Download Genetic and molecular mapping of the Hmt region of mouse and more Lecture notes Genetics in PDF only on Docsity!
The EMBO Journal vol.8 no.12 (^) pp.3749-3757, 1989
Genetic and molecular mapping of the Hmt region of
mouse
Sue Richards', Maja Bucan2'5, Kurt Brorson3,
Michael C.Kiefer'", Stephen W.Hunt, 1113.6,
Hans Lehrach2'7 and Kirsten Fischer Lindahl"
'Howard Hughes Medical Institute and Departments of Microbiology and Biochemistry, University of Texas Southwestern Medical Center, Dallas, TX 75235-9050, USA, 2European Molecular Biology Laboratories, D-6900 Heidelberg 1, FRG, 3Division of Biology, California Institute of Technology, Pasadena, CA 91125, USA, 4Basel Institute for Immunology, (^) CH-4058 Basel, Switzerland 5Present address: Wistar Institute, Spruce at 36th, Philadelphia, PA 19104, USA 6Present address: Division of (^) Rheumatology and Immunology, University of North (^) Carolina, Chapel Hill, NC 27599-7280, USA 7Present address: Imperial Cancer Research Foundation, London WC2A 3PX, UK Communicated by H.Lehrach
We have mapped a new (^) region of the mouse major
histocompatibility complex (NHC) that contains the
nuclear gene, Hmt, for the maternally transmitted antigen, Mta. The Hmt region of chromosome (^17) lies between a recombinational breakpoint distal to (^) Tla and another proximal to Tpx-1, thus including Pgk-2. A novel MHC class I gene fragment, R4B2, was cloned and mapped to this region as was another new class I gene,
Thy19.4. Both lie proximal to Pgk-2, within the distal
inversion in t-haplotypes. The presence of several other MHC class I genes in the Hmt region is predicted from analysis of the recombinants that define the region. Key words: genome organization/maternally transmitted antigen (Mta)/MHC class I genes/mouse chromosome 17/ t-complex
Introduction
Mta, the (^) maternally transmitted antigen of the mouse, is an unusual (^) transplantation antigen (Chan and Fischer Lindahl, 1985): it^ shows^ maternal inheritance, and is detected by cytotoxic T^ lymphocytes (CTL) that are not H-2 restricted (Fischer Lindahl et al., 1980). Since CTL remains the only reagent available for^ the characterization of Mta, the antigen has proved difficult to isolate. We have chosen to
analyze its components by molecular methods after they were
defined (^) by classical (^) genetics.
Three genes are required for the expression of Mta: Mtf,
the maternally transmitted factor (Fischer Lindahl and Burki,
1982; Fischer Lindahl and Hausmann, 1983); Hmt, a
chromosomal gene closely linked to H-2 (Fischer Lindahl
et al., 1983); and B2m, the gene encoding 02 microglobulin
(32m) (Fischer Lindahl and Langhorne, 1981; Fischer
Lindahl et al., 1988). The epitopes of^ Mta^ are^ determined
jointly by Hmt and Mtf (Fischer Lindahl^ et^ al., 1986). Our
working hypothesis is that^ Hmt encodes^ a^ class^ I^ MHC
molecule which, associated with ,B2m, binds the small peptide encoded by Mtf, and presents it on the cell surface to the cytotoxic T cells (Fischer Lindahl et al., 1983). Hmt is (^) considered a class I (^) gene because Mta-specific killer T cells are not H-2 (^) restricted; an active B2m gene is required for Mta expression; and Hmt maps at the end of a string of class I genes on chromosome 17 distal to H-2 (Fischer Lindahl et (^) al., 1983). Mtf has long been thought to be a mitochondrial (^) gene (Ferris et al., 1983; Smith et al., 1983; Huston et (^) al., 1985; Hirama and Fischer Lindahl, 1985). By sequencing three mitochondrial genomes, representing allelic forms of Mtf, we could identify a candidate for the gene (H.Yonekawa, E.Hermel and K.Fischer Lindahl, in preparation; Wang et al., to be published). We have recently shown that a peptide of 17 amino acids, when added to cells, can recreate the epitopes normally determined by Mtf (B.E.Loveland, C.-R.Wang, E.Hermel, H.Yonekawa and K.Fischer Lindahl, in preparation). Our strategy for the isolation of Hmt has been first to define by genetic recombination, and as narrowly as possible, the region encompassing Hmt, then find DNA probes from this (^) region to be (^) used as starting points for chromosome walking and clone (^) any new (^) class I (^) gene located in this region. Candidates for Hmt can be tested by transfection into a cell line that carries a null allele of (^) Hmt: an active (^) Hmt gene is expected to make this cell line (^) susceptible to cytotoxic T cells specific for Mta. Here we report success in all three stages of the strategy.
