Chronic multifocal osteomyelitis, a new recessive mutation on Chromosome 18 of the mouse

GENOMICS

11,

794-798

(1991)

Chronic Multifocal Osteomyelitis, a New Recessive Mutation on Chromosome 18 of the Mouse LINDA BYRD, * MICHAEL *Laboratory

GRossrmtw,

-/ MICHAEL

POTTER,* AND GRACE L. C. SHEN-ONG**’

of Genetics, National Cancer Institute, National Institutes of Health, Bethesda, Maryland and t Guggenheim Mayo Medical Center, Rochester, Minnesota 55901 Received

April

15, 1991;

M. musculus domesticus (CLA) mice from Centreville, MD and the C.D2 congenic mouse strain and backcross progeny were generated and maintained in our laboratory under NC1 Contract NOl-CB-71085 at Hazleton Laboratories. The C.D2-Qu-2+N6 congenic stock was developed by introgressive backcrossing of the DBA/2N-Qu-2 gene onto a BALB/cAnPt background (Potter et aZ., 1984). The C.D2-Qo-2+ mice at the sixth backcross generation (N6) were mated to generate a homozygous C.D2-Qu-2+/Qu-2+ congenic strain. This strain was maintained by filial matings. At F15 generation, the cmo mutation was first observed. The C.D~-QU-~+/QU-~+-C~.~/C~~ mutant line was then generated and maintained by brother-sister mating.

Osteomyelitis is defined as an inflammation of the marrow, trabeculae, and cortex of bone. In most cases, the causative agent is bacterial, although fungal and viral diseases can also play a role. Gideon et al. (1972) first described a syndrome termed chronic recurrent multifocal osteomyelitis (CRMO) . No pathogens were isolated from affected areas. The inflammation of the bones in these cases had both subacute and chronic features, with lesions localized predominantly in the metaphyses of the long bones. So far the CRMO syndrome has not been shown to have a genetic etiology. A spontaneous mutation was observed in a BALB/c.DBA/2 (C.D2) congenic mouse litter. Mice with tail kinks and deformities in their lower extremities appeared in an F15 litter from brother-sister mating of normal parents of the C.D2-Qu-2+N6 congenie stock. Some of the abnormalities resembled the CRMO syndrome (manuscript in preparation). Accordingly, the new mutation was named chronic multifocal osteomyelitis and the symbol selected was cno.

Southern Blot Analysis

High-molecular-weight DNAs prepared from mouse livers (Sambrook et al, 1989) were digested overnight with restriction endonucleases and electrophoretically separated on either 0.8% horizontal agarose gel OP 3% Nu-Sieve (3:l) agarose (FMC) gel at voltage gradients of approximately 1 V/cm in 40 m&f Tris-acetate, 20 mM sodium acetate, 1 mM EDTA, pH 7.4. The DNAs in gels were stained with ethidium bromide, photographed, and transferred to HybondN nylon membranes (Amersham). Blots were hybridized at 42°C for 16 h in 50% formamide, 5X SSC, 0.1 mg/ml salmon sperm DNA, 1X Denhardt’s solution, 10% dextran sulfate, and one of the following probes:

be ad20892.

794 Inc. reserved.

AND METHODS

Mice

INTRODUCTION

0888-7543/91$3.00 Copyright 0 1991 by Academic Press, All righh of reproduction in any form

7, 1991

MATERIALS

IN.

requests should Bethesda, MD

August

This is the first report that describes the cmo mouse mutant. In this paper, we present results that establish cmo as a single autosomal recessive mutation. We used a backcross to establish the linkage of the cmo gene with molecular markers on mouse Chromosome 18.

Mice with tail kinks and deformities in their lower extremities were observed in a litter of C.D24&-2+N6F16 mice. A mutant line that exhibits this phenotype in 100%of its offspring was establishedby subsequentbreeding. The abnormalities resembledto somedegree those found in a human syndrome termed chronic recurrent multifocal osteomyelitis (CRMO). Accordingly, we namethe new mutation chronic multifocal osteomyelitis (cmo).Breeding analysis showed that the defect was determined by a single autosomal recessive gene. Restriction fragment length polymorphism (RFLP) analysis of progeny from a backcrossbetween Mus musculusdome&ells (CLA) and C.D2Qa-2+-cmo/cmo indicated that the cmo gene resides on mouseChromosome18. o iooi Academic~rene.

1 To whom correspondence and reprint dressed at: Bldg. 37, Rm. 2B23, NCI, NIN,

revised

20892;

CHROMOSOMAL

FIG. 1.

Morphological

appearance

of a 12-month-old

LOCALIZATION

homozygous

cma mouse.

