Comparison of structural and functional variation in class I HLA molecules: the role of charged amino acid substitutions

Functional and structural heterogeneity of HLA class I molecules was sought among five donors serologically identical for A2, A3, B7, and Bw44. Functional differences were identified by cell-mediated lympholysis (CML) after allogeneic mixed lymphocyte reaction among the five donors. Structural differences were characterized by high resolution two-dimensional electrophoretic maps of the class I HLA proteins synthesized by peripheral blood lymphocytes of these donors. Cells from three donors showed no CML-defined differences from one another; their HLA protein maps were identical. The cells of one donor recognized an A3-associated target antigen on the cells of all the other donors; her HLA map revealed a unique protein with altered isoelectric point. Another donor's cells differed by two CML-detected antigens: one was identified as a variant of Bw44 ("44.2") and the other was associated with Cw4. This donor's two-dimensional HLA map showed two novel charged proteins. By using these data, a two-dimensional map locating HLA-A2, -A3, -A3', -B7, -Bw44.1, -Bw44.2, and -Cw4 was prepared. Because each of three CML-detected antigens was correlated with a protein of distinctive charge, our results and the available published data raise the possibility that amino acid substitution producing charge variation may be a particularly important mechanism in the generation of CML-detectable HLA diversity.     

The HLA system of the Uzbek population in the Ferghana Valley. HLA antigens, genes and haplotypes of the Uzbek population in relation to its ethnogenesis

Pecularities of distribution of 40 HLA antigens within the Uzbek population of the Ferghana Valley have been studied. The frequencies of these antigens are subdivided into three main groups having frequencies characteristic of the following populations: Caucasoid population (A9, B40, Bw22 etc.); Mongoloid population (B8, B7, B12); Middle Asian population (Aw31, B13, B16, Bw35), probably. The Uzbek population contains haplotypes both of Europeoid (Aw30, B13; A3, Bw35) and Mongoloid origin (A9, B40) and, probably, of the local origin (A1, B14; A1, Bw53; Aw32, B40). The data on the HLA genetics of the Uzbeks confirm the historical, linguistic and anthropological information concerning the role of inhabitants of the Central Asia in ethnogenesis and formation of the modern Uzbek population.     

Quantification of soluble HLA class I gene products by an enzyme linked immunosorbent assay

A simplified enzyme linked immunosorbent assay utilizing an HLA class I framework-specific monoclonal antibody and a polyclonal enzyme linked beta-2 microglobulin specific antiserum has been established for the quantitative measurement of soluble HLA class I molecules. A total of 219 unrelated healthy individuals and 137 members of 28 families typed for HLA were analyzed for their non-membrane bound, i.e. soluble HLA-A,B,C antigens (sHLA-A,B,C). As reported by others, we observed associations of higher or lower sHLA-A,B,C values to particular HLA antigens: High plasma values were observed in probands positive for HLA-A23, A24, A29, Aw33, Bw65, and Cw8 and low values in HLA-B27 and B37 positive individuals. However, as shown by family studies, levels of sHLA-A,B,C were apparently not controlled by the MHC haplotypes alone, since no significant difference between HLA identical siblings and two haplotype different individuals could be detected. Thus, additional non-MHC linked gene(s) may be involved in the release of class I gene products.     

An interesting case of defective HLA phenotype

In a 27 years old man whose parents had HLA phenotypes A 9/B 7, Bw 16 and Al/B 17, B 18 only HLA-B 18 could be reliably identified in lymphocytes by means of microlymphocytotoxic tests. It is probable that HLA-Bw 16 was also present in the lymphocytes of this test person. Even by applying the neutralisation test, the HLA antigens A 1 and a 9 could not be detected in the test person's serum. The cause for the existence of the defective HLA phenotype is discussed.     