Results
Castaneus recombinants Hmt was first mapped by CTL typing for Mta of F2 hybrids produced in a cross of strains which differed at Hmt: C3H/HeJ (C3H) and Mus musculus castaneus (Fischer Lindahl et (^) al., 1983). C3H is H-2k, Qa-2b, Thb Qa-Jb,
Hmt', and the^ M.^ m. castaneus parent, CAS3, is H-2c3,
Qa-2a, T7ac3, Qa-Jc3, Hmtb. Three^ recombinants, RI, R4-e,
and (^) R4-l, between the CAS3 and C3H (^) haplotypes have (^) been particularly informative^ for^ the^ mapping of^ Hmt (Figure 1). The crossover in RI occurred between (^) Qa-2 and Tla. The CAS3 parent donated the (^) proximal end of chromosome 17 including H-2 and Qa-2, and the C3H (^) parent, Tla, lImt, and the more distal regions. The RI haplotype has since been backcrossed onto C57BL/lOSnJ (BlO), and, after 10
generations of^ backcrossing, it is^ now^ established in a
congenic strain.
In the original, 'early' R4 (R4-e) recombinant, the H-
through Tia region of chromosome 17 was of the (^) C3H parent, while the CAS3 parent donated the Hmt and more distal regions. During four subsequent backcrosses to
C3H.SW (H-2b, Tlae, Hmta), a second recombination
occurred between Hmt and (^) Tpx-J with the C3H.SW (^) parent
S.Richards et al.
-^ ..../^ F^ I
II
..
R 4 ......
Fig. 1. The major histocompatibility complex in the castaneus^ recombinants.^ The^ origin of the^ H-2,^ Qa-2, and^ Tia genes^ as^ determined by serotyping, Hmt and Qa-J by CTL assay and Tpx-J by RFLP^ analysis.
Tla Region
C B
Q (^) T18 37 T17 T16 T15 T14 T13 T12 Tll T10 T9 T8 T7 T Iri (^) rlA= n r-1A n r (^) mr- r-7- 15.3 (^) -- ,, 47.1 , 22.1 1 1 2.2 - pH2d-
A
T5 T rI
T3 T2 TI r___ , r,, A
Fig. 2. The Tla region in the BALB/c mouse.^ A, B and C^ are^ clusters of^ genes represented in^ contiguous cosmid clones^ (Steinmetz et^ al., 1982; Fisher et al., 1985; Transy et al., 1987). In this^ figure, the^ orientation^ and location of A is^ based^ on^ data from Passmore and Romano^ (1988), B is oriented the same way to reflect the^ extensive^ homologies with^ A that^ presumably arose^ by gene duplication and C could^ have been located anywhere. Tia lies between the^ proximal Qa (Q) and^ the distal^ Hmt^ regions. The^ genes of the^ Tia^ region are^ indicated^ by boxes. The cosmid clones used in this study are shown^ below^ the^ genes, and the^ probes made from them^ are^ indicated^ by black^ triangles.
donating the region distal to Hmt to the 'late' R4 (^) (R4-l) haplotype (Figure 1). The R4-l haplotype was subsequently backcrossed to both C3H.SW (^) (mice designated CSW.R4-l) and to C57BL/6J (B6) (mice designated B6.R4-l).
Mapping Hmt with Tia region probes The Tia region encompasses at least 19 genes in BALB/c
(1la') (Figure 2) and 15 genes in B6 and BlO^ (Th7a) (Brown
et al., 1988). It^ was therefore possible that^ Hmt mapped within the Tla region but distal to the (^) gene encoding the TL antigen in^ CAS3^ (corresponding to^ gene T13c). The organization and orientation of the Tia region is not completely established. At least three gene clusters (A, B and C), separated by gaps of unknown length, have been reported. Figure 2 represents one construct on current data. To map Hmt with reference to the Tia region, we used DNA probes that hybridize to the ends of each Tia cluster as shown in (^) Figure 2. The Ti (^7) probe also (^) hybridizes to (^79) (S.W.Hunt, K.A.Brorson, H.Cheroutre, A.Matsuura, F.-W. Shen and L.Hood, in preparation) as seen by doublet bands in Table I. The gene 37 probe cross-hybridizes with TJO (Transy et al., 1987). The sizes (or the deletion) of restriction fragments detected by these probes show that RI and R4-e share the C3H, rather than CAS3, equivalents of the BALB/c genes Ti, 79, T10, Til, T17 and (^) T18 and 37 (Table I). If the orientation and order is unchanged for those class I genes common to Tia haplotypes, Hmt must be distal to all known class I (^) genes within the Tia region of BALB/c. This result
is consistent with our failure to render a BIO.CAS2 (Hmtb)
cell line Mta positive by transfection with any cosmid from the BALB/c lZa region (K. Fischer Lindahl and M.Steinmetz, unpublished).
