Reduced local immunity in HPV-related VIN: Expression of chemokines and involvement of immunocompetent cells



Usual type VIN is a premalignant disorder caused by persistent HPV infection. High prevalence of VIN in immuno-suppressed women suggests that a good innate and adaptive immune response is important for defense against HPV. Here, we explored expression levels of chemokines and related these to the presence or absence of immuno-competent cells (dendritic and T-cells) in affected (HPV-positive VIN) and non-affected (HPV-negative) vulvar tissues from the same patients. Combining microarray data with quantitative real-time RT-PCR, it was observed that several important chemokines were differentially expressed between VIN and control samples (up-regulation of IL8, CXCL10, CCL20 and CCL22 and down-regulation of CXCL12, CCL21 and CCL14). Furthermore, an increased number of mature dendritic cells (CD208+) seemed to be bottled up in the dermis, and although a T-cell response (increased CD4+ and CD8+ cells) was observed in VIN, a much larger response is required to clear the infection. In summary, it seems that most mature dendritic cells do not receive the proper chemokine signal for migration and will stay in the dermis, not able to present viral antigen to naive T-cells in the lymph node. Consequently the adaptive immune response diminishes, resulting in a persistent HPV infection with increased risk for neoplasia. © 2008 Wiley-Liss, Inc.

During the last decade, vulvar intraepithelial neoplasia (VIN), a premalignant disorder, has been diagnosed with increasing frequency in relatively young women in western countries.1 This rise in incidence is presumably due to the rise in human papilloma virus (HPV) infections. Recently, more than 100 types of HPV are identified and the life-time risk for infection with one of these types is around 80%.2, 3 The different types of HPV are subdivided into low-risk (non-oncogenic, e.g., HPV 6 and 11) and high-risk HPV (oncogenic, e.g., HPV 16, 18 and 33). Around 40% of young, sexually active women are infected with the latter, a high-risk HPV.4 Fortunately, most women are able to clear this infection, and less than 10% of infected women develop a persistent HPV infection, which is one of the leading causes of preneoplastic and neoplastic lesions in the female genital tract, including VIN.5

The host immune response is of critical importance in determining the progression or regression of HPV-related VIN, because being immuno-compromised is a risk factor. For example, VIN is observed more often in women infected with HIV, and in women who are on systemic steroids to treat autoimmune diseases or to prevent rejection after transplantation.6, 7 Smoking is also a risk factor, because it results in a decreased local immune response in the epithelium by reducing the number of intraepithelial Langerhans cells.8, 9

After infection, the host develops a virus-specific cell-mediated immune response, which, in many instances, will lead to clearance of the virus within 1 year.10 This immunological defense system consists of 2 components, namely a fast first-line innate immune response, which is followed by a more sustained adaptive immune response.

The innate immune response is initiated immediately or within hours after the first exposure to the virus. It is a non-specific reaction in which multiple immuno-competent cells, like natural killer cells (NKs), mast cells, eosinophils, basophils, macrophages and neutrophils are involved. Of special importance for the innate immune response are dendritic cells (DCs). These cells are responsible for the first recognition of viral antigens by toll-like receptors (TLRs) at their cell surface. CpG-rich regions in viruses, including HPV, are detected primarily by TLR7 and TLR9.11

The adaptive immune response, in which T-cells play a central role, regulates the destruction of infected cells. After DCs have presented viral antigens to naive T-cells in the draining lymph nodes, these naive T-cells will differentiate into so-called effector cells. There are 3 types of effector cells: T-helper cells, cytotoxic T-cells and regulatory T-cells (Treg cells) (respectively identified by the surface markers CD4, CD8 and CD25/HLA-DR). During a successful response to a viral infection, the infected squamous epithelium will be invaded by large infiltrates of T-cells.

