MATERIALS AND METHODSChromosomal Localization of Insert in Transgenic Mice. Transgenic mice. The transgenic mice used in this study were derived from microinjection of cloned genomic DNA fragments containing the gene for the human major histocompatibility complex class I molecule HLA-B7. As described in detail elsewhere (9), line 18A was derived from microinjection of a 6.0-kb EcoRI/BamHI DNA fragment (fragment a), containing the wild-type gene and 0.66 kb of 5' and 2.0 kb of a 3' flanking DNA. Lines 30, 47, and 50 were generated with a 6.0-kb EcoRI/BamHI fragment (k) identical to fragment a except that it contained a 4-bp insertional mutation in a 5' cis-active regulatory sequence (the a site) located at approximately -0.11 kb.Transgenic animals with insert sizes of >20 kb were selected for study to facilitate visualization of loops. A transgenic mouse with a large inset of approximately 1000 copies of the mouse globin gene was also used to study the changes in morphology of the chromatin packaging of the foreign inserts as the chromosome progressed through the first meiotic division.Abbreviations: FISH, fluorescence in situ hybridization; SC, synaptonemal complex. tTo whom reprint requests should be addressed.

We introduced the human genes HLA-B7 and B2M encoding the heavy (HLA-B7) and light [beta 2-microglobulin (beta 2m)] chains of a human major histocompatibility complex class I antigen into separate lines of transgenic mice. The tissue-specific pattern of HLA-B7 RNA expression was similar to that of endogenous class I H-2 genes, although the HLA-B7 gene was about 10-fold underexpressed in liver. Identical patterns of RNA expression were detected whether the HLA-B7 gene contained 12 or 0.66 kilobase(s) (kb) of 5' flanking sequence. The level of expression was copy number dependent and as efficient as that of H-2 genes; gamma interferon enhanced HLA-B7 RNA expression in parallel to that of H-2. In addition to the mechanism(s) responsible for gamma interferon-enhanced expression, there must be at least one other tissue-specific mechanism controlling the constitutive levels of class I RNA. Tissue-specific human beta 2m RNA expression was similar to that of mouse beta 2m, including high-level expression in liver. Cell surface HLA-B7 increased 10- to 17-fold on T cells and on a subset of thymocytes from HLA-B7/B2M doubly transgenic mice compared to HLA-B7 singly transgenic mice. The pattern of expression of HLA-B7 on thymocytes resembled that of H-2K as opposed to H-2D. These results confirm that coexpression of both human chains is required for efficient surface expression and that HLA-B7 may share a regulatory mechanism with H-2K, which distinguishes it from H-2D.

The major histocompatibility complex (MHC) class I HLA-B7 transgene carrying a 660-bp upstream sequence is expressed in the mouse with tissue specificity that parallels that of the expression of endogenous mouse MHC class I (H-2) genes. We have performed in vivo genomic footprinting for the HLA-B7 transgene and the endogenous HN2Kb gene. We show that the upstream region of both the transgene and the endogenous gene was extensively occupied in spleen tissue, where these genes are expressed at high levels. In contrast, no occupancy was detected in brain tissue, where expression of these genes is virtually absent. Sites exhibiting in vivo protection correspond to cis elements previously shown to bind to nuclear factors in vitro, including the constitutive enhancer region I and the interferon response element. The strongest tissue-specific protection was detected at site a, located downstream from the interferon response element. Site of bound a constitutively expressed nuclear factor(s) in vitro that exhibited an overlapping specificity which may involve a nuclear hormone receptor, RXR, and an AP-1-related factor. Site a was functional in vivo, as it enhanced MHC class I transcription in lymphocytes. These results show that the tissue-specific occupancy of the MHC class I regulatory sequences in vivo correlates with their expression and suggest that in vivo occupancy is controlled by a mechanism other than the mere presence of factors capable of binding to these sites. Our results suggest that a sequence present in the 660-bp upstream region in a human leukocyte antigen gene directs tissue-specific occupancy of MHC class I genes in vivo, independently of their position and copy number, illustrating a potential advantage of using a transgene for delimitation of the sequence requirement for in vivo occupancy.

Mature T cells express either CD4 or CD8 on their surface. Most helper T cells express CD4, which binds to class II major histocompatibility complex (MHC) proteins, and most cytotoxic T cells express CD8, which binds to class I MHC proteins. In the thymus, mature CD4+CD8- and CD4-CD8+ T cells expressing alpha beta T-cell antigen receptors (TCR) develop from immature thymocytes through CD4+CD8+ alpha beta TCR+ intermediates. Experiments using mice transgenic for alpha beta TCR suggest that the specificity of the TCR determines the CD4/CD8 phenotype of mature T cells. These results, however, do not indicate how a T cell differentiates into the CD4 or CD8 lineage. Here we show that the CD4 transmembrane region and/or cytoplasmic tail mediates the delivery of a specific signal that directs differentiation of T cells to a CD4 lineage. We generated transgenic mice expressing a hybrid molecule composed of the CD8 alpha extracellular domains linked to the CD4 transmembrane region and cytoplasmic tail. We predicted that this hybrid molecule would bind to class I MHC proteins through the extracellular domains but deliver the intracellular signals characteristic of CD4. By crossing our transgenic mice with mice expressing a transgenic alpha beta TCR specific for a particular antigen plus class I MHC protein, we were able to express the hybrid molecule in developing thymocytes expressing the class I MHC-restricted TCR. Our results show that the signal transduced by the hybrid molecule results in the differentiation of immature thymocytes expressing a class I-restricted TCR into mature T cells expressing CD4.

