Cell surface expression of hybrid murine/human MHC class II beta alpha dimers. Key influence of residues in the amino-terminal portion of the beta 1 domain

To aid in the identification of key residues responsible for the control of class II MHC beta-alpha dimer assembly and expression, a series of cotransfections of human plus mouse beta- and alpha-genes was performed. The resulting expression data were correlated with the sequences of the relevant proteins to identify residues that played critical roles in these processes. For the I-E/DR homologues good expression was seen for both E beta DR alpha and DR beta E alpha combinations involving several allelically variable beta-chains of each species. These results are consistent with the sequence conservation seen for I-E and DR gene products, and indicate that the species-specific differences that do exist play little role in controlling dimer formation or transport. For A beta chains, a more complex picture was seen. A beta d, but not A beta k or A beta b, was found to coexpress with human alpha-chains. Not only did A beta d show expression with the homologous DQ alpha-chain, but it also was expressed with DR alpha and DP alpha. These data indicate that species-specific residues do not control dimer expression under these conditions and confirm that allelically polymorphic residues have a crucial role in this process. Mapping studies using recombinant A beta genes established the importance of the residues in the amino-terminal half of the beta 1 domain in the differences observed among the A beta alleles. Sequence comparison of DR beta, DP beta, DQ beta, E beta, and A beta chains in this region revealed a single residue (position 12) conserved in most chains and differing in a nonconservative fashion between A beta d vs A beta b or k. A beta d has the conserved lysine at this position, whereas A beta b has methionine and A beta k has glutamine. To test whether this residue actually was important physiologically, a lysine codon was created in a recombinant A beta gene possessing the amino-terminal sequence of the kappa haplotype, and the ability of this mutant chain to be expressed with various mouse A alpha-chains was examined. This mutant chain was shown to gain the ability to be efficiently expressed with A alpha d without losing its ability to be expressed with A alpha k. These data reemphasize the special role played by allelically polymorphic residues in Ia expression and identify one such polymorphic site as position 12.(ABSTRACT TRUNCATED AT 400 WORDS)     

An MHC interaction site maps to the amino-terminal half of the T cell receptor alpha chain variable domain.

We have used cloned T cell receptor (TCR) genes from closely related CD4 T cell lines to probe the interaction of the TCR with several specific major histocompatibility complex (MHC) class II ligands. Complementarity determining region 3 (CDR3) equivalents of both alpha and beta TCR chains are required for antigen-MHC recognition. Our data provide novel information about the rotational orientation of TCR-MHC contacts in that exchange of the amino terminal portion of the TCR alpha chain containing the putative CDR1 and CDR2 regions results in both gain and loss of MHC class II specificity by the resulting receptor. These two TCRs differ primarily in recognition of polymorphisms in the second hypervariable region of the MHC class II alpha chain. These results document the involvement of CDR1 and/or CDR2 of the TCR alpha chain in MHC recognition and suggest a rotational orientation of this TCR to its MHC ligand.     

Loss of a glycine in the alpha2 domain affects MHC peptide binding but not chaperone binding.

Prior to the binding of peptide in the endoplasmic reticulum (ER), the major histocompatibility complex (MHC) class I heavy chain associates with an assembly complex that includes the transporter associated with antigen processing (TAP). The proximity of a part of the MHC class I alpha2 domain alpha-helix to areas previously shown to influence assembly complex binding suggests that this region might also be involved in chaperone association. Position 151, found in this part of the alpha2 domain alpha-helix, has a side chain that points up, away from direct contact with peptide, and is occupied by a glycine in all murine MHC class I heavy chains. We found that substitution of this glycine in H-2L(d) with a histidine substantially increased the proportion of peptide-free forms, although TAP binding was not abrogated. Thus, interaction of the heavy chain with peptides, but not with the assembly complex, is influenced by this glycine.     