The (^) distal limit of the Hmt region Among (C3H x CAS3)F2 progeny we found several recombinants between Hmt and Upg-J (Fischer Lindahl
Table I. Mapping Hmt by Tia region probes
Probe: T18 37 T17 Til TIO Ti Enzyme: HindIlI BglI BamHI HindIII BamHI BamHI
Strain Tla Estimated fragment size (kb) BALB/c c 3.8 14.2, 3.0 4.5, 4.2 3.4 9.8, 4.3^ 10. CAS3 c3 1.9, 1.6 8.8, 6.6 6.2, 5.1^ 3.4^ 7.3, 6.9^ d C3H b 3.8 14.2, 7.6 4.5, 3.8 d^ d^ 10. RI b 3.8 14.2, 7.6 4.5, 3.8^ d^ d^ 10. R4-e b^ 3.8 14.2, 7.6^ 4.5, 3.8^ d^ d^ 10.
d =^ deleted.
-.-@0-
,~~~~~~~..qj_*
Fig. 3.^ Demonstration^ of^ the^ 'late'^ R4^ recombination.^ The^ origin of the RI and R4 (^) haplotypes is described in the text. Liver DNA from the strains indicated was (^) digested with Pstl or Bcll and (^) hybridized to probe 1-1-1H4 (Kasahara et al., 1987). The (^) highest mol. wt band in CSW.R4-I DNA^ cut with Bcll represents a partial (^) digestion.
Hmt
.-.^ 7:, .1 !i.-,
S.Richards et (^) al.
Table IV. (^) Mapping Th7y19.4 and R4B2 (^) by Passmore's (^) recombinants
Tla Hmt Strain 12.21 pTLA.5 R4B2 and (^) Thy19.4 Pgk-2 Ce-
MA.RI D2 D2 D2 MA (^) MA MA.R2 D2 D2 D2 D2 (^) MA MA.R5 MA MA MA MA (^) D BlO.DRI D2 D2 D2 D2 K
Vertical lines indicate crossover (^) points.
Chromosomal (^) origins (D2 = (^) BlO.D2, MA = (^) MA/MyJ, K = (^) BlOK) were taken from Passmore and Romano (^) (1988) or determined in this study for R4B2 and (^) 7h7y19.4. R4B2: (^) BglII RFLP detected (^) by P7.500, where D2 = 11.2 (^) kb, MA (^) and K = 8.4 kb. 7hy19.4: HindIll RFLP detected (^) by Thyl9.4 5'-end probe, where D2 = 6.6 kb, MA and K = 11.2 kb.
Fig. 6. (^) Mapping of (^) probe P7.500 (^) to the Hmt region. Liver DNA from the strains indicated was (^) digested with (^) BglII and the Southern blot was probed with P7.500, the probe for the 3' end of R4B2. Fragment sizes are given in kb.
Table III. Size (^) (in kb) of restriction (^) enzyme fragments of (^) class I genes in the Hmt (^) region of strains (^) C3H (a) and R4 (^) (b)
Probes PstI (^) BglII HindlIl (^) EcoRI KpnI C3H R4 C3H R4 C3H R4 (^) C3H R4 (^) C3H R
Multi copy pH-2IIa (exon 4) 9* 7? 17?? 5 7? (^6) Single copy P7.500 (R4B2 3' end) ns ns 9 17* 23 23 8 7* 23 ns Thyl9.4 3' end 9* 3* ns ns 13* 9* 3 3 9* 9* Thyl9.4 5' end 3 3* 1 1 13* 9* 1 1 9* 9*
?= fragment not identifiable. ns = no (^) signal (small fragments run off the gel). *In each (^) column, pairs of fragments known (PstI 9 kb and (^) BglII (^17) kb) or (^) presumed to (^) be identical are marked.
R4B2 is (^) more similar at the amino acid level (79-81 %) to
exon 4 of genes from the H-2, Qa-2 and Tia region than
to exon 4 of Thy19.4 (75%), the other class I gene in the
Hmt region, and MbJ (65 %), a newly described class I gene
distal to Qa-2 (Singer et (^) al., 1988). The two cysteines at positions 21 and 77 are conserved as (^) expected of a functional class I gene. The exon is bounded 5' (^) by AG and 3' by GT, suggesting functional splice sites. There (^) is no glycosylation site in^ exon 4 of R4B2. A (^) single-copy probe, P7.500, was subcloned (^) from R4B as a (^500) bp PstI fragment. It is downstream of (^) the gene (Figure 4), and its (^) sequence shows no similarity to any class I gene. This region appears conserved (^) with limited restriction
site polymorphism, but using BglII we could confirm the
mapping of P7.500 and R4B2 to the Hmt region (Figure 6).
As expected of an Hmt gene, the R4-e and R4-4 recombinants
have the 17 kb fragment of the CAS3 parent, while RI has the 9 kb fragment of the C3H parent. More class I genes from the Hmt region (^) are currently being cloned using the RFLP approach. By comparing the fragment sizes detected by pH-2IIa and by P7.500 or the Thy 19. 5'-end and 3'-end probes, (^) we find at least one additional
class I gene in both Hmta (EcoRI 5 kb from C3H) and Hmt"
strains (PstI 7 kb from R4) (Table III).
