Cytokine and Chemokine Profiles following Vaccination with Human Papillomavirus Type 16 L1 Virus-Like Particles

Study design.A phase II clinical trial of the HPV16 L1 VLP vaccine without adjuvant (Novavax, Rockville, MD) was conducted with a total of 220 healthy, adult, female volunteers 18 to 25 years of age, as previously described (10, 21). Briefly, subjects were enrolled at The Johns Hopkins University Center for Immunization Research (Baltimore). Participants received three intramuscular doses of either 50 μg of HPV16 L1 VLP vaccine or a placebo (0.5 ml of saline). Blood was collected before the initial vaccination (at month 0) and 1 month after each of the subsequent vaccinations (months 2 and 7). The Johns Hopkins University Institutional Review Board approved the protocol for this study.

Twenty-six subjects were randomly selected (19 vaccine recipients [mean age ± standard deviation, 20.4 ± 2.1 years; median age, 21.9 years; race, 3 black and 16 white] and 7 placebo recipients [mean age, 21.9 ± 1.9; race, 1 black and 6 white]) among those with enough cryopreserved cells available. Month 0 and month 2 samples were selected for this study. Cytokine responses to L1 VLP at month 2 (1 month following the second dose of vaccine) are in general close to peak levels (21).

Cytokine induction assay.Cryopreserved PBMCs were thawed and cultured (2.0 × 10⁶ cells/ml) for 3 days as previously described (21). The following conditions were used in vitro: media; HPV16 L1 VLP (2.5 μg/ml); influenza A virus (Flu, H3N2, 1:100; ATCC); and Sf9/baculovirus insect cell lysate (Bac, 0.1 μg/ml; Novavax). Flu was used as a positive control. Media and the baculovirus cell lysate were used as negative controls. Cell-free supernatants were harvested, aliquoted, and frozen at −80°C.

Cytokine determinations.Culture supernatants were tested in duplicate for 22 cytokine/chemokines (interleukin-1α [IL-1α], IL-1β, IL-2, IL-4, IL-5, IL-6, IL-7, IL-8, IL-10, IL-12, IL-13, IL-15, IL-17, gamma interferon [IFN-γ], tumor necrosis factor alpha [TNF-α], granulocyte colony-stimulating factor [G-CSF], granulocyte-macrophage colony-stimulating factor [GM-CSF], monocyte chemoattractant protein-1 [MCP-1], Eotaxin, macrophage inflammatory protein-1α [MIP-1α], IFN-γ-inducible 10-kDa protein [IP-10], and RANTES [Linco Research, Inc.]). Data acquisition and analysis were performed using Bioplex manager software version 2.0 (Bioplex; Bio-Rad). The lower levels of detection for all the cytokines were 1.6 pg/ml. Specimens with levels below the detection limit were assigned a value of one-half of this limit (20). All IL-7 and IL-12 levels were below detection level and were excluded from analysis. Cytokines were grouped into Th1 type (G-CSF, GM-CSF, IL-2, IL-15, IL-17, IFN-γ, TNF-α), Th2 type (IL-4, IL-5, IL-6, IL-10, IL-13), and inflammatory/chemokine type (IL-1α, IL-1β, IL-8, MIP-1α, MCP-1, IP-10, Eotaxin, RANTES).

Neutralization assay.HPV16-neutralizing antibody titers were determined using a secreted alkaline phosphatase reporter gene pseudovirus-based neutralization assay performed as described previously (19). Neutralizing titers were defined as the reciprocal of the highest dilution that caused at least 50% reduction in secreted alkaline phosphatase activity. Undetectable antibody levels were considered 0.

At month 0, only one participant (a placebo recipient) had detectable neutralizing antibodies, and she maintained detectable levels throughout the study. All other subjects were negative. At month 2, all vaccine recipients had detectable antibody levels (mean, 1,050; median, 664). Two out of the seven placebo recipients had detectable anti-HPV16 antibody titers at month 2 (month 2 titers, 93 and 209, respectively), presumably due to natural infection.

Statistical analysis.Wilcoxon signed-rank test was used to determine significance in changes of cytokine expression after stimulation with L1 VLP before and after vaccination. In addition, Bonferroni's correction for multiple comparisons was applied in the analysis of vaccine-induced changes in cytokine levels.

