| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Division of Immunogenetics, University of Göttingen, Göttingen, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2 Mb, and present a
physical map of this MHC region. Forty-five class I exon 4-hybridizing
BamHI fragments were detected, including the previously
known rat class I genes RT1-E, RT-BM1, RT1-N, RT1-M2,
RT1-M3, and RT1-M4. Twenty-six non-class I genes
known to map to the corresponding part of the human and mouse MHC were
tested and could be fine mapped in the RT1-C/E/M region at orthologous
position. Four previously known microsatellite markers were fine mapped
in the RT1-C/E/M region and found to occur in multiple copies. In
addition, a new, single-copy polymorphic microsatellite has been
defined. The expression profiles of several class I genes and the 26
non-class I genes were determined in 13 different tissues and exhibited
restricted patterns in most cases. The data provide further molecular
information on the MHC for analyzing disease susceptibility and
underline the usefulness of the rat model. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and 
, TNF-), complement
components (C2, C4, BF), and heat-shock proteins (hsp70–1, hsp70–2).
Others have no apparent function in the immune response. Their
expression and function are mostly not well understood, and their
involvement in disease control is unclear (10, 11, 12).
To understand the role of the MHC in controlling disease susceptibility
and graft rejection, a detailed knowledge of this gene complex
is necessary. The human MHC has been mapped physically and sequenced
recently (12). The overall structure of the rat MHC, the
RT1 complex, is similar to the homologous systems in mice and humans
(Fig. 1
). A characteristic difference between mouse and rat MHC on the
one hand and the human MHC on the other hand is the presence of an
additional region of classical class I (class Ia) genes centromeric to
the class II region (13). The telomeric class I region of
the rat MHC, RT1-C/E/M, is homologous to the region containing the
HLA-A, B, C, E, F, G genes in humans and to the H2-D/Q/T/M region
in mice.
|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The PAC library (RPCI-31), encompassing 10 genome equivalents of
BN rat (RT1n) origin (14), was
supplied as filters by the Resource Center of the German Human Genome
Project (Berlin, Germany). The filters were screened under stringent
conditions with two probes of the rat class I gene
RT1-Au (15), containing exon
2, introns 1 and 2 (1 probe), and exon 4 (3 probe) (see Table I), respectively. Positive clones were
digested with various restriction enzymes, notably BamHI and
EcoRI, and hybridized with the screening probe for
verification and then with further class I as well as non-class I
probes (Table I). The non-class I probes were generated from known
sequences of other MHC genes (12), H2 markers
(16), or on the basis of sequences obtained from the PAC
clones. Clones were ordered according to restriction fragment overlap,
hybridization patterns with various probes, and sequence data.
Sequencing of PAC clone ends was performed as previously described
(13) and analyzed with an ABI310 sequencer.
|
Total RNA was prepared (17) from various organs of
3-mo-old male BN/Gun rats and day 12 embryos (LEW.1W/Gun strain) bred
in our own colony and from BN lymphocytes after 3 days of stimulation
with Con A. RNA from three BN rats were pooled and analyzed by Northern
blot (washing conditions 2x SSC, 60°C for 2 x 10 min). In most
cases, probes were derived by PCR from the 3'-untranslated region (utr)
using DNA from the corresponding rat PAC clones, and -actin served
as loading and hybridization control. The primers were designed
according to rat gene or EST (expressed sequence tag) sequences when
available, or on the basis of mouse and human sequence information
(Table I
). Hybridization signals were scanned (Epson Scanner, software
ScanPack; Biometra, Göttingen, Germany), and the ratio of gene to
-actin signal was determined.
