (lane c; lane d in Panel C), male C57BL/6J (lane d; lane c in Panel C), male CAST/Ei (lane e), male CBA/J (lane f), female DBA/2J (lane g), male MOLF/Ei (lane h), male MEV/1Ty (lane i), male SM/J (lane j), female SPRET/Ei (lane k), and male SWR/J (lane 1). DNA fragments from C57BL/6J and

627

MOLF/Ei (an inbred strain derived from Mus musculus molossinus; lane h, Fig. 1), or SPRET/Ei (lane k, Fig. 1). For example, C3H/HeJ and C57BL/6J have the same sized BamHl fragments (lanes b and d, Fig. IA); they differ by only two of approximately 16 PvulI fragments (lanes b and d, Fig. 1B) and by only six of approximately 22 TaqI fragments (lanes b and c, Fig. 1C). In contrast, C57BL/6J and CAST/Ei differ by at least six of approximately 13 BamH1 fragments (lanes d and e, Fig. 1A), eight of approximately 16 PvulI fragments (lanes d and e, Fig. 1B), and 11 of approximately 22 TaqI fragments (lanes c and e, Fig. 1C). The segregation of Hmg14 and six Hmg14-related sequences (Hmg14-rsl, -rs2, -rs3, -rs4, -rs6, and -rs7) among progeny from the (C57BL/6J x SPRET/Ei) x SPRET/Ei interspecific backcross is illustrated in Fig. 2A for PvulI-digested DNA. BamH1 polymorphisms (lane d, Fig. IA) could be used to score two additional loci (Hmg14-rs5 and -rs8) in this cross. Locus genotypes were scored as the presence or absence in backcross mice of a particular DNA fragment that originated from the genome of the nonrecurrent backcross parent (C57BL/6J). The segregation of Hmgl4 and ten Hmg14-related sequences (Hmg14-rsl, -rs2, -rs4, -rs6, -rs7, -rslO, -rs11, -rsl2, -rs13, and -rsI4) among progeny from the (C57BL/6J x CAST/Ei) x C57BL/6J intersubspecific backcross is shown in Fig. 2B for TaqIdigested DNA. Two additional loci (Hmg14-rs5 and -rs9) could be scored as BamH1 polymorphisms (lane

CAST/Ei that were genetically mapped and identified as the loci

Hmg14-rs1, -rs2, -rs3, -rs4, -rs5, -rs6, -rs7, -rs8, -rs9, -rslO, -rs11, -rs12, -rs13, and -rs14 are indicated by corresponding numbers and arrows. DNA fragments containing intron I sequences of the functional Hmg14 gene are identified by the letter " f . " Differently sized DNA fragments originating from the same mouse strain were given the same gene designations if they cosegregated among all backcross mice, DNA fragments from C57BL/6J and CAST/Ei were given the same gene designations if they mapped to nearly identical locations in both backcrosses. Molecular size markers are shown on the right.

628

K.R. Johnson et al.: Mapping of the HMG-14 multigene family

Fig. 2. Segregation of Hmgl4 and Hmg14-related RFLVs among progeny of the (C57BL/6J x SPRET/Ei) x SPRET/Ei interspecific backcross (panel A; PvulI digests) and among progeny of the (C57BL/6J x CAST/Ei) x C57BL/6J intersubspecific backcross (panel B; TaqI digests). Digested genomic DNA from each backcross mouse was hybridized with the mouse Hmgl4 cDNA probe.

The last two lanes contain control DNAs from the parental strains. Genotypes of the backcross mice were scored as the presence or absence of DNA fragments originating from the nonrecurrent backcross parent. An arrow points to each of these diagnostic fragments in the control lane, and the locus each identifies is shown to the right. Molecular size markers are shown on the left.