Interesting relations in the HLA system

There exists no absolute binding between the antigens HLA-Cw 2, Cw 3 and Cw 4, on the one hand, and HLA-B 27, HLA-B 15 and HLA-Bw 35, on the other hand. Even if 91% of human beings with HLA Cw 4 will simultaneously have the antigen HLA-Bw 35, another antigen as HLA-B 27 or HLA-B 15 can be identified in approximately 55% of individuals with HLA-Cw 2 and Cw 3. In this connection, the joint presence of some pairs of cross-reacting HLA antigens (A 2 and A 28, B 5 and Bw 35, B 7 and B 27, B 8 and B 14, B 12 and Bw 2) could be proved and their frequency be determined. 2 cases of a simultaneous presence of two subtypes of HLA-A 10 antigen, A 25 and A 26, could be found in family examinations. Moreover, two atypical bindings of anti-HLA-Bw 4 and anti-HLA-Bw 6 cytotoxins with HLA antigens could be identified: 7,49% of HLA-Bw 35 positive lymphocytes no positive response with anti-HLA-B 4 and 1,69% of HLA-B 12 with anti-HLA Bw 6. The importance of the findings for determining HLA in practice is discussed.     

A public HLA antigen associated with HLA-A9, Aw32, and Bw4

The HLA complex codes for three distinct 44000 dalton molecules associated with beta2 microglobulin--HLA-A, B and C--each with its own multiallelic series of private antigens. The HLA-B molecule is exceptional in that it also carries a diallelic system, Bw4 and Bw6. One of these, Bw4, is often associated with the A-locus specificity A9. This finding has usually been ascribed to linkage disequilibrium between A- and B-locus antigens. We have shown, however, that an epitope called LHe is actually shared by A-locus and B-locus molecules. This epitope is found on all HLA-B molecules bearing the Bw4 determinant and is also found on all HLA-A molecules carrying the A9 (Aw23 and Aw24) or Aw32 specificities. We consider this a "public" HLA antigen; the possible molecular basis for both subtypic and public antigens on a single glycoprotein is discussed.     

KIR/HLA combination associated with the risk of complications in celiac disease

The pathogenesis of celiac disease (CD) is associated with polymorphisms in human leukocyte antigen (HLA) genes; however, compelling evidence suggests that additional non-HLA genes are associated with CD and related complications. The present study investigated whether killer cell immunoglobulin-like receptor (KIR)/HLA gene combinations are associated with CD and its clinical complications in the population of northeast Italy. The study included 61 adults affected by CD: 48 patients were at first diagnosis and 13 patients had CD-related complications (8 with refractory CD and 5 with cancer). Controls were 69 blood donors genotyped for KIR and HLA. Several statistically significant differences emerged between CD patients and blood donors. The results herein presented show that susceptibility to CD with refractory disease or cancer is associated with various genotypes including the 2DS2/2DL2+C1, 2DS3, 3DL1, and 2DL5B genes. In addition, the absence of the Bw4 ligand may be a predisposing factor for cancer. These results suggest that a KIR haplotype and HLA ligands may be involved in the susceptibility to important clinical CD complications such as tumors or refractoriness as a result of a gluten-free diet.     

Human leukocyte antigen profile in HIV-1 infected individuals and AIDS patients from Chongqing, China

