A C TA Obstetricia et Gynecologica

MAIN RESEARCH ARTICLE

Cell-mediated immune response to human papillomavirus 16 E7 peptide pools in patients with cervical neoplasia
YONG SEOK LEE1, CHUNG WON LEE1, MIN JONG SONG1, EUN MI HO1, CHAN JOO KIM1, TAE CHUL PARK1, TAI GYU KIM2 & JONG SUP PARK1
Department of Obstetrics and Gynecology, 2Department of Microbiology, College of Medicine, The Catholic University of Korea, Seoul, Korea
1

Key words Cell-mediated immunity, cervical cancer, cervical intraepithelial neoplasia, enzyme-linked immunosorbent assay, E7, HPV, loop electrosurgical excision procedure, peptide pools Correspondence Jong Sup Park, Department of Obstetrics and Gynecology, The Catholic University of Korea, 505 Banpo-dong, Seocho-gu, Seoul, 137 040, Korea. E-mail: jspark@catholic.ac.kr Conict of interest The authors have stated explicitly that there are no conicts of interest in connection with this article. Received: 27 February 2011 Accepted: 4 September 2011 DOI: 10.1111/j.1600-0412.2011.01277.x

Abstract Objective. To identify characteristics of the cell-mediated immune (CMI) response to human papillomavirus-16 (HPV) E7 viral peptide pools to help the formulation of therapeutic vaccines. Design. Prospective study. Population. Korean women. Setting. University hospital. Methods. From December 2008 to August 2010, 33 HPV-16positive patients, seven patients exhibiting a high-risk HPV infection other than HPV-16 with grade 2/3 cervical intraepithelial neoplasm (CIN2/3), and nine healthy control donors were enrolled. Main Outcome Measures. CMI response to synthetic HPV-16 E7 overlapping peptide pools using the IFN- ELISPOT assay. Results. The E7 sequence comprising amino acids 1655 was a major immunogenic region. The CMI response to HPV-16 E7 is highly type-specic. The follow-up CMI response may last longer than expected after the lesion is resected. Conclusions. We found that the E7 sequence comprising amino acids 1655 is a major immunogenic region that is critical for the T-cell-mediated immune response with CIN2/3 or cervical cancer. The identication of CMI responses to HPV-16 E7 peptide pools may provide insight into therapeutic vaccine trials for the control of HPV-associated diseases. Abbreviations: aa, amino acids; CIN, cervical intraepithelial neoplasia; CMI, cellmediated immunity; ELISPOT, enzyme linked immunosorbent spot; HPV, human papillomavirus; LEEP, loop electrosurgical excision procedure; NC, negative control; PBMC, peripheral blood mononuclear cell; SD, standard deviation.

Introduction
Cervical cancers are caused by persistent infection with oncogenic strains of the human papillomavirus (HPV) (1). HPV16 is the most common type found in cervical cancer and accounts for more than 50% of cases worldwide (2). Currently, prophylactic vaccines are used successfully to induce antibody responses in HPV-negative patients. Despite their remarkable success, prophylactic vaccines have some limitations, such as high price, need for repeated vaccinations, and no therapeutic effects in HPV-positive cervical neoplastic conditions. Therefore, there is still a need to develop therapeutic vaccines to protect against persistent infection and to treat HPV-positive patients with cervical neoplastic disease. Two human HPV antigens, E6 and E7, are early viral proteins that are can induce and maintain the malignant transformation of cells and they are the only proteins that are ex-

pressed at high levels during persistent infection (3). As they are completely foreign proteins, they have high antigenicity. Therefore, E6 and E7 function as major antigenic proteins of the human immune system, and most therapeutic vaccination strategies aim to induce cell-mediated immune (CMI) responses against these antigens. Various forms of HPV therapeutic vaccines have been explored, including viral or bacterial vector-based vaccines, peptide/protein-based vaccines, dendritic-cell-based vaccines, and nucleic-acid-based vaccines (48). Preliminary clinical trials are ongoing to conrm that premalignant lesions of cancer can be cured immunologically (9,10). To develop therapeutic vaccination strategies, it is important to characterize HPV antigen-specic immune responses, particularly T-cell-mediated immune responses. This is necessary for determining the efcacy of HPV vaccines administered to patients and in identifying immunological

