Peptide length preferences for rat and mouse MHC class I molecules using random peptide libraries

MHC class I molecules bind short peptides for presentation to CD8⁺ T cells. The determination of the three-dimensional structure of various MHC class I complexes has revealed that both ends of the peptide binding site are composed of polar residues conserved among all human and murine MHC class I sequences, which act to lock the ends of the peptide into the groove. In the rat, however, differences in these important residues occur, suggesting the possibility that certain rat MHC class I molecules may be able to bind and present longer peptides. Here we have studied the peptide length preferences of two rat MHC class I a molecules expressed in the TAP2-deficient mouse cell line RMA-S: RT1-A1^c, which carries unusual key residues at both ends of the groove, and RT1.A^a which carries the canonical residues. Temperature-dependent peptide stabilization assays were performed using synthetic random peptide libraries of different lengths (7 – 15 amino acids) and successful stabilization was determined by FACS analysis. Results for two naturally expressed mouse MHC class I molecules revealed different length preferences (H2-K^b, 8 – 13-mer and H2-D^b, 9 – 15-mer peptides). The rat MHC class Ia molecule, RT1-A^a, revealed a preference for 9 – 15-mer peptides, whereas RT1-A1^c showed a more stringent preference for 9 – 12-mer peptides, thereby ruling out the hypothesis that unusual residues in rat MHC molecules allow binding of longer peptides.     

Extensive major histocompatibility complex class I binding promiscuity for Mycobacterium tuberculosis TB10.4 peptides and immune dominance of human leucocyte antigen (HLA)‐B*0702 and HLA‐B*0801 alleles in TB10.4 CD8+ T‐cell responses

The molecular definition of major histocompatibility complex (MHC) class I‐presented CD8⁺ T‐cell epitopes from clinically relevant Mycobacterium tuberculosis (Mtb) target proteins will aid in the rational design of T‐cell‐based diagnostics of tuberculosis (TB) and the measurement of TB vaccine‐take. We used an epitope discovery system, based on recombinant MHC class I molecules that cover the most frequent Caucasian alleles [human leucocyte antigen (HLA)‐A*0101, A*0201, A*0301, A*1101, A*2402, B*0702, B*0801 and B*1501], to identify MHC class I‐binding peptides from overlapping 9‐mer peptides representing the Mtb protein TB10.4. A total of 33 MHC class I‐binding epitopes were identified, spread across the entire amino acid sequence, with some clustering at the N‐ and C‐termini of the protein. Binding of individual peptides or closely related peptide species to different MHC class I alleles was frequently observed. For instance, the common motif of xIMYNYPAMx bound to six of eight alleles. Affinity (50% effective dose) and off‐rate (half life) analysis of candidate Mtb peptides will help to define the conditions for CD8⁺ T‐cell interaction with their nominal MHC class I‐peptide ligands. Subsequent construction of tetramers allowed us to confirm the recognition of some of the epitopes by CD8⁺ T cells from patients with active pulmonary TB. HLA‐B alleles served as the dominant MHC class I restricting molecules for anti‐Mtb TB10.4‐specific CD8⁺ T‐cell responses measured in CD8⁺ T cells from patients with pulmonary TB.     

Identification of naturally processed peptides bound to the class I MHC molecule H-2Kd of normal and TAP-deficient cells

This report details the biochemical features of natural peptides selected by the H-2Kd class I MHC molecule. In normal cell lines, the length of the naturally processed peptides ranged from 8 to 18 amino acids, although the majority were 9-mers (16% were longer than nine residues). The binding motif for the 9-mer peptides was dominated by the presence of a tyrosine at P2 and an isoleucine/leucine at the P9 position. The P2 residue contributed most towards binding; and the short peptides bound better and formed longer-lived cell surface complexes than the long peptides, which bound poorly and dissociated rapidly. The longer peptides did not exhibit this strictly defined motif. Trimming the long peptides to their shorter forms did not enhance binding and conversely, extending the 9-mer peptides did not decrease binding. The long peptides were present on the cell-surface bound to H-2Kd (Kd) and were not intermediate products of the class I MHC processing pathway. Finally, in two different TAP-deficient cells the long peptides were the dominant species, which suggested that TAP-independent pathways selected for long peptides by class I MHC molecules.     