Wild (^) type Chr. tf (^) Glo-i H-2 Tla * Pgk- o-,; A t- (^) haplotype
- ITla H-2 Gl-l (^) tf Pgk-
tw
- (^) tf Glo-I H-2 Tla * (^) Pgk-
Fig. 7. Schematic comparison of (^) wild-type and t (^) complex chromosomes. (^) Straight and (^) wavy lines indicate (^) wild-type and t (^) haplotype chromosomes respectively, and circles a centromere. The breakpoints of the distal inversion in the t (^) haplotypes are (^) represented by vertical bars. The asterisk refers to the loci DI87Leh9, D17Leh and (^) DJ7Leh525. The brackets mark the region duplicated as a result of the crossover (^) (indicated by dashed arrows).
*vv
Fig. 8. Mapping of the loci DI (^) 7Leh89, Dl 7Leh467 and DI 7Leh525 to the Hmt region. Liver DNA from the six strains (^) was digested with the enzymes shown, and Southern blots were (^) hybridized with the probes listed.
:-, (^) I.) r L.r-
..: 46-ma-^ **'W Abuml- u (^) w
Map of the mouse Hmt region
PV..r
;.
(00cn 1 m < aJcmm m<
Fig. 10. Mapping of^ R4B2^ and^ ThyJ9.4 within^ the^ distal^ t^ inversion. DNA was prepared from^ homozygous, wild-type strains C3H/HeJ (C3H) and^ LT.MA-Glo-1b (LT.MA) and from^ heterozygous mice^ with the t^ haplotypes tw5, tw8 tvI8 or^ tvi32^ and^ the^ LT.MA-Glo-Jb^ wild-type (+) chromosome 17. Stul^ digests were^ probed with^ P7.500^ (A) and the (^) ThyI19.4 5' end (^) probe (B). Specific t (t) and (^) wild-type bands and their size in kb are indicated.
Fig. 9. Probe Tu108 detects DNA fragments on^ both sides of the^ R4-e crossover. Liver DNA from the strains indicated was^ digested^ with TaqI and the Southern blots were probed with Tul08.
Thy19.4 and^ R4B2^ are^ proximal^ to^ Pgk- Passmore has produced a^ set^ of^ recombinants^ in^ the^ Na^ to Ce-2 interval between^ strains^ BIO.D2^ and^ MA/MyJ^ or BIO.K (^) (Passmore and Romano, 1988). As the parental strains are all Hmta, Hmt could not be mapped with these recombinants, but RFLPs have allowed us to map Thy19. 4 and R4B2 (Table IV). The recombinant MA.R1 was particularly informative. Its H-2 through Ta regions are derived from BlO.D2, MA donating Pgk-2^ and^ the^ distal^ part^ of^ chromosome^ 17.^ Since the (^) probes for Thy19. 4 and R4B2 revealed fragments of the B1O.D2 size, these genes can be placed proximal to Pgk-2. The other recombinants are consistent with this interpretation.
The Hmt region includes part of the t-complex Among the^ characteristics^ of^ t^ haplotypes^ of^ chromosome 17 are two (^) large inversions, the distal one including tf and H-2:Tla (Figure 7), and the proximal one including T^ and qk. Pgk-2 is believed to lie outside^ the^ inverted^ region (Nadeau, 1983; J.-Y.Tsai and^ L.Silver, personal communication). Rare^ crossovers^ between^ wild-type^ and^ t chromosomes produce partial t^ haplotypes. If^ the recombination occurs^ within^ an^ inverted^ region,^ as^ shown in Figure 7 for^ tw18, it^ will^ result^ in^ a^ duplication^ of^ one end of the inverted region and a deletion of^ the other^ end. Several DNA clones isolated from^ the^ proximal portion of mouse chromosome 17 by microdissection^ and^ micro-
cloning (Rohme et^ al., 1984) detect loci^ that^ are^ duplicated
in tWI8^ with both^ wild-type and^ t^ haplotype^ alleles^ present
(Bucan et al., 1987). Thus, the loci^ DJ7Leh89,^ D^ 7Leh467, and D 7LehS25 could be mapped to the end of the distal inversion between Tia (not duplicated) and the inversion breakpoint. These^ same^ probes^ detect^ bands^ that^ are^ shared by R4-e^ and^ CAS3^ on^ the^ one^ hand and by^ RI^ and^ C3H on the other, mapping them distal to Tla (Figure 8). Accordingly, the end of the distal inversion in t haplotypes lies in the Hmt region, and that part of the Hmt region is
duplicated in tW18.