e, Fig. 1A) and one additional locus (Hmgl4-rsS) could be scored as a PvulI polymorphism (lane e, Fig. 1B). In this case, the diagnostic DNA fragments that were mapped originated from the CAST/Ei genome. To unambiguously identify which DNA fragments corresponded to the functional Hmg14 gene in both backcrosses, we used the first intron of the mouse Hmg14 gene as a probe. Only a single locus was detected with this probe as shown in Fig. 3. A one-to-one correspondence of allelic segregation was observed between the DNA fragments detected with this intron probe and those fragments indicated with the letter " f " in Fig. 1 and as Hmg14 in Fig. 2. PvulI fragment sizes containing the first intron of Hrngl4 were 1.8 kb in C57BL/6J (lane d, Fig. 1B and Fig. 2A) and 2.0 kb in CAST/Ei (lane e, Fig. 1B). TaqI fragment sizes were 1.4 kb in C57BL/6J (lane c, Fig. 1C and Fig. 3), 1.0 kb in CAST/Ei (lane e, Fig. 1C and Fig. 2B), and 0.8 kb in SPRET/Ei (lane k, Fig. 1C and Fig. 3). Genetic linkage of Hmg14 and Hmgl4-related sequences was detected with 28 prepositioned reference loci on ten mouse chromosomes (Table 2). Gene order was determined by maximum likelihood three-point analysis; recombination estimates between ordered loci are given in Table 2. The functional Hmgl4 gene mapped to the distal end of mouse Chr 16. We detected a single recombinant between Hrng14 and Ets-2 among 141 interspecific backcross mice typed for both loci; based on haplotype analysis of the recombinant mouse, the most likely position for Hmgl4 is 0.7 cM

distal to Ets-2 (Table 2A). In 124 intersubspecific backcross mice examined, we found no recombinants between Hmg14 and Mx-1 (Table 2B). Ets-2 and Mx-1 are located 57 and 58 cM, respectively, from the Chr 16 centromere on the composite genetic map of the mouse (GBASE, 1992).

Hmg14

intron probe Taql (C57BL/6J X SPRET/Ei) X SPRET/Ei

k

tllele

1,

C57BL/6J

0

SPRET/Ei Six Backcross Progeny

Fig. 3. Segregation of TaqI fragments associated with the functional Hmg14 gene among six progeny of the (C57BL/6J x SPRET/Ei) x SPRET/Ei interspecific backcross. Genomic DNA from each backcross mouse was digested with TaqI and hybridized with the probe derived from the first intron of mouse Hmg14. Only a single locus was detected with this probe. The C57BL/6J allele was identified as a 1.4 kb TaqI fragment and the SPRET/Ei allele as a 0.8 kb fragment; backcross mice were either homozygous for the SPRET/Ei allele or heterozygous for both alleles.

K . R . J o h n s o n et al.: M a p p i n g o f the H M G - 1 4 m u l t i g e n e f a m i l y

In addition to the functional Hmgl4 gene, we were able to genetically map eight Hmgl4-related sequences in the C57BL/6J genome (Table 2A) and 13 in the CAST/Ei genome (Table 2B). The map locations of seven Hmg14-related sequences in the C57BL/6J genome were nearly identical with the locations of seven sequences in the CAST/Ei genome; these sequences were presumed to represent the same loci in both genomes and were therefore given the same locus designations. The 14 different Hmgl4-related loci that we mapped are widely distributed on ten different chromosomes: Chrs 3, 5, 7, 9, 11, 12, 16, 17, 19, and X (Table 2, Fig. 4).

629

Discussion

1990). The putative role of HMG-14 in transcriptional modulation and the location of the human gene in the Down syndrome region of Chr 21 led Pash and colleagues (1990) to speculate that dosage aneuploidy involving HMG-14 may be a contributing factor in the etiology of the syndrome. Mice fully or partially trisomic for the distal region of Chr 16 can serve as animal models for Down syndrome (Gearhart et al. 1986). The ability to identify the functional Hmgl4 gene and knowledge of its precise location on mouse Chr 16 will aid in the analysis of these animal model systems. Even if the HMG-14 protein is not directly involved in the etiology of the syndrome, the localization of the Hmgl4 gene provides an additional genetic reference point for the analysis of candidate genes in this region.

