Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Feb 1.
Published in final edited form as: Hum Immunol. 2009 Oct 25;71(2):212. doi: 10.1016/j.humimm.2009.10.009

Activating KIR receptors 3DS1 and 2DS1 protect against developing the severe form of recurrent respiratory papillomatosis

Vincent R Bonagura 1,2,3, Zeying Du 5, Elham Ashouri 5, Lihui Luo 5, Lynda J Hatam 1, James A DeVoti 1, David W Rosenthal 1,2,3, Bettie M Steinberg 1,3,4, Allan L Abramson 1,4, David W Gjertson 5, Elaine F Reed 5, Raja Rajalingam 5
PMCID: PMC2815039  NIHMSID: NIHMS155366  PMID: 19861144

Abstract

The polymorphic killer-cell immunoglobulin-like receptors (KIR) control natural killer (NK) cell response against viral infection and tumor transformation. Here we investigated if select KIR genes are associated with recurrent respiratory papillomatosis (RRP), a rare disease of the larynx and upper airway caused by human papilloma viruses (HPV)-6/11. DNA from 66 RRP patients and 195 healthy controls were characterized for KIR and HLA gene polymorphism. Patients lacking activating KIR genes 3DS1 and 2DS1 were more common in severe RRP compared to mild-moderate RRP (78.8% vs. 48.5%, p=0.019). Further, patients carrying any of the known susceptible HLA-DRB1/DQB1 alleles were more frequently negative for KIR3DS1 (p=0.006), KIR2DS1 (p=0.003) or KIR2DS5 (p=0.004) compared to controls carrying any of these HLA allotypes. Nearly 80% of the severe patients were missing the protective HLA-DQB1*0602 allele as well as both KIR3DS1 and KIR2DS1 genes. Phenotyping of papilloma-infiltrating mononuclear-cells revealed an elevated numbers of NK cells and CD57+CD4+ T-cells in KIR3DS1−KIR2DS1− patients compared to patients carrying either one or both of these KIRs. Our data suggest that NK cells expressing activating receptors KIR3DS1 and KIR2DS1 may be necessary to trigger an effective early immune response against HPV-infected targets to establish resistance to RRP development.

Keywords: Killer cell immunoglobulin-like receptors, natural killer (NK) cells, recurrent respiratory papillomatosis, human papilloma viruses (HPV)-6 and 11, HLA association

Introduction

Recurrent respiratory papillomatosis (RRP) is a rare disease characterized by recurrent benign tumor formation within the larynx and upper airway, with an incidence in the United States of 4.3/105 children and 1.8/105 adults [1]. Although a benign disease, the outcome of RRP can be life threatening because of airway obstruction, distal airway spread, and malignant transformation. Patients with RRP often require repeated surgeries and close clinical surveillance. Almost all RRP lesions are caused by small non-enveloped, double-stranded DNA human papilloma viruses (HPV) types 6 and 11 that replicate within the nuclei of keratinocytes [2].

Impaired cell-mediated immunity is largely responsible for the development of RRP [3]. Overrepresentation of CD8+CD28− T-cells [3], increased expression of TH2-like cytokines in papillomas and by peripheral blood mononuclear cells responding to autologous papilloma tissue [4], and aberrant CD4+ T-cell responses to HPV-6 and -11 peptides [5], have been shown to contribute to the compromised cell-mediated immune responses made against HPV proteins in RRP patients. Select alleles of polymorphic human leukocyte antigen (HLA) class II genes (DRB1*0102, DRB1*0301, DQB1*0201 or DQB1*0202) were also shown to be the critical host genetic factors responsible for HPV-specific hyporesponsiveness [5, 6]. However, not everyone carrying these disease-susceptibility HLA alleles develops RRP following HPV infection, indicating that additional host susceptibility factors are important in the RRP development.

Discovery of HLA class I-specific killer-cell immunoglobulin-like receptors (KIR) led to a series of epidemiological studies that implicated KIR genes in the susceptibility or resistance to various human diseases [7]. KIRs are the key receptors of human natural killer (NK) cells [8], a subset of lymphocytes that trigger early innate immune responses against infection and tumors [9]. In addition to NK cells, CD8+ cytotoxic T-cells with a memory phenotype also express KIRs, and therefore KIRs can further modulate antigen-specific adaptive immune responses [10].

Fourteen KIR genes encoding receptors with either inhibitory (3DL1-3, 2DL1-3, 2DL5) or activating functions (3DS1, 2DS1-5), or both inhibitory and activating functions (2DL4), and two non-functional pseudogenes (2DP1, 3DP1) have been identified [8, 11, 12]. Inhibitory KIR receptors recognize distinct HLA class I molecules and trigger signals that stop NK cell function. KIR2DL2 and 2DL3 bind HLA-C allotypes containing an Asparagine80 (group C1 alleles), KIR2DL1 binds HLA-C allotypes with Lysine80 (group C2 alleles), KIR3DL1 binds HLA-B allotypes containing the Bw4 epitope, and KIR3DL2 binds HLA-A3/11. The ligands for activating KIRs have yet to be discovered. Studies in mice predict that activating KIRs may recognize cellular targets expressed on the surface of infected-cells that trigger NK cytolysis [13, 14]. Based on genetic epidemiological studies [15], in vitro activation [16], and weak binding of KIR-Fc proteins [17] or tetramers [18], three activating KIR receptors are thought to bind the same HLA class I ligands, as do their homologous inhibitory counterparts at low affinity. KIR3DS1 shares 97% sequence similarity with KIR3DL1 in the extracellular Ig-domains and is thought to bind the HLA-Bw4 ligand [15]. Similarly, KIR2DS1 (homologue of 2DL1) and 2DS2 (homologue of 2DL2) are thought to bind weakly to HLA-C2 and HLA-C1 respectively [17, 18].