Cytokines were clustered (R version 2.2.0) to visualize their expression patterns using Euclidean distance metric and average linkage. Samples were clustered according to the log difference in response pre- and postvaccination. Only cytokines with an unadjusted P value of <0.05 (comparing pre- and postvaccination specimens among vaccine recipients) were included (GM-CSF, IL-1α, IL-17, IL-8, TNF-α, IL-6, IL-13, IL-5, IL-4, IFN-γ, IL-10, IL-2, IP-10, and MIP-1α).

Cytokine profile induced by L1 VLP upon vaccination.Twenty-two different cytokines were evaluated in supernatants from VLP-treated PBMCs before and after vaccination in vaccine and placebo recipients. A total of eight cytokines were significantly upregulated in L1 VLP cultures following vaccination compared to levels in cultures taken before vaccination (Fig. 1A). These include Th1-type cytokines (GM-CSF, median n-fold increase, 1.7-fold, P = 0.0002; IL-2, 4.2-fold, P = 0.0002; IFN-γ, 3.1-fold, P = 0.0007), Th2-type cytokines (IL-4, 5.2-fold, P = 0.0002; IL-5, 4.95-fold, P = 0.0007; IL-10, 3.8-fold, P = 0.0017; IL-13, 4.7-fold, P = 0.0001), and the chemokine IP-10 (2.8-fold, P = 0.0021). An increase in IL-1α (1.1-fold), IL-6 (3.0-fold), IL-8 (1.7-fold), IL-17 (1.5-fold), TNF-α (2.5-fold), and MIP-1α (2.6-fold) was observed but did not reach statistical significance after adjusting for multiple comparisons.

• Open in new tab
• Download powerpoint

FIG. 1.

Cytokine profiles in supernatants of treated PBMCs collected before and after vaccination among L1 VLP (V; n = 19) and placebo (P; n = 7) recipients. Supernatants were tested for cytokine content using multiplex cytokine analysis as described in the text. Results are presented as median cytokine levels in picograms per milliliter. Only cytokines that exhibited a statistically significant increase in response to L1 VLP before adjusting for multiple comparisons are shown. Only significant P values are indicated. A nonparametric Wilcoxon test was used to determine statistical significance. (A) Cells from vaccine and placebo recipients were treated with HPV16 L1 VLP (2.5 μg/ml) for 72 h. P values shown were obtained from the comparisons between responses to L1 VLP from pre- and postvaccination samples and adjusted for multiple comparisons. A P value of <0.0025 was considered significant for the primary hypothesis. (B) Cells from vaccine and placebo recipients were left untreated (media) for 72 h. (C) Cells from vaccine and placebo recipients were incubated with baculovirus extract (0.1 μg/ml) for 72 h. (D) Cells from vaccine and placebo recipients were incubated with flu (1:100) for 72 h.

No significant changes in cytokine levels were observed in untreated (Fig. 1B) or control (Fig. 1C and D) cultures (baculovirus or flu cultures) when prevaccination (month 0) and postvaccination (month 2) samples were compared for vaccine or placebo recipients. In addition, no significant changes in cytokine levels were observed in L1 VLP-stimulated cultures from placebo recipients when month 0 and month 2 samples were compared (Fig. 1A).

Identification of patterns of cytokine response to vaccination.Cluster analysis was performed to determine the patterns of response induced by vaccination in vaccine recipients compared to the placebo group (Fig. 2). Subjects clustered into three groups, when their pre- and postvaccination levels of response to L1 VLP were compared. Two of these groups include only vaccine recipients, and one group contains all placebo recipients plus six vaccine recipients.

• Open in new tab
• Download powerpoint

FIG. 2.

Two-dimensional cluster analysis of cytokine/chemokine profiles produced in response to incubation with HPV16 L1 VLP. The top dendrogram represents the similarity between cytokine profiles of different subjects, and the left dendrogram represents the similarity between measured cytokines/chemokines. Differences in relative levels of gene expression are indicated in color, where red indicates upregulation and green indicates downregulation relative to the corresponding gene expression in prevaccination samples. Each row represents the values for a cytokine, and columns correspond to patients (blue label, placebo; pink label, vaccine recipient).