OF

(Lnsert)

795

cmo

Close-up

views

of the affected

hindfeet

( not to

scale).

glucocorticoid reductase-1 (G&l) [the 2.3-kb BgllI(nucleotide 1)-Hind111 (nucleotide 2347) insert of pSV2Wrec (Danielson et al., 1986)], & adrenergic receptor (A&b-2) [the 1.3-kb BumHI-Hind111 insert of the mouse genomic clone (Allen et al., 1988)], Ia associated invariant chain (li) [the 1.3-kb EcoRI insert (Richards et aZ., 1985)] and myelin basic protein (Mbp) [the 1.65kb EcoRI insert of pMBP-1 clone (Hood Group number 8C-1, Roach et al., 1983)]. Each radioactive probe was generated by nick translation. After hybridization, filters were rinsed twice in 2~ SSC at room temperature and for 30 min in 0.1X SSC at 65°C. Statistical Analysis Maximum likelihood estimates of recombination probabilities (3, their variance (Vd), and standard errors (5’;) among backcross progeny were calculated a rding to Green (1981): c”= r/n, V, = 6(1-6)/n, Sd = in a r Vd, where r is the number of recombinations sample of size n. RESULTS

cmo is a Single Autosomal Recessive Mutation Homozygous cmo/cmu mice can first be clearly distinguished from their normal siblings at about 6-8 weeks of age by the presence of tail kinks. Subsequently both hind feet and to a lesser extent the front paws are also affected. The feet become markedly swollen and red initially, and severely deformed by the time the mice are 3 months of age. Figure 1 shows

the phenotypes of a typical homozygous cmo/cmo mouse. Aside from the crippling effect of the mutation, these mice have a normal life span. A more detailed histopathological study of these mutants will be reported elsewhere. The results of breeding tests with C.D2-&a-2+-cmo/cmo mice with either normal BALB/cAnPt or M. musculus domesticus (CLA) mice showed that the mutant phenotype is recessive and fully penetrant on different genetic backgrounds (Table 1). Both female and male offspring are affected to the same degree. The heterozygous Fl offspring of both crosses 1 and 4 were all normal. About onefourth of the F2 offspring and one-half of the Nl backcross progeny were homozygous cmo/cmo mice. x2 analysis indicates that only a single autosomal recessive gene is involved. RFLP

Analyses of Backcross Mice

To identify the chromosomal location of the cmo gene, the cosegregation of cmo mutation with genes on various mouse chromosomes (including 4, 5, 6, 7, 11,12,15,16,17, and 18) was examined in F2 and Nl crosses (5 and 6 listed in Table 1). Only genes on mouse Chromosome 18 were found to have less than 10% recombination events and were considered to be linked to the cmo gene (see below). Genes on the other chromosomes displayed 40% or greater frequency of recombination with the cmo allele (data not shown). We therefore focused our mapping studies on genes

796

BYRD

cmo is a Single Cross number 1 2 3 4 5 6 Note.

ET

TABLE

1

Gene

Recessive

Mutation

Total number of offspring

Matings cmo/cmo (cmo/cmo (cmo/cmo cmo/cmo (cmolcmo (cmolcmo

AL.

x BALB/c = F, X BALB/c)F, X (cmolcmo X BALB/c)F, X BALB/c)F, X cmo/cmo = N, x CLA = F, X CLA)F, X (cmo/cmo X CLA)F, = F, X CLA)F, X cmo/cmo = N,

= F,

117 193 136 32 158 150

X2 (n = 1)

0 23 46 0 22 50

P

0.44 0.74

0.5 0.3-0.5

0.68 0.00

0.3-0.5 >0.99

P, probability.

that have previously been localized on mouse Chromosome 18. The four gene markers used are glucocorticoid reductase-1 (G&l), & adrenergic receptor (A&b-2), Ia associated invariant chain (Ii), and myelin basic protein (Mbp) (Francke and Gehring, 1980; Sundaresan and Francke, 1989; Richards et al., 1985; Roach et aZ., 1985). For each probe, appropriate restriction fragment length polymorphisms (RFLPs) were identified by digesting DNAs from CLA and C.D2-@z-2+-cmo/cmo with a range of restriction enzymes. The RFLPs detected for each probe are detailed in Table 2. A panel of 112 backcross mice (69 with cmo/cmo phenotype and 43 with normal phenotype) from cross number 6 listed in Table 1 was analyzed for the segregation and recombination frequencies of the cmo/cmo phenotype with the probes listed. Except for recombinants, all mice with cmo/cmo phenotype were found to have inherited both C.D2-Qa-2+ alleles for the probes examined. In contrast, backcross mice with normal phenotype carried one CLA and one C.D2-Qu-2+ allele. The linkage map of mouse Chromosome 18 in Fig. 2 shows four genetic markers in relation to the cmo locus. Gene placement was determined by minimization of crossover events (Table