KIR genes are tightly clustered at chromosome 19q13.4 [19]. The number and type of KIR genes show considerable variability between haplotypes, resulting in substantial differences in KIR gene content between individuals [20]. Since KIR genes at chromosome 19q13.4 and HLA genes at chromosome 6p21.3 are polymorphic, and independent segregation of these unlinked gene families produce a wide diversity in the number and type of KIR-HLA gene combinations inherited in individuals [21]. To test the possibility that certain KIR-HLA gene combinations confer risk for developing RRP and disease severity, we characterized both KIR and HLA genes in 66 RRP patients and, compared with those we previously characterized in 195 healthy controls [21].

Materials and Methods

Study subjects and index of RRP disease severity

Sixty six randomly selected RRP patients of varying disease severity were studied from a large RRP population that receive care at the Long Island Jewish Medical Center which is a major national referral center for this disease. The study population is not of local origin and includes most patients from throughout the United States. All patients enrolled signed informed consent approved by the Institutional Review Board of North Shore-Long Island Jewish Health System. Fifty-two males and 14 females, a male:females ratio characteristic of RRP [22] were studied. Twenty-six patients had juvenile onset RRP and 40 had adult onset RRP. Fifty patients were Caucasians (75.7%), 11 were African Americans (16.7%) and 5 were Hispanics (7.6%). All patients had either HPV type 6 or 11 infection, documented by HPV typing [23]. The demographics and the disease severity scores of the RRP are listed in table 1.

Table 1.

Demographics and the disease severity scores of patients with RRP.

Patients1 Number Juvenile Onset/
Adult Onset
Caucasian# All 55 18/36
Severe 16 4/12
Mild Moderate 30 8/21
Tracheal
Involvement* 		9 6/3
Black All 11 8/3
Severe 6 4/2
Mild Moderate 3 2/1
Tracheal
Involvement* 		2 2/0
Open in a new tab
*

Patients with tracheal involvement are by definition, a subset of the severe group.

#

Five Hispanics are included in the Caucasian group.

Clinical severity was determined by extent of disease at time of surgery and frequency of recurrence. At each surgery, the number of disease sites, the anatomic surface area of disease, and extent of luminal obstruction was documented to yield a composite score as described [24]. This composite score is divided by the number of days elapsed since the previous surgery, yielding a growth rate that is a measure of disease severity. The mean growth rate from multiple surgeries defined the overall severity score for an individual patient. An overall disease severity score of ≥ 0.06, or tracheal extension, was defined as severe. An overall severity score of < 0.06, and no tracheal extension was defined as mild-moderate disease [24–26]. This scoring system was used throughout all of our clinical studies and was developed using a much larger number of patients (n >150) [24]. Minimal variation in patients’ scores was observed over time, and using the mean score further improved the reliability of this variable [27]. Thirty three patients had severe disease and 33 others had mild-moderate disease.

One hundred and ninety-five unrelated Caucasian blood donors from throughout the United States were included in this study. All these samples were typed for both KIR and HLA class I genes [21], 133 were typed for HLA-DRB1 alleles [6], and 112 were typed for HLA-DQB1 alleles [6].

Determination of KIR-HLA gene combinations

Genomic DNA samples from all RRP patients were PCR-typed for presence and absence of 16 KIR genes using the multiplex KIR-SSO typing kit (Tepnel Lifecodes, Stamford, CT). This kit consists of (i) two PCR reaction mixtures for the amplification of KIR-exons 4, 5, 7, 8 and 9; (ii) a mixture of color coded microspheres (Luminex Corporation, Austin, TX) each conjugated with locus-specific oligonucleotide probes. Each DNA sample was PCR amplified and the PCR product hybridized with the probe-labeled microspheres according to the manufacturer. The hybridized plate was read in a Luminex instrument and analysis was performed according to the manufacturer. A set of 10 DNA standards comprising most commonly occurring KIR genotypes from the UCLA KIR exchange reference program were processed as standards along with patient samples, and the results were identical to the known typing results.

HLA-A, -B, and -C typing was performed by either sequence-specific primer-directed PCR amplification (PCR-SSP) or sequence-specific oligonucleotide (SSO) hybridization methods using commercial kits (One Lambda, Canoga Park, CA) as previously described [6, 21]. The KIR-binding HLA class I epitopes were predicted from the HLA typing results. If the HLA typing results were ambiguous, we PCR-SSP typed the KIR-binding HLA motifs as described [21]. The allele-level typing of HLA-DRB1 and DQB1 loci for all patients and most control subjects were performed using sequencing-based typing method as described [6]. All KIR3DS1-positive patients and controls were further typed for the unexpressed null allele 3DS1*049N as described [28]. Only those carrying a functional allele were considered as positive for the KIR3DS1 gene. The presence of KIR-HLA gene combination, encoding the following receptor-ligand pairs were analyzed in each individual: 2DL1+HLA-C2, 2DL2/3+HLA-C1, 3DL1+HLA-Bw4, 3DL2+HLA-A3/11, 2DS1+HLA-C2, 2DS2+HLA-C1, and 3DS1+Bw4.

Phenotyping of papilloma-infiltrating mononuclear cells

Papilloma biopsies were minced, filtered, and plated in tissue culture medium 1640, supplemented with 100 units/mL penicillin, 100 µg/mL streptomycin, 2 mM glutamine (Gibco, Gand Island, NY), 10% bovine serum (Hyclone Laboratories, Inc., Logan UT), and 25 units/mL IL-2 (Boehringer Mannheim, Germany). This medium was replaced every 3–4 days and cultures were harvested at day 14 and immunophenotyped as previously described [3]. The yield from each papilloma ranged from 2 × 104 to 8.6 × 108 (x̄ = 10.7 × 106 ± 2.06 × 106) mononuclear cells. The success rate in obtaining sufficient mononuclear cells for immunophenotyping from a given papilloma was 72%. At day 14, cell surface marker analysis was performed on mononuclear cells using combinations of FITC-, PE-, or PerCP-conjugated antibodies specific for CD4, CD8, CD14, CD19, CD28, CD45, CD56, CD57, TCRγ, TCRδ, TCRα, or TCRβ (Becton-Dickinson Immunocytometry Systems, San Jose, CA). Lymphocyte subsets were enumerated by flow cytometry by FACS analysis as described [29].