Another probe, Tu108, detects several bands, all of which can be mapped distal to the Qa-2 region and the recombinational breakpoint in strain B6.K2 (Figure 9). B6.KI and B6.K2 display the AKR pattern, B6.Tla" and BlO.A have a different one, and A.7Tab and C57BL/6 yet another. Since the polymorphic bands belong to^ the differential segment in the Tia^ congenic strains, they must be proximal to Upg-1, but^ could be distal^ to^ Pgk-2 (Klein et al., 1982). The loci detected by TulO8^ are^ not^ included in the tWJ8^ duplication (M.Bucan and^ H.Lehrach, unpublished). When used to probe DNA from the CAS3 recombinants, TulO8 identifies the RI bands with C3H, not^ CAS3, confirming their location distal to Qa-2 (Figure 9). R4-e gives a composite pattern: it shares two^ polymorphic bands with C3H and two with^ CAS3, and^ lacks^ two^ C3H^ and^ three CAS3 bands. Similar^ results^ were^ obtained^ with^ EcoRI,^ MspI and BamHI. We conclude that some of the bands detected by Tu108^ are^ located^ proximal^ to^ the^ R4-e^ crossover,^ in^ the Tia region, and others are distal, possibly inside^ the^ Hmt region.
J
i. i
Map of the^ mouse^ Hmt^ region
are deleted in C3H or CAS3. Since C3H is TL (^) negative (Tiab), one could argue that it should be mapped relative to BlO, which is also Tiab. Therefore, BALB/c (^) probes which also detect regions of homology in BlO were used. A recent comparison of the T7ac (BALB/c) and 7Tab (BIO/B6) regions revealed both gene duplications and deletions. Despite extensive reorganization, these (^) haplotypes share the following homologous (^) regions: T3C_ (^77) and
T5b-T8b; 79b/C T1dO/C; TJ4c-TJT and 77'_ TJd' (Brown
etal., 1988). The genes TJlb-TJSb (>35 kb) have no counterpart in BALB/c. As probes for these genes were not available, we were unable to determine whether the Hmt
region included Tlb"- T15b. Since Hmt is distal to all the
Tia probes tested, we conclude that it must be distal to all genes in^ that region of BALB/c, whatever its orientation.
New class I genes in the Hmt region Several class I genes, found in genomic (^) and cDNA libraries, were missed in the original screening of BALB/c (^) (Steinmetz et al., 1982) and BiO (Weiss et al., 1984) cosmid libraries. Some have lower homology to other class I genes (^) (Mbl) and may have been missed because the screening stringency was too high. Others may have been overlooked because
the DNA is not easily cloned into cosmids (Thy19.4). Genes
37 and TJlb- T15b have been mapped to the Tla region,
proximal of (^) Hmt; Mbl is known to be distal to Qa-2, but it (^) remains to be mapped with respect to the Hmt region.
We have mapped Thy19.4, a complete class I gene
(Brorson et (^) al., 1989), and R4B2, which may be a full-length class I (^) gene, to the Hmt region. Based on comparison of their exon (^4) sequences, both of these genes show greater
similarity to expressed class I genes of the H-2, Qa and Tia
region than to Mbl. They were mapped proximal of Pgk-2, both by their inclusion in the distal t (^) inversion and by Passmore's recombinant MA.R1. We are continuing the cloning of class I genes from the Hmt region using the RFLP approach. The Hma candidates include an EcoRI 5 kb fragment, which differs in size from
both Thyl9.4 and R4B2 (Table III). The Hmt' candidates
include the PstI 7 kb and KpnI 6 kb fragments.
Could any of the new class I genes be Hmt?
Mta, and hence Hmt, is expressed on all nucleated cells tested
(including fibroblasts and lymphoid, myeloid, epithelial and endothelial cells) (Rodgers et al., 1986). Mbl is (^) unlikely to be Hmt, because there is no evidence of expression at the RNA level. The same argument rules out 7hyl9.4, which is (^) expressed only at extremely low levels, predominantly in
the thymus. Furthermore, transfection with the Thy19.
genomic clone^ from an Hmta strain failed to cause
expression of^ Mta on an Hmtb cell line (Loveland et al.,
unpublished). Since the (^) epitopes of Mta are (^) highly conserved (^) among
inbred strains, the gene must be present in all Hmia strains;
yet Mbl is missing in Hmta mice derived from the H-2k
haplotype. Similarly, the Tilb-TJSb genes can be ruled out
because they have no parallel in Hmta BALB/c mice. Gene
37, with its wide tissue distribution and high degree of
conservation (Lalanne et al., 1985; Transy et al., 1987),
seemed a promising candidate for Hmt, but it clearly (^) lies
in the Tia rather than the Hmt region.
Our probes for R4B2 and Thy]9.4 react with all strains
we have tested, including Hmtb. Since (^) Hmt1 is immunologically a null (^) allele, we cannot test (^) R4B2 or any other class I (^) gene cloned from (^) Hmit mice (^) by transfection
followed by a killer assay for a new Mta antigen. Rather,
we must first use a (^) probe from the (^) Hmt' gene to clone the Hmt' (^) homolog. An (^) Hmta counterpart of R4B2, which we cloned from C3H as a 9 kb (^) BglI fragment binding P7. and exon (^4) probes, has turned out to contain only the 3'
end of a class I gene. Neither could we obtain a full-length
clone when (^) screening an A/J genomic library with P7.500. The genetic divergence of the a and b alleles, which allowed us to detect Hmt in the first place, is reflected in (^) the difference in their restriction maps, and makes it more difficult to clone corresponding Hmt region genes.