The interaction of beta 2-microglobulin (β2m) with mouse class I major histocompatibility antigens and its ability to support peptide binding. A comparison of human and mouse β2m

The function of major histocompatibility complex (MHC) class I molecules is to sample peptides derived from intracellular proteins and to present these peptides to CD8⁺ cytotoxic T lymphocytes. In this paper, biochemical assays addressing MHC class I binding of both peptide and β₂-microglobulin (β₂m) have been used to examine the assembly of the trimolecular MHC class I/β₂m/peptide complex. Recombinant human β₂m and mouse β₂m² have been generated to compare the binding of the two β₂m to mouse class I. It is frequently assumed that human β₂m binds to mouse class I heavy chain with a much higher affinity than mouse β₂m itself. We find that human β₂m only binds to mouse class I heavy chain with slightly (about 3-fold) higher affinity than mouse β₂m. In addition, we compared the effect of the two β₂m upon peptide binding to mouse class I. The ability of human β₂m to support peptide binding correlated well with its ability to saturate mouse class I heavy chains. Surprisingly, mouse β₂m only facilitated peptide binding when mouse β₂m was used in excess (about 20-fold) of what was needed to saturate the class I heavy chains. The inefficiency of mouse β₂m to support peptide binding could not be attributed to a reduced affinity of mouse β₂m/MHC class I complexes for peptides or to a reduction in the fraction of mouse β₂m/MHC class I molecules participating in peptide binding. We have previously shown that only a minor fraction of class I molecules are involved in peptide binding, whereas most of class I molecules are involved in β₂m binding. We propose that mouse β₂m interacts with the minor peptide binding (i.e. the “empty”) fraction with a lower affinity than human β₂m does, whereas mouse and human β₂m interact with the major peptide-occupied fraction with almost similar affinities. This would explain why mouse β₂m is less efficient than human β₂m in generating the peptide binding moiety, and identifies the empty MHC class I heavy chain as the molecule that binds human β₂m preferentially.     

MUC1 peptide epitopes associated with five different H-2 class I molecules

We have previously described the induction of murine CD8⁺ major histocompatibility complex (MHC) class I-restricted cytotoxic T cells (CTL) recognizing the 20-amino acid repeat region of the human mucin 1 (MUC1) variable number of tandem repeats region (VNTR), a mucin greatly increased in expression in breast cancer and proposed as a target for immunotherapy. In that study, CTL could detect MUC1 peptides associated with the MHC of all nine strains examined, and we now report the different epitopes presented by five different MHC class I molecules. The epitopes were defined in CTL assays using peptide-pulsed phytohemagglutinin blasts or MHC class I-transfected L cells as targets; in addition, peptide binding assays and T cell proliferation studies were performed. Within the 20-amino acid VNTR, nine potential epitopes could be defined. The epitopes for the four MHC class I molecules [K^b (three epitopes), D^d, L^d and K^k] were closely related, all containing the amino acids PDTRPAP. For D^b, three epitopes were identified, all containing APGSTAP. Most of the epitopes did not contain a consensus motif for the particular MHC class I allele, and bound with low ‘affinity’, compared with known high-affinity peptides. CD8⁺ T cell proliferation also occurred to the same MHC class I-presented epitopes. Finally, when conventional anchor residues were introduced into the peptides, peptide binding increased, whereas CTL recognition was either retained (K^b) or lost (D^b) depending on the epitope.     

Qa-1 interaction and T cell recognition of the Qa-1 determinant modifier peptide

The peptide-binding properties of the nonclassical major histocompatibility complex (MHC) class 1b molecule Qa-1 were investigated using a transfected hybrid molecule composed of the α1 and α2 domains of Qa-1^b and the α3 domain of H-2D^b. This allowed the use of a monoclonal antibody directed against H-2D^b whilst retaining the peptide-binding groove of Qa-1^b. By comparison with classical MHC class I molecules, intracellular maturation of the chimeric molecule was inefficient with weak intracellular association with β2-microglobulin. However, at the cell surface the hybrid molecules were stably associated with β2-microglobulin and were recognized by cytotoxic T lymphocyte (CTL) clones specific for the Qa-1^b -presented peptide Qdm (AMAPRTLLL). A whole-cell binding assay was used to determine which residues of Qdm were important for binding to Qa-1^b and CTL clones served to identify residues important for T cell recognition. Substitutions at position 1 and 5 did not reduce the efficiency of binding and had little effect on CTL recognition. In contrast, substitutions at position 9 resulted in loss of MHC class I binding. Mass spectrometric analysis of peptides eluted from immunopurified Qa-1^b/D^b molecules indicated that Qdm was the dominant peptide. The closely related peptide, AMVPRTLLL, which is derived from the signal sequence of H-2D^k, was also present, although it was considerably less abundant. The mass profile suggested the presence of additional peptides the majority of which consisted of eight to ten amino acid residues. Finally, the finding that a peptide derived from Klebsiella pneumoniae can bind raises the possibility that this non-classical MHC class I molecule may play a role in the presentation of peptides of microorganisms.     