HIV/AIDS is currently the leading cause of infectious disease mortality around the world. Since many alleles and/or haplotypes of HLA have been reported to be associated with progressive HIV infection, more detailed information on the HLA profile in HIV‐1 infected individuals in Chongqing, southwest China would facilitate further understanding of HIV‐1 infection, help AIDS vaccine design and the planning of effective preventive strategies. In this study, we performed 4‐digit resolution HLA‐A, B, DRB1 genotyping of 759 HIV‐1 seropositive individuals using PCR‐SSO methods. Six alleles were found at more than 10% high frequency: A*1101, A*0201, A*2402, B*4601, B*4001 and DRB1*0901. The most common 2‐ and 3‐locus haplotypes were A*0201‐B*4601, A*1101‐B*4001, A*1101‐B*4601, A*3303/1‐B*5801, A*0201‐B*4601‐DRB1*0901, A*1101‐B*4601‐DRB1*0901 and A*3303/1‐B*5801‐DRB1*0301. 690 HIV‐1 seropositive individuals with records of CD4 counts were divided into two groups: an AIDS patient group comprising 216 subjects with AIDS‐defining symptoms and CD4 counts below 200 cells/mm³ and an asymptomatic, HIV seropositive group of 474 subjects with a stable CD4 count of no less than 200 individuals. In the AIDS patient group, A*3303/1 and B*5801 alleles and the A*3303/1‐B*5801 haplotype were significantly underrepresented as compared to the HIV‐infected group, whereas A*1101‐B*4001, A*1101‐B*1502, A*2402‐B*4801 haplotypes and five common haplotypes from two groups were significantly overrepresented. HLA‐A or B and HLA‐Bw6‐Bw6 homozygotes were also overrepresented in the AIDS patients group. Our observations suggest that the presence of the B*3501 allele, A*2402‐B*4801, common 2‐locus and 3‐locus haplotypes, HLA‐A or B and Bw6‐Bw6 homozygosity may predict a poor disease outcome in HIV‐1 infection. However, HIV‐1 infected individuals who have B*5801 alleles, A*3303/1‐B*5801 haplotype and are heterozygous for Bw4‐Bw6 are more likely to be resistant to progression of AIDS in this Chinese population.     

Frequency of HLA-A*03 associates with HIV-1 infection in a Chinese cohort

During the early mid-1990s, a number of rural farmers across central China were employed to the unregulated plasmaselling-activity and many of them were infected by HIV-1. However, AIDS progression in the former blood donors (FBDs) is various. The aim of this study is to assess human leukocyte antigen (HLA) class I allele distribution in FBDs and evaluate its association with HIV-1 infection and disease progression. A total of 353 FBDs were enrolled in the cohort including 294 ART naïve HIV-1 seropositive and 59 HIV-1 seronegative age-matched subjects. The viral load and CD4/CD8 T cell counts were assessed in all subjects. Compared with HIV-seropositive group, the frequency of HLA-A*03 in control was significantly higher. After classifying the HLA-B alleles of the subjects according to the presence of Bw4/Bw6 serological epitopes, detrimental effect of HLA Bw6/ Bw6 homozygosity was also confirmed in the HIV-seropositive subjects. This study provides novel evidence on HLA class I allele distribution and association of HLA-A*03 frequency with HIV-1 infection and viremia in the HIV-1 infected FBDs, which may throw light on intervention strategy for the HIV-1 infection and our understanding how host immunity and genetic background affect HIV infection and AIDS progression.     

Immunofixation for C2 typing: C2 allotypes in Spaniards in relation to HLA,Bf and C4

Summary C2 typing is performed by immunofixation with anti-C2 antiserum instead of by a hemolytic overlay. This method gives sharp band definition, is less cumbersome than the hemolytic overlay, gel files are easily made, and it also enables one to describe putative new nonhemolytic variants. C2 allele frequencies were studied in a sample of the normal Spanish population and were found to be similar to other Caucasoids. HLA-Bw62,-Cw3, and-DR4 were significantly associated with C2 B. Concordantly, the only C2^*B extended HLA haplotype found in family material was Bw62-Cw3-Bw6-(DR4)-Bf^*S-C2^*B-C4A^*3 B^*2-(GLO^*1). C4A^*4 B^*2 and C4A^*4 B^*4 are not found within the same haplotype together with C2^*B and Bw62 or Bw22 respectively, nor do other C2^*B haplotypes occur with common HLA-B alleles. These results may favour the hypothesis that the Bw62-C2^*B haplotype is produced by one mutation arising in the Bw62-C2^*C haplotype and that subsequent crossovers can explain other C2^*B haplotypes (including Bw22-C2^*B).     