Microsatellite analysis
The D20Arb2 (18, 19), D20Mgh3
(18), D20Wox4 (18),
D20Wox5 (20), RT1-M4
(21), and D20Img2 (this study; Table I)
microsatellites were analyzed using primers and PCR conditions
according to the respective references. The PCR profile for
D20Img2 was 30 cycles of 94°C, 30 s; 55°C, 45
s; 72°C, 60 s. Genomic DNA of inbred rat strains BN/Gun,
LEW/Ztm, LEW.1A/Gun, LEW.1W/Gun, bred in our colony, was isolated
according to standard methods. Amplificates were analyzed on 2%
agarose gels. Additionally, the 5' primer of D20Arb2 was
6-FAM labeled, and analysis was conducted on an ABI310 with GeneScan
software.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The structure of the rat MHC is schematically shown in Fig. 1. The analysis focuses on the RT1-C/E/M
class I region. Probes 1 and 3, representing class I exons 2 and
4, respectively, were used for initial screening of the PAC filters.
Ninety-eight clones were obtained and verified to carry class I
sequences. These PAC clones were analyzed by Southern blot with various
class I and non-class I gene probes. In a first step, 31 PAC clones
could be ordered into four separate contigs (22) on the
basis of overlapping restriction fragments and hybridization patterns.
Class I gene-carrying clones that mapped to the RT1-A region have been
reported elsewhere (13).
The four initial contigs were then assembled into a single contig by
screening the PAC library with either non-class I probes known to map
to the respective part of the MHC or probes derived from information of
PAC end sequencing (sequence-tagged site (STS) markers). A single PAC
contig of 2 Mb could thus be constructed for the RT1-C/E/M region,
based on a usually deep coverage with overlapping clones (Fig. 2). The contig starts with the
Tnf and Bat1 genes and is thereby anchored in the
RT1 class III region. The contig ends beyond the RT1-M3b
gene. As is evident from Fig. 2, four subregions with class I genes are
separated by three subregions of non-class I genes.
|
The individual BamHI restriction fragments that
hybridized with the 1 and/or 3 class I probes (symbolized by
rectangles in Fig. 2) will be provisionally designated according to
their size. In total, 45 fragments hybridized with the 3 probe,
which represents the most conserved part of a class I gene. In
addition, in the second class I subregion, 24 consecutive
BamHI fragments were detected that hybridized only with the
1, but not the 3 probe.
In the first class I cluster, between Bat1 and
Pou5f1, 15 class I genes may be present, if one assumes that
a single BamHI fragment that hybridizes with both the 1
and 3 probes, or two neighboring fragments hybridizing with 1 and
3 probes, respectively, represent a complete class I gene. Most of
the class I genes in the first subregion (but none in the other class I
subregions) also hybridized with promoter and 3'-utr probes (Table I)
derived from the class Ia RT1-Au gene,
indicating that indeed complete class I genes are present and that they
are very similar to class Ia genes. Several class I fragments have been
partially sequenced. By comparison with published data, the class I
exon 2 sequence of BamHI fragment 1 is nearly identical with
the RT1-Cc clone cc1 (23), and
fragment 4 was identified as the RT1-E gene (24, 25). Exon 2 sequences of fragments 2, 19, 23, and 10 revealed
highest similarity (above 90%) to class Ia genes of the RT1-A region,
but not to H2-D or H2-Q genes, which reside in
the corresponding part of the mouse MHC.
In the second class I cluster, extending between Gln1 and
Znf173 (Fig. 2), the RT-BM1 gene
(26, 27, 28), which is identical to RT1-S3
(29), could be assigned to fragment 4.4 on the basis of
specific hybridization and partial sequence information obtained from
the 3'-utr of fragment 4.4. Similarly, the RT1-N1 gene
(30) was identified according to specific hybridization
and partial sequence data and assigned to BamHI
fragment 4.6.
Sequence analysis of the 1-cross-hybridizing fragment 2.1 (PAC clone
H12587) revealed the presence of 81 nucleotides from the 3' half of a
class I exon 2 without any neighboring class I sequences. Most likely,
the 24 consecutive 1-cross-hybridizing fragments (starting with
1-fragment 8 and ending with 1-fragment 5 in the second class I
subregion, Fig. 2) are duplications of the same module, since they
hybridize with the 3' half of exon 2, but not with the 5' half (not
shown). The presence of the rat homologue of the mouse STS marker
255D16T (31) is of interest because it maps to the
beginning of the H2-M region in the mouse (16).