The Hmt gene Finally, the case for our working hypothesis that Hmt is an MHC class I gene. Although Hmt has not yet been located, we have shown that there are indeed several new class I genes in the Hmt region. Hmtb mice do not express a detectable Mta antigen, yet the number of class I genes found in the Hmt region of (^) b does not differ from that of a mice.
It is quite possible that the Hmtb gene product is only
immunologically silent. Immunization of the CSW.R4-l mice with (^) C3H gives rise to two kinds of H-2 unrestricted CTL (^) (B.E.Loveland and K.Fischer Lindahl, to be published). One kind reacts (^) only with Hmt (^) a:Myf, mice, as do standard anti-Mta CTL raised in combinations that differ for the mitochondrial factor (^) My. But the second kind of CTL reacts with all Hmin mice,
irrespective of their Mtftype, and thus appears to recognize
Hmt as an alloantigen, as one would expect for an MHC class I antigen. A short peptide that mimics the Mtfproduct and a single amino acid difference that accounts for the allelic forms of Mtf (Loveland et al., in preparation) strongly support our view of Hmt as a restriction element that presents a peptidic ligand to the CTL. At the same time , this finding rules out several more fanciful models of Mta (Rodgers et al., 1986; Han et al., 1987). It remains (^) to be seen whether Mtf is the only ligand for Hmt, and (^) why Hmt rather than H-2K (^) or H-2D presents Mtf.
Materials and methods
Mice Standard inbred strains. Mice were bred at the Basel Institute for Immunology or^ they were^ purchased from the Jackson Laboratory (Bar Harbor, ME). The (^) particular sublines used were BALB/cJ, BlO.D2/nSnJ, C3H/HeJ, C3H.SW/SnJ, C57BL/6J and C57BL/1OSnJ.
Castaneus haplotypes. These mice were all bred in the (^) colony of KFL (^) at the Basel Institute for (^) Immunology or at the (^) University of Texas Southwestern Medical School.
Passnore's recombinants. Genomic DNA from the strains MA/MyJ, BlO.K, MA.Rl, MA.R2, MA.R5, and BlO.DRI was a gift from H.C.Passmore (Rutgers University, Piscataway, NJ).
t haplotypes. Genomic DNA from strain (^) LT.MA_Glo-Jb and its Fl (^) hybrids with t haplotypes tWS, tw8, twIS, and tw32 was a gift from J.H. Nadeau (Jackson Laboratory, Bar Harbor, ME). DNA from mice with t haplotypes t"5, t12, and t x12was a^ gift from K.Arzt (Department of Zoology, University of Texas, Austin, TX).
S.Richards et al.
Southern blot analysis Mouse liver DNA was extracted (^) using the methods of (^) Blin and Stafford (1976). The DNA was (^) digested for 4-16 h with various restriction enzymes (IBI), and 10 (^) lg of the digested DNA loaded (^) per lane on (^) a 0.7% agarose gel, to^ be electrophoresed at 60 V for 17 h in TBE buffer (^) (Maniatis et al., 1982). The gel was then placed in 1.5 M (^) NaCI, 0.5 M (^) NaOH for 1 h. The transfer to (^) Hybond membrane (Amersham) was done (^) overnight, using the same buffer. After neutralization in 50 mM sodium (^) phosphate buffer, pH 6.5, for 2^ min, the membrane was baked for 2 h in a vacuum (^) oven at (^) 80°C. Prehybridization in (^) 0.5 M sodium phosphate (^) buffer, pH 7.2, 7% SDS, 1 mM^ EDTA, 0.1 mg/ml salmon sperm DNA at (^) 65°C for 4 h (^) was followed (^) by hybridization with 32P-labeled probes (109 (^) c.p.m./A4g) overnight in^ the^ same^ solution. Restriction fragments used as probes were isolated and labeled with (^) [a-32P]CTP (3000-4000 Ci/mmol, (^) NEN) by the method of (^) Feinberg & (^) Vogelstein (1984). Following hybridization the filter was washed for 5 min at room (^) temperature and for 2 x (^40) min at 65°C in 40 mM sodium (^) phosphate buffer, pH 7.2, 1 % SDS. The filter was then exposed for^ 12-36^ h^ to^ KODAK^ X-OMAT XAR 5 film, with two intensifying screens at^ -80°C. In Table I the (^) protocol of Steinmetz et al. (^) (1986) was followed, (^) except that the (^) hybridization solution contained 1 M (^) NaCl and 1.0% SDS and no mouse (^) genomic DNA. For (^) Figures 8 and (^) 9, the (^) protocol of Bucan et al. (1987) was^ followed. Molecular (^) probes l-l-1H4. This (^) probe, a (^) gift from M.Kasahara and J.Klein (^) (Department of Microbiology, (^) University of (^) Florida, Miami, FL), is a (^950) bp HindlIl fragment isolated from cosmid clone (^) 1-1-1, derived from mouse (^) chromosome (^17) (Kasahara et al., (^) 1987), that contains the (^) Tpx-i gene and (^) maps between Pgk-2 and Mep-J (Kasahara et (^) al., 1989).
