Abstract
KIR-ligands are HLA molecules that can be grouped into 3 major categories based on the amino acid sequence determining the KIR-binding epitope in HLA-C and HLA-B alleles. Almost all HLA-C alleles are of the C1 or C2 group defined by single nucleotide polymorphisms (SNPs) found at amino acid positions 77 and 80. C1 is designated S77 and N80, while C2 is designated N77 and K80. Most HLA-B alleles can be classified as either Bw4 or Bw6, defined by SNPs located at amino acid position 77 and 80–83. Bw6 allele types which are invariant and consistently S77 and N80 are not KIR-ligands, in contrast to Bw4 which is a KIR-ligand and can have multiple amino acid sequences at these positions. Killer-immunoglobin receptors KIR2DL1, KIR2DL2 and KIR3DL1 bind KIR-ligands C1, C2 and Bw4 respectively, resulting in inhibition of NK cell mediated lysis. Recent transplant strategies based on KIR-ligand mismatch to predict NK cell alloreactivity have resulted in less relapse and better survival in patients with AML. Although allele level high-resolution HLA-typing is the gold standard for accurately determining KIR-ligand status, it is not performed at some centers and is cost prohibitive for retrospective cohorts. In these settings, KIR-ligand assignment is being extrapolated from serologic or low-resolution HLA data. This is only accurate in about 80% of donor/recipient pairs, because misclassifications based on the frequency of less common alleles can result in assignment to the opposite KIR-ligand group (Bw4 versus Bw6 or C1 versus C2). Our aim was to develop a high-throughput assay for determining KIR-ligand status which is accurate, inexpensive and rapid.
Pyrosequencing is a relatively new method for sequencing DNA and is especially useful in detecting SNPs when most of the sequence is already known. Sequencing is based on a single strand of biotinylated DNA which is used as a template for a sequencing primer specific to the region of interest. As nucleotides are added base by base from the sequencing primer, the pyrosequencing apparatus (PSQ MA 96) can detect which base is added and in what quantity. This information can be used to determine if the sample is homozygous or heterozygous. We hypothesized that pyrosequencing would be a viable alternative to high resolution HLA-typing for KIR-ligand status determination. It directly sequences the ligand epitopes, thus avoiding misclassifications encountered with low resolution HLA-typing alone. This high throughput system would be of particular interest in analysis of banked RNA and DNA tissue samples for retrospective cohorts when high resolution typing is not available. KIR-ligand status was assigned for 34 samples by testing with both pyrosequencing and high-resolution HLA-typing. Initially we found a discrepancy rate of 9% between the two methods. To investigate discrepant samples, PCR products from these reactions were sequenced. We concluded that the initial PCR primer set was designed over a polymorphic region which did not amplify all known HLA-B or HLA-C alleles. The PCR primers were redesigned and the samples retested by pyrosequencing, resulting in full concordance with high resolution HLA data. Pyrosequencing is a sensitive, specific, high-throughput and inexpensive screening technique to rapidly determine KIR-ligand status for evaluating potential alloreactive NK cell or transplant donors.
Author notes
Corresponding author