1350

Acta Obstetricia et Gynecologica Scandinavica

C 2011 The Authors 2011 Nordic Federation of Societies of Obstetrics and Gynecology 90 (2011) 13501356

Y. S. Lee et al.

Human papillomavirus and immune response

parameters that are relevant to the clinical outcomes. Antibody titers obtained using virus-like particle-specic enzyme-linked immunosorbent assay correlate well with neutralizing activity (11). Although the technology for detection of antibodies is well established, the measurement of Tcells is very cumbersome and standardization of assays is difcult. The enzyme-linked immunosorbent spot (ELISPOT) assay is currently the preferred method for studies involving T-cells, as it detects the presence of activated T-cells in a quantitative manner and combines high sensitivity with a relatively low T-cell demand (12). Therefore, in this study, we used an interferon (IFN)- ELISPOT assay to test the immune responses to E7 peptides. Ultimately, our goal is to identify the pattern of CMI response to HPV-16 E7 peptides, such as the type-specicity and persistence of the immune memory after treatment of cervical lesions.

cal specimens using a DNA isolation kit (Intron Biotech. Inc., Seoul, Korea), and target HPV DNA was amplied by PCR with consensus GPd5+/d+6 primers. Betaglobin was amplied using PCR with PC03/PC04 primers as internal controls. Amplied DNA was labeled by indocyanine-dUTP (NEM Life Science Products, Inc., Boston, MA, USA). The PCR product was hybridized onto the chip at 40C for two hours, and washed with 3SSPE and with 1SSPE for 2 minutes each. Hybridized signals were visualized with DNA Chip Scanner (GSI Lumonics, Scanarray Lite, Ottawa, Canada).

In vitro stimulation of PBMCs with the E7 overlapping peptides and IFN- ELISPOT assay
Eighteen HPV-16 E7 15-mer peptides that overlapped by 10 amino acids (aa) and spanned the full length of the E7 protein were synthesized using PeptrEXTM (Peptron Inc., South Korea). The identity of the peptides was validated using mass spectrometric analysis, and their purity was conrmed using high-performance liquid chromatography. Table 1 lists the peptides used in the present study. The 18 15-mer HPV-16 E7 overlapping peptides were divided into six pools: pool 1, E7.1E7.3 (E7 aa 125); pool 2, E7.4E7.6 (E7 aa 1640); pool 3, E7.7E7.9 (E7 aa 3155); pool 4, E7.10E7.12 (E7 aa 4670); pool 5, E7.13E7.15 (E7 aa 6185); and pool 6, E7.16E7.18 (E7 aa 7698) (Table 1). The IFN- ELISPOT assay was performed using an IFN ELISPOT assay kit from BD Bioscience Inc. (San Diego, CA, USA). Briey, 5104 to 1105 in vitro-stimulated peripheral blood mononuclear cells (PBMCs) collected from
Table 1. 18 HPV-16 E7 15-mer overlapping peptides spanning the fulllength E7 protein. Pool number 1 Peptide name E7.1 E7.2 E7.3 E7.4 E7.5 E7.6 E7.7 E7.8 E7.9 E7.10 E7.11 E7.12 E7.13 E7.14 E7.15 E7.16 E7.17 E7.18 aa, amino acid. Position of amino acid E7 aa 115 E7 aa 620 E7 aa 1125 E7 aa 1630 E7 aa 2135 E7 aa 2640 E7 aa 3145 E7 aa 3650 E7 aa 4155 E7 aa 4660 E7 aa 5165 E7 aa 5670 E7 aa 6175 E7 aa 6680 E7 aa 7185 E7 aa 7690 E7 aa 8195 E7 aa 8698 Peptide sequence MHGDTPTLHEYMLDL PTLHEYMLDLQPETT YMLDLQPETTDLYCY QPETTDLYCYEQLND DLYCYEQLNDSSEEE EQLNDSSEEEDEIDG SSEEEDEIDGPAGQA DEIDGPAGQAEPDRA PAGQAEPDRAHYNIV EPDRAHYNIVTFCCK HYNIVTFCCKCDSTL TFCCKCDSTLRLCVQ CDSTLRLCVQSTHVD RLCVQSTHVDIRTLE STHVDIRTLEDLLMG IRTLEDLLMGTLGIV DLLMGTLGIVCPICS TLGIVCPICSQKP