3). Our gene mapping results further confirm that cm0 represents a single recessive gene mutation. DISCUSSION

The chronic multifocal osteomyelitis locus is the first described mutation of the mouse that causes inflammation of the bone at multiple sites. We do not presently know if this can be a model system for any particular human disease. It resembles human CRMO only as a chronic inflammatory process involving multiple osseous sites, and that no known pathogen can be isolated from affected areas. In this

Loci

Probes and mesticus from 2+cmo+ Mice Enzyme

2

MapDistance

from cmo

Gskl T f.cH

TABLE

Locus

% cmo/cmo offspring

21.423.9

(24/l

12)

9.8+2.8 9.8+2.6

(I l/l

12)

(I l/l

12)

RFLPs Detected in M. musculus doCentreville, MD (CLA) and C.D2-Qa-

CLA

alleles

(kb)

Grl-1

Hi&I

Adrb-2 Zi

DrcaI PstI

1.2, 1.0,0.7,0.5, 0.3*, 0.2, 0.1 2.7* 13.0*, 1.9, 1.5, 0.6

Mb

Tap1

3.7*,

1.9, 1.4

Note. Hybridizing fragments that are unique C.D2-Qa-2+cmo+ mice are marked by *.

t

C.D2-Qa-2+cmo+ alleles (kb) 1.2, 1.0,0.7, 0.5, 0.2, 0.1 3.5* 6.0*, 1.9, 1.5, l.O*, 0.6 1.9, 1.4 in either

CLA

7.1~2.4

or

(WI 12)

FIG. 2. Molecular genetic linkage map of mouse Chromosome 18. Genetic distances in relation to cmo are shown. In each case, the genetic distance -+ standard error is given in centimorgans (CM). The number of recombinant progeny observed, over the total number progeny, is given in parentheses.

CHROMOSOMAL

LOCALIZATION

TABLE Summary

of the Results

Parentals

Loci C C C C C

Grl-1 Adrb-2 Ii g No. of mice

C C C

c

,“”

L C C C 1 I

797

cmo

3

of the Backcross Single

C

OF

Analysis

recombinants

Double

C

C

C

C

C

C

C C

C

c

C

C C C

C

C

C

C

C C C C

C 1

C

C

C

C

4

1

4

1

F! C 6

C

h C C

I

81

recombinant

i-5

30

1

Recombination Locus

Pair

G&l, Adrb-2 Adrb-2, Zi Zi, cm0 cmo, Mbp Note. A total of 112 animals CLA, and c, with both alleles

% Recombination

r/n 11/112 2/112 11/112 81112 were tvoed for the loci listed from C:D2-Qa-2+.

9.8 1.8 9.8 7.1

at the left. The

report, we show that the mutation is inherited as an autosomal recessive gene on mouse Chromosome 18. We have also established a molecular genetic linkage map of genes surrounding the cmo locus on mouse Chromosome 18 (see Fig. 2). Breeding and linkage analyses indicate that the Qa-2 gene of DBA origin does not play a role in the cmo phenotype of the C.D2Qu-2+-cmo/cmo congenic mice. Further RFLP analysis showed that both the Ii and Mbp genes that flank the cmo gene on the C.D2-Qa-2+ chromosome are of BALB/c origin (data not shown), suggesting that the spontaneous cmo mutation probably involved a gene of BALB/c origin. To characterize the cmo mutation, we need to obtain other more closely linked markers. Our gene mapping study reported here establishes the gene order of the Chromosome 18 markers (see Table 3). Three of the markers used in this study are proximal to cmo, and have been mapped in humans to a conserved syntenic group of loci on the long arm of human Chromosome 5. They are the Grl-1 locus (Gehring et al., 1985), the Ii locus (Claesson-Welsh et al., 1984) and the A&b-2 locus (Kobilka et al., 1987). On the other end, the Mbp locus located at the distal end of mouse Chromosome 18 has been localized to human Chromosome 18q22-qter (Saxe et al., 1985). Mapping of additional loci common to mouse Chromosome 18 and human Chromosomes 5 and 18q should provide useful markers flanking the cmo locus. Our long-term goal is to identify the mutation that is responsible for the cmo/cmo phenotype. To that end, we have already obtained a mouse Chromosome 18

hybridizing

fragments

+ + -e f

are designated

(*SE) 2.8 1.3 2.8 2.4 as C, with

an allele

from

marker that resides less than 1 cm from the cmo locus (unpublished data) and will utilize reverse genetic approaches to clone the cm0 gene. ACKNOWLEDGMENTS We thank Drs. N. A. Jenkins and N. G. Copeland for helpful discussions and communication of unpublished data; Drs. G. M. Ringold and J. M. Allen for their kind gifts of DNA probes; Dr. K. Huppi and M. Krall for providing prepared DNA probes; Dr. B. A. Mock for her help in data analyses and critical reading of the manuscript; and S. Wong for technical assistance. The mice were generated under NC1 Contract NOl-CB-71085.