For the immunological defense system a specific group of cytokines, namely chemokines (cytokines with chemotactic activities) are important. Locally produced chemokines (such as CXCL12, CCR1, CCR2, CCR5, CCR7) stimulate DCs and T-cell migration to and from affected tissue. Studies describing this trafficking of immunocompetent cells in VIN are scarce, but nevertheless it is established that large numbers of T-cells, both T helper-cells (CD4+) and cytotoxic T-cells (CD8+), are present in HPV-related VIN.12, 13 Recently, our group investigated the distribution patterns of immunocompetent cells in VIN lesions compared to control tissue. It was observed that DCs and T-cells specifically migrated towards the dermis of a VIN lesion, suggesting that the cellular-immune response upon viral HPV infection occurs mainly in the dermis (van Seters et al., unpublished observation).

The concept that the immune response in HPV-related VIN seems to be insufficient and has lead to new treatment options in which the immunomodulator imiquimod plays a role. Imiquimod binds to TLR7 at the surface of DCs and induces plasma DCs to express and secrete multiple cytokines, such as TNF-α, type-1 IFN and IL-12.14 Recently, our group performed a randomized, placebo-controlled, double-blind trial to investigate the effectiveness of imiquimod 5% cream in VIN. It was observed that in 81% of cases treatment with imiquimod led to a reduction of lesion size, in comparison with 0% in the placebogroup (p < 0.001)15. In addition to this, the reduction of lesion size was correlated with partial normalization of the numbers of immunocompetent cells. Furthermore, it was also shown that arousal of HPV16-specific Type 1 cellular immunity by induction of local inflammation by imiquimod was involved in regression of HPV-related VIN lesions.16

In the current study, we are further exploring the local immune responses in VIN. Using earlier produced gene expression profiles of HPV-related VIN tissues and healthy vulvar control tissues,17 here we analyzed the expression of cytokines and cytokine receptors. In addition, the expression of specific cytokines (CCL20, CCL21, CCL22, CXCL10, IL12), the Treg cell marker FOXP3 and the inflammation-inducer PPARγ was analyzed in 14 new VIN patients. From each of these 14 patients, 1 affected (HPV-positive VIN lesion) and 1 non-affected (HPV-negative) vulvar tissue was obtained for this study. Subsequently, data were related to the presence or absence of immuno-competent cells (DCs and T-cells) in affected and non-affected vulvar tissues.


CIN, cervical intraepithelial neoplasia; DCs, dendritic cells; HPV, human papilloma virus; NK, natural killer; TLR, toll-like receptor; Treg cells, regulatory T-cells; VIN, vulvar intraepithelial neoplasia.

Material and methods

Patient samples

All patients (n = 14) were histologically and clinically diagnosed with high-grade VIN. From these 14 women we collected a 4-mm-punch biopsy of the affected vulvar skin (HPV-positive VIN lesion) and a 4-mm-punch biopsy from contra-lateral nonaffected, healthy vulvar skin (HPV-negative). These 2 biopsies were taken at the same time and were directly frozen in liquid nitrogen and stored at −80°C until further analysis. All samples (affected and non-affected) were reviewed by a pathologist (P.C.E.) for histological diagnosis and were analyzed for the presence of HPV DNA by using a standard GP5+/6+ PCR enzyme immunoassay followed by reverse line blot analysis, as described previously.18 The medical Ethical Committees approved our study design and all women voluntarily gave written informed consent15.

Microarray data analysis

Microarray data were obtained from an earlier study, (where we compared HPV-related VIN to healthy vulvar skin obtained from women visiting a medical center for plastic cosmetic surgery for reduction of the labia minora.17 To identify differences in cytokine and cytokine receptor expression between VIN and control tissue, genes involved in the cytokine and cytokine-receptor interaction pathway (KEGG PATHWAY database: were selected and used as a starting point for our current investigations. Raw and normalized microarray data have been deposited in the GEO repository at NCBI under accession GSE5563.