Data analysis and statistical methods

Statistics were performed using Stata Statistical Software: Release 10.1 (Stata Corporation, College Station, TX). The percentage of each KIR gene was determined by direct counting (number of individuals positive for the gene divided by the number of individuals tested per group × 100%). Differences between study groups in the distribution of each KIR gene and genotype profile were assessed using Fisher’s exact test (two-tailed), and p<0.05 was considered to be statistically significant. Lymphocyte subsets (expressed as percentages) were compared by separate variance t-tests using a Welch correction (GraphPad Software: San Diego, CA). When phenotyping was performed on multiple samples per patient over time, one sample was randomly selected by algorithm (an atmospheric noise randomness program at www.random.org) for analysis.

Results

Patients lacking the activating KIR genes 3DS1 and 2DS1 developed severe RRP

To determine if select KIR genes were associated with the risk of developing RRP, we PCR typed 16 known KIR genes in 66 RRP patients (Figure 1). No difference was observed in the frequency of any KIR gene between RRP patients and a randomly collected population of 195 healthy Caucasian controls. However, when the patients were grouped on the basis of their disease severity, one of the activating receptors, KIR2DS1 was significantly underrepresented in patients with the severe disease compared to patients with mild-moderate disease (21.2% vs. 48.5%, p=0.038). KIR3DS1 and KIR2DS5, the genes that are in strong linkage disequilibrium with KIR2DS1 were also underrepresented in patients with the severe disease compared to patients with mild-moderate disease. However, these differences were not statistically significant.

Figure 1. Genotypes missing both KIR3DS1 and KIR2DS1 are prevalent in RRP patients with severe disease.

Figure 1

Open in a new tab

Twenty-six distinct genotypes differed from each other by the presence (shaded box) and absence (white box) of 16 KIR genes in RRP patients compared to those found in healthy controls. Percent population of each genotype was defined as the number of individuals (N) having the genotype divided by the number of individuals studied (n) in the given study group. The genotype of most patients with severe RRP lacked both KIR3DS1 and KIR2DS1 genes (genotype#1–16). KIR2DS1 occurred at a significantly lower frequency in severe RRP patients compared to those with mild-moderate disease (p=0.038).

The KIR gene content of a given individual is called the “KIR genotype”. Within the group of RRP patients, there were 26 distinct KIR genotypes observed that differed in gene content (Figure 1), of which 11 were found to be unique in patients. The majority of the KIR genotypes identified in healthy controls (21 of 36, 58.3%) were not detected in RRP patients. The overall difference in the distribution of KIR genotypes between controls and patients was not statistically significant. However, genotypes lacking activating KIR genes 3DS1 and 2DS1 (#1–16) were significantly overrepresented in RRP patients with severe disease (78.8% in severe RRP vs. 48.5% in mild-moderate RRP, p=0.019). Of these 16 genotypes, six (#4, 7–11) were observed only in severe patients and five (#12–16) were observed only in controls. Eleven of these 16 genotypes (#1–6, 8, 10, 12, 15 and 16) further lacked an additional activating gene, KIR2DS5. Consequently, 78.8% of patients with severe RRP disease were negative for both KIR3DS1 and KIR2DS1 genes while only 48.5% (p=0.019) of the patients with mild-moderate disease lacked both of these KIR loci (Figure 2). KIR3DS1*049N, the null allele-specific typing revealed that all patients and controls typed here that were positive for KIR3DS1 gene carried a functional KIR3DS1 allele. To avoid a possible ethnic-specific deviation caused by the increased frequency of KIR3DS1−KIR2DS1− carriage in African Americans [21], we also performed a selective analysis within the Caucasian patients only, representing 74.2% of the total patient cohort. Within the Caucasian RRP patients (n=50), 75% of those with severe disease lacked both KIR3DS1 and KIR2DS1 genes, while only 44.8% (p=0.045) of the patients with mild-moderate disease lacked both of KIR genes (Figure 2). Of the 11 African American patients with RRP, all seven of the African American patients with severe RRP disease (100%) lacked both KIR3DS1 and KIR2DS1 genes, as did 2 of 3 patients with mild moderate RRP compared to 77.6% of the reported frequency in African American controls [21]. The one remaining African American patient with mild-moderate RRP carried only one of these activating KIR genes, KIR2DS1.

Figure 2. Patients with severe RRP lack both KIR3DS1 and KIR2DS1 genes.

Figure 2

Open in a new tab

Individuals carrying either KIR3DS1, KIR2DS1, or both genes, and individuals lacking both KIR3DS1 and KIR2DS1, were equally distributed in healthy controls and in mild-moderate patients. In contrast, the majority of the patients with severe RRP lacked both KIR3DS1 and KIR2DS1 genes compared to patients who carried one or both of these genes. NS = Not significant.

RRP patients expressing “disease-susceptible" HLA-DRB1/DQB1 alleles lack activating KIR genes 3DS1, 2DS1 and/or 2DS5

To determine the status of KIR genes in patients carrying disease-susceptibility or disease-protective HLA-DRB1/DQB1 alleles [5, 6], we assessed the cognate combinations of KIR genes, DRB1 alleles, and DQB1 alleles, by a three way association test using a best log-linear model (Figure 3 and Figure 4). Patients carrying at least one of the four disease-susceptible HLA class II allele (DRB1*0102, DRB1*0301, DQB1*0201 and DQB1*0202) more frequently lacked either KIR3DS1 (p=0.006), KIR2DS1 (p=0.003) or KIR2DS5 (p=0.004), compared to controls expressing any of these disease-susceptibility HLA alleles (Figure 3). Although differences were not statistically significant for Caucasians only, a similar trend was observed (data not shown). Of note, there was also a trend in patients who lacked expression of the disease-protective HLA-DQB1*0602 allele and the absence of KIR3DS1 and KIR2DS1 genes in RRP patients who had severe disease, p=0.06 (Figure 4).