REFERENCES 1.

2.

3.

4.

5.

ALLEN, J. M., BAETGE, E. E., ABFLASS, I. B., AND PALMITER, R. D. (1988). Isoproterenol response following transfection of the mouse beta 2-adrenergic receptor gene into Yl cells. EMBO J. 7: 133-138. CLAESSON-WELSH, L., BARKER, P. E., LARHAMMAR, D., RASK, L., RUDDLE, F. H., ANLI PETERSON, P. A. (1984). The gene encoding the human class II antigen-associated gamma chain is located on chromosome 5. Immunogenetics 20: 8993. DANIELSON, M., NORTHROP, J. P., ANO RINGOLD, G. M. (1986). The mouse glucocorticoid receptor: Mapping of timetional domains by cloning, sequencing and expression of wild-type and mutant receptor proteins. EMBO J. 6: 25132522. FRANCKE, U., AND GEHRING, U. (1980). Chromosome assignment of a murine glucocorticoid receptor gene (Grl-1) using intraspecies somatic cell hybrids. Cell 22: 657-664. GEHRING, U., SEGNITZ, B., FOELLMER, B., AND FRANcKE, U. (1985). Assignment of the human gene for the glucocorticoid

798

BYRD receptor to chromosome 3751-3755.

5. Proc.

Natl.

Acad.

Sci.

USA

GIDEON, A., HOLTHUSEN, W., MASEL, L. F., AND VISCHER, D. (1972). Subacute and chronic “symmetrical” osteomyelitis. Ann. Radiol. 15: 329-342.

7.

GREEN, E. L. (1981). Linkage, recombination, and mapping. In “Genetics and Probability in Animal Breeding Experiments” (E. Green, Ed.), pp. 77-113, MacMillian, New York,

8.

KOBILKA, B. K., DIXON, R. A. F., FRIELLE, T., DOHLMAN, H. G., BOLANOWSKI, M. A., SIGAL, I.-S., YANG-FENG, T.-L., FRANCKE, U., CARON, M. G., AND LEFKOWITZ, R. J. (1987). cDNA for the human beta 2-adrenergic receptor: A protein with multiple membrane-spanning domains and encoded by a gene whose chromosomal location is shared with that of the receptor for platelet-derived growth factor. Proc. Natl. Acad. Sci. USA 84: 46-50.

10.

11.

12.

Chromosomal assignment and segregation in recombinant inbred strains. Immunogenetics 22: 193-199. ROACH, A., BOYLAN, K., HORVATH, S., PRUSINER, S. B., AND HOOD, L. E. (1983). Characterization of cloned cDNA representing rat myelin basic protein: Absence of expression in brain of shiverer mutant mice. Cell 34: 799-806.

D.

ROACH, A., TAKAHASHI, N., PRAVTCHEVA, D., RUDDLE, F., AND HOOD, L. (1985). Chromosomal mapping of mouse myelin basic protein gene and structure and transcription of the partially deleted gene in shiverer mutant mice. Cell 42: 149155. SAMBROOK, J., FRITSCH, E. F., AND MANZATIS, T. (1989). “Molecular Cloning, a Laboratory Manual,” 2nd ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. SAXE, D. F., TAKAHASHI, N., HOOD, L., AND SIMON, M. I. (1985). Localization of the human myelin basic protein gene (MBP) to region 18q22-qter by in situ hybridization. Cytogenet. Cell Genet. 39: 246-249.

C., RUDDLE, chain gene:

SUNDARESAN, S., AND FRANCKE, U. (1989). Genes for beta-2 adrenergic receptor and platelet-derived growth factor receptor map to mouse chromosome 18. Somat. Cellkfol. Genet. 15: 367-371.

POTTER, M., HARTLEY, J. W., WAX, J. S., AND GALLAHAN, (1984). Effect of MuLV-related genes on pbsmacytomagenesis in BALB/c mice. J. Exp. Med. 160: 435-440. RICHARDS, J. E., PRAVTCHEVA, D. D., DAY, F. H., AND JONES, P. P. (1985). Murine invariant

AL.

82:

6.

9.

ET

13.

14.