Immunohistochemical staining

Immunohistochemical staining was performed for the following markers: CD1a, classic marker for Langerhans cells; CD207, marker for immature Langerhans Cells expressing Langerin; CD208, marker for mature DCs; CD94, marker for NK cells; CD4, marker for T-helper cells; CD8, marker for cytotoxic T-cells; and CD25/HLA-DR, marker for Treg cells. For plasmacytoid DCs, characterized by the presence of CD123 and absence of CD11c, both markers were used. Staining and light microscopic evaluation of the obtained data were performed as described earlier by van Seters.15

Quantitative real-time RT-PCR

Total RNA was isolated from tissue samples using Trizol (Invitrogen, Life Technologies, Philadelphia, PA) and the quality and quantity was assessed on the Agilent 2100 Bioanalyzer (Agilent, Palo Alto, CA). RNA was considered of sufficient quality with an RNA Integrity number (RIN-value) of 7.5 or higher. Accordingly, cDNA was generated from 1 μg total RNA from 18 samples (from each of 9 patients, 1 affected and 1 non-affected vulvar tissue was obtained) using T7 oligo d(T) primers (Invitrogen) and SuperScript II reverse transcriptase (Invitrogen) according to the Affymetrix protocol for first strand cDNA synthesis (Affymetrix, Santa Clara, CA). Real-time PCR (RT-PCR) was performed in duplicate using the Opticon I (Applied Biosystems, Foster City, CA) and SYBR Green I™ (Applied Biosystems). Ten ng of the cDNA samples were amplified with 0.5 μM of primer pairs specific for the genes, in a 25 μl reaction with 12.5 μl SYBR Green PCR master mix (Applied Biosystems). The housekeeping gene β-actin was used for normalization and all PCR primers were designed to be intron spanning (Table I). PCR reactions were performed as follows: 38 cycles of denaturation at 95°C (15 sec), annealing at 59–62°C (30 sec) and extension at 72°C.

Table I. Primers Used for Real-Time PCR
GeneForward primer 5′–3′Reverse primer 5′–3′

The comparative CT method (Applied Biosystems) was used to determine the relative quantitation of gene expression for each gene compared to the β-actin control. The difference in cycle time ΔCT, was determined as the difference between the tested gene and the reference housekeeping gene, β-actin. ΔΔCT was then found by obtaining the difference between each sample compared to the mean of the expression values of the control group. The relative fold change was calculated as FC = 2−ΔΔCT.


For the selected cytokine and cytokine receptor mRNA levels, the Mann–Whitney test was used for evaluation of differences in signal intensity between the normalized microarray data (VIN versus healthy controls). A p-value <0.05 was considered statistically significant. For the immuno-competent cell counts and quantitative real-time RT-PCR, Wilcoxon Signed Ranked Test was used to calculate the significant differences between affected and nonaffected vulvar tissues. A p-value <0.05 was considered statistically significant.


Patient characteristics

Biopsies specimens were obtained from 14 women, aged 32–58 years (median age: 43 years). From each of these 14 patients, 1 affected (HPV-positive VIN lesion) and 1 contra-lateral nonaffected (HPV-negative) vulvar tissue was obtained. Patient characteristics of those 14 women are described in Table II.

Table II. Patient Characteristics
PatientAgeHistologySmokingHPV genotype in VIN lesionHPV genotype in non-affected tissueImmunostainingRT-PCR

Expression of cytokines and cytokine receptors in VIN (microarray data)

In the study by Santegoets et al.17 we observed in the VIN tissue a 11-fold decreased expression of PPARγ, a receptor molecule which has been implicated in maturation of DCs.19 Here, we have reanalyzed these microarray data to review differential expression of genes involved in initiating and maintaining an immune response. From the KEGG database we have obtained a comprehensive scheme ( on “cytokine–cytokine receptor interaction”. In Figure 1, this scheme was used as a background for our own microarray data. In Figure 1, all significantly upregulated (red) or downregulated (green) “cytokine–cytokine receptor interaction” genes in VIN as compared to control tissue are shown.

Figure 1.

Cytokine and cytokine-receptor expression in VIN lesions. From the KEGG database ( the scheme on “cytokine–cytokine receptor interaction” was used to overlay our significantly differentially expressed genes between VIN and control.15 The Mann–Whitney test was used for evaluation of differences in signal intensity of the normalized microarray data (VIN versus healthy controls). A p-value of 0.05 was considered statistically significant. Red boxes represent genes with a significantly higher expression in VIN compared to controls; green boxes represent genes with a significantly lower expression in VIN compared to controls. [Color figure can be viewed in the online issue, which is available at]

Up-regulation of the following genes was observed in VIN lesions: Chemokines + receptors: IL8, CXCL10, CCL20, CCL22 and CCR7; Hematopoietins + receptors: IL2RG; PDGF-family + receptors: VEGF, IFNGR1 and IL28RA; TNF-family + receptors: TNFSF10, TNFRSF10A, TNFRSF10B, TNFRSF25 and TNFRSF7; TGFbeta-family + receptors: TGFA, INHBA, ACVR1B and BMP; IL17-family + receptors: IL17RB; and IL1-family + receptors: IL1F9 and IL1F5.