Material and methods

From December 2008 to August 2010, 33 patients affected with CIN2/3 and invasive cervical cancer and exhibiting HPV-16 positivity, seven patients with CIN2/3 exhibiting high-risk HPV infection other than HPV-16, and nine healthy control donors were enrolled prospectively from the Department of Obstetrics and Gynecology of the Seoul St. Marys Hospital, the Catholic University of Korea, using a research protocol approved by our Institutional Review Board. All of the participants provided informed consent, and they underwent screening cervical cytologic tests and HPV genotyping, performed by an expert gynecologist. When initial cervical screening or colposcopy revealed suspicious cervical lesions, cervical biopsies under colposcopic guidance were performed. Nine women, ranging in age from 26 to 34 years, served as controls: six were young women who had not experienced sexual intercourse and three were women who had had intercourse but in whom there was no evidence of cervical cytological abnormalities or HPV infection. All infected women underwent standardized treatment for their disease, which included surgical procedures such as loop electrosurgical excision procedure (LEEP) and radical hysterectomy. Initial samples were collected at the time of diagnosis and follow-up samples were collected six months after diagnosis.

HPV genotyping
HPV genotyping was performed using the HPV DNA oligonucleotide chip (Mygene Co., Seoul, Korea), a PCRbased DNA microarray system which is approved by the Korea Food and Drug Administration. The HPV DNA chip contains 22 type-specic probes, including 15 in the highrisk group (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, 68, 69) and seven in the low-risk group (6, 11, 34, 40, 42, 43, 44). Specimens were prepared and tested according to the manufacturers instructions. Briey, DNA was isolated from cerviC

2011 The Authors

C

Acta Obstetricia et Gynecologica Scandinavica

2011 Nordic Federation of Societies of Obstetrics and Gynecology 90 (2011) 13501356

1351

Human papillomavirus and immune response

Y. S. Lee et al.

HPV-16-positive women and healthy controls were seeded in the wells of 96-well polyvinylidene uoride plates coated with an anti-human IFN- monoclonal antibody (Capture Antibody) at a concentration of 10 g/mL. HPV-16 E7 15-mer overlapping peptide matrix pools were added to the cells at a concentration of 10 g/mL and were incubated at 37C in the presence of 5% CO2 for over 12 hours. PBMCs were separated by density centrifugation using polysucrosesodium metrizoate. The cultures were then plated at a density of 5104 to 1105/well. The 18 15-mer HPV-16 E7 overlapping peptides were added to the PBMC cultures at a concentration of 10 g/mL and were incubated at 37C in the presence of 5% CO2. Cells were removed by washing three times with phosphate-buffered saline and 0.05% Tween-20. The captured IFN- was then detected using a biotin-conjugated anti-human IFN- monoclonal antibody, which was followed by incubation with enzyme-conjugated streptavidin. The spots that formed were developed by adding 3-amino-9ethyl-carbazole, and the reaction was stopped by rinsing the wells with water. Spots were quantied and the number of spots analyzed.