Figure 3. RRP patients expressing disease-susceptible HLA-DRB1/DQB1 alleles frequently lack activating KIR genes 3DS1, 2DS1 and/or 2DS5.

Figure 3

Open in a new tab

Severe RRP patients carrying at least one of the four disease-susceptible HLA class II allele (DRB1*0102, DRB1*0301, DQB1*0201 and DQB1*0202) more frequently lacked either KIR3DS1, KIR2DS1 or KIR2DS5, compared to controls expressing any of these disease-susceptibility HLA alleles.

Figure 4. Patients with severe RRP lack protective HLA and KIR genes.

Figure 4

Open in a new tab

Most patients with severe RRP were lacked the protective genes of HLA (DQB1*0602) and KIR (3DS1 and 2DS1) loci. NS = Not significant.

KIR-HLA class I gene combinations were comparable between patients and controls

Since KIRs on chromosome 19q13.4 and HLA on chromosome 6p21.3 are substantially polymorphic, their independent segregation produces diverse KIR+HLA class I gene combinations among individuals [21]. To investigate whether certain combinations of KIR-HLA combinations confer risk for RRP, we identified the well-defined inhibitory KIR+HLA class I ligand gene combinations in each subject (Table 2). No difference was observed in the frequencies of KIR2DL1+HLA-C2, KIR2DL2/3+HLA-C1 and KIR3DL1+HLA-Bw4. However, the KIR3DL2+HLA-A3/11 combination was significantly reduced in RRP patients compared to the controls (21.2% vs. 35.4%, p=0.023). In contrast, all three predicted activating KIR+HLA class I ligand combinations (KIR3DS1+HLA-Bw4, KIR2DS1+HLA-C2 and KIR2DS2+HLA-C1) were distributed equally between patients and controls (Table 2).

Table 2.

Frequency of KIR-HLA combinations in RRP patients and controls.

RRP patients Controls
KIR-HLA combination (n=66) (n=195) p-value
% N+ % N+
Inhibitory KIR-HLA class I ligand
KIR2DL1+HLA-C2 66.7 (44) 55.9 (109) 0.321
KIR2DL2/3+HLA-C1 81.8 (54) 80.0 (156) 0.858
KIR3DL1+HLA-Bw4 60.6 (40) 55.4 (108) 0.778
KIR3DL2+HLA-A3/11 21.2 (14) 35.4 (69) 0.023
Activating KIR-HLA class I ligand
KIR3DS1+HLA-Bw4 24.2 (16) 21.5 (42) 0.866
KIR2DS1+HLA-C2 28.8 (19) 22.5 (44) 0.403
KIR2DS2+HLA-C1 57.6 (38) 49.7 (97) 0.579
Open in a new tab

Papillomas from KIR3DS1− KIR2DS1− RRP patients contained more NK cells and CD57+CD4+ T cells

To determine if the activating KIR genes 3DS1 and 2DS1 affect the lymphocyte subpopulations in papillomas from patients with RRP, we retrospectively reviewed our immunophenotyping data obtained from patients we previously described [6]. Of interest, the number of CD3− CD56+ NK cells and the CD57+CD4+ T cell subsets were significantly (p<0.05) elevated in papillomas of RRP patients who lacked both KIR3DS1 and KIR2DS1 genes compared to patients who expressed one or both of these KIR genes (Figure 5). No other subpopulation of mononuclear cells in papillomas was significantly different in patients who did or did not carry these activating KIR genes (data not shown).

Figure 5. NK cells and CD57+CD4+ T-cells are enriched in KIR3DS1−/KIR2DS1− papillomas.

Figure 5

Open in a new tab

The percent of CD3−CD56+ NK cells (panel A) and CD57+CD4+ T-cells (panel B) in infiltrating mononuclear cells found in papillomas from patients with RRP who carry (n=12), or lack (n=24) KIR3DS1-KIR2DS1 genes. Data are depicted by box plots showing the median and 75th and 25th percentiles. Whiskers represent the 90th and 10th percentiles, and outliers outside the 95th and 5th percentiles are shown (where applicable). Means are represented by the dotted lines and medians by solid lines.

Discussion

Natural killer cells are an integral component of innate immunity and provide a crucial initial defense against pathological organisms during the time (0–5 days) when the adaptive immune system is processing antigen [30]. NK cell responses to virally-infected target cells are mediated by the interplay of polymorphic inhibitory receptors, activating receptors and their corresponding ligands [31]. Activating NK cell receptor LY49H in mice recognizes mouse cytomegalovirus (MCMV)-encoded determinants expressed on the surface of infected cells and trigger the NK activation cascade leading to infected cell cytolysis [13, 14]. Although such direct receptor-ligand interactions have not been determined in humans, epidemiological studies revealed that the presence of activating receptor KIR3DS1 slowed down the AIDS progression in HIV-1 infected patients, compared to patients lacking KIR3DS1, thus indicating a potential anti-HIV effect of KIR3DS1 [15]. The KIR3DS1 receptor was also shown to be weakly protective to hepatitis-C virus (HCV) infection [32]. Analogous to these findings, we report here that patients with HPV-6/11 infection who lacked KIR3DS1 and KIR2DS1 genes were prone to develop severe RRP, compared to the patients carrying one or both of these KIR genes, thus indicating the anti-HPV effect of these activating KIR receptors. Taken together, the KIR receptors 3DS1/2DS1 on NK cells presumably recognize select viral-encoded/-induced determinants expressed on the surface of infected target cells, and enhance NK cell response against viral infection (Figure 6). The activating KIR genes 3DS1, 2DS1 and 2DS5 are linked together as a cluster in the telomeric half of the KIR gene complex [19]. Underrepresentation of this cluster in RRP patients also suggests the possibility that other unknown but linked factors may contribute resistance of developing RRP.

Figure 6. A model depicting how the compound KIR-HLA genotypes may confer the risk of developing RRP.