In the third class I cluster between
Tctex4/Tctex5 and Mog, the
RT1-M4 gene was identified by a RT1-M4-specific
microsatellite (21). RT1-M3 (32)
was assigned to a fourth class I cluster telomeric to Ubd by
hybridization of the specific probe and by partial sequence data.
Different from the mouse, three RT1-M3 genes, designated
provisionally RT1-M3a, M3b, and M3c, are found
(Fig. 2), as well as three copies of the Leh525 homologous
marker. The RT1-M3c gene, together with the
RT1-M2 gene (33), is localized on a contig that
could not yet be linked to the main contig directly, but has been
assigned to the RT1 region by fluorescence in situ hybridization (K.
Helou and L. Walter, unpublished data). This finding is in accord with
the identification of three H2-M3 homologous genes on a
single clone of a rat yeast artificial chromosome (YAC) library
(34).
No cross-hybridization signal with MHC class I-related probes could be found for any PAC clone of the contig, confirming negative genomic Southern blot results (not shown).
Non-class I genes
In the RT1-C/E/M-corresponding parts of the human and mouse MHC,
represented by the HLA class I and the H2-D/Q/T/M regions,
respectively, clusters of conserved non-class I genes (also designated
framework genes; 16) have been described extending from
Bat1 to Ubd (10, 11, 12, 16). Of the
34 non-class I genes (excluding known pseudogenes) reported
(10, 11, 12), 26 have been included in the PAC analysis. These
genes could be assigned to the RT1-C/E/M region (Fig. 2, genes
symbolized by triangles), and their order, as far as determined, is the
same as in the HLA (10, 11, 12) and H2 complex (16, 35). The first three class I subregions are interspersed at
orthologous positions in the three species, as defined by framework
genes Bat1, Pou5f1, Gnl1 (Gna-rs1), Znf173,
Tctex4, and Mog (Fig. 2).
A syndrome of decreased male fertility, dwarfism, and increased
susceptibility to chemical carcinogenesis has been mapped to the RT1-C
region of the RT1r16 haplotype (36).
The respective genes ft, dw3, and rcc are assumed
to be part of a growth and reproduction complex (Grc) in the MHC. The
Grc, defined at the molecular level as a region hybridizing multiply
with the grc1.4 probe (37), could now be assigned with
this probe by PAC hybridization to the second class I subregion
(Fig. 2).
Microsatellite markers
Besides the RT1-M4 microsatellite in the RT1-M region
(21), four known microsatellite markers can be localized
in the RT1-C/E/M region. The D20Mgh3 and D20Wox4
markers (18) are both derived from the class I gene
RT1A-4 (38) and are identical, except for an
incomplete overlap of the 3'-primer (see
http://www.nih.gov/niams/scientific/ratgbase/data/ARBPR20.htm). Both
markers have now been mapped to the first class I subregion (Fig. 2).
They do not occur in single copy, but are repeated at least five times,
as is documented by their presence on three nonoverlapping PAC clones
from which one or two amplification products can be obtained (Fig. 3, A and B). A
composite of these amplificates is seen in BN genomic DNA (Fig. 3, A and B). The D20Wox5 microsatellite
(20), derived from the RT1-O gene
(39), is present in at least three copies in the second
class I subregion (Fig. 2). One and two amplificates are detectable
with two nonoverlapping PAC clones, respectively (Fig. 3C).
The BN genomic DNA shows both PCR product lengths (Fig. 3C).