Table 2. Characteristics of the 33 HPV-16-positive patients and 7 CIN2/3 patients exhibiting other high-risk HPV infections. Patient no. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 33 32 34 35 36 37 38 39 40 Age 27 44 33 32 31 42 36 32 27 35 32 62 31 29 41 46 39 36 35 38 53 34 61 46 57 47 41 52 40 44 33 56 49 50 36 45 36 42 30 29 HPV type 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16, 53 16, 40 16, 31 16, 58 16, 70 16, 33, 35 16, 54, 58 16 16 16 16 16 16 16 16 16 16 16, 53 16 70 40, 44, 53 56, 58 58 31 31 53 Treatment LEEP LEEP LEEP LEEP LEEP LEEP LEEP LEEP LEEP LEEP LEEP LEEP LEEP LEEP LEEP LEEP LEEP LEEP LEEP LEEP LEEP R/H R/H R/H R/H R/H R/H R/H R/H R/H R/H R/H R/H LEEP LEEP LEEP LEEP LEEP LEEP LEEP Final diagnosis CIN2 CIN2 CIN2 CIN2 CIN2 CIN2 CIN3 CIN3 CIN3 CIN3 CIN3 CIN3 CIN3 CIN3 CIN2 CIN3 CIN3 CIN3 CIN3 CIN3 CIN3 ADC SCC SCC SCC SCC SCC SCC SCC SCC SCC SCC SCC CIN3 CIN2 CIN3 CIN3 CIN3 CIN2 CIN3

Statistical analysis
Data were evaluated using sas Windows version 13.0. Comparisons between experimental groups were performed using the independent Students t-test, and the MannWhitney test. All p-values reported are two-sided and signicance was set at p<0.05.

Results
The clinical characteristics, including HPV status, of the women are listed in Table 2. The mean age of women with HPV-16 infection was 39 years (range 2762). Among these, 21 with CIN2/3 had LEEP and 12 cancer women were treated with radical hysterectomy based on a standard treatment protocol. The median age of women with CIN2/3 and invasive cancer was 37 (range 3460) and 47 (range 2762) years, respectively (p<0.05). No other variables between the two groups were different. The analysis of the effect of the HPV infection status in all patients including the control women is presented in Figure 1. A signicant number of IFN- spot-forming cells were observed when the PBMCs were incubated (stimulated) with peptide pools 2 and 3 in the HPV-16-positive women. These results were not inuenced by HPV infection status (i.e. single or multiple) and there was no difference in CMI response to HPV-16 E7 peptide pool between HPV-16-positive women only and other type-positive women. The seven women who were infected with high-risk types other than HPV-16 and diagnosed with CIN2/3 underwent LEEP. They exhibited weak immune responses compared with HPV-16-positive women and did not show higher levels with pools 2 and 3.

SCC, squamous cell carcinoma; ADC, adenocarcinoma; R/H, radical hysterectomy; LEEP, loop electrosurgical excision procedure; CIN, cervical intraepithelial neoplasm.

There were no signicant differences in CMI response to synthetic HPV-16 E7 overlapping peptides between CIN2/3 and cervical cancer patients. The positive responses to pools 2 and 3 revealed by the IFN- ELISPOT assay suggested that the E7 sequence of aa 1655 is critical for the T-cell immune response. Although the absolute values were different individually, the results showed consistently higher levels for pools 2 and 3 in all women regardless of clinical variables (such as nal pathological diagnosis and stage of disease) and patient characteristics (such as age, parity and menopausal status).
C

2011 The Authors

1352

Acta Obstetricia et Gynecologica Scandinavica

2011 Nordic Federation of Societies of Obstetrics and Gynecology 90 (2011) 13501356

Y. S. Lee et al.

Human papillomavirus and immune response

Figure 1. Graphical presentation of the IFN- ELISPOT analysis used to characterize E7 peptide-specic T-cell responses in 21 HPV-16-positive CIN2/3 patients, 12 cervical cancer patients, seven CIN2/3 patients exhibiting other high-risk HPV infections, and nine noninfected healthy donors. SFC, spot-forming cells; NC, negative control; P, pool. center, mean; bars, SD. The mean number of SFCs of IFN- ELISPOT analysis to

positive control (phytohemagglutinin) were 238.71 (SD=27.09) in HPV16-positive CIN2/3 patients, 240.0 (SD=55.2) in HPV-16-positive cervical cancer patients, 312.0 (SD=29.1) in other types of HPV-positive patients and 264.6 (SD=32.5) and 264.6 (SD=32.5) in noninfected healthy donors (data not shown).