Figure 6

Open in a new tab

NK cells expressing activating KIR receptors 3DS1 and 2DS1 are presumably well equipped to recognize and kill HPV-infected keratinocytes, and mount an effective early NK cell response. In contrast, NK cells that do not express KIR3DS1/2DS1 receptors may not have the ability of sensing HPV-infected targets cells, and thus, the infection may spread to neighboring cells. If KIR3DS1−KIR2DS1− individuals carry one of the susceptible HLA-DRB1/DQB1 alleles, the skewed TH2 response may further block eradication of infection and support the development of clinical disease. The expression of the protective HLA-DQB1*0602 type may help protect in individuals who lack KIR3DS1/2DS1 genes. HPV-p – Human pappiloma viral protein, APC – Antigen presenting cell.

Of interest to RRP, the presence of the activating receptor KIR3DS1 was shown to associate with cervical neoplasia progression to cervical cancer, especially in the absence of protective inhibitory KIR3DL1 and HLA-Bw4 ligend combination [7, 33]. The difference in clinical disease expression induced by different HPVs in KIR3DS1+ patients could be due to other host and viral genetic factors, such as (i) differences in the HPV strains that causes cervical cancer (HPV-16 and 18) vs. RRP (HPV-6 and 11), and thus the interaction of KIR3DS1 with putative HPV strain-specific ligands leading to either killing of HPV-6/11-infected cells keeping the host from developing RRP, or leading to inappropriate tissue-specific hyperresponsiveness promoting growth of cervical cancer, vs. benign respiratory papillomas, (ii) differences in the HLA alleles associated with cervical cancer (protection: DQB1*03, DRB1*1501/DQB1*0602, DRB1*13 and DQB1*0603) [34] vs. RRP disease (susceptibility: DRB1*0102, DRB1*0301, DQB1*0201 and DQB1*0202, protection: DQB1*0602) [5, 6] with different HLA class II restrictions diversifying the CD4+ T-cell immune response to HPV strain-specific antigens favoring malignant vs. benign HPV-induced disease development, and (iii) differences in other immunoregulatory, and malignancy-related genes in papillomas [35] that are important in directing the outcome of HPV infection that may alter CD4+ T-cell immune responses made to HPV strain-specific antigens.

In addition to KIR3DS1-mediated activation, decreased inhibition triggered by HLA-C and HLA-B binding inhibitory KIR receptors has been implicated in the development of cervical neoplasia [33] and HCV resolution [32]. Although no difference was observed in the distribution of the HLA-C and -B binding inhibitory KIR-ligand combinations, a decrease in HLA-A binding inhibitory receptor-ligand pair (KIR3DL2+HLA-A3/11) in RRP patients suggests a possible protective effect by this interaction against RRP development. Furthermore, the majority of the RRP patients with severe disease carried at least one of the known disease-susceptibility HLA class II alleles (DRB1*0102, DRB1*0301, DQB1*0201 and/or DQB1*0202) but lacked both KIR3DS1 and KIR2DS1 genes. These observations suggest that certain combination of KIR and HLA genes could provide a permissive immune microenvironment that fosters chronic HPV-6/11 infection and ultimately tumor development.

In addition to the gene content variation, each KIR gene exhibits considerable sequence diversity [36]. Eleven alleles of KIR3DS1 and five alleles of KIR2DS1 have been described to date (http://www.ebi.ac.uk/ipd/kir/index.html). The sequence polymorphisms that distinguish the alleles of KIR3DS1 and KIR2DS1 may influences expression, ligand binding, cytolytic, and cytokine secretion as described for other KIR receptors [37–39]. Furthermore, as the KIR receptors are clonally expressed on NK cells in a stochastic manner such that each NK cell clone expresses only a portion of the genes within the genotype [40], a substantial fraction of NK cells in RRP patients may not express KIR3DS1 and KIR2DS1 receptors even if patients carry the corresponding genes; thus NK cells from these patients may still ignore HPV-infected targets leading to RRP. Future studies of KIR gene polymorphism and NK clonal expression to investigate the possible dominance of unexpressed and underexpressed variants of KIR3DS1/2DS1 in RRP patients who carry these activating KIR genes are needed.

It was surprising that there was an increased frequency of NK cells and CD57+CD4+ T-cells in papillomas from RRP patients who lacked both activating KIR3DS1 and KIR2DS1, compared to patients carrying one or both of these genes. It is possible that the larger fraction of NK cell infiltration in KIR2DS1− KIR2DS3− papillomas represents a compensatory response to counteract the absence of these activating KIR genes by generating an effective NK response against HPV-infected keratinocytes using receptors other than KIR3DS1/2DS1. CD57+CD4+ T cells have been reported to represent: 1) a chronically activated CD45+ memory subset of immunoregulatory cells that express IL-4 as described in Mycobacterium tuberculosis [41], 2) senescent T-cells undergoing apoptosis in HIV infection [42], or 3) a subset of cells marking advanced chronic lymphocytic leukemia [43]. It is possible that the predominance of this T cell subset in KIR2DS1− KIR2DS3− papillomas represents immunoregulatory T-cells that express IL-4. This would be consistent with our previous studies that showed increased IL-4, but not IFN-γ expression in papillomas [3], and that severe disease was associated with a greater IL-4/IFN-γ expression ratio compared to mild-moderate disease [4]. Thus, it is possible that this CD57+CD4+ T-cell subpopulation is in part responsible for enhanced IL-4 expression in RRP, generated by chronic activation induced by HPV-6/11 infection, supporting the development of a TH2-like, regulatory microenvironment that is characteristic of RRP [3, 4, 6]. Alternatively, recent studies suggest that these cells can express granzyme and perforin, and thus, may function as cytolytic T-cells [44]. If this is the case, then they may also represent a compensatory immune mechanism to counteract defective, HPV-specific, NK cytolysis in RRP patients who lack KIR3DS1 and KIR2DS1 genes. Further investigation of the functional of these interesting subpopulations of T- and NK cells is ongoing in our laboratory.