The structural basis of MHC control of collagen-induced arthritis; binding of the immunodominant type II collagen 256 – 270 glycopeptide to H-2Aq and H-2Ap molecules

The A^q major histocompatibility complex (MHC) class II molecule is associated with susceptibility to murine collagen-induced arthritis (CIA), whereas the closely related H-2A^p molecule is not. To understand the molecular basis for this difference, we have analyzed the ability of H-2A^q and H-2A^p molecules (referred to as A^q and A^p) to bind and present collagen type II (CII)-derived glycosylated and non-glycosylated peptides. T cell clones specific for the immunodominant CII 256 – 270 peptide and restricted to both A^q and A^p molecules were identified. When these clones were incubated with CII protein and either A^q- or A^p-expressing antigen-presenting cells (APC), only A^q-expressing APC were able to induce stimulation. With the use of A_β transgenic mice this could be shown to be solely dependent on the MHC class II molecule itself and to be independent of other MHC- or non-MHC genes. Peptide binding studies were performed using affinity-purified MHC class II molecules. The CII 256 – 270 peptide bound with lower affinity to the A^p molecule than to the A^q molecule. Using a set of alanine-substituted CII 256 – 270 peptides, MHC class II and T cell receptor (TCR) contacts were identified. Mainly the side chains of isoleucine 260 and phenylalanine 263 were used for binding both the A^q and A^p molecule, i. e. the peptide was orientated similarly in the binding clefts. The major TCR contact amino acids were lysine 264, which can be posttranslotionally modified, and glutamic acid 266, which is the only amino acid in the heterologous peptide which differs from the mouse sequence. Glycosylation at positions 264 and 270 of the CII 256 – 270 peptide did not change the anchor positions used for binding to the A^q or A^p molecules. The autologous form of the peptide (with aspartic acid at position 266) bound with lower affinity to the A^q molecule as compared with the heterologous peptide. The variable affinity displayed by the immunodominant CII 256 – 270 peptide for different MHC class II molecules, the identification of MHC and TCR contacts and the significance of glycosylation of these have important implications for the understanding of the molecular basis for inherited MHC class II-associated susceptibility to CIA and in turn, for development of novel treatment strategies in this disease.     

Translocation of long peptides by transporters associated with antigen processing (TAP)

The major histocompatibility complex (MHC)-encoded transporters associated with antigen processing (TAP) translocate peptides from the cytosol into the lumen of the endoplasmic reticulum (ER) where they associate with MHC class I molecules. The length of class I-binding peptides is usually 8–11 amino acids, but examples of significantly longer peptides have been described. The preferred lengths and upper and lower size limits for peptides translocated by TAP have not been determined in detail because in the currently used test systems, peptides are subject to proteolytic degradation. In the present study, three sets of individual peptides or partially randomized peptide libraries ranging between 6 and 40 residues were used that contained a radiolabeled tyrosine and a consensus sequence for ER-specific N-glycosylation at opposite ends, thus ensuring that only nondegraded peptides were monitored in the transport/glycosylation assay. For three different transporters, rat TAP1/2^a, rat TAP1/2^u and hTAP, the most efficient ATP-dependent transport was observed for peptides with 8–12 amino acids. Hexamers and longer peptides of up to 40 amino acids were also translocated, albeit less efficiently. For two of the three sets of peptides analyzed, rat TAP1/2^a showed a less stringent length selection than rat TAP1/2^u and human TAP. The superior transport of the decamer of the TNKT . Y series was not due to faster degradation or less efficient glycosylation of shorter or longer length variants. A binding assay with TAP-containing microsomes revealed a high affinity for the radiolabeled decamer (K_D = 580 nM), while other length variants were clearly inferior in their binding affinities. Thus, TAP binds and preferentially translocates peptides with a length suitable for binding to MHC class I molecules, but peptides that are considerably longer may also be substrates. About 10⁵ peptide binding sites per cell equivalent of microsomes were determined, providing an estimate for the number of TAP complexes in the ER membrane.     