Recognition of HLA-A2 mutant and variant target cells by an HLA-A2 allospecific human cytotoxic T lymphocyte line

HLA-A2 specific human cytotoxic T lymphocytes (CTL) cell lines have been developed using T cell growth factor and coculture of peripheral blood lymphocytes with selected allogeneic target cell lines. The CTL-8 line showed specificity for human leukocyte antigens (HLA)-A2 bearing target cells after 5 weeks in culture when tested against a panel of 14 lymphoblastoid cell lines in a 51Chromium (51Cr) release assay. Purified anti-human leukocyte antigens (HLA) monoclonal antibodies W6/32 and PA2.1 inhibited cytolysis by 85% and 60%, respectively. The CTL-8 line lysed non-HLA-A2 target cells in the presence of lectins concanavalin A (Con A) or phytohemagglutinin-P lectin (PHA-P) indicating the specificity of cytolysis was not due to nonspecific resistance of target cells to the CTL-lytic mechanism. The T5-1 HLA-A2 mutant cell series were tested as targets for the CTL-8 line. Cell clones 8.18.1, 8.21.1 and 8.6.1, which express altered HLA-A2 molecules as determined by their decreased reactivity with allospecific monoclonal antibodies, were lysed by the CTL-8 line as efficiently as the T5-1 wild type. These cell lines also acted as efficient cold target competitors for a normal HLA-A2 target cell. The 8.14.1 cell clone expressed a lower amount of HLA-A2 alloantigen and showed a corresponding decreased reactivity with CTL-8 in direct cytolytic and cold target competitive inhibition assays. In contrast, the M7 and DK1 HLA-A2 variant cell lines, which express normal HLA-A2 serological determinants, were inefficiently lysed by CTL-8 and did not act as competitive inhibitors of normal HLA-A2 target cells. These results support the concept that the alloantigenic determinant(s) recognized by T cells and antibodies occur at separate regions on the HLA-A2 molecule.     

A recombinant C-terminal fragment of staphylococcal enterotoxin A binds to human MHC class II products but does not activate T cells.

Binding of staphylococcal enterotoxin A (SEA) to MHC class II encoded proteins is a prerequisite for its subsequent activation of a large fraction of T lymphocytes through interaction with variable segments of the TCR-beta chain. We cloned SEA in Escherichia coli and produced four recombinant fragments covering both the N- and C-terminal regions. These fragments were used to analyze the interaction between SEA and the human MHC class II products. A C-terminal fragment of SEA, representing amino acids 107-233 bound to HLA-DR and HLA-DP but did not activate T cells. The three other fragments (amino acids 1-125, 1-179 and 126-233) neither bound to MHC class II Ag nor activated T cells. SEA apparently bind to HLA-DR and HLA-DP through its C-terminal part, whereas T cell activation is dependent on additional parts of the protein.     

Recognition of the alpha-1 and alpha-2 domains of H-2 molecules by allospecific cloned T cells

Previously we had shown that allospecific bulk cultures of cytolytic T lymphocytes lysed the products of cloned class I major histocompatibility genes expressed after DNA-mediated gene transfer. In these experiments, performed by using cloned allospecific T cell effectors, a T cell hybridoma, and recombinant DNA technology, we have been able to map determinants recognized by these T cell clones to the alpha-1 domain of H-2Dd and the alpha-2 domain of H-2Ld (four of eight clones). Target cells used were L cells (H-2k), expressing wild type or hybrid H-2 antigens of H-2d origin. Thus, for the first time determinants recognized by cloned T cells are found in the recombined alpha-1 and alpha-2 domains.     

The amino-terminal one-third of pseudorabies virus glycoprotein gIII contains a functional attachment domain, but this domain is not required for the efficient penetration of Vero cells.

We have examined the attachment and penetration phenotypes of several glycoprotein gIII mutants of pseudorabies virus (PRV) and have identified the first one-third of gIII as a region that mediates efficient virus attachment to PK15 and Vero cells. This portion of gIII, amino acids 25 through 157 of the wild-type sequence, appeared to support attachment by binding to heparinlike molecules on cell surfaces. Virions containing the first one-third of gIII were sensitive to heparin competition and showed greatly reduced infectivity on cells treated with heparinase. PRV virions lacking the first one-third of the mature glycoprotein exhibited only residual binding to cells if challenged by vigorous washing with phosphate-buffered saline at 2 h postinfection at 4 degrees C. This residual binding was resistant to heparin competition, and strains lacking the first one-third of gIII were able to infect cells treated with heparinase as effectively as untreated cells. When we determined the penetration phenotypes for each strain, we found that gIII-mediated virus attachment was necessary for timely penetration of PK15 cells but remarkably was not required for efficient virus penetration of Vero cells. Moreover, wild-type PRV was actually prohibited from rapid penetration of Vero cells by a gIII-heparan sulfate interaction. Our results indicate that initial virus binding to heparan sulfate via glycoprotein gIII is not required for efficient PRV infection of all cell types and may in fact be detrimental in some instances.     