The D20Arb2 microsatellite (18, 19), derived
from the RT1-N1 gene (30), also maps to the
second class I subregion at multiple positions (Fig. 2). An
amplification product of the same length can be obtained with three
nonoverlapping PAC clones (Fig. 3, D and E). Up
to five D20Arb2 amplification products of different length
can be obtained in other RT1 haplotypes (Fig. 3E). By end
sequencing of PAC N0751 a CA dinucleotide repeat and corresponding
flanking primers (Table I) could be established (Fig. 2). This
microsatellite, 288 bp long (RT1n haplotype),
designated D20Img2, occurs only once in the RT1-C/E/M region
and is polymorphic in different strains (Fig. 3F).
|
Northern blot analysis was conducted with total RNA from 11
organs, lymphoblasts, and day 12 embryo using 3'-utr probes (Table I)
from each of the 26 framework and four class I genes of the RT1-C/E/M
region (Table II). Only
few genes such as Tubb are ubiquitously expressed. The class
I genes tested and most framework genes exhibit differential expression
profiles, often in an organ-specific manner, such as Mog and
Gabbr1 showing brain-specific expression and
Tctex4 being testis specific. The expression of genes such
as Ubd is induced in mitogen-stimulated lymphocytes, in
accord with stimulation of this gene by IFN-
(40). For some genes, no positive Northern blot signal
could be detected, because expression might be too weak or
restricted to tissues not included in the panel. In the case of
Spr1 and Rnf9, expression was checked by RT-PCR
with RNA from liver, small intestine, lung, lymphoblasts, spleen, and
embryo, but no amplification products could be obtained. In general,
the expression patterns obtained are in accord with the relative tissue
representation of corresponding ESTs (expressed sequence tags; see
http://www.ncbi.nlm.nih.gov/UniGene/query.cgi and
http://ratest.eng.uiowa.edu/cgi-bin/map-info?chr=20).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In total, four class I clusters have been found to alternate with three
non-class I gene clusters in between (Fig. 2). This organization is
very similar to that described for the H2-D/Q/T/M region in mice
(16) and the class I region in humans (10, 11, 12, 16). The boundaries of the first three class I gene clusters are
marked by orthologous framework genes in the three species. An
exception is the fourth class I gene cluster containing the
RT1-M3 genes. It is found telomeric to the Mog
gene, similar to the mouse, whereas no class I gene has been described
in the corresponding part of the HLA complex. Each of the 26 framework
genes tested in the rat has been found in the RT1-C/E/M region at an
orthologous position, and none of them mapped into a class I subregion.
In this context, the RANBP1 (TC4) sequence found
in the second class I cluster of the HLA complex (10, 11, 12)
is of interest, because it is not present in the rat RT1-C/E/M region
(own unpublished data). Inspection of the sequence indicates that it
might be a processed pseudogene, and as such would not be expected to
be conserved between rat and human in contrast to expressed framework
genes.
The number of class I genes in the RT1-C/E/M region of the
RT1n haplotype will only be known when the
complete sequence of this region is available. On the basis of
hybridization with the conserved exon 4 probe, 45 class I genes
could be assumed. This number is lower than the 61 (42)
and 62 class I genes (43) extrapolated from screening
cosmid libraries of the RT1av1 and
RT1r21 haplotypes, respectively. Haplotype
differences in the number of class I genes are a familiar phenomenon of
the MHC, also in the rat (13). It is unknown at present
how many of the class I genes are indeed expressed. Among the class I
genes described in the RT1-C/E/M region before, RT1-E
(24, 25), RT-BM1 (26, 27, 29),
RT1-N1 (30), RT1-M3
(32), and RT1-M4 (21) have been
fine mapped now, and for some of them restricted tissue distribution
could be shown (Table II). Genes of the first class I cluster, such as
RT1-E, as well as class I sequences obtained from this
region, are more similar to rat class Ia genes than to other rat class
I genes or mouse H2-D, Q, T, M genes. Thus, whereas the localization of
this subregion immediately telomeric to Bat1 corresponds to
the H2-D/L/Q region of the mouse, the individual genes are not
orthologous. The RT-BM1 (RT1-S3) gene in the
second class I gene cluster is assumed to be orthologous to the mouse
H2-T23 gene (27, 28), which encodes the Qa-1
molecule, and the RT1-N1 gene is reported to be orthologous
to H2-T10 and T22 (30). Furthermore,
the Grc probe, defining the Grc region in the second cluster, has been
shown to cross-hybridize to sequences flanking the
H2-T10d/11d
and
H2-T22d/23d
genes in the mouse (44). Consequently, the second class I
subregion appears to correspond to the mouse H2-T region. Since in the
mouse the 255D16T marker maps to the beginning of the H2-M
region, the position of the homologous sequence in the second class I
subregion (Fig. 2) could mark the beginning of the homologous region,
RT1-M, in the rat, and the 3-cross-hybridizing fragments telomeric
to this marker therefore might represent homologues of mouse H2-M
genes. The H2-M4 orthologous gene RT1-M4
(21) maps to the third class I subregion, and the
H2-M3 orthologous gene RT1-M3 (32)
to the fourth subregion. Thus, the telomeric part of the whole contig
corresponds to the mouse H2-M region, confirming and extending
previously published genetic data (33).