The functional Hmgl4 gene

Hmgl4-related sequences

We have identified the functional mouse Hmgl4 gene from among more than 15 Hmgl4-related sequences and mapped it to the distal region of mouse Chr 16, a region known to have conserved linkage with the Down syndrome region of human Chr 21 (Reeves et al. 1991). We detected tight linkage of Hmgl4 with Ets-2 and Mx-1 on mouse Chr 16 (Table 2); the human homologs of these three genes are all assigned to the same region of human Chr 21, 21q22.3 (Cox and Shimizu 1991). HMG-14 is an abundant and ubiquitous chromosomal protein whose binding to nucleosomal core particles probably alters their conformation (Bustin et al.

The Hmg14-related sequences hybridized with Hmg14 cDNA but not with intron I of the Hmgl4 gene; therefore, these sequences apparently lack introns, a hallmark of processed pseudogenes (Vanin 1985). The dispersed nature of the 14 Hmgl4-related sequences on ten different mouse chromosomes is additional evidence that these sequences are processed pseudogenes that have undergone retrotransposition and not tandemly duplicated gene copies. The inbred strains of mice examined, including those derived from different subspecies (M.m. castaneus and M.m. molossinus) and species (M. spretus) contained approximately the same number of Hmg14-

Table 2. Genetic linkage of Hmg14 and Hmg14-related sequences with prepositioned reference loci: three-point analysis of gene order and estimates of recombination frequencies.

Likelihood differences b for alternative orders

Best gene order~ Chr

Locus 1

Locus 2

Locus 3

2-1-3

1-3-2

Percent recombination Locus 1--Locus 2 n

A. (C57BL/6J x SPRET/Ei) x SPRET/Ei interspecific b a c k c r o s s - - C 5 7 B L / 6 J DNA fragments mapped 3 Fire-3 Hmg14-rs5 Amy-1 7.09 6.81 117 7 Pmv-4 Hmg14-rs2 Gpi-1 0.00 22.97 66 7 Tyr Odc-rs7 Hmg14-rs7 3.87 19.53 88 11 Xm v-42 Hmg 14-rs4 Es-3 24.85 19.08 134 12 Mpmv-11 Hmgl4-rs3 Aat 2.30 31.88 98 16 Pit-1 Ets-2 Hmg14 47.73 1.35 119 17 Pim-1 Hmg14-rs1 Hmg14-rs8 3.12 17.89 100 17 Hmg14-rs I Hmg14-rs8 C3 20.68 10,72 85 X Hprt Hmg14-rs6 Pgk-1 20.00 17.09 107 B. (C57BL/6J x CAST/Ei) x C57BL/6J intersubspecific backcross--CAST/Ei DNA fragments mapped 3 Fire-3 Hmg14-rs5 Amy-1 10.40 20.81 136 5 Pgy-1 Hmg14-rs9 Pgm-I 17.61 29.57 141 7 Hmg14-rs2 Gpi-1 Hbb 1.91 38.60 142 7 Gpi-1 Hbb Hmg14-rs7 55.30 0.30 143 7 Hbb Hmg14-rs7 Hmgl4-rslO 8.88 40.62 143 9 Hmg14-rs13 Tpi-4 Mpi-1 14.26 28.93 133 11 ll-4 Hmg l4-rs4 Es-3 39.89 5.50 140 16 Prm-I Hmg14-rs14 Pit-1 1.39 57.67 122 16 Pit-1 Mx-1 Hrngl4 44.83 0.00 123 17 Tcp-lOb Pim-1 Hmg14-rsI 23.24 14.59 143 17 Pim-1 Hmgl 4-rs 1 Hmg14-rs8 13.73 22.89 140 19 Cyp2c Hrng14-rs12 Adra-2 4.23 23.18 121 X Hmg14-rsll Hprt Hmg14-rs6 3.14 16.25 123 X Hprt Hmgl 4-rs6 Odc-rsl3 22.06 12.70 123