Calnexin expression does not enhance the generation of MHC class I-peptide complexes

We investigated the requirement for calnexin in the biogenesis of MHC class I molecules. Mutant human cells lacking calnexin were infected with recombinant vaccinia viruses encoding mouse MHC class I molecules, K ^d , K^b , K^k , D ^d , D^b , and L^d . Flow cytometry indicated that each of the six MHC class I allomorphs was transported to the cell surface at similar rates in calnexin-deficient cells and transfectants expressing calnexin. For K^b and K ^d , the calnexin-independent biogenesis occurred regardless of whether the MHC class I molecules contained human or mouse β2-microglobulin. Also addressed was the effect of calnexin on the surface expression of K^b molecules bearing the immunodominant peptide from ovalbumin (OVA_{257 – 264} ). This was detected with a recently described monoclonal antibody specific for the K^b/peptide complex. Calnexin expression had no significant effect on the formation of K^b /peptide complexes generated from full-length OVA, cytosolic OVA_{257 – 264} , or endoplasmic reticulum-targeted OVA_{257 – 264} , which was expressed in the presence of the herpes simplex virus ICP47 protein to ensure detection of TAP-independent peptide-MHC class I complexes. Complementary results were obtained with TAP-independent formation of K ^d /peptide complexes. These findings indicate that calnexin is not required for the efficient assembly of MHC class I molecules with TAP-dependent or independent peptides.     

Binding of radioiodinated influenza virus peptides to class I MHC molecules and to other cellular proteins as analyzed by gel filtration and photoaffinity labeling

In order to determine how T cell-presented peptides associate with the antigen binding sites (desetopes) of class I major histocompatibility complex (MHC) molecules and how they might be scavenged from an endogenous processing pathway for transfer to those molecules, we characterized the binding of two synthetic peptides restricted by HLA-B37 or HLA-A2 to class I MHC molecules and to cellular proteins of histotyped cell lines, by gel filtration and photo-affinity labeling techniques. In gel filtration binding studies, each peptide associated with immunopurified class I MHC molecules from cells with its restricting, histotype, but little was bound to class I MHC molecules from cells without the restricting histotype and none was bound to bovine serum albumin. After crosslinkage of a radioiodinated photoreactive derivative of influenza virus nucleoprotein peptide NP(336-355Y) and immunoprecipitations with antibodies to class I MHC molecules, that peptide was found to bind to immunopurified class I MHC molecules from HLA-B37+ but not HLA-B37- cells. Binding of the [125I]NP peptide increased from 6 to 12 hr of incubation and was competed by unlabeled, NP peptide but not by HLA-A2-restricted, influenza virus matrix MA(57-73). The principal microsomal membrane proteins binding [125I]NP were about 65, 45 and 33 kD.     

Unconventional cytotoxic T lymphocyte recognition of synthetic peptides corresponding to residues 1—23 of Ras protein

A peptide corresponding to amino acids 1 through 23 of Ras protein containing a mutation at position 12 was used to induce cytotoxic T lymphocytes (CTL) in mice. Although the CTL were CD8⁺ and expressed α, β T cell antigen receptors (TCR), their major histocompatibility complex (MHC)-restriction was unconventional. They recognized peptide-treated murine cells of different H-2 haplotypes, but not MHC class I-negative cells. Human HLA class I molecules did not present Ras peptides and hybrid human/mouse MHC molecules revealed that all three extracellular domains α1, α2 and α3 were required for recognition by peptide-specific CTL. Shortening the 23-mer peptide by 5 residues at either the amino or carboxy terminus resulted in loss of CTL recognition. This demonstrates an unusual form of antigen recognition by mouse CTL in which peptide presentation requires murine H-2 class I molecules but is not class I allele restricted, and the peptides recognized are much larger than peptides in conventional class I-restricted responses.     