Pgk-2. Probe (^) pcAB12EHI.4, a (^) gift from (^) M.McBurney (Departments of Medicine (^) and Biology, (^) University of (^) Ottawa, Ottawa, Ontario), is a 1.4 kb fragment just downstream of the (^) Pgk-2 gene, which was cloned from BALB/c (Pgk-2a) testis cDNA (Boer et (^) al., 1987).
Tu89, Tu108, Tu467and Tu525. These are (^) 1.7, 0.4, 1.7 and 2.0 kb (^) EcoRI clones derived (^) by microdissection of chromosome (^17) (Rohme et (^) al., 1984).
pH-21Ia. A 440^ bp SacI^ -HhaI fragment from a cDNA clone which detects exon 4 of all known class I (^) genes (Steinmetz et al., (^) 1981a).
Ti -^ T18. (^) Single or low (^) copy probes derived from cosmids of the BALB/c (Tla') mouse (^) (Winoto et (^) al., 1983), were (^) provided by K.Minard, A.Winoto, and L.Hood^ (Division of^ Biology, California Institute of Technology, (^) Pasadena, CA). (i) Ti: 2 kb (^) KpnI-SmaI fragment of (^) cosmid 66.1 cut out with (^) EcoRI, Hindml and subcloned in (^) M13mp8; (ii) TJO: 2.2 (^) kb EcoRI (^) fragment of cosmid 12.2 cut out with BamHI and (^) subcloned in M13mp8; (iii) Tll: 2 kb BamHI (^) fragment of cosmid 22.1 cut out with BamHI and subcloned with (^) M13mp8; (iv) T17: 2.6 kb (^) XAo-HpaI fragment of cosmid 47.1 (^) cut out with (^) EcoRl, HindlIl and subcloned in (^) M13mp8; v) T18:^3 kb^ XhoI^ fragment of^ cosmid 15.3.
- A (^250) bp PstI (^) fragment from (^) clone (^) pH-2d-37, isolated from a DBA/ (H-2d) liver^ cDNA^ library (Lalanne et^ al., 1985), came from J.L.Lalanne (Institut (^) Pasteur, Paris, France).
7hyJ9.4 3' end. A (^500) bp EcoRI-BglIl fragment subcloned from (^) a BALB/c thymus cDNA^ clone (^) (Hunt et (^) al., in (^) preparation). The (^) probe was (^) used to clone a (^) complete and new class I (^) gene as a 7.2 kb Hindlll (^) fragment from a BALB/c (^) genomic band (^) library, and it detects the 3' untranslated (^) region of this (^) gene (Brorson et (^) al., 1989).
7hy19.4 5' end. A 1.2 kb EcoRI (^) fragment subcloned from the Thy 19. genomic clone described above. This (^) single-copy probe detects exons 1, 2 and 3 of the (^) gene.
P7.500. A (^500) bp PstI (^) subclone, located downstream of the R4B2 gene (see below), acts as a (^) single-copy probe in genomic blots.
Cloning of R4B Genomic DNA (^) (500 (^) Ag) from strain (^) CSW.R4-1 was digested with BglII and (^) separated on (^) a 10-25% sucrose (^) gradient (Maniatis et al., 1982). The fraction (^) containing the desired size range was cloned in the EMBL3 vector (Stratagene). The band^ library was^ screened^ using pH-2IIa and washed in
2 x^ SSC at 65°C to detect the class I gene, R4B2. The insert was cloned
into the Sall site of (^) Bluescribe Vector (^) (Strategene) for restriction (^) mapping.
Two (^) fragments were (^) sequenced by the (^) dideoxy method (^) (Sanger et (^) al., 1977). The (^) sequences were (^) analyzed and (^) compared to known class I genes with (^) programs from the (^) University of Wisconsin Genetics (^) Computer (^) Group.