We performed another IFN- ELISPOT analysis for six (E7.4E7.9, pools 2 and 3, Table 1) of the 18 E7 peptides with specic T-cell responses in ve HPV-16-positive women randomly selected from the 33 HPV-16-positive women. They included three CIN and two invasive cancer patients (patients nos 7, 10, 16, 28 and 31). Among these six peptides, E7.6 and E7.9, which comprised aa 2640 and aa 4155, respectively, yielded relatively higher responses compared with other peptides; however, the differences between each peptide were not signicant (Supporting Information Figure S1). The follow-up samples were collected from the six patients selected from the 33 HPV-16-positive patients. They included three CIN and three invasive cancer patients (patients nos 7, 10, 16, 26, 28, and 31). The delay between the initial ELISPOT assay and the follow-up varied. Four of the six women were followed-up after a six-month interval, one was seen after an eight-month interval, and another after a 15-month interval. The IFN- ELISPOT analysis of the follow-up samples also showed the presence of consistently higher levels for pools 2 and 3 in all three women who were examined during the follow-up period (Figure 2). None of the women included

in the follow-up analysis showed recurrence of the disease during this period.

Discussion
Considerable efforts are being made to develop efcient vaccination strategies aimed at reducing the incidence of cervical cancer. The rationale for the therapeutic vaccines that stimulate CMI was suggested initially when an increased risk of HPV-associated disease was observed among immunocompromised women, in particular those with CMI deciency but with normal humoral immune function (13,14). Since then, many studies have shown other evidence that supports the relation between CMI and HPV (1517). The E7 protein, which has been studied most extensively, comprises several T-helper and cytotoxic T-cell epitopes that have been identied in both experimental animals and humans (1821). Recently, many researchers have reported that T-cell responses to specic peptides of HPV-16 are related to regression of HPV-associated lesions (18,2223). These results indicate that there is an association between signicant cellular immune responses specic to synthetic peptides from the E6

2011 The Authors

C

Acta Obstetricia et Gynecologica Scandinavica

2011 Nordic Federation of Societies of Obstetrics and Gynecology 90 (2011) 13501356

1353

Human papillomavirus and immune response

Y. S. Lee et al.

Figure 2. Follow-up results of the IFN- ELISPOT analysis to characterize E7 peptide-specic T-cell responses in six HPV-16-positive patients compared with initial data. SFC: spot-forming cells, NC: negative control, P: pool; center, mean; bars, SD.

and E7 oncoproteins of HPV-16, and disease-free survival in women with HPV-induced CIN. In contrast, the humoral responses against the E6 and/or E7 oncoproteins may not play a protective role against HPV-associated cervical neoplasms (24). Therefore, E6 and E7 can be regarded as key factors for the development of therapeutic vaccines. To develop effective vaccines, it is important to determine the most immunogenic portions of the E7 peptide sequence and to identify the immunological responses that are consistently relevant to the clinical outcomes. The IFN- ELISPOT assay performed in this study revealed that peptide pools 2 and 3 produce a relatively strong response, suggesting that the E7 sequence comprising aa 1655 is a major immunogenic region critical for the T-cell immune response. This result diverges from the regions previously reported by most investigators [aa 7185 (Peng et al., 23), aa 6798 (de Gruijl et al., 25) and aa 4172 (van der Burg SH et al., 19)] but is similar to the region reported by Kadish et al. (22) as a target for immune therapy (aa 3754). There are several possible explanations for these differences. First, ethnic differences should be considered. To the best of our knowledge, this study is the rst report on the cellular immune response against peptides of the HPV-16 E7 protein in Korea. The fact that the ethnic variability of specic immunological responses can be wider than expected was demonstrated previously by the Getinib (IRESSA) clinical trial for lung cancer. This trial revealed a much more favorable response rate in Asian populations (including Korean individuals) than in other ethnic groups (26,27). Secondly,