Efficient MHC class I-peptide binding is required but does not ensure MHC class I-restricted immunogenicity

Cytotoxic T lymphocyte (CTL) epitopes are naturally processed peptides bound and presented by major histocompatibility (MHC) class I molecules. Since they are expressed at the cell surface in sufficient amounts to be recognized by CTL, it is generally believed, and in some cases demonstrated, that they bind efficiently to MHC class I molecules in vivo. Based on this knowledge, candidate CTL epitopes are now searched for by identifying peptides that efficiently bind to MHC class I molecules in vitro. We analysed whether this approach is valid by systematically investigating the relationship between MHC class I-peptide binding and peptide immunogenicity. Fifteen peptides that represent known CTL epitopes were tested for their MHC class I binding ability. In a comparative study with 83 peptides that bear the appropriate MHC class I allele-specific motifs but do not contain known CTL epitopes, the CTL epitope-bearing peptides showed the highest binding affinity for MHC class I. This was true for two MHC class I alleles in two different assay systems that monitor peptide-MHC class I binding. Furthermore, selected motif-bearing K^b binding peptides were used to induce peptide-specific CTL responses in mice. Only a subset of the high affinity K^b binding peptides induced reproducible peptide-specific CTL responses, whereas none of the low affinity K^b binding peptides induced a response. Taken together, these results indicate that efficient peptide-MHC class I binding is required for immunogenicity. Vice versa, immunogenicity is not guaranteed by efficient peptide-MHC class I binding, implying that additional factors are involved. Nevertheless, selection of candidate CTL epitopes on the basis of MHC class I binding seems valid. Our data indicate that, although an excess of peptides might be selected, the chance of missing immunogenic peptides is minimal.     

The assembly of H2-Kb class I molecules translated in vitro requires oxidized glutathione and peptide

Association of the mouse major histocompatibility complex (MHC) class I heavy chain H2-K^b with mouse β₂-microglobulin (β₂m) was studied in an in vitro translation system. Formation of stable class I complexes was found to be dependent on the presence of presentable peptides and oxidized glutathione, which promotes the formation of disulfide bridges. Translocation of peptides into microsomes was demonstrated by showing that a radioiodinated peptide containing an N-glycosylation acceptor site became glycosylated. Class I complex formation was observed only when heavy chains and β₂m were translated simultaneously, and thus occurs in the microsomes and not after their solubilization. However, peptide binding takes place only after solubilization of the microsomes. The class I complexes translated in vitro show the same specificity and length preference for peptides as their counterparts in RMA-S cells. Assembly of in vitro translated class I complexes was found to occur also in the absence of peptides, resulting in the formation of unstable molecules that are stabilized by incubation with peptides.     

Cytotoxic T lymphocyte responses to wild-type and mutant mouse p53 peptides

Cytotoxic T lymphocytes (CTL) recognize peptides presented at the cell surface in association with major histocompatibility complex (MHC) class I molecules. The finding that peptides binding to MHC class I molecules share common amino acid motifs renders feasible the selection of antigenic peptides by simply scanning protein sequences, and thus, provides the possibility of inducing CTL to pre-defined specificities. Tumor cells possess antigens known to generate MHC class I-restricted CD8⁺ CTL responses. Thus, these antigens represent good targets to induce tumor-specific immunity. Among these antigens, the p53 tumor suppressor gene product is an attractive candidate for cancer immunotherapy. Mutations in the p53 gene have been found to be very frequently associated with a malignant transformation and often lead to p53 protein overexpression. Thus, we investigated the possibility of inducing CTL to wild-type or mutant p53 peptides in a BALB/c (H-2^d) mouse model. Peptides possessing the H2-K^d binding motif were selected and tested for binding to the H-2K^d molecules in vitro. Synthetic peptides p53_122–130 wild-type or “mutant” (Lys → Glu substitution at position 129) were shown to be the best binder peptides and were tested for their immunogenicity in mice. H-2K^d-restricted p53-specific CD8⁺ CTL were generated following immunization of mice with either wild-type (wt) p53_122–130 or mutant (mut) p53_122–130 (E129) peptides. Only low-affinity CTL can be obtained by immunization with the wt sequence. In contrast, CTL elicited with the mut peptide recognized the mut sequence at a 10–100-fold lower concentration. This indicates that CTL elicited with the mut peptide recognized the mut sequence very efficiently, whereas the wt sequence is poorly recognized, if at all. Taken together, these results thus suggest that p53-specific tumor immunotherapy may be successful only if the mutated protein is taken into consideration.     