Acknowledgements We wish to (^) acknowledge the (^) many people who have assisted us (^) and contributed to this (^) study. Ms I.Carlo, Ms K.S.Katz and Mr G.R.Dastoornikoo provided technical^ assistance, and Mrs C.Speck helped type the (^) manuscript. We thank Drs (^) M.Kasahara, M.McBurney, J.L. (^) Lalanne, M. Steinmetz and Ms K.Minard for (^) gifts of (^) probes and Drs (^) H.C.Passmore, J.H.Nadeau and K.Arzt for (^) purified mouse DNA. We thank Drs (^) B.E.Loveland, (^) C.-R.Wang, K.Arzt and D.Bennett for (^) helpful discussions, and Drs C.Steinberg and S.Fazekas de St.Groth for critical (^) reading of the (^) manuscript. The Basel Institute for (^) Immunology was founded and is (^) supported by F.Hoffmann- La Roche and Co. (^) Ltd, Basel, Switzerland. S.W.H. is a (^) special fellow of the Leukemia (^) Society of America.
References
Blin,N. and^ Stafford,D.W. (1976) Nucleic Acids Res., 3, 2303-2308. Boer,P.H., Adra,C.N., Lau,Y.F. and^ McBurney,M. (1987) Mol. and Cell. Biol., (^) 7, 3107-3112. Brorson,K.A., Richards,S., Hunt,S.W.LH, Cheroutre,H., Fischer Lindahl,K. and (^) Hood,L. (1989) Immunogenetics, in press. Brown,G.D., Choi,Y., Egan,G. and^ Meruelo,D. (1988) (^) Immunogenetics, 27, 239-251. Bucan,M., (^) Herrmann,B., Frischauf,A.M., Bautch,V.L., Bode,V., Silver,L.M., Martin,G.R and (^) Lehrach,H. (1987) Genes Dev., 1, 376-385. Chan,T. and Fischer Lindahl,K. (^) (1985) Transplantation, 39, 477-480. Feinberg,A.P. and Vogelstein,B. (1984) Anal. Biochem., (^) 137, 266-267. Ferris,S.D., Ritte,U., Fischer Lindahl,K., Prager,E.M. and (^) Wilson,A.C. (1983) Nucleic Acids Res., 11, 2917-2926. Fischer (^) Lindahl,K. (1986) Curr. Top. Microbiol. Immunol., (^) 127, 272 -278. Fischer (^) Lindahl,K. and Burki,K. (1982) Proc. Natl. Acad. Sci. USA, (^) 79, 5362-5366. Fischer (^) Lindahl,K. and (^) Hausmann,B. (1983) Genetics, 103, 483 -494. Fischer (^) Lindahl,K. and (^) Langhorne,J. (1981) Scand. J. Immunol., 14, 643 -654. Fischer (^) Lindahl,K., Bocchieri,M. and (^) Riblet,R. (1980) J. Exp. Med., 152, 1583-1596. Fischer (^) Lindahl,K., Hausmann,B. and (^) Chapman,V.M. (1983) Nature, 306, 383 -385. Fischer Lindahl,K., (^) Hausmann,B., Robinson,P.J., Gudnet,J.-L., Wharton,D.C. and (^) Winking,H. (1986) J. (^) Exp. AMfed., 163, 334-346. Fischer Lindahl,K., Loveland,B. and Richards,S. (^) (1988) In (^) David,C.S. (ed.), Major Histocompatibility Genes and Their Roles in (^) Immune Function. Plenum Press, New York, pp. 327-338. Fisher,D.A., Hunt,S.W. and Hood,L. (1985) J. (^) Exp. Med., 162, 528-545. Han,A.C., Rodgers,J.R. and Rich,R.R. (^) (1987) Immunogenetics, 25, 234-240. Hirama,M. and Fischer Lindahl,K. (1985) In Quagliarello,E., (^) Slater,E.C., Palmieri,F., Saccone,C. and Kroon,A.M (eds), Achievements and Perspectives in Mitochondrial Research, Vol. II: Biogenesis. (^) Elsevier, Amsterdam, pp. 445-454. Huston,M.M, Smith,R.III, (^) Hull,R., Huston,D.P. and Rich,R.R. (^) (1985) Proc. Natl. Acad. Sci. USA, (^) 82, 3286-3290. Kasahara,M., Figueroa,F. and (^) Klein,J. (1987) Proc. Natl. Acad. Sci. USA, 84, 3325-3328. Kasahara,M., Passmore,H.C. and Klein,J. (1989) Immunogenetics, 29, 61 -63. Klein,D., Tewarson,S., Figueroa,F. and Klein,J. (1982) (^) Immunogenetics, 16, 319-328. Lalanne,J.L., Transy,C., Guerin,S., Darche,S., Meulie,P. and (^) Kourilsky,P. (1985) Cell, 41, 469-478. Maniatis,T., Fritsch,E.G. and (^) Sambrook,J. (1982) Molecular Cloning. A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold (^) Spring Harbor, NY. Moore,K.W., Sher,B.T., Sun,Y.H., Eakle,K.A. and Hood,L. (^) (1982) Science, 215, 679-682. Nadeau,J.H. (1983) Genet. Res., 42, 323-333. Passmore,H.C. and Romano,J. (1988) In (^) David,C.S. (ed.), Major Histocompatibility Genes and Their Role in Immune Functions. (^) Plenum Publishing, NY, pp. 49-60.