Locus 2

Locus 3

r

SE

n

r

SE

23.9 0.0 4.5 13.4 4.1 16.8 7.0 15.3 16.8

3.9 -2.2 2.9 2.0 3.4 2.6 3.9 3.6

136 92 128 142 139 141 85 110 141

25.7 10,9 12.5 11.3 22,3 0.7 15.3 10.0 17.7

3.7 3.2 2.9 2.7 3.5 0.7 3.9 2.9 3.2

19.1 12.1 12.7 41.3 6.3 14.3 33.6 2.5 14.3 8.4 5.7 1.7 11.4 18.7

3.4 2,7 2.8 4.1 2.0 3.0 4.0 1.4 3.2 2.3 2.0 1.2 2.9 3.5

137 142 143 143 142 140 140 121 124 140 121 134 123 113

26,3 17.6 41.3 6.3 19.0 23.6 11.4 36.4

3.8 3.2 4.1 2.0 3.3 3.6 2,7 4.4

0.0

5.7 9.1 7.5 18.7 13.3

--

2.0 2.6 2,3 3.5 3,2

a Genes are listed from centromere to telomere in the most likely order as determined by maximum likelihood analysis of the three-point data. b Log likelihood differences between the best gene order and an alternative order: for example, a value of 7.09 means that the best gene order is 107`09 times more likely than the alternative order.

K.R. Johnson et al.: Mapping of the HMG-14 multigene family

630 7

~-Hmgl4-rsl3

!Pm~4-rs2 Pgy-1

-4"

14.3 ,+ 3.d 12.7 ,+ 2.;

12.1_+2.7

Fir~3

14

23.9

19.(•

Z

3.9

10.9 + 3,2

11-

-Gpi-1 -

11

9

0.0

q

Hmg14-rs9

?

!

-

Tpi.4

12 --

23.6-+3.6

17.6,+3.2

~

-Mpi.1

34 -- - Pgn~l

"Hmgl4-rs5

"30

II.4

~

33.6 +_4.0

26.3 _+ 3,8

25.7 ,+ 3.7

]

|

I

1 -~

4.5+-2.2

L-m-49 -]]-Hbb 6.3 +- 2.0

12.5 -+ 2.9

]

. Hmg l 4-rs7 -Amy-1

68

-Xmv.42

47

"7

19.0 ,+ 3.:

13.4 ,+ 2.9

- Hmgl4-rs4

11.4 +- 2.7 75 "

11.3 + 2.7

-Es-3

-Hmg14-rs10

17

16

12

4 7 --

2,5+--1,4 --

!Jrm- 1

19

X

)

q

Fop- 10b

-Hm4114-r,a l 4

--

Flmgl4-rsll

8.4 Z 2~ 17

P/m-I

-

7.0 ,+ 2.6

5.7_+ 2.0

-- ,' yp2d 2~ L7 Z L2 - ~-flmg14-rs12

11.4 + 2.9

-Hmgl4-rsl 36.4 +_4.4

9.1+_2.( 30

23--

- Hprt

7.5+2.3 15.3 + 3.9

~.7•

Mpmv. l l 4.1 +_2.0

-Hmgl4-rs8

-Hmgl4-rs3 22~3 + 3.5

35-

16.8+3.6

-Adra.2

-- -Hmg14-rs6 ~ t ]

10.0 _+2.9

39 -

-Pit.1

28

-

(7,3

12.3,+3.2

17.7 _+ 3.2

44 -- -Odc.rs13 14.3

49

16.8

"Aat

57 58

48

Pgk-1

J

_Ets-2 .~.~..0.7 + 0.7 - Mx-1, Hmg14

Fig. 4. Chromosomal locations of the Hmgl4 functional gene and related sequence genes. The recombination percentages and standard errors for the (C57BL/6J • CAST/Ei) • C57BL/6J intersubspecific backcross are shown on the left of the chromosomes, and those for the (C57BL/6J • SPRET/Ei) • SPRET/Ei interspecific backcross on the right. Distances from the centromeres to anchor loci are given to the left of these loci in smaller print and are taken from the GBASE composite genetic map (GBASE 1992). Except for two chromosomes (17 and X), the distance estimates between

anchor loci in our two crosses were not significantly different from the distances on the GBASE composite map. Compared with the GBASE map and our intersubspecific backcross, our interspecific backcross exhibited increased recombination in the distal region of Chr 17 between Pim-1 and ~C3, as we have reported previously (Johnson et al. 1991). The only other significant discrepancy that we detected is on Chr X, where our estimates of recombination between Hprt and Ode-rsl3 or Pgk-1 are greater than those given on the composite GBASE map.