No human/rat orthology can be established for the class I genes in the various class I clusters. This is in accord with the general lack of interorder class I orthology. It is of note that HLA-MHC class I-related-like genes are missing in the rat MHC (as in the mouse; 45), whereas H2-T-like as well as H2-M-like genes, which are missing in the human MHC, are present in the rat. Therefore, with respect to class I genes, the rat RT1-C/E/M region cannot serve as a model for the HLA complex, whereas the non-class I genes are clearly orthologous.
The function of the RT1-C/E/M class I genes is not well defined. It is known that they can act as targets of CTL (23) and as stimulatory targets of alloreactive NK cells (46). Notably, the RT1-E gene product has been shown to stimulate alloreactive NK cells (47). RT1-C/E/M incompatibility has been shown to induce skin and pancreas graft rejection (48) and to modulate the fate of MHC class II-mismatched heart grafts (49). The individual RT1-C/E/M genes that are responsible for the histoincompatibility reaction have not yet been identified. Ag presentation by RT1-C/E/M-encoded class I molecules could not be shown (50) except one report (51). Nevertheless, since H2-Qa1 and H2-M3, for example, are able to present particular peptides in the mouse, a similar function could be assumed for RT-BM1 and RT1-M3, respectively. Also, the class Ia similarity of genes in the first class I subregion could indicate a peptide presentation function.
The RT1-C/E/M region is involved in controlling disease susceptibility. Analysis of RT1 recombinants has shown that severity of collagen-induced arthritis is associated with the RT1-C/E/M genotype (52). Similarly, the course of anti-myelin/oligodendrocyte glycoprotein (MOG)-induced encephalomyelitis is modulated by genes in the RT1-C/E/M region (7, 8).
The function of the non-class I genes, as far as known, is diverse
(Table I). Some genes appear to encode transcription factors
controlling cell proliferation or DNA repair, and have been suggested
to be involved in cancer development (11). An example is
the Ppp1r10 gene (alias designations Fb19 and
Pnuts) that has been detected because of its expression in
pancreatic cancer (53). In this context it is noteworthy
that the Grc controls susceptibility to diethylnitrosamine-induced
hepatocellular carcinomas (36). Unexpectedly, the Grc, as
far as defined by the grc1.4 probe (37), maps into the
second class I subregion without including known framework genes.
To implement RT1-C/E/M genes into whole genome analyses of physiological or pathological traits, microsatellite markers are extremely useful. The microsatellites D20Arb2, D20Mgh3, D20Wox4, and D20Wox5, known to localize to the RT1-C/E/M region, were fine mapped. Surprisingly, PAC analysis revealed that these markers occur in several copies, which are not always distinguishable by length polymorphism. This multicopy nature must be taken into account in genetic analyses. Since the four microsatellite markers are associated with class I genes, multiple occurrence presumably reflects the amplification of the respective genes. A new polymorphic microsatellite marker, D20Img2, that is found in single copy, could be identified in the second framework gene region.