Down-regulation of the following genes was observed in VIN lesions: Chemokines + receptors: IL8RB, CXCL12, CCL21 and CCL14; Hematopoietins + receptors: IL11RA, LIFR, LEPR and IL13RA; PDGF-family + receptors: PDGFA, PDGFB, PDGFC, PDGFD, PDGFRB, VEGFB, VEGFC, KDR, EGFR, CSF1, CSF1R, KITLG, KIT, IFNGR2 and IL20RA; TNF-family + receptors: TNFSF12 and TNFSF13; TGFbeta-family + receptors: TGFB1, TGFB4, INHBB, TGFBR3, ACVR2 and BMPR2; IL17-family + receptors: IL17D, IL17RC and IL17RD; and IL1-family + receptors: IL1R2.

Detection of immunocompetent cells

The above observations showed that during a persistent HPV infection, several immuno-modulatory proteins are regulated, suggesting an ongoing immune response. In order to investigate this further, we have measured the presence of immuno-competent cells in affected (HPV-positive VIN lesion) and nonaffected (HPV-negative) vulvar tissue obtained from 14 patients. Overall, immuno-competent cells could easily be detected by immuno-histochemical staining in affected and non-affected vulvar tissues (Fig. 2) and in general, more cells were present in the dermis than in the epidermis.

Figure 2.

Immuno-competent cells in VIN and control tissue. Immuno-histochemical staining was performed for different markers indicated at the Y-axis of each figure. Positive cells were counted in the epidermis as well as in the dermis and were represented per mm2. Median values are indicated by horizontal lines in the box plots. Wilcoxon Signed Ranked Test was used to calculate significances. White boxes represents control tissue and grey boxes VIN samples from the same patients (n = 14).


For (immature) Langerhans Cells (CD1A+ and CD207+) higher numbers were observed in the epidermis, and no significant differences were observed between affected and nonaffected vulvar tissues. For mature (CD208+) and plasmacytoid DCs (CD123+/CD11c) the situation was almost opposite: higher numbers of these DCs were observed in the dermis and a significant increase in cell numbers was observed in affected as compared to non affected vulvar tissues for mature DCs. For plasmacytoid DCs the difference in cell counts in the dermis did not reach significance between affected and nonaffected vulvar tissues (p = 0.088).

NK cells

NK cells (CD94+) were present in approximately similar numbers in epidermis and dermis and no significant differences were observed in NK cell-counts between affected and non-affected vulvar tissues.

T cells

Compared to the epidermis, T-cells were enriched in the dermis. Furthermore, the most abundant T-cell type was the T-helper cell (CD4+), which was found in significantly higher numbers in the dermis of VIN as compared to nonaffected vulvar tissue. The distribution of cytotoxic T-cells (CD8+) was comparable to that of T-helper cells: Cytotoxic T-cells were also enriched in the dermis and significantly higher numbers were observed in VIN lesions as compared to nonaffected tissues. For Treg cells (CD25+/HLA-DR) low numbers were found in both epidermis and dermis. The Treg cell-numbers in the dermis were higher than in the epidermis, but no differences were observed between VIN lesions and nonaffected vulvar tissues.

Real-time PCR

Based on the finding that CD208+, CD8+ and CD4+ cells were more abundant in the dermis of affected vulvar skin, and based on reports in literature,20–22 a number of specific cytokines (CCL20, CXCL10, IL12A, CCL22, CCL21), the Treg cell marker FOXP3 and the inflammation-inducer PPARγ were chosen for quantitative real-time RT-PCR.