Comparison of one-dimensional IEF patterns for serologically detectable HLA-A and B allotypes

A new mouse monoclonal antibody (MoAb) 4E, which detects an epitope shared by HLA-B locus antigens, together with the MoAb W6/32, detecting a common HLA, B, C, determinant, and the MoAb 4B, detecting HLA-A2 and A28, were used to isolate HLA-A and -B antigens in sequential immunoprecipitation. The HLA antigens obtained from metabolically labeled cell extracts of B-lymphoblastoid cell lines or from phytohemagglutinin (PHA) activated peripheral blood lymphocytes were compared by one-dimensional isoelectric focusing (1D-IEF). The IEF banding patterns obtained with native HLA antigens segregated in a family with HLA. Neuraminidase treatment of isolated antigens reduced the number of bands to one or two, simplifying the analysis of characteristic patterns. Thus, we have cataloged IEF banding patterns for the majority of the serologically recognized HLA-A and -B allotypes obtained from 57 unrelated American Caucasians. While no HLA-A locus or HLA-B locus specific banding patterns were observed, the HLA-A antigens had, in general, slightly higher pl values than the HLA-B antigens. HLA-C antigens could not be detected in this assay system. The polymorphism detected by IEF banding patterns was as extensive as the serologically detected polymorphism identified by classical HLA serology. Moreover, variants for some HLA allotypes could be detected by this method. In addition to previously recognized A2 variants, new variants were identified for HLA-A1, A26, and Bw44. Each A1 and Bw44 variant was associated with particular haplotypes. The HLA-A2 antigens occurred on 43 HLA haplotypes in the unrelated Caucasian population. Only one of each A2 variants was identified in this population.     

KIR2DS1-positive NK cells mediate alloresponse against the C2 HLA-KIR ligand group in vitro

The inhibitory 2DL1 and activating 2DS1 killer Ig-like receptors (KIR) both have shared ligand specificity for codon sequences in the C2 group HLA-Cw Ags. In this study, we have investigated NK cell activation by allogeneic target cells expressing different combinations of the HLA-KIR ligand groups C1, C2, and Bw4. We demonstrate that fresh NK cells as well as IL-2-propagated NK cells from 2DS1-positive donors that are homozygous for the C1 ligand group are activated in vitro by B lymphoblastoid cell lines expressing the C2 group. This response is, in part, due to the absence of C1 group recognition mediated by the inhibitory receptor 2DL2/3. This "missing self" alloresponse to C2, however, is rarely observed in NK cells from donors lacking 2DS1. Even in presence of 2DS1, the NK alloresponse is dramatically reduced in donors that have C2 group as "self." Analysis of selected NK clones that express 2DS1 mRNA and lack mRNA for 2DL1 demonstrates that activation by the C2 ligand and mAb cross-linking of 2DS1 in these clones induces IFN-gamma. Furthermore, this C2 group-induced activation is inhibited by Abs to both HLA class I and the receptor. Collectively, these studies demonstrate that NK cells from 2DS1-positive donors are activated by target cells that express the C2 group as an alloantigen. This leads to increased IFN-gamma-positive fresh NK cells and induces NK allocytotoxicity in IL2-propagated polyclonal NK cells and NK clones. This study also provides support for the concept that incompatibility for the HLA-KIR ligand groups C1, C2, and Bw4 dominates NK alloactivation in vitro.     