K.R. Johnson et al.: Mapping of the HMG-14 multigene family

related sequences (15-22). The similarity of map positions in both the C57BL/6J and CAST/Ei genomes of at least seven Hmgl4-related sequences (Fig. 4) is evidence that these two genomes share many pseudogene integration sites and that these common sites probably represent the same ancestral retrotransposition events. Taken together, the similarities in Hmgl4 pseudogene numbers and integration sites many inbred mouse strains suggest that many of the ancestral retrotransposition events occurred before these strains diverged. In contrast, the content of endogenous murine leukemia proviruses varies greatly among mouse strains (Frankel et al. 1989a,b, 1990), suggesting that these proviruses were either more recently acquired or are less stably integrated than Hmgl4 pseudogenes. Although much of the Hmg14 pseudogene variation observed among mouse strains can be attributed to DNA sequence divergence, some mouse strains may contain unique Hmgl4 pseudogenes. Interstrain differences in pseudogene content have been shown for other pseudogene families (Adra et al. 1988; Yamaguchi et al. 1991; Johnson et al. 1992; van Deursen et al. 1992). Some pseudogenes appear to be present in most or all mouse strains, while others, presumably more recently acquired, are present only in a particular strain or in closely related strains.

Hmg14-related sequences as genomic reference markers Dispersed members of multigene families such as endogenous proviruses (Kozak et al. 1987; Taylor and Rowe 1989; Frankel et al. 1990, 1992) and pseudogenes (Siracusa et al. 1991; Richards-Smith and Elliott 1992) can serve as useful reference markers for analyzing the mouse genome. Siracusa and co-workers (1991) described the many advantages of using DNA probes that detect multiple loci for mapping mutations and multilocus traits and in identifying protoonocogenes. Hmgl4 and the 14 Hmgl4-related sequences that we have mapped comprise yet another multigene family that can be used for these purposes. The widely dispersed members of the Hmgl4 multigene family can be efficiently detected with a single cDNA probe. For example, 11 polymorphic loci on seven chromosomes can be typed from a single Southern blot of TaqIdigested genomic DNA from progeny of the (C57BL/ 6J x CAST/Ei) x C57BL/6J intersubspecific backcross (Fig. 2B). In addition to the Hmgl4-related loci, our laboratory is actively characterizing and mapping individual members of several other multigene families, including the gene family for high-mobility-group protein 17 and more than ten ribosomal protein gene families. We are concentrating on mapping members of multigene families in the genome of M.m. castaneus rather than M. spretus as described by Siracusa et al. (1991), because of the advantages associated with using M.m. castaneus as a linkage testing stock. Both male and female F 1 hybrids between M.m. castaneus and the

631

standard inbred mouse strains are fertile (in hybrids with M. spretus, only females are fertile), thereby permitting intercrossing when backcrossing is not possible because of inviability or reduced fertility. M.m. castaneus hybrids also produce more progeny per litter than do M. spretus hybrids. Ft hybrids with M.m. castaneus are heterozygous at many loci; we have tested more than 50 single-locus DNA probes and found that over 90% detect RFLPs between C57BL/6J and CAST/Ei (our unpublished data). As shown in this study for Hmg14 (Figs. 1A-C, 2B), RFLPs between the standard inbred strains and CAST/El can also be detected for most members of multigene families.

Acknowledgments. We thank the following for their generous gifts of DNA probes: Dr. Sylvie Gisselbrecht for Fim-3, Dr. Rosemary Elliott for Odc, Dr. Peter D'Eustachio for Aat, Dr. Roger Reeves for Ets-2, Dr. Joe Nadeau for Pim-l, and Dr. Don Lehn for intron I of Hmgl4. We thank Linda Neleski for help in preparing the manuscript and Drs. Ben Taylor and Joe Nadeau for their critical review of the manuscript. This research was supported by National Institutes of Health grants RR01183, CA34196, and GM46697.

