Primary Culture And In Vitro Study Of Biological Behavior Of Human High-Grade Cervical Intraepithelial Neoplasia Keratinocytes

Cervical cancer is the most common gynecological cancer in developing countries and the incidence significantly decrease with the wide application of cervical screening test in recent years. It was found out that cervical cancer usually proceeded by a long phase of pre-malignant disease. These pre-malignant changes represent a spectrum of histological abnormalities ranging from cervical intraepithelial neoplasia (CIN) I (mild dysplasia) to CIN2(moderate dysplasia) to CIN3(severe dysplasia/carcinoma in situ). CINI will regress spontaneously in most cases and only0.3%may progress to cervical cancer, whereas CIN II and III will persist and progress to cervical cancer in about20%-45%of untreated CIN II/III lesions. Therefore CINII and CINIII are regarded as high-grade lesions. Fortunately, it is appears that the mean interval for progression of cervical precursors to invasive cancer is about10years, therefore we have enough time to detect and manage this precancerous lesion, and cervical cancer may be prevented subsequently. It is well established that almost all high-grade CIN cases are caused by persistent oncogenic human papillomavirus (HPV) infection and long latency period of viral infection is needed for CIN progression. However, viruses do not have their own metabolism, and require a host cell to make new products. Therefore, HPV does not culture in a laboratory outside a host cell. Culture of naturally HPV-infected CIN keratinocytes has significant value in exploring biological behavior of high-grade CIN in vitro and in vivo then, as well as the transforming process of neoplastic cervical keratinocytes.Studies on normal uterine cervix (NUC) and CIN keratinocyte cultures, as well as characteristics of CIN keratinocytes, are poorly available due to the numerous technical and methodological problems associated with the in vitro culturing of these cells. Dysplasia is characterized by four major pathological microscopic changes including cells of unequal size, abnormally shaped, hyperchromatism and presence of mitotic figures. The primary basis for assessing the specific grade of CIN is the ratio of nucleus to cytoplasm in a given cell, and higher ratios are associated with higher grades of CIN. However, it is very difficult to differentiate dysplastic cells from malignant cells in morphology to some extent. As we all known, loss of contact-dependent growth inhibition, migration and invasion ability are biological behavior of malignant cells, and we wonder if high grade CIN keratinocytes have such characteristics? Why such CIN keratinocytes are limited to intraepithelial and not broken through into the stroma? In this study, we plan to establish a novel method for culturing normal human cervical keratinocytes and naturally HPV-infected high-grade CIN keratinocytes starting from small pieces of neoplastic cervical tissues. Additionally, we intend to explore the in vitro biological behavior of high-grade CIN keratinocytes and provide an experimental basis for the further study of precancerous lesion of the cervix and cervical carcinoma.Part I:Establishment of a Novel Method for Primary Culture of Normal Human Cervical KeratinocytesObjectives:1. To investigate a modified culture medium which could improve cervical keratinocytes attachment and constrain the fibroblast growth as well.2. To explore a special culture plastics bottle which may help to improve cervical keratinocytes attachment.3. To explore a method which could obtain highly purified keratinocytes from normal human cervical epithelium in vitro.Methods 1. Normal cervical epithelial tissue pieces were obtained and digested with type I collagenase to dissociate the cells and a single cell suspension produced.2. The cells were seeded on the following culture plastics:(1) Uncoated plastic tissue culture bottles containing keratinocyte serum-free medium (K-SFM) without5%fetal bovine serum (FBS).(2) Uncoated plastic tissue culture bottles containing modified K-SFM supplemented with5%FBS.(3) Plastic tissue culture bottles coated with collagen type I from rat tail containing K-SFM without5%FBS.