Characterization of the amino-terminal activation domain of peroxisome proliferator-activated receptor alpha. Importance of alpha-helical structure in the transactivating function

The transactivating function of the A/B region of mouse peroxisome proliferator-activated receptor alpha (PPARalpha; NR1C1) was characterized. The truncated version of PPARalpha lacking the A/B region had 60-70% lower transactivating function than full-length PPARalpha in both the presence and absence of the peroxisome proliferator ciprofibrate. When tethered to the yeast Gal4 DNA-binding domain, the A/B region exhibited the significant ligand-independent transactivating function, AF-1 activity. The first 44 amino acid residues were necessary for maximal transactivation, and the minimally essential region was further delimited to amino acids 15-44. This region is highly enriched with acidic residues, but mutational analyses showed that the protein structure, rather than the negative charge itself, was important for the AF-1 activity. An alpha-helical configuration was predicted for this region, and a CD spectrum analysis of the synthetic peptides showed that mutant sequences with higher AF-1 activity have higher helical contents and vice versa. The most active mutant, in which Met(31) was replaced with Leu, was approximately 5-fold more potent than the wild-type A/B region. These findings indicate that the AF-1 region of PPARalpha is an acidic activation domain and that the helix-forming property is implicated in the transactivating function.     

The loss of class II MHC antigen expression byras-transformed murine fibroblasts passaged as tumours correlates with increased tumorigenicity but is not mediated by T cells

Summary In several murine tumour systems, expression of class II major histocompatibility complex (MHC) antigens by tumour cells, either constitutive or inducible, correlates with reduced tumorigenicity as compared with equivalent class-II-negative cells, and CD4 phenotype T cells together with interferon (which induces the expression of class II) may be involved in the control of the proliferation of class-II-expressing tumours. This implies a potential T-cell-mediated selection pressure against class II expression. To test this possibility, we have repeatedly passaged as tumours in euthymic, syngeneic miceras-transformed murine fibroblast lines, which are class-II-inducible, to determine whether class-II-non-inducible variants are selected. We examined the expression of both class I and class II antigen in tumour cells re-established in vitro. It was found that the inducibility of class II, but not class I, expression rapidly diminished, correlating with augmented tumorigenicity. However, this loss of class II inducibility occurred in athymic as well as euthymic mice. Therefore, despite the fact that the tumorigenicity of these lines is augmented in euthymic mice depleted of CD4 T cells or interferon , we found no evidence of T-cell-mediated selection against class II expression. The loss of class II expression observed must be due to mechanisms other than immune selection. The possibility that this might result from other soluble factors modulating the response to interferon in vivo is discussed.     

A 48 kDa collagen-binding phosphoprotein isolated from bovine aortic endothelial cells interacts with the collagenous domain, but not the globular domain, of collagen type IV.

We have identified collagen-binding proteins in detergent extracts of metabolically labelled bovine aortic endothelial cells (BAEC) by collagen type IV-Sepharose affinity chromatography. The major collagen type IV-binding protein identified by SDS/PAGE had a molecular mass of 48 kDa, which we term the 'collagen-binding 48 kDa protein' (CB48). The pI of CB48 was 8.0-8.3 in a two-dimensional gel system, running non-equilibrium pH gel electrophoresis in the first dimension and SDS/PAGE in the second dimension. Under these conditions CB48 separated into two major (a and b) and one minor isoform (c); a was the most basic of the three isoforms. Two-dimensional chymotryptic peptide maps derived from each individual isoform were virtually identical. The charge differences between the isoforms were due in part to differential H3(32)PO4 incorporation by the protein. CB48 bound to intact collagen type IV and the collagenous region of collagen type IV, but not to the globular NC1 domain. Cell-surface labelling and indirect immunofluorescence experiments localized the bulk of CB48 intracellularly in the endoplasmic reticulum Golgi region, with a minor population of molecules on the cell surface. A specific rabbit polyclonal anti-CB48 serum did not inhibit the attachment or spreading of BAEC to collagen type IV in an 'in vitro' adhesion assay, suggesting that the cell-surface population of CB48 is not involved in BAEC adhesion. We conclude that CB48 is a collagen-binding phosphoprotein that interacts with the collagenous domain of collagen type IV and may be involved in intracellular transport of collagen molecules.     

An antigenic HIV-1 peptide sequence engineered into the surface structure of transferrin does not elicit an antibody response.