Of interest was the finding that 10 of the 11 African American patients with RRP lacked both KIR3DS1 and KIR2DS1, and the 11th lacked the KIR3DS1 allele, but had KIR2DS1. Of note, the 11th patient also had mild-moderate disease. While the absence of these activating KIR genes is more common in African Americans [21, 45–47], the fact that all of our RRP patients showed complete or partial absence of these KIR genes is intriguing. It is known that RRP is more common in African Americans and in juvenile onset disease [48]. It is thought that the increased frequency of RRP in African Americans is associated lower socioeconomic conditions with increased birth canal HPV colonization [49, 50]. However, while the number of African American patients in our study is small, it is possible that another risk factor of developing RRP in these individuals may be related to their higher incidence of the absence of activating KIR gene carriage that further predisposes HPV exposed African American infants to develop RRP.

In summary, NK cells expressing activating KIR receptors 3DS1 and 2DS1 are likely to be critical in removing HPV-infected keratinocytes. Thus individuals carrying and expressing these genes may better mount an effective early NK cell response against HPV-6/11, thereby clearing HPV early after infection (Figure 6), or suppressing latent virus activation. In contrast, NK cells that do not express KIR3DS1/2DS1 may not recognize HPV-infected keratinocytes efficiently, leading to a failure to contain HPVs. If these KIR3DS1−KIR2DS1− individuals carry one of the susceptible HLA-DRB1/DQB1 alleles, the skewed TH2 response may further prevent recognition of HPV-infected keratinocytes, thus enhancing the development of clinical disease. The presence of the protective HLA-DQB1*0602 gene likely further protects individuals who fail to carry or express KIR3DS1 and KIR2DS1. Finally, our findings may provide a rationale to develop new and better treatments that target cells involved in innate as well as in adaptive immune responses in patients with severe RRP.

Acknowledgements

This work was supported by start-up funds from the Department of Pathology and Laboratory Medicine (to R.R.) and by NIH grants R01 DE017227 (to V.R.B.). Elham Ashouri was supported by a fellowship from the Ministry of Health and Medical Education, The Islamic Republic of Iran.