Direct analysis of peptide binding to cell-associated MHC class I molecules.

Exogenously added synthetic peptides can mimic endogenously produced antigenic peptides recognized on target cells by MHC class I-restricted cytolytic T lymphocytes. While it is assumed that exogenous peptides associate with class I molecules on the target cell surface, direct binding of peptides to cell-associated class I molecules has been difficult to demonstrate. Using a newly developed binding assay based on photoaffinity labeling, we have investigated the interaction of two antigenic peptides, known to be recognized in the context of H-2Kd or H-2Db, respectively, with 20 distinct class I alleles on living cells. None of the class I alleles tested, with the exception of H-2Kd or H-2Db, bound either of the peptides, thus demonstrating the exquisite specificity of peptide binding to class I molecules. Moreover, peptide binding to cell-associated H-2Kd was drastically reduced when metabolic energy, de novo protein synthesis or protein egress from the endoplasmic reticulum was inhibited. It is thus likely that exogenously added peptides do not associate with the bulk of class I molecules expressed at the cell surface, but rather bind to short-lived molecules devoid of endogenous peptides.     

Minute quantities of a single immunodominant foreign epitope are presented as large nested sets by major histocompatibility complex class II molecules

The processing and presentation of immunogenetic peptides is an obligate event in the generation of an immune response. However, the degree of complexity with which an immunogenic foreign epitope is presented is still unclear. This question was addressed by analyzing the naturally processed peptides generated from exogenously-derived hen egg white lysozyme (HEL) bound to the murine major histocompatibility complex (MHC) class II molecule, H-2A^k. Using reversed-phase chromatography (RPC), T cell hybridomas and mass spectrometry, 16 peptides were identified that contain the minimal MHC binding epitope 52–61. These peptides exhibited substantial N- and C-terminal extensions and ranged from 13–28 amino acids in length. In contrast, MHC class I molecules present peptides of 8–11 residues and each foreign epitope appears to be represented by only a single peptide. The data here also show that only ～ 0.8% of the total bound peptide was derived from this single HEL epitope. These findings provide direct evidence that relatively small amounts of processed peptide are required to stimulate an effective T cell response.     

Expression of β2m-Free HLA Class I Heavy Chains in Neuroblastoma Cell Lines

Flow cytometry with the specific monoclonal antibody (MoAb) L31 was used to analyse the expression of HLA class I heavy chains not bound with β₂-microglobulin (β₂m) by neuroblastoma (NB) cell lines IMR-32 and LA-N-1. The cells, which express barely detectable amounts of β₂m-free (L31-positive molecules) and β₂m-complexed HLA class I antigens (W6.32- and BBM. I-reactive molecules), expressed MHC class I molecules not bound to light chains upon differentiation with either retinoic acid or serum starvation. The expression was not accompanied by an increase of surface heterodimers. Conversely, recombinant interferon-γ (rIFN-γ) treatment led IMR-32 and LA-N-1 cells to almost exclusively express β₂m-complexed HLA class I heavy chains. Surface β₂m-free MHC class I molecules displayed a molecular mass of ~45 kDa and did not bind exogenously added β₂m. No changes in the synthesis of either HLA class I and β₂m mRNAs or of L31 proteins were observed in differentiated NB cells, thus suggesting that the surface exposure of unusual HLA class I antigens is regulated post-translationally. These findings indicate that, in addition to activated lymphocytes, the surface expression of β₂m-free class I heavy chains is a feature of other cell types, such as NB cells.     