Although HLA transgenic mice (HLA TgM) could provide a powerful approach to investigate human MHC-specific T cell responsiveness, the extent to which these molecules are recognized by the mouse immune system remains unclear. We established TgM expressing HLA class I alleles A2, B7, or B27 in their fully native form (HLAnat) or as hybrid molecules (HLAhyb) of the HLA α1/α2 domains linked to the H-2Kb α3, transmembrane, and cytoplasmic domains (i.e., to maintain possible species-specific interactions). Comparison of each as xeno- (i.e., by non-TgM) vs allo- (i.e., by TgM carrying an alternate HLA allele) transplantation Ags revealed the following: 1) Although HLAhyb molecules induced stronger xeno-CD8+ T cell responses in vitro, additional effector mechanisms must be active in vivo because HLAnat skin grafts were rejected faster by non-TgM; 2) gene knockout recipients showed that xenorejection of HLAnat and, unexpectedly, HLAhyb grafts doesn’t depend on CD8+ or CD4+ T cells or B cells; 3) each HLAhyb strain developed tolerance to “self” but rejected allele- (-B27 vs -B7) and locus- (-B vs -A) mismatched grafts, the former requiring CD8+ T cells, the latter by CD8+ T cell-independent mechanisms. The finding that recognition of xeno-HLAhyb does not require CD8+ T cells while recognition of the identical molecule in a strictly allo context does, demonstrates an α1/α2 domain-dependent difference in effector mechanism(s). Furthermore, the CD8+ T cell-independence of locus-mismatched rejection suggests the degree of similarity between self and non-self α1/α2 determines the effector mechanism(s) activated. The HLA Tg model provides a unique approach to characterize these mechanisms and develop tolerance protocols in the context of human transplantation Ags.

Although mice transgenic (Tg) for human MHC (HLA) class I alleles could provide an important model for characterizing HLA-restricted viral and tumor Ag CTL epitopes, the extent to which Tg mouse T cells become HLA restricted in the presence of endogenous H2 class I and recognize the same peptides as in HLA allele-matched humans is not clear. We previously described Tg mice carrying the HLA-B27, HLA-B7, or HLA-A2 alleles expressed as fully native (HLAnat) (with human β2-microglobulin) and as hybrid human/mouse (HLAhyb) molecules on the H2b background. To eliminate the influence of H2b class I, each HLA Tg strain was bred with a H2-Kb/H2-Db-double knockout (DKO) strain to generate mice in which the only classical class I expression was the human molecule. Expression of each HLAhyb molecule and HLA-B27nat/human β2-microglobulin led to peripheral CD8+ T cell levels comparable with that for mice expressing a single H2-Kb or H2-Db gene. Influenza A infection of Tg HLA-B27hyb/DKO generated a strong CD8+ T cell response directed at the same peptide (flu nucleoprotein NP383–391) recognized by CTLs from flu-infected B27+ humans. As HLA-B7/flu epitopes were not known from human studies, we used flu-infected Tg HLA-B7hyb/DKO mice to examine the CTL response to candidate peptides identified based on the B7 binding motif. We have identified flu NP418–426 as a major HLA-B7-restricted flu CTL epitope. In summary, the HLA class I Tg/H2-K/H2-D DKO mouse model described in this study provides a sensitive and specific approach for identifying and characterizing HLA-restricted CTL epitopes for a variety of human disease-associated Ags.

We previously reported that genomic major histocompatibility complex class I human leukocyte antigen (HLA)-B7 gene constructs with as little as 0.66 kb of 5'-and 2.0 kb of 3'-flanking DNA were expressed efficiently and appropriately in transgenic mice. To identify and characterize the relevant cis-acting regulatory elements in more detail, we have generated and analyzed a series of transgenic mice carrying native HLA-B7 genes with further 5' truncations or intronic deletions and hybrid constructs linking the 5'-flanking region of B7 to a reporter gene. We were unable to detect a specific requirement for sequence information within introns 2 to 7 for either appropriate constitutive or inducible class I expression in adult animals. The results revealed the presence of cis-acting regulatory sequences between -0.075 kb and -0.66 kb involved in driving efficient copy number-dependent constitutive and gamma interferon-enhanced tissue-specific expression. The region from -0.11 to -0.66 kb is also sufficient to prevent integration site-specific "position effects," because in its absence HLA-B7 expression is frequently detected at significant levels at inappropriate sites. Conserved sequence elements homologous to the H-2 class I regulatory element, or enhancer A, and the interferon response sequence are located between about -151 and -228 bp of the B7 gene. Our results also indicate the existence of sequences downstream of -0.11 kb which can influence the pattern of tissue-specific expression of the HLA-B7 gene and the ability of this gene to respond to gamma interferon.

No abstract