Two subgroups of HLA Bw44 defined by cell-mediated lympholysis that differ in Bw44 expression on platelets and in patterns of genetic linkage disequilibrium

Possible immunogenic heterogeneity of the HLA-Bw44 antigen was investigated using cytotoxic T lymphocytes (CTL) generated between donors identical for HLA-A2,3,-B7,w44. Highly discriminatory CTL combinations were identified that defined two subgroups of Bw44, designated 44.1 and 44.2. Out of 47 Bw44-positive donors tested in a population study, 30 were lysed by the CTL defining 44.1, and 19 were lysed by the CTL defining 44.2. All Bw44 cells could be typed as either 44.1 or 44.2, except two Bw44-positive cells that were phenotypically homozygous for the serologically defined Bw44 antigen and were lysed by both CTL. No Bw44-negative donors (zero out of 37) expressed either 44.1 or 44.2, although cold target blocking was required to eliminate a contaminating reactivity of one CTL population on Bw35 and some Bw45 cells. CTL were also raised between responder/stimulator combinations mismatched for Bw44. These CTL lysed all Bw44-positive target cells, indicating a CML antigen shared by all Bw44 cells. But clear discrimination of the 44.1 and 44.2 subgroups was obtained when appropriate cold target blocking cells were added. All donors with 44.2 expressed high levels of serologically detectable Bw44 on their platelets, and all with 44.1 expressed low levels (p less than 0.005). Furthermore, population studies indicate that 44.1 is in positive linkage disequilibrium with HLA-A2 and possibly DR4, whereas 44.2 is in positive linkage disequilibrium with HLA-DR7 and possibly HLA-A23, -A26, and -A29. These data suggest the existence of two genetically and functionally different subgroups of Bw44 antigens.     

Polymorphic sites away from the Bw4 epitope that affect interaction of Bw4+ HLA-B with KIR3DL1

KIR3DL1 is a polymorphic, inhibitory NK cell receptor specific for the Bw4 epitope carried by subsets of HLA-A and HLA-B allotypes. The Bw4 epitope of HLA-B*5101 and HLA-B*1513 is determined by the NIALR sequence motif at positions 77, 80, 81, 82, and 83 in the alpha(1) helix. Mutation of these positions to the residues present in the alternative and nonfunctional Bw6 motif showed that the functional activity of the Bw4 epitopes of B*5101 and B*1513 is retained after substitution at positions 77, 80, and 81, but lost after substitution of position 83. Mutation of leucine to arginine at position 82 led to loss of function for B*5101 but not for B*1513. Further mutagenesis, in which B*1513 residues were replaced by their B*5101 counterparts, showed that polymorphisms in all three extracellular domains contribute to this functional difference. Prominent were positions 67 in the alpha(1) domain, 116 in the alpha(2) domain, and 194 in the alpha(3) domain. Lesser contributions were made by additional positions in the alpha(2) domain. These positions are not part of the Bw4 epitope and include residues shaping the B and F pockets that determine the sequence and conformation of the peptides bound by HLA class I molecules. This analysis shows how polymorphism at sites throughout the HLA class I molecule can influence the interaction of the Bw4 epitope with KIR3DL1. This influence is likely mediated by changes in the peptides bound, which alter the conformation of the Bw4 epitope.     

HLA antibody responses in HLA class I transgenic mice

In a previous report we described how cross-immunizations of pairs of transgenic mice expressing different HLA class I antigens led to the production of antibodies directed exclusively at polymorphic epitopes. This was ascribed to self-tolerance of HLA that prevents immune responses to monomorphic epitopes and focuses responses on polymorphic ones. In the present report we extend our findings and demonstrate that immunizations of class I transgenic mice with HLA transfected mouse fibrosarcoma as well as with human lymphoblastoid cells also preferentially yield antibodies to polymorphic epitopes. This was the case whether or not immunizations were carried out across locus barriers [e.g., Tg (HLA-A *0201) or Tg (HLA-Cw*0301) transgenic mice immunized with HLA-B27 transfectants] or within the same locus [e.g., Tg (HLA-B*1302) transgenic mice immunized with HLA-B27 transfectants or B27-expressing lympho-blastoid cell]. Use of an extended immunization protocol with four or more booster injections favored antibodies of IgG isotype with affinities high enough to lyse normal peripheral blood lymphocytes (PBLs) in complement-dependent cytotoxicity assays and to immunoprecipitate HLA antigens. The specificities covered by the monoclonal antibodies (mAbs) could be either broad or narrow, depending on the genetic distance of the HLA antigens or alleles involved. For instance, a Tg(HLA-B*1302) transgenic mouse immunized with B27 produced both broad B7/B27-specific antibodies, Bw4-specific antibodies, and one antibody reacting with all B alleles except B13 and with some C alleles. On the other hand, a Tg(HLA-B*1302) transgenic mouse immunized with Bw47 transfectants responded narrowly with an antibody to Bw60 and Bw47. Thus it appears that by choosing appropriate recipient mice and closely related or more distant HLA antigens, antibodies of a programmed specificity can be generated. Address correspondence and offprint requests to: U. Hämmerling.     