Poor loading of major histocompatibility complex class II molecules with endogenously synthesized short peptides in the absence of invariant chain

In normal antigen-presenting cells, newly synthesized major histocompatibility complex (MHC) class II molecules associate with the invariant chain (Ii) glycoprotein in the endoplasmic reticulum (ER). They are loaded with peptides only after proteolytic removal of the Ii in post-Golgi endocytic vesicles. Since the Ii inhibits peptide binding to MHC class II molecules, this association could protect MHC class II molecules from being loaded with endogenous peptides early after biosynthesis. If this were an important function of the Ii in vivo, MHC class II molecules synthesized in cells lacking the Ii should be loaded efficiently with short endogenous peptides in the ER; such peptides are known to be present there due to TAP-mediated import from the cytosol. To examine this possibility, we have studied peptide loading in HeLa transfectants expressing murine H-2A^k MHC class II molecules either alone or together with an excess of Ii. Endogenous peptides could readily be extracted from conformationally intact A^k αβ dimers of biosynthetically labeled Ii⁺ cells, whereas peptide loading was greatly (> 95%) diminished in the absence of Ii. Significant amounts of sodium dodecyl sulfate-(SDS) stable 55-kDa peptide: A^k complexes were only found in the Ii⁺ transfectants. In the absence of Ii, the MHC class II molecules instead formed stable complexes with long (20 and 50 kDa) polypeptides. Known A^k-binding peptides bound stably to A^k molecules on Ii^? cells, could be co-purified with them, and were resistant to release in SDS, suggesting that poor recovery of endogenous peptides was not due to decreased stability of A^k: peptide complexes in the absence of Ii. We conclude that protection of MHC class II molecules from endogenous short peptides does not appear to be a quantitatively important function of the Ii molecule, because peptide loading is inefficient in its absence.     

Major histocompatibility complex class I-binding peptides are recycled to the cell surface after internalization

Cytotoxic T lymphocytes (CTL) recognize target antigens as short, processed peptides bound to major histocompatibility complex class I (MHC-I) heavy and light chains (β₂-microglobuhn; β₂ m).The heavy chain, which comprise the actual peptide binding α-1 and α-2 domains, can exist at the cell surface in different forms, either free, bound to β₂m or as a ternary complex with β₂m and peptides. MHC-I chains are also known to internalize, and recycle to the cell surface, and this has been suggested to be important in peptide presentation. Whether MHC-I-bound peptides also can recycle is not known. We have investigated this by using both peptide transporter mutant RMA-S cells and EL4 cells loaded with D^b-binding peptides, by two different approaches. First, peptides were covalently linked with galabiose (Galα4Gal) at a position which did not interfere with D^b binding or immunogenicity, and peptide recycling tested with Gal₂-specific monoclonal antibodies. By flow cytometry, a return of Gal₂ epitopes to the cell surface was found, after cellular internalization and cell surface clearance by pronase treatment. This peptide recycling could be discriminated from free fluid-phase uptake and was inhibited by methylamine, chloroquine and low temperature (18°C) but not by leupeptin. Second, specific CTL were reacted with peptide-loaded target cells after complete removal of surface D^b molecules by pronase, and after different times of incubation at 37C to allow reexpression. By this procedure, reappearance of target cell susceptibility was confirmed. The results are in agreement with a model for optimizing peptide presentation by recycling through an intracellular compartment similar to early endosomes in certain antigen-presenting cells.     

Influence of glycosylphosphatidylinositol-linked H-2Dd molecules on target cell protection and natural killer cell specificity in transgenic mice

The expression of certain major histocompatibility complex (MHC) class I ligands on target cells is one important determinate of their susceptibility to lysis by natural killer (NK) cells. NK cells express receptor molecules that bind to MHC class I. Upon binding to their MHC class I ligand, the NK cell is presumed to receive a signal through its receptor that inhibits lysis. It is unclear what role the MHC class I molecules of the effector and target cells play in signaling to the NK cell. We have investigated the role of the cytoplasmic and transmembrane domains of MHC class I molecules by producing a glycosylphosphatidylinositol (GPI)-linked H-2D^d molecule. The GPI-linked H-2D^d molecule is recognized by H-2D^d-specific antibodies and cytotoxic T lymphocytes. Expression of the GPI-linked H-2D^d molecule on H-2^b tumor cells resulted in protection of the tumor cells after transplantation into D8 mice (H-2^b, H-2D^d) from rejection by NK cells. In addition, NK cells from mice expressing the GPI-linked H-2D^d molecule as a transgene were able to kill nontransgenic H-2^b lymphoblast target cells. The GPI-linked MHC class I molecule was able to alter NK cell specificity at the target and effector cell levels. Thus, the expression of the cytoplasmic and transmembrane domains of MHC class I molecules are not necessary for protection and alteration of NK cell specificity.