the relatively small number of specimens enrolled in each of these studies may also have contributed to the differences observed. For example, other factors, such as HLA type, may inuence results. Lastly, we cannot exclude the possibility of subtle variability among laboratory techniques used in the aforementioned studies, which may be attributed to the lack of standardized quantication methods of the cellular immune response. For instance, van der Burg et al. (19) used PBMC cultures, whereas de Gruijl et al. (25) estimated interleukin (IL)-2 production after peptide stimulation. Peng et al. (23) used an IFN- ELISPOT assay that was similar to the one described in our study. Whether the peptides from other high-risk types of HPV are homologous to E6/E7 is of interest for the development of vaccines. In our study, the seven women that contracted highrisk-type HPV other than HPV-16 exhibited weak immune response to HPV-16 E7 peptide pools, suggesting that there was not a signicant cross-reaction of the E7 peptide between HPV-16 and other HPV types detected in the seven women. Four of the seven women with specic HPV types that had a phylogenetic relation with HPV-16, such as HPV-31, 33, 35, 52, and 58, also did not show cross-reaction in our study (28). However, as very few patients with other HPV types were analyzed, it is difcult to comment on the generality of the responses we observed. Future studies should take into consideration the taxonomy of the viruses. There are few studies on the duration of immune memory or on the persistence of CMI response to HPV. Nakagawa et al. (29) reported that seven of eight women (88%) had detectable HPV-16 E6 and/or E7 CTL responses after a mean follow-up of 58 months. However, their study population was small and the authors reported on the CMI response of women with low-grade cervical lesions that did not need immediate treatment (such as surgical resection). In our study, the follow-up samples also exhibited a consistently higher level with pools 2 and 3; this was observed in all women included in the follow-up for at least six months after treatment. This suggests that the CMI response may be longer than expected in the absence of continuous stimulation of the immune system by HPV. It also suggests that the efcacy of therapeutic vaccines based on CMI may be sustained for quite some time. However, our study has some limitations. First, we did not analyze the human leukocyte antigen (HLA) complex and its correlation with immune response. An effective immune response may require optimal peptide presentation by both class I and class II molecules to activate efcient helper and effector T-cell responses to HPV. Subtle changes of the HLA complex are an important variable in the determination of the overall cellular immune response to HPV infection and, furthermore, CIN and cervical cancer susceptibility (30). We performed random HLA genotyping in 19 of the 33 women with HPV-16 infection, (Supporting Information Table S1).
C

2011 The Authors

1354

Acta Obstetricia et Gynecologica Scandinavica

2011 Nordic Federation of Societies of Obstetrics and Gynecology 90 (2011) 13501356

But in our study, each HLA haplotype seemed to be distributed indiscriminately and it was not possible to comment on any statistical relation of particular HLA haplotypes with ELISPOT results, presumably due to the small number of specimens analyzed. Moreover, we did not characterize the populations of T-immune cells (CD4+, CD8+), which would alsobehelpful tounderstand immune response toHPV. Thus it is important to further characterize the E7 peptide-specific T-cell immune response using more samples. In summary, our study showed that peptide pools 2 and 3 from HPV-16 E7 exhibited a relatively strong response, suggesting that the E7 sequence comprising aa 1655 is a major immunogenic region that is critical for the T-cell-mediated immune response in Korean women with CIN2/3 or cervical cancer. Many aspects of the immunologic response in HPV remain unclear. However, we hope our identification and characterization of CMI responses to HPV E7 peptide pools in Korean womenmay provide insight into novel therapeutic vaccine trials for the control of HPV-associated diseases.