Associations between restriction fragment length polymorphisms detected with a probe for human 21-hydroxylase (21-OH) and two clinical forms of 21-OH deficiency

Summary DNAs from unrelated healthy individuals and unrelated individuals affected with 21-hydroxylase deficiency (congenital and late-onset adrenal hyperplasia) were digested with seven restriction enzymes and hybridized with a cDNA probe specific for human 21-hydroxylase genes. Associations were found between restriction fragments and the two forms of the disease: (i) The late onset form is associated with a double dose of a 14 kb fragment generated by Eco RI and with a triple dose of a 3.2 kb fragment generated by TaqI in patients with HLA B14 haplotypes; (ii) The classical congenital form is negatively associated with the 14 kb fragment and with a 3.7 kb fragment generated by TaqI in patients with HLA Bw47 haplotypes. A 3.2 kb TaqI fragment is negatively associated with the HLA B8 haplotypes. The other five enzymes tested give no polymorphisms or polymorphisms without correlation with the two forms of the disease.     

HLA genotypes and HLA-linked genetic markers in Italian patients with classical 21-hydroxylase deficiency

Summary HLA genotype and HLA-linked marker data for 40 unrelated patients from central Italy and 2 unrelated patients from Sardinia with congenital adrenal hyperplasia due to 21-hydroxylase deficiency (21-OH-def) were analyzed. The results confirm that the HLA-linked 21-OH-def gene is associated with several different HLA determinants and complete HLA haplotypes, although the only determinant with significantly increased frequency was the complement C2 allele C2B. The HLA antigens B8 and DR3 were found in significantly decreased frequencies. The haplotype A3, Cw6, Bw47, BfF, DR7, which is exceptionally rare in the general population but which has been found in many other 21-OH-def patients from diverse geographical origins, was also found in one of the Italian patients. This and other HLA haplotype associations found among the Italian patients may represent mutations that have occurred on HLA haplotypes with genetic linkage disequilibrium or, alternatively, may represent mutations that have not yet had time to become randomly associated with different HLA complex determinants. The marked negative associations with B8 and DR3 could, however, result from an interaction between the gene products of the HLA complex and the 21-OH-def phenotype.     

Rhesus macaque KIR bind human MHC class I with broad specificity and recognize HLA-C more effectively than HLA-A and HLA-B

Human killer cell immunoglobulin-like receptors (KIR) recognize A3/11, Bw4, C1, and C2 epitopes carried by mutually exclusive subsets of human leukocyte antigen (HLA)-A, -B, and -C allotypes. Chimpanzee and orangutan have counterparts to HLA-A, -B, and -C, and KIR that recognize the A3/11, Bw4, C1, and C2 epitopes, either individually or in combination. Because rhesus macaque has counterparts of HLA-A and -B, but not HLA-C, we expected that rhesus KIR would better recognize HLA-A and -B, than HLA-C. Comparison of the interactions of nine rhesus KIR3D with 95 HLA isoforms, showed the KIR have broad specificity for HLA-A, -B, and -C, but vary in avidity. Considering both the strength and breadth of reaction, HLA-C was the major target for rhesus KIR, followed by HLA-B, then HLA-A. Strong reactions with HLA-A were restricted to the minority of allotypes carrying the Bw4 epitope, whereas strong reactions with HLA-B partitioned between allotypes having and lacking Bw4. Contrasting to HLA-A and -B, every HLA-C allotype bound to the nine rhesus KIR. Sequence comparison of high- and low-binding HLA allotypes revealed the importance of polymorphism in the helix of the α₁ domain and the peptide-binding pockets. At peptide position 9, nonpolar residues favor binding to rhesus KIR, whereas charged residues do not. Contrary to expectation, rhesus KIR bind more effectively to HLA-C, than to HLA-A and -B. This property is consistent with major histocompatibility complex (MHC)-C having evolved in hominids to be a generally superior ligand for KIR than MHC-A and MHC-B.     