References Adra, C.N., Ellis, N.A., and McBurney, M.W.: The family of mouse phosphoglycerate kinase genes and pseudogenes. Somat Cell Mol Genet 14: 69-81, 1988. Bustin, M., Lehn, D.A., and Landsman, D.: Structural features of the HMG chromosomal proteins and their genes. Biochim Biophys Acta 1049: 231-243, 1990. Cox, D.R. and Shimizu, N.: Report of the committee on the genetic constitution of chromosome 21. Cytogenet Cell Genet 58: 800826, 1991. Cuypers, H.T., Selten, G., Quint, W., Zijlstra, M., Maandag, E.R., Boelens, W., Van Wezenbeek, P., Melief, C., and Berns, A.: Murine leukemia virus-induced T-cell lymphomagenesis: Integration of proviruses in a distinct chromosomal region. Cell 37: 141150, 1984. DeLorenzo, R.J. and Ruddle, F.H.: Genetic control of two electrophoretic variants of glucosephosphate isomerase in the mouse (]flus musculus). Biochem Genet 3: 151-162, 1969. D'Eustachio, P.: A genetic map of mouse chromosome 12 composed of polymorphic DNA fragments. J Exp Med 160: 827-838, 1984. Frankel, W.N., Stoye, J.P., Taylor, B.A., and Coffin, J.M.: Genetic analysis of endogenous xenotropic murine leukemia viruses: association with two common mouse mutations and the viral restriction locus Fv-1. J Virol 63: 1763-1774, 1989a. Frankel, W.N., Stoye, J.P., Taylor, B.A., and Coffin, J.M.: Genetic identification of endogenous polytropic proviruses by using recombinant inbred mice. J Virol 63: 3810-3821, 1989b. Frankel, W.N., Stoye, J.P., Taylor, B.A., and Coffin, J.M.: A linkage map of endogenous murine leukemia proviruses. Genetics 124: 221-236, 1990. Frankel, W.N., Lee, B.K., Stoye, J.P., Coffin, J.M., and Eicher, E.M.: Characterization of the endogenous nonecotropic murine leukemia viruses of NZB/BINJ and SM/J inbred strains. Mammalian Genome 2: 110-122, 1992. GBASE: The genomic database for the mouse maintained at The Jackson Laboratory by DP Doolittle, LJ Maitais, AL Hillyard, JN Guidi, MT Davisson, and TH Roderick, 1992. Gearhart, J.D., Davisson, M.T., and Oster-Granite, M.L.: Autosomal aneuploidy in mice: generation and developmental consequences. Brain Res Bull 16: 789-801, 1986. Johnson, K.R.: Improved oligonucleotide labeling and hybridization assay for endogenous nonecotropic murine leukemia proviruses. Mammalian Genome 1: 260--262, 1991. Johnson, K.R., Cook, S.A., and Davisson, M.T.: Chromosomal localization of the murine gene and two related sequences encod-

632 ing high-mobility-group I and Y proteins. Genomics 12: 503-509, 1992. Kaplan, R.D., Chapman, V., and Ruddle, F.H.: Electrophoretic variation of a-amylase in two inbred strains of Mus musculus. J Hered 64: 155-157, 1973. Kozak, C.A., Peters, G., Pauley, R., Morris, V., Michalides, R., Dudley, J., Green, M., Davisson, M., Prakash, O., Vaidya, A., Hilgers, J., Verstraeten, A., Hynes, N., Diggelmann, H., Peterson, D., Cohen, J.C., Dickson, C., Sarkar, N., Nusse, R., Varmus, H., and Callahan, R.: A standardized nomenclature for endogenous mouse mammary tumor viruses. J Viro161: 1651-1654, 1987. Lander, E.S., Green, P., Abrahamson, J., Barlow, A., Daly, M.J., Lincoln, S.E., and Newburg, L.: MAPMAKER: An interactive computer package for constructing primary genetic linkage maps of experimental and natural populations. Genomics 1: 174-181, 1987. Landsman, D. and Bustin, M.: Mouse non-histone chromosomal protein HMG-14 cDNA sequence. Nucleic Acids Res 18: 5311, 1990. Landsman, D., Srikantha, T., Westermann, R., and Bustin, M.: Chromosomal Protein HMG-14: Complete human cDNA sequence and evidence for a multigene family. J Biol Chem 261: 16082-16086, 1986. Landsman, D., McBride, O.W., Soares, N., Crippa, M.P., Srikantha, T., Bustin, M.: Chromosomal protein HMG-14. Identification, characterization, and chromosome localization of a functional gene from the large human multigene family. J Biol Chem 264: 3421-3427, 1989. Martin, J.E. and Petras, M.L.: Es-3 esterases in erythrocytes in erythrocytes of Mus musculus. Can J Genet Cytol 13: 777-781, 1971. Natsuume-Sakai, S., Hayakawa, J-I., and Takahashi, M.: Genetic polymorphism of murine C3 controlled by a single co-dominant locus on chromosome 17. J Immunol 121: 491-498, 1978. Nichols, E.A., Chapman, V.M., and Ruddle, F.H.: Polymorphism and linkage for mannosephosophate isomerase in Mus musculus. Biochem Genet 8: 47-53, 1973. Pash, J., Popescu, N., Matocha, M., Rapoport, S., and Bustin, M.: Chromosomal protein HMG-14 gene maps to the Down syndrome region of human chromosome 21 and is overexpressed in mouse trisomy 16. Proc Natl Acad Sci USA 87: 3836-3840, 1990. Reeves, R.H., Gallahan, D., O'Hara, B.F., Callahan, R., and Gear-