Abbreviations

RRP

Recurrent respiratory papillomatosis,

HPV

Human papilloma viruses,

HLA

Human leukocyte antigen,

KIR

Killer-cell immunoglobulin-like receptors,

NK cell

Natural killer cell,

MCMV

Mouse cytomegalovirus,

HCV

Hepatitis-C virus,

SSP

Sequence-specific primer-directed amplification,

SSO

Sequence-specific oligonucleotide hybridization.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • 1.Derkay CS, Wiatrak B. Recurrent respiratory papillomatosis: a review. Laryngoscope. 2008;118:1236. doi: 10.1097/MLG.0b013e31816a7135. [DOI] [PubMed] [Google Scholar]
  • 2.Auborn KJ, Wang H, Vaccariello MA, Taichman LB. Kinetics of HPV11 DNA replication after infection of keratinocytes with virions. Virus Res. 1996;43:85. doi: 10.1016/0168-1702(96)01322-6. [DOI] [PubMed] [Google Scholar]
  • 3.Bonagura VR, Hatam L, DeVoti J, Zeng F, Steinberg BM. Recurrent respiratory papillomatosis: altered CD8(+) T-cell subsets and T(H)1/T(H)2 cytokine imbalance. Clin Immunol. 1999;93:302. doi: 10.1006/clim.1999.4784. [DOI] [PubMed] [Google Scholar]
  • 4.DeVoti JA, Steinberg BM, Rosenthal DW, Hatam L, Vambutas A, Abramson AL, et al. Failure of gamma interferon but not interleukin-10 expression in response to human papillomavirus type 11 E6 protein in respiratory papillomatosis. Clin Diagn Lab Immunol. 2004;11:538. doi: 10.1128/CDLI.11.3.538-547.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Gelder CM, Williams OM, Hart KW, Wall S, Williams G, Ingrams D, et al. HLA class II polymorphisms and susceptibility to recurrent respiratory papillomatosis. J Virol. 2003;77:1927. doi: 10.1128/JVI.77.3.1927-1939.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Bonagura VR, Vambutas A, DeVoti JA, Rosenthal DW, Steinberg BM, Abramson AL, et al. HLA alleles, IFN-gamma responses to HPV-11 E6, and disease severity in patients with recurrent respiratory papillomatosis. Hum Immunol. 2004;65:773. doi: 10.1016/j.humimm.2004.05.014. [DOI] [PubMed] [Google Scholar]
  • 7.Khakoo SI, Carrington M. KIR and disease: a model system or system of models? Immunol Rev. 2006;214:186. doi: 10.1111/j.1600-065X.2006.00459.x. [DOI] [PubMed] [Google Scholar]
  • 8.Lanier LL. NK cell recognition. Annu Rev Immunol. 2005;23:225. doi: 10.1146/annurev.immunol.23.021704.115526. [DOI] [PubMed] [Google Scholar]
  • 9.Biron CA, Brossay L. NK cells and NKT cells in innate defense against viral infections. Curr Opin Immunol. 2001;13:458. doi: 10.1016/s0952-7915(00)00241-7. [DOI] [PubMed] [Google Scholar]
  • 10.Phillips JH, Gumperz JE, Parham P, Lanier LL. Superantigen-dependent, cell-mediated cytotoxicity inhibited by MHC class I receptors on T lymphocytes. Science. 1995;268:403. doi: 10.1126/science.7716542. [DOI] [PubMed] [Google Scholar]
  • 11.Vilches C, Parham P. KIR: diverse, rapidly evolving receptors of innate and adaptive immunity. Annu Rev Immunol. 2002;20:217. doi: 10.1146/annurev.immunol.20.092501.134942. [DOI] [PubMed] [Google Scholar]
  • 12.Rajalingam R. Killer cell immunoglobulin-like receptors influence the innate and adaptive immune responses. Iran J Immunol. 2007;4:61. [PubMed] [Google Scholar]
  • 13.Arase H, Mocarski ES, Campbell AE, Hill AB, Lanier LL. Direct recognition of cytomegalovirus by activating and inhibitory NK cell receptors. Science. 2002;296:1323. doi: 10.1126/science.1070884. [DOI] [PubMed] [Google Scholar]
  • 14.Smith HR, Heusel JW, Mehta IK, Kim S, Dorner BG, Naidenko OV, et al. Recognition of a virus-encoded ligand by a natural killer cell activation receptor. Proc Natl Acad Sci U S A. 2002;99:8826. doi: 10.1073/pnas.092258599. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Martin MP, Gao X, Lee JH, Nelson GW, Detels R, Goedert JJ, et al. Epistatic interaction between KIR3DS1 and HLA-B delays the progression to AIDS. Nat Genet. 2002;31:429. doi: 10.1038/ng934. [DOI] [PubMed] [Google Scholar]
  • 16.Chewning JH, Gudme CN, Hsu KC, Selvakumar A, Dupont B. KIR2DS1-positive NK cells mediate alloresponse against the C2 HLA-KIR ligand group in vitro. J Immunol. 2007;179:854. doi: 10.4049/jimmunol.179.2.854. [DOI] [PubMed] [Google Scholar]
  • 17.Biassoni R, Pessino A, Malaspina A, Cantoni C, Bottino C, Sivori S, et al. Role of amino acid position 70 in the binding affinity of p50.1 and p58.1 receptors for HLA-Cw4 molecules. Eur J Immunol. 1997;27:3095. doi: 10.1002/eji.1830271203. [DOI] [PubMed] [Google Scholar]
  • 18.Stewart CA, Laugier-Anfossi F, Vely F, Saulquin X, Riedmuller J, Tisserant A, et al. Recognition of peptide-MHC class I complexes by activating killer immunoglobulin-like receptors. Proc Natl Acad Sci U S A. 2005;102:13224. doi: 10.1073/pnas.0503594102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Wilson MJ, Torkar M, Haude A, Milne S, Jones T, Sheer D, et al. Plasticity in the organization and sequences of human KIR/ILT gene families. Proc Natl Acad Sci U S A. 2000;97:4778. doi: 10.1073/pnas.080588597. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Uhrberg M, Valiante NM, Shum BP, Shilling HG, Lienert-Weidenbach K, Corliss B, et al. Human diversity in killer cell inhibitory receptor genes. Immunity. 1997;7:753. doi: 10.1016/s1074-7613(00)80394-5. [DOI] [PubMed] [Google Scholar]
  • 21.Du Z, Gjertson DW, Reed EF, Rajalingam R. Receptor-ligand analyses define minimal killer cell Ig-like receptor (KIR) in humans. Immunogenetics. 2007;59:1. doi: 10.1007/s00251-006-0168-4. [DOI] [PubMed] [Google Scholar]
  • 22.Lindeberg H, Oster S, Oxlund I, Elbrond O. Laryngeal papillomas: classification and course. Clin Otolaryngol Allied Sci. 1986;11:423. [PubMed] [Google Scholar]
  • 23.Maran A, Amella CA, Di Lorenzo TP, Auborn KJ, Taichman LB, Steinberg BM. Human papillomavirus type 11 transcripts are present at low abundance in latently infected respiratory tissues. Virology. 1995;212:285. doi: 10.1006/viro.1995.1486. [DOI] [PubMed] [Google Scholar]
  • 24.Abramson AL, Shikowitz MJ, Mullooly VM, Steinberg BM, Amella CA, Rothstein HR. Clinical effects of photodynamic therapy on recurrent laryngeal papillomas. Arch Otolaryngol Head Neck Surg. 1992;118:125. doi: 10.1001/archotol.1992.01880010029011. [DOI] [PubMed] [Google Scholar]
  • 25.Steinberg BM, Gallagher T, Stoler M, Abramson AL. Persistence and expression of human papillomavirus during interferon therapy. Arch Otolaryngol Head Neck Surg. 1988;114:27. doi: 10.1001/archotol.1988.01860130031010. [DOI] [PubMed] [Google Scholar]