One novel approach for the biological delivery of peptide drugs is to incorporate the sequence of the peptide into the structure of a natural transport protein such as human serum transferrin (HST). However, a potential drawback is that the HST may increase the immunoreactivity of the peptide, in the same way that carrier proteins can be used to generate highly immunogenic peptide hapten conjugates. In this study we have generated a recombinant HST carrier protein that contains a peptide substrate of HIV-1 protease (VSQNYPIVL). The protein retained native HST function, and the peptide was surface exposed since it was immunoreactive in native dot blots, and was cleaved by HIV-1 protease. Immunisation of rabbits with the recombinant protein elicited only a very poor anti-peptide immune response. In contrast, strong anti-peptide immune responses were raised against both the peptide alone, and a chemical conjugate of the peptide with HST. These data demonstrate that it is possible to attenuate the immune response normally directed against an immunogenic peptide sequence by engineering into a surface exposed loop of HST. These findings may have an important impact on the future design of peptide delivery systems.     

T cell receptor junctional regions and the MHC molecule affect the recognition of antigenic peptides by T cell clones.

Nine independent pigeon cytochrome c-specific T cell clones were analyzed by using a panel of antigenic peptide analogs presented in association with three allelic IE-encoded MHC glycoproteins. Eight of the T cell clones expressed a TCR composed of a unique alpha- and beta-chain amino acid sequence, and concordantly, each of these T cell clones exhibited a unique Ag specificity. This was true for several clones which differed only in TCR V-J junctional regions. Interestingly, for a given clone, the response to some of the peptide analogs depended to a large extent on the allelic form of the presenting MHC molecule. A simple interpretation of these data would suggest that certain positions of the peptide Ag are most important for Ag-MHC molecule interactions, and that these specific interactions can influence the antigenic epitope recognized by the TCR. We suggest that an antigenic peptide binds to an MHC glycoprotein in a distinct way, but may retain a measure of flexibility.     

Inhibition of an allospecific T cell hybridoma by soluble class I proteins and peptides: estimation of the affinity of a T cell receptor for MHC

To investigate the molecular basis of the interaction between the T cell receptor and the MHC class I antigen in an allogeneic response, a soluble counterpart of the murine class I molecule, H-2Kb, was genetically engineered. Cells secreting this soluble molecule, H-2Kb/Q10b, inhibited stimulation of an H-2Kb-reactive T cell hybridoma by cells transfected with H-2Kbm10, a weak stimulus, but not by H-2Kb- or H-2Kbm6-transfected cells. Soluble purified H-2Kb/Q10b protein also blocked T cell stimulation. In addition, a peptide from the wild-type H-2Kb molecule spanning the region of the bm10 mutation specifically inhibited activation of the T cell hybridoma by H-2Kbm10 cells, thus suggesting that amino acid residues 163-174 of H-2Kb define a region important for T cell receptor binding. An estimate for the Kd of the T cell receptor for soluble H-2Kb/Q10b was 10(-7) M, while the Kd for soluble peptide 163-174 was 10(-4) M.     

TCR beta chain influences but does not solely control autoreactivity of V alpha 14J281T cells.

CD1d-dependent accumulation of alphabeta T cells bearing a canonical Valpha14Jalpha281 alpha-chain (Valpha14+ T cells) is thought to model positive selection of lipid-specific T cells, based on their ability to recognize CD1d-presented self glycolipid(s). However, it has been difficult to demonstrate self ligand specificity in this system, as most Valpha14+ T cells do not exhibit significant autoreactivity despite high reactivity to alpha-galactosylceramide presented by CD1d (alpha-GalCer/CD1d). To assess the role of TCRbeta chain in determining the alpha-GalCer/CD1d vs autoreactive specificity of Valpha14+ T cells, we conducted TCRalpha or TCRbeta chain transduction experiments. In this study we demonstrate, by combining different TCRbeta chains with the Valpha14 alpha-chain in retrovirally transduced T cell lines, that the Valpha14 alpha-chain plays a primary role, necessary but not sufficient for imparting alpha-GalCer/CD1d recognition. beta-Chain usage alone is not the sole factor that controls the extent of autoreactivity in Valpha14+ T cells, since transduction of TCRalphabeta chains from a high CD1d autoreactive Valpha14+ T cell line conferred the alpha-GalCer/CD1d specificity without induction of autoreactivity. Thus, heterogeneity of Valpha14+ T cell reactivity is due to both beta-chain diversity and control mechanism(s) beyond primary TCR structure.     

Adenovirus type 5 fiber knob binds to MHC class I alpha2 domain at the surface of human epithelial and B lymphoblastoid cells.