Susceptibility of major groups of Tamil Nadu to diseases: 1. Psoriasis vulgaris

Eighty-three patients with psoriasis vulgaris, living in Madurai, Tamil Nadu, India, were studied for HLA-A, HLA-B, HLA-C, HLA-DR and HLA-DQ antigen frequencies and compared with seventy-seven controls studied using the same batch of reagents. A highly significant increase of frequency of HLA-Bw57, a split of HLA-B 17, was found in the patients; Bw58, another split of B17, was absent. Relative risk was high for A1, B17, Bw57 and DR7 individuals; it was highest for Bw57. Frequencies of the haplotypesAl-Bw57 andDR7-DQw3 were also significantly higher in patients. Analysis of the HLA data based on ethnic differences identified as major groups revealed high relative risk for B17, Bw57 and DR7 only in major group III, a Western brachycephal Armenoid group, but not in major group II, a Mediterranean one thought to be an earlier settler of this region. Analysis of the data based on age and sex subgroups yielded interesting information. The age at onset of the disease in the total patient sample showed a bimodal distribution. The two sexes differed in their age-at-onset distributions: females showed a preponderance of early onset of the disease (< 30 years of age, 68%), while the majority of males had late onset (>30 years of age, 71%). HLA data for the early-onset patients indicate very high relative risk for B17, Bw57 and DR7. This suggests that psoriasis may be influenced by sex, and that the early-onset and late-onset forms of the disease may be of different aetiopathogenesis. These observations stress the importance of considering the ethnic origin or composition of samples, and age, sex and other parameters in HLA and disease association studies.     

The alpha 1/alpha 2 domains of class I HLA molecules confer resistance to natural killing

The expression of transfected HLA class I Ag has previously been shown to protect human target cells from NK-mediated conjugation and cytolysis. In this same system, transfected H-2 class I Ag fail to impart resistance to NK. In this study, we have mapped the portion of the HLA class I molecule involved in this protective effect by exploiting this HLA/H-2 dichotomy. Hybrid class I genes were produced by exon-shuffling between the HLA-B7 and H-2Dp genes, and transfected into the class I Ag-deficient B-lymphoblastoid cell line (B-LCL) C1R. Only those transfectants expressing class I Ag containing the alpha 1 and alpha 2 domains of the HLA molecule are protected from NK, suggesting the "protective epitope" is located within these domains. Since a glycosylation difference exists between HLA and H-2 class I Ag within these domains (i.e., at amino acid residue 176), the role of carbohydrate in the class I protective effect was examined. HLA-B7 mutant genes encoding proteins which either lack the normal carbohydrate addition site at amino acid residue 86 (B7M86-) or possess an additional site at residue 176 (B7M176+) were transfected into C1R. Transfectants expressing either mutant HLA-B7 Ag were protected from NK. Thus, carbohydrate is probably not integral to a class I "protective epitope." The potential for allelic variation in the ability of HLA class I Ag to protect C1R target cells from NK was examined in HLA-A2, A3, B7, and Bw58 transfectants. Although no significant variation exists among the HLA-A3, B7, and Bw58 alleles, HLA-A2 appears unable to protect. Comparison of amino acid sequences suggests a restricted number of residues which may be relevant to the protective effect.