  • 26.Kashima HK, Lenventhal B, Mounts P. papillomatosis: Molecular and Clinical Aspects. New York: Alan R. Liss; 1985. Papilloma Study Group, Scoring system to assess severity and course in recurrent respiratory papillomatosis. [Google Scholar]
  • 27.Bonagura VR, Siegal FP, Abramson AL, Santiago-Schwarz F, O'Reilly ME, Shah K, et al. Enriched HLA-DQ3 phenotype and decreased class I major histocompatibility complex antigen expression in recurrent respiratory papillomatosis. Clin Diagn Lab Immunol. 1994;1:357. doi: 10.1128/cdli.1.3.357-360.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Luo L, Du Z, Sharma SK, Cullen R, Spellman S, Reed EF, et al. Chain-terminating natural mutations affect the function of activating KIR receptors 3DS1 and 2DS3. Immunogenetics. 2007;10:779. doi: 10.1007/s00251-007-0239-1. [DOI] [PubMed] [Google Scholar]
  • 29.Bonagura VR, Cunningham-Rundles S, Edwards BL, Ilowite NT, Wedgwood JF, Valacer DJ. Common variable hypogammaglobulinemia, recurrent Pneumocystis carinii pneumonia on intravenous gamma-globulin therapy, and natural killer deficiency. Clin Immunol Immunopathol. 1989;51:216. doi: 10.1016/0090-1229(89)90021-4. [DOI] [PubMed] [Google Scholar]
  • 30.Yokoyama WM, Scalzo AA. Natural killer cell activation receptors in innate immunity to infection. Microbes Infect. 2002;4:1513. doi: 10.1016/s1286-4579(02)00034-5. [DOI] [PubMed] [Google Scholar]
  • 31.Lanier LL. Natural killer cell receptor signaling. Curr Opin Immunol. 2003;15:308. doi: 10.1016/s0952-7915(03)00039-6. [DOI] [PubMed] [Google Scholar]
  • 32.Khakoo SI, Thio CL, Martin MP, Brooks CR, Gao X, Astemborski J, et al. HLA and NK cell inhibitory receptor genes in resolving hepatitis C virus infection. Science. 2004;305:872. doi: 10.1126/science.1097670. [DOI] [PubMed] [Google Scholar]
  • 33.Carrington M, Wang S, Martin MP, Gao X, Schiffman M, Cheng J, et al. Hierarchy of resistance to cervical neoplasia mediated by combinations of killer immunoglobulin-like receptor and human leukocyte antigen loci. J Exp Med. 2005;201:1069. doi: 10.1084/jem.20042158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Hildesheim A, Wang SS. Host and viral genetics and risk of cervical cancer: a review. Virus Res. 2002;89:229. doi: 10.1016/s0168-1702(02)00191-0. [DOI] [PubMed] [Google Scholar]
  • 35.DeVoti JA, Rosenthal DW, Wu R, Abramson AL, Steinberg BM, Bonagura VR. Immune dysregulation and tumor-associated gene changes in recurrent respiratory papillomatosis: a paired microarray analysis. Mol Med. 2008;14:608. doi: 10.2119/2008-00060.DeVoti. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Garcia CA, Robinson J, Guethlein LA, Parham P, Madrigal JA, Marsh SG. Human KIR sequences 2003. Immunogenetics. 2003;55:227. doi: 10.1007/s00251-003-0572-y. [DOI] [PubMed] [Google Scholar]
  • 37.Yawata M, Yawata N, Draghi M, Little AM, Partheniou F, Parham P. Roles for HLA and KIR polymorphisms in natural killer cell repertoire selection and modulation of effector function. J Exp Med. 2006;203:633. doi: 10.1084/jem.20051884. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.VandenBussche CJ, Dakshanamurthy S, Posch PE, Hurley CK. A single polymorphism disrupts the killer Ig-like receptor 2DL2/2DL3 D1 domain. J Immunol. 2006;177:5347. doi: 10.4049/jimmunol.177.8.5347. [DOI] [PubMed] [Google Scholar]
  • 39.O'Connor GM, Guinan KJ, Cunningham RT, Middleton D, Parham P, Gardiner CM. Functional polymorphism of the KIR3DL1/S1 receptor on human NK cells. J Immunol. 2007;178:235. doi: 10.4049/jimmunol.178.1.235. [DOI] [PubMed] [Google Scholar]
  • 40.Valiante NM, Uhrberg M, Shilling HG, Lienert-Weidenbach K, Arnett KL, D'Andrea A, et al. Functionally and structurally distinct NK cell receptor repertoires in the peripheral blood of two human donors. Immunity. 1997;7:739. doi: 10.1016/s1074-7613(00)80393-3. [DOI] [PubMed] [Google Scholar]
  • 41.Jimenez-Martinez MC, Linares M, Baez R, Montano LF, Martinez-Cairo S, Gorocica P, et al. Intracellular expression of interleukin-4 and interferon-gamma by a Mycobacterium tuberculosis antigen-stimulated CD4+ CD57+ T-cell subpopulation with memory phenotype in tuberculosis patients. Immunology. 2004;111:100. doi: 10.1111/j.1365-2567.2003.01785.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Huang KH, Loutfy MR, Tsoukas CM, Bernard NF. Immune correlates of CD4 decline in HIV-infected patients experiencing virologic failure before undergoing treatment interruption. BMC Infect Dis. 2008;8:59. doi: 10.1186/1471-2334-8-59. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Serrano D, Monteiro J, Allen SL, Kolitz J, Schulman P, Lichtman SM, et al. Clonal expansion within the CD4+CD57+ and CD8+CD57+ T cell subsets in chronic lymphocytic leukemia. J Immunol. 1997;158:1482. [PubMed] [Google Scholar]
  • 44.Chattopadhyay PK, Betts MR, Price DA, Gostick E, Horton H, Roederer M, et al. The cytolytic enzymes granyzme A, granzyme B, and perforin: expression patterns, cell distribution, and their relationship to cell maturity and bright CD57 expression. J Leukoc Biol. 2009;85:88. doi: 10.1189/jlb.0208107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Denis L, Sivula J, Gourraud PA, Kerdudou N, Chout R, Ricard C, et al. Genetic diversity of KIR natural killer cell markers in populations from France, Guadeloupe, Finland, Senegal and Reunion. Tissue Antigens. 2005;66:267. doi: 10.1111/j.1399-0039.2005.00473.x. [DOI] [PubMed] [Google Scholar]
  • 46.Norman PJ, Carrington CV, Byng M, Maxwell LD, Curran MD, Stephens HA, et al. Natural killer cell immunoglobulin-like receptor (KIR) locus profiles in African and South Asian populations. Genes Immun. 2002;3:86. doi: 10.1038/sj.gene.6363836. [DOI] [PubMed] [Google Scholar]
  • 47.Single RM, Martin MP, Gao X, Meyer D, Yeager M, Kidd JR, et al. Global diversity and evidence for coevolution of KIR and HLA. Nat Genet. 2007;39:1114. doi: 10.1038/ng2077. [DOI] [PubMed] [Google Scholar]
  • 48.Stanley MA. Human papillomavirus vaccines. Rev Med Virol. 2006;16:139. doi: 10.1002/rmv.498. [DOI] [PubMed] [Google Scholar]
  • 49.Parikh S, Brennan P, Boffetta P. Meta-analysis of social inequality and the risk of cervical cancer. Int J Cancer. 2003;105:687. doi: 10.1002/ijc.11141. [DOI] [PubMed] [Google Scholar]
  • 50.Leung R, Hawkes M, Campisi P. Severity of juvenile onset recurrent respiratory papillomatosis is not associated with socioeconomic status in a setting of universal health care. Int J Pediatr Otorhinolaryngol. 2007;71:965. doi: 10.1016/j.ijporl.2007.03.009. [DOI] [PubMed] [Google Scholar]

RESOURCES