The first group includes 11 vaccine recipients and is characterized by a considerable increase for most cytokines. The n-fold increase for the majority of cytokines in this group (10 out of 14; namely, IL-2, IL-4, IL-5, IL-6, IL-10, IL-13, TNF-α, IFN-γ, IP-10, and MIP-1α) was higher than fourfold. The remaining four cytokines (GM-CSF, IL-1α, IL-17, and IL-8) showed a weaker increment in response postvaccination, but the responses observed were higher than those observed for the other two groups identified.

The second group was composed of six vaccine recipients, presenting a heterogeneous cytokine expression pattern, with weaker cytokine responses than the first group. A twofold increase or higher was observed in at least four out of six vaccine recipients for IL-4, IL-5, IL-6, IL-8, IFN-γ, and IP-10 and in three out of six vaccinees for IL-2, IL-10, IL-13, and IL-17. Flu-induced responses for these cytokines were not significantly different between the vaccine recipients within these two clusters (data not shown).

Finally, the third group contains two vaccine recipients for whom a decrease in cytokine expression was detected for most of the cytokines following vaccination.

Comparison of neutralizing antibody titers with cytokine levels.Antibody titers from individuals from each cluster were also evaluated (Table 1). Neutralizing anti-HPV16 titers were not detectable at the first visit for vaccine recipients (month 0) and increased after vaccination, with peak titers at month 7. Although the number of individuals in each cluster is small and differences in neutralizing antibody titers did not reach statistical significance, vaccine recipients in cluster 1 had higher titers than those in cluster 2 (for cluster 1, the month 2 mean was 1,426 and the median was 928; for cluster 2, the month 2 mean was 595 and the median was 548; P = 0.23). In general, the level of overall cytokine response to VLP tends to mirror the neutralizing antibody titer pattern at month 2. However, no significant correlations between individual cytokine levels and neutralizing antibody titers were detected.

Cytokine production induced by HPV16 L1 VLP prior to vaccination.Cytokine levels obtained after in vitro treatment of PBMCs with L1 VLP were compared to media cultures in all 26 subjects (Fig. 3). L1 VLP induced a statistically significant increase in the level of inflammatory cytokines or chemokines (IL-1α, IP-10) and in Th1-type (IL-2, IFN-γ) and Th2-type (IL-4, IL-10) cytokines. The relative increase in cytokine expression ranged from 1.1-fold for IL-1α to a 5.4-fold increase in the expression of IP-10. IFN-γ expression increased 4.0-fold, and the induction observed for IL-2, IL-4, and IL-10 was 2.4-fold, 1.7-fold, and 2.2-fold, respectively. No significant changes were observed when baculovirus-treated cultures were compared to unstimulated cells.

• Open in new tab
• Download powerpoint

FIG. 3.

Cytokine profiles in supernatants of treated PBMCs collected before vaccination (n = 26). Cells were left untreated (media) or incubated with HPV16 L1 VLP, baculovirus extract, or flu (1:50) for 72 h. Supernatants were tested for cytokine content using multiplex cytokine analysis as described in the text. Results are presented as median cytokine levels in picograms per milliliter. P values were obtained from comparisons between media and stimulated cultures. A nonparametric Wilcoxon test was used to determine statistical significance. A P value of <0.05 was considered significant and is indicated with an asterisk (*). Only cytokines that exhibited a statistically significant increase in response to any stimulus are shown.

A wider spectrum of cytokines was induced by flu with significantly increased levels of 14 out of the 22 cytokines, compared to media cultures. The relative increase in cytokine levels following flu stimulation ranged from 1.13-fold for IL-1α to 294-fold for IFN-γ. Significant increases were detected in inflammatory cytokines/chemokines (IL-1α, median n-fold increase, 1.13-fold; MIP-1α, 241-fold; IP-10, 31-fold; RANTES, 2.3-fold; MCP-1, 6.2-fold), Th1-type cytokines (GM-CSF, 2.4-fold; IL-2, 18.1-fold; IL-15, 1.5-fold; IFN-γ, 294-fold; TNF-α, 7.9-fold), and Th2-type cytokines (IL-4, 7.4-fold; IL-6, 5.9-fold; IL-10, 15.4-fold; IL-13, 2.5-fold).