Adenovirus serotype 5 (Ad5) fiber receptor was investigated using reverse antibody biopanning of a phage-displayed hexapeptide library, and virus-neutralizing monoclonal antibodies (mAbs 1D6.3 and 7A2.7) raised against recombinant Ad5 fiber knob. Both mAbs inhibited attachment of Ad5 to HeLa cells. Mimotopes of 1D6.3 showed homology with the C-terminal segment of the alpha2 domain of the heavy chain of human MHC class I molecules (MHC-I alpha2), and mimotopes of 7A2.7 were consensus to human fibronectin type III (FNIII) modules. In vitro, GST-fused MHC-I alpha2- and FNIII-derived oligopeptides interacted with recombinant fibers in a subgroup-specific manner. In vivo, the MHC-I alpha2 synthetic icosapeptide RAIVGFRVQWLRRYFVNGSR showed a net neutralization effect on Ad5 in HeLa cells, whereas the FNIII icosapeptide RHILWTPANTPAMGYLARVS significantly increased Ad5 binding to HeLa cells. Daudi cells, which lack surface expression of HLA class I molecules, showed a weak capacity for Ad5 binding. In beta2-microglobulin-transfected Daudi cells, Ad5 attachment and permissivity were restored to HeLa cell levels, with 4000 receptors per cell and a binding constant of 1.4x10(10)/M. The results suggested that the conserved region of MHC-I alpha2-domain including Trp167 represents a high affinity receptor for Ad5 fiber knob, whereas ubiquitous FNIII modules would serve as auxiliary receptors.     

Cell surface sulfhydryls are required for the cytotoxicity of diphtheria toxin but not of ricin in Chinese hamster ovary cells.

A previous study on cleavage of disulfide bonds in endocytosed model compounds had shown that an initial phase of cleavage was totally inhibited by membrane-impermeant sulfhydryl inhibitors and thus was mediated by cell surface sulfhydryls (Feener, E. P., Shen, W.-C., and Ryser, H. J.-P. (1990) J. Biol. Chem. 265, 18780-18785). This paper uses the same inhibitors (5,5'-dithiobis(2-nitrobenzoic acid) and p-chloromercuriphenylsulfonic acid) to examine the role of surface sulfhydryls in the cytotoxicity of diphtheria toxin (DT). Since the interchain disulfide of endocytosed DT must be cleaved prior to translocation of chain A from endosomes to cytoplasm, it was postulated that surface sulfhydryls might mediate the cleavage of that disulfide bond as well. Both sulfhydryl blockers did indeed markedly inhibit DT cytotoxicity. This effect was not due to inactivation of unbound DT, inhibition of receptor-mediated endocytosis, or impairment of acidification of endosomes. We conclude that cell surface sulfhydryls susceptible to blockage by 5,5'-dithiobis(2-nitro-benzoic acid) and p-chloromercuriphenylsulfonic acid are required for the cytotoxicity of DT and, most likely, for the reductive cleavage of DT's interchain disulfides. Ricin cytotoxicity was not decreased; this is consistent with the view that ricin reaches the cytoplasm from a late endocytic structure and with the finding that endocytosed disulfides are also cleaved in a cell fraction containing elements of the Golgi apparatus (Feener, E. P., Shen, W.-C., and Ryser, H. J.-P. (1990) J. Biol. Chem. 265, 18780-18785).     

An extensive region of an MHC class I alpha 2 domain loop influences interaction with the assembly complex.

Presentation of antigenic peptides to CTLs at the cell surface first requires assembly of MHC class I with peptide and beta 2-microglobulin in the endoplasmic reticulum. This process involves an assembly complex of several proteins, including TAP, tapasin, and calreticulin, all of which associate specifically with the beta 2-microglobulin-assembled, open form of the class I heavy chain. To better comprehend at a molecular level the regulation of class I assembly, we have assessed the influence of multiple individual amino acid substitutions in the MHC class I alpha 2 domain on interaction with TAP, tapasin, and calreticulin. In this report, we present evidence indicating that many residues surrounding position 134 in H-2Ld influence interaction with assembly complex components. Most mutations decreased association, but one (LdK131D) strongly increased it. The Ld mutants, with the exception of LdK131D, exhibited characteristics suggesting suboptimal intracellular peptide loading, similar to the phenotype of Ld expressed in a tapasin-deficient cell line. Notably, K131D was less peptide inducible than wild-type Ld, which is consistent with its unusually strong association with the endoplasmic reticulum assembly complex.