(4) Plastic tissue culture bottles coated with collagen type I from rat tail containing modified K-SFM supplemented with5%FBS.3. After attachment, the medium were replaced with K-SFM without FBS.4. The expression of basal keratins (K) of the ectocervical epithelium, K5, K14and K19were assayed by immunofluorescence with monoclonal antibodies to identify the cell purity.Results1. Cells attached to the culture plastic bottles more quickly in K-SFM supplemented with5%FBS than in K-SFM alone2. Cells attached to tissue culture plastic bottles coated with collagen type I more quickly than uncoated plastic bottles.3. The modified medium composed of K-SFM and5%FBS combined with a specific tissue culture plastic bottles coated with collagen type I from rat tail was the best method for culture of normal cervical epithelial cells.4. Diffuse and strong cytoplasmatic immunostaining for keratin14and19was obtained in nearly all cultured cervical epithelial cells. But these cells stained weakly for K5.Conclusions:1. In the modified medium, low concentration FBS could improve cervical keratinocyte attachment, whereas K-SFM may constrain the fibroblasts growth.2. The modified medium composed of K-SFM and5%FBS combined with a specific tissue culture plastic coated with collagen type I from rat tail was a novel, simple and effective method for culture of normal cervical epithelial cells in vitro. 3. We could rapidly obtain highly purified keratinocytes from normal human cervical epithelium in this study. Part II:Primary Culture and Identification of High-grade Cervical Intraepithelial Neoplasia Keratinocytes from Human Neoplastic Cervical BiopsiesObjectives:To investigated a novel method for the in vitro culturing of naturally HPV-infected high-grade CIN keratinocytes from small-sized human neoplastic cervical biopsies. To provide an experimental basis for different research purposes, particularly the further study of the biological characteristics of high-grade CIN keratinocytes.Methods:1. A small tissue fraction from a CIN II-III sample was excised for DNA extraction.2. Small pieces of neoplastic tissue were digested with0.2%type I collagenase and single cell suspensions were obtained.3. The isolated cells were seeded into T25tissue culture plastic bottles, which was coated overnight with2μg/cm2collagen type I from the rat tail, in medium containing K-SFM supplemented with5%FBS.4. On day3, the serum-supplemented K-SFM medium was replaced with K-SFM medium, and the unattached cells in group1were recovered, reseeded into new T25tissue culture plastic bottles coated with collagen type I from the rat tail, and allowed to attach for three days. The medium was replaced with K-SFM alone on day6, and the unattached cells in group2were discarded. The medium was changed twice a week.5. For the HPV genotyping of the sub-cultured CIN keratinocytes, a small number of the third-generation cells were subjected to DNA extraction.6. The expression of basal keratins of ectocervical epithelium including keratin K14and K19; markers for cervical stem cell including P63and K17were assayed by immunofluorescence with monoclonal antibodies to identify the presence of CIN keratinocytes.Results:1. Approximately20%,60%, and80%of the high-grade CIN keratinocytes were attached to the culture plastic bottles on days1,3, and6, respectively. Although the viability of the unattached cells on the third day was more than80%, only30%of the remaining cells attached to the new culture plastic bottles, and the attached cells were unable to survive well because of the low cell density in group1.2. The passaged CIN keratinocytes maintained their original unequally sized, abnormally shaped morphology and did not undergo differentiation.3. The passaged CIN keratinocytes exhibited the same HPV genotype detected in the original primary cells.4. K14and K19were found to be expressed in nearly all of the normal and CIN keratinocytes, whereas K17and P63were expressed only in high-grade CIN keratinocytes.Conclusions:1. We combine "explant tissue culture method" with "digestion culture method" in this study, and obtain plenty of cells.2. The6th day may be appropriate time for the first medium change for primary cultures of high-grade CIN keratinocytes.3. Our study proposes a simple and practical method for rapidly obtaining highly purified naturally HPV-infected high-grade CIN keratinocytes from small-sized neoplastic cervical tissues, and provides an experimental basis for the further study of the biological characteristics of high-grade CIN keratinocytes. Part Ⅲ In Vitro Study of Biological Behavior on Primary Cultured Human High-grade Cervical Intraepithelial Neoplasia Keratinocytes Objectives: Compared with NUC keratinocytes and Caski cells, we sought to explore the in vitro biological behavior of high-grade CIN keratinocytes including cell growth curve, cell cycle and DNA aneuploidy, expression of p16and Ki-67, and migration and invasion ability. This study could serve as a basis for the further study of cervical cancer and precancerous lesion of the cervix.Methods:1. High-grade CIN cells growth curve was measured with MTT, NUC keratinocytes and Caski cells were used as controls.2. Contact growth inhibition assay