RNA integrity was guaranteed by only using samples with an RNA Integrity Number (RIN) of 7.5 or higher. Consequently, 9 of 14 patients were included. We performed real-time PCR analysis in VIN lesions and their corresponding non-affected vulvar tissues (Fig. 3 and Table III). Analysis of these data by comparing the 2 groups (affected versus non-affected vulvar tissues) showed that only PPARγ was significantly down-regulated in VIN lesions (p = 0.001). The other genes were clearly not regulated significantly in VIN (FOXP3, CXCL10 and IL12A) or significance was not reached (CCL20, p = 0.115; CCL21, p = 0.151; and CCL22, p = 0.084). In Table III, differences per patient are being evaluated. Interestingly in Figure 1, showing microarray data from our earlier study,17CCL20 and CCL22 are indeed significantly upregulated in VIN lesions while CCL21 is significantly down-regulated in VIN lesions. This is in accordance with the observed trends toward significance in our RT-PCR experiment (Fig. 3 and Table III).

Figure 3.

Real-time RT-PCR results. On the X-axis the different investigated genes are indicated. On the Y-axis the ratio between the measured expression level in VIN lesions versus control tissues from the same patients is indicated. All values have been corrected for the expression of the household gene β-Actin. Median values are indicated by horizontal lines in the box plots. Wilcoxon Signed Ranked Test is used to calculate significances.

Table III. Quantitative Real-Time RT-PCR Results Per Patient
  1. Data are expressed as Fold Change differences, calculated by dividing the VIN expression value by the control expression value. Upregulated genes are marked in gray.

  2. 1Describes in how many patients the gene is upregulated or downregulated.-2Wilcoxon signed ranked test is used to calculate the significances between all VIN samples in comparison with all control samples.

inline image


Forty percent of women are, during their lifetime, infected by any of the high-risk HPV types, and although the majority of HPV-infections are cleared from the system, the HPV virus sometimes presents a problem for the immune system. HPV exclusively infects and multiplies in keratinocytes, which are located distant from lymph nodes. Furthermore, keratinocytes have a short lifespan, and because of this, the virus does not need to lyse the cell, which avoids inflammation as a potent trigger for the first-line innate immune response. Also, HPV downregulates the expression of TLR-9, which is 1 of the HPV-binding receptors in DCs, and HPV reduces the production of interferon, which plays an important role in the adaptive immune response.23

This article examines the immune response in HPV-positive VIN lesions in comparison to HPV-negative, non-affected vulvar tissues in the same patient. More particularly, the expression of different cytokines in relation to immuno-competent cell numbers into the affected region was assessed, to obtain better understanding of the local immune response in patients with VIN.

The innate immune response (Fig. 4)

CXCL12 is an important chemoattracting agent, which is produced by stromal cells in response to inflammation.24 In our microarray analysis, the expression of CXCL12 was significantly downregulated, suggesting that the process of migration of DCs into the affected area is significantly disturbed. To investigate this further, we studied the presence of antigen-presenting DCs in the dermis and epidermis of HPV-related VIN in comparison to control vulvar tissues from the same patient. Overall, we measured higher levels of DCs in the dermis as compared to the epidermis. Furthermore, in VIN-affected tissue, the cell numbers seemed to decrease in the epidermis while there was a slight increase in DCs in the dermis (Fig. 2, CD208+, p = 0.020). These observations were in good accordance with earlier findings of our group (van Seters et al., unpublished observation) and suggest that, after recognition of viral antigens, DCs are stimulated to move toward the dermis, while new DCs are hampered to fill up the epidermal niche.

Figure 4.

Cells and cytokines involved in the innate and adaptive immune response.