To establish expression profiles, rat-specific probes of 30 RT1-C/E/M genes were used to analyze total RNA from the BN strain by Northern blot. Characteristic patterns were detected, although more refined tissue representation, also from diseased organs, is necessary to get information about coordinate or clustered expression patterns. Preliminary results indicate that RT1 haplotype-related expression differences exist, for example for Mog (L.W., R.D., E.G., unpublished data). DNA chips containing the RT1-C/E/M gene probes are being produced to speed up this analysis.
The physical map of the RT1-C/E/M region and the preliminary expression data presented in this work will help to characterize the role of this MHC region in controlling immune responsiveness, graft rejection, and disease susceptibility. Furthermore, the contig provides the basis for sequencing this part of the rat MHC.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 The nucleotide data reported in this work have been assigned the database accession numbers AJ294759, AJ294760, and AJ294761 (see Table I).
3 S.I. and L.W. contributed equally.
4 Address correspondence and reprint requests to Dr. Eberhard Günther, Division of Immunogenetics, University of Göttingen, Heinrich-Düker-Weg 12, 37073 Göttingen, Germany.
5 Abbreviations used in this paper: PAC, P1-derived artificial chromosome; Grc, growth and reproduction complex; STS, sequence-tagged site; utr, untranslated region.
Received for publication October 18, 2000. Accepted for publication January 4, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and tumor necrosis factor-. Eur. J. Immunol. 29:4030.[Medline]
This article has been cited by other articles:
|
|
 |
|
  C. Naper, K.-Z. Dai, L. Kveberg, B. Rolstad, E. C. Niemi, J. T. Vaage, and J. C. Ryan Two Structurally Related Rat Ly49 Receptors with Opposing Functions (Ly49 Stimulatory Receptor 5 and Ly49 Inhibitory Receptor 5) Recognize Nonclassical MHC Class Ib-Encoded Target Ligands J. Immunol., March 1, 2005; 174(5): 2702 - 2711. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Roos, R. Dressel, B. Schmidt, E. Gunther, and L. Walter The Rat Expresses Two Complement Factor C4 Proteins, but Only One Isotype Is Expressed in the Liver J. Immunol., January 15, 2005; 174(2): 970 - 975. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Tuzun, J. A. Bailey, and E. E. Eichler Recent Segmental Duplications in the Working Draft Assembly of the Brown Norway Rat Genome Res., April 1, 2004; 14(4): 493 - 506. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Hurt, L. Walter, R. Sudbrak, S. Klages, I. Muller, T. Shiina, H. Inoko, H. Lehrach, E. Gunther, R. Reinhardt, et al. The Genomic Sequence and Comparative Analysis of the Rat Major Histocompatibility Complex Genome Res., April 1, 2004; 14(4): 631 - 639. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Amadou, R. M. Younger, S. Sims, L. H. Matthews, J. Rogers, A. Kumanovics, A. Ziegler, S. Beck, and K. Fischer Lindahl Co-duplication of olfactory receptor and MHC class I genes in the mouse major histocompatibility complex Hum. Mol. Genet., November 15, 2003; 12(22): 3025 - 3040. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Takada, A. Kumanovics, C. Amadou, M. Yoshino, E. P. Jones, M. Athanasiou, G. A. Evans, and K. F. Lindahl Species-Specific Class I Gene Expansions Formed the Telomeric 1 Mb of the Mouse Major Histocompatibility Complex Genome Res., April 1, 2003; 13(4): 589 - 600. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Naper, S. Hayashi, L. Kveberg, E. C. Niemi, L. L. Lanier, J. T. Vaage, and J. C. Ryan Ly-49s3 Is a Promiscuous Activating Rat NK Cell Receptor for Nonclassical MHC Class I-Encoded Target Ligands J. Immunol., July 1, 2002; 169(1): 22 - 30. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