was used to determine the proliferation of high-grade CIN keratinocytes and its controls. When the cultures reached eighty percent confluence, the cells were cultured for further3-5days to test the existence of contact growth inhibition.3. Flow cytometry analysis was used to detect the cell cycles of high-grade CIN keratinocytes group and its controls.4. DNA ploidy analysis was used to detect the DNA aneuploidy among high-grade CIN keratinocytes group and its controls.5. Ki-67and p16expression were examined by immunofluorescence assays among high-grade CIN keratinocytes group and its controls.6. Competitive inhibition enzyme linked immunosorbent assay (ELISA) and solid phase ELISA was used to determine the concentrations of MMP-2and TIMP-2respectively among high-grade CIN keratinocytes group and its controls.7. Transwell assay was employed to test cell migration and invasion of high-grade CIN keratinocytes and the controls.8. Quantitative data were expressed as means±standard deviation (SD), the significance of the difference between groups was evaluated with analysis of variance (ANOVA). Cell growth curve was analyzed with repeated measures ANOVA. All calculations were performed using SPSS13.0statistical software.Results:1. Compared with NUC cells, the cells growth speed of high-grade CIN keratinocytes was significantly fast; however, it was significantly slower than Caski cells (P<0.001). After2days of incubation, high-grade CIN keratinocytes entered the logarithmic growth phase which was2days earlier than NUC keratinocytes. No apparent lag phase was observed in the growth curve of Caski cells.2. Cell contact growth inhibition was identified in high-grade CIN keratinocytes, and Caski cells exhibited a character of uncontrolled cell proliferation with loss of cell contact inhibition.3. Compared with NUC cells, the S phase fraction (SPF) of high-grade CIN keratinocyte group was significantly high, but was lower than Caski cell group (P＜0.01).4. DNA aneuploidy ratios of high-grade CIN keratinocyte group and Caski cell group were significantly higher than NUC keratinocyte group (P＜0.01). However, no significant difference was seen between high-grade CIN keratinocyte group and Caski cell group (P=0.392).5. Nuclear immunostaining for Ki-67was obtained in nearly all of the cultured high-grade CIN keratinocytes and the controls incuding NUC cells and Caski cells. In addition, p16were found to be expressed in nearly all of the high-grade CIN keratinocytes and the controls nuclear and cytoplasm.6. The concentration of extracelluar matrix metalloproteinase2(MMP-2) in Caski cell group was higher than high-grade CIN and NUC keratinocyte groups (P<0.05), but no significant difference was seen between high-grade CIN and NUC keratinocyte groups (P>0.05). Additionally, the concentrations of extracelluar tissue inhibitors of metalloproteinase2(TIMP-2), intracellular MMP-2and TIMP-2, and the ratio of MMP-2/TIMP-2were insignificantly different among high-grade CIN keratinocyte, NUC keratinocyte and Caski cell groups (P>0.05).7. Cell migration and invasion assays indicated that the migration ability of high-grade CIN keratinocytes was much higher than NUC keratinocytes, but it was significantly lower than Caski cells (P＜0.001). High-grade CIN keratinocytes had the ability of invasion, but the invasion power was much lower than Caski cells (P<0.001). NUC keratinocytes had no ability of invasion.Conclusions: 1. Proliferation ability of high-grade CIN keratinocytes was much higher than NUC keratinocytes, but cell contact growth inhibition was still remained.2. Plenty of DNA aneuploidy occurred during the course of proliferation of high-grade CIN keratinocytes and Caski cells, suggesting that high-grade CIN keratinocytes had the ability of high proliferation. Increased S phase fraction in cell cycle of high-grade CIN keratinocytes further supported this point.3. Ki-67and p16were expressed not only in high-grade CIN keratinocytes and Caski cells, but also in NUC keratinocytes, suggesting that both Ki-67and p16were probably only markers of cell proliferation.4. The relatively equal concentrations of extracelluar TIMP-2, intracellular MMP-2and TIMP-2among high-grade CIN keratinocyte group and it controls, the lower concentrations of extracelluar MMP-2and the mild ability of invasion in high-grade CIN keratinocytes, suggested that extracelluar MMP-2play an important role in the course of invasion. However, the basal membranes still keep intact in high-grade CIN lesions. Our study suggested that during the progression from carcinoma in situ to invasive cancer, the interaction of parenchyma and stroma play an important role in addition to to the malignant potential of the cell itself.