K.R. Johnson et al.: Mapping of the HMG-14 multigene family hart, J.D.: Genetic mapping of Prm-1, Igl-1, Smst, Mtv-6, Sod-l, and Ets-2 and localization of the Down syndrome region on mouse Chromosome 16. Cytogenet Cell Genet 44: 76-81, 1987. Reeves, R.H., Miller, R.D., and Riblet, R.: Mouse Chromosome 16. Mammalian Genome 1 (Suppl): $269-$279, 1991. Richards-Smith, B.A. and Elliott, R.W.: Mapping of the mouse ornithine decarboxylase-related sequence family. Mammalian Genome 2: 215-232, 1992. Shows, T.B., Ruddle, F.H., and Roderick, T.H.: Phosphoglucomutase electrophoretic variants in the mouse. Biochem Genet 3: 2535, 1969. Siracusa, L.D., Jenkins, N.A., and Copeland, N.G.: Identification and applications of repetitive probes for gene mapping in the mouse. Genetics 127: 169-179, 1991. Sola, B., Simon, D., Mattei, M-G., Fichelson, S., Bordereaux, D., Tambourin, P.E., Gutnet, J.-L., and Gisselbrecht, S.: Fim-1, Fim-2/c-fms, and Fim-3, three common integration sites of Friend murine leukemia virus in myeloblastic leukemias, map to mouse chromosomes 13, 18, and 3, respectively. J Virol 62: 3973-3978, 1988. Srikantha, T., Landsman, D., and Bustin, M.: Retropseudogenes for human chromosomal protein HMG-17. J Mol Biol 197: 405413, 1987. Srikantha, T., Landsman, D., and Bustin, M.: A single copy gene for chicken chromosomal protein HMG-14b has evolutionarily conserved features, has lost one of its introns and codes for a rapidly evolving protein. J MoI Biot 211: 49-61, 1990. Taylor, B.A. and Rowe, L.: A mouse linkage testing stock possessing multiple copies of the endogenous ecotropic murine leukemia virus genome. Genomics 5: 221-232, 1989. van Deursen, J., Schepens, J., Peters, W., Meijer, D., Grosveld, G., Hendriks, W., and Wieringa, B.: Genetic variability of the murine creatine kinase B gene locus and related pseudogenes in different inbred strains of mice. Genomics 12: 340-349, 1992. Vanin, E.F.: Processed pseudogenes: characteristics and evolution. Annu Rev Genet 19: 253-272, 1985. Whitney, J.B. III: Simplified typing of mouse hemoglobin (Hbb) phenotypes using cystamine. Biochem Genet 16: 667-672, 1978. Yamaguchi, M., Hayashi, Y., Hirose, F., Matsuoka, S., Moriuchi, T., Shiroishi, T., Moriwaki, K., and Matsukage, A.: Molecular cloning and structural analysis of mouse gene and pseudogenes for proliferating cell nuclear antigen. Nucleic Acids Res 19: 24032410, 1991.