Under influence of different signals, DCs will mature and migrate toward local lymphoid tissues to present antigens to naive T cells (Fig. 4). This trafficking process has been described as resulting from a switch in chemokine receptor expression at the DC cell surface.25 Migration of DCs toward the draining lymph node is stimulated by, on the 1 hand, downregulation of the receptors of inflammatory chemokines (CCR1, CCR2, CCR5 and CCR6) and on the other hand, up-regulation of the receptors for lymphoid chemokines (CCR7). In VIN a number of observations suggests that DC maturation is taking place. First, upon measuring CD208+ cells (CD208 is a marker for mature DCs), we found significantly increased numbers of mature DCs in the dermis (Fig. 2). Second, our microarray as well as quantitative real-time RT-PCR data showed a strong and significant down-regulation of PPARγ, which has been implicated in DC functional maturation. More specifically, Klotz et al. studied PPARγ-deficient DCs and showed that absence of this gene leads to increased DC immunogenicity.19 And last, we found a significant increase in CCR7 expression, which is expressed in mature DCs. These findings indicate that in VIN, mature DCs are present at the affected site and ready to migrate towards the lymph node. However, this migration process seems to be disturbed, since 1 of the most important ligands of the CCR7 receptor, namely CCL21, is significantly downregulated. Because of this lack of accurate chemokine signaling, mature DCs seem to be bottled up in the dermis (Fig. 2, CD208+, p = 0.020).

In conclusion, it seems that most mature DCs do not receive the proper chemokine signal for migration and will stay in the skin, not able to present the viral antigen to naive T cells in the lymph node. Perchance this may be one of the reasons of an inaccurate initiation of the adaptive immune response.

The adaptive immune response (Fig. 4)

Homing of effector T-cells to sites of infection is dependent on different chemokines. For example CXCL10 is known to be the chemotactic and proliferation factor for certain T-helper cells (Th1) and chemotactic for DCs.26 We found an upregulation of CXCL10 and in line with this observation we found higher numbers of CD4+ (T-helper cells) and CD8+ cells (cytotoxic T-cells) in the dermis of VIN. Furthermore, a significant correlation (Pearson correlation, p = 0.001) was observed between numbers of T-helper cells and the expression of CXCL10 (Table SI, supplementary data). Gul et al. examined low- and high-grade VIN and found only a 3-fold increase in T-helper cells and cytotoxic T-cells, which is in accordance to our results.13 Interestingly, Bourgault et al. could show that a successful clearance of high-grade VIN was accompanied by a strong epidermal and dermal T-helper and cytotoxic T-cell infiltration, something which was not observed in the current investigations.12 Combining our current data with data from literature, it seems that although a T-cell response is observed in persistent VIN, a much stronger response is required in order to clear the infection.

Investigations into the working mechanism of the immunomodifying agent imiquimod in the treatment of VIN also supports this hypothesis: Todd et al.27 showed that application of imiquimod increased the magnitude of the cytotoxic T-cell response, and when imiquimod was used in the treatment of actinic keratosis, it stimulates a cutaneous immune response characterized by increases in activated DCs and T-helper and cytotoxic T cells.28 In addition, upon reviewing imiquimod-responding patients, van Seters also observed that imiquimod treatment resulted in influx of cytotoxic T-cells into the epidermis of VIN lesions.15

T-cell activation results in the secretion of different cytokines that help and regulate other immunocompetent cells. The pattern of cytokine expression is dependent on the pathway that is activated. In general, 2 main responses are initiated by T-helper cells: a Th2 type response, in which IL-4 and IL-10 are the key players; or a Th1 type response, which is important to fight viral infections (Fig. 4). Van Poelgeest et al. could show that the Th1-type response plays an important role in the protection against progressive HPV-related VIN by recognizing the HPV early antigens E2, E6 and E7.16 One of the produced cytokines during a Th1-type response is IFN-γ, which is crucial for an effective innate as well as adaptive immune response. IFN-γ is known to act as a potent effector in limiting viral replication and increasing resistance to infection. In this study, we did not observe increased expression of IFN-γ mRNA in VIN tissues. This finding is in accordance with observations in cervical intraepithelial neoplasia (CIN), where high-grade CIN is associated with decreased expression of the Th1-type cytokines, tumor necrosis factor-α and IFN-γ.29–31 In the absence of a sufficient Th1-type response, cytotoxic T-cells will not migrate and differentiate, which will result in reduced antiviral and antitumor immunity.

In summary, analysis of the immune response during HPV-related VIN revealed that an ineffective innate immune response may be an important factor causing a significantly reduced adaptive immune response. Stimulation of innate as well as adaptive immunity may therefore represent an important instrument to treat persistent HPV infections, which may otherwise develop further to neoplasia.