Germ cell-specific heat shock protein 70-2 is expressed in cervical carcinoma and is involved in the growth, migration, and invasion of cervical cells

Authors


Abstract

BACKGROUND:

Cervical cancer is a major cause of death among women worldwide, and the most cases are reported in the least developed countries. Recently, a study on DNA microarray gene expression analysis demonstrated the overexpression of heat shock protein 70-2 (HSP70-2) in cervical carcinoma cells (HeLa). The objective of the current study was to evaluate the association between HSP70-2 expression in cervical carcinogenesis and its potential role in various malignant properties that result in disease progression.

METHODS:

HSP70-2 expression was examined in various cervical cancer cell lines with different origins and in clinical cervical cancer specimens by reverse transcriptase-polymerase chain reaction (RT-PCR), flow cytometry, and immunohistochemistry (IHC) analyses. A plasmid-based, short-hairpin RNA approach was used specifically to knock down the expression of HSP70-2 in cervical tumor cells in vitro and in vivo to examine the role of HSP70-2 on various malignant properties.

RESULTS:

RT-PCR and IHC analyses revealed HSP70-2 expression in 86% of cervical cancer specimens. Furthermore, knockdown of HSP70-2 expression significantly reduced cellular growth, colony formation, migration, and invasion in vitro and reduced tumor growth in vivo. A significant association of HSP70-2 gene and protein expression was observed among the various tumor stages (P = .046) and different grades (P = .006), suggesting that HSP70-2 expression may be an indicator of disease progression.

CONCLUSIONS:

The current findings suggested that HSP70-2 may play an important role in disease progression in cervical carcinogenesis. Patients who had early stage disease and low-grade tumors had HSP70-2 expression, supporting its potential role in early detection and aggressive treatment modalities for cervical cancer management. Cancer 2010. © 2010 American Cancer Society.

Heat shock protein 70-2 (HSP70-2) is a germ cell-specific protein and is a member of the HSP70 chaperone protein family, which plays an important role in protein-protein interactions that result in the proper folding, confirmation, transport, and assembly of proteins in the cytoplasm, mitochondria, and endoplasmic reticulum.1, 2 It is noteworthy that HSP70-2 is expressed at high levels in testis with no or undetectable expression in other somatic tissues,3, 4 and it plays an important role in the progression of meiosis in the germ cells during spermatogenesis.5 Recently, HSP70-2 overexpression was associated with cell survival and cellular growth in various cancer cell lines.6 In addition, a recent DNA microarray gene expression analysis reported the overexpression of HSP70-2 in cervical carcinoma cells (HeLa).6 Therefore, we believed it was important to investigate HSP70-2 expression in patients with cervical cancer and to examine its role in various malignant properties of cancer cells.

Cervical cancer is the major cause of death in women worldwide, and most cases are reported in less developed countries because of limited awareness and the absence of medical support.7 Furthermore, it has been well documented that, if they are diagnosed early, then patients with cervical cancer have better treatment modalities available and a higher survival rate,8 whereas patients who are diagnosed with advanced-stage disease have poor survival rates.9 Therefore, it is important to identify and characterize the tumor-specific proteins involved in early stages and grades of cervical cancer that may contribute to cervical carcinogenesis.

In the current study, we investigated the relation of HSP70-2 expression to various early stages and different grades of cervical squamous cell carcinoma (SCC). We also used a short-hairpin RNA (shRNA) approach to selectively knock down the expression of HSP70-2 in various cervical tumor cell lines and investigate its potential role in cellular motility and growth, which are key characteristics of disease progression.

MATERIALS AND METHODS

Patient Specimens

Tissue specimens from 76 patients who were diagnosed with cervical SCC were obtained from Safdarjung Hospital and Vardhman Mahavir Medical College under a protocol approved by the institutional ethics committees. Matched, adjacent noncancerous tissue (ANCT) specimens also were collected and were available from only 60 cancer patients. Tumor tissues were stored in both formalin and RNALater and were used for immunohistochemistry and reverse transcriptase-polymerase chain reaction (RT-PCR) analysis. Informed consent was obtained from each patient.

Cell Culture

Human cervical SCC cell lines (SiHa, CaSki, and C-33 A) were gifts from Dr. Nicholas Denko of Stanford University School of Medicine (Stanford, Calif), and HeLa cervical adenocarcinoma cells were purchased from the American Type Culture Collection (Rockville, Md). All 4 cell lines were cultured in Dulbecco modified Eagle medium supplemented with 10% heat-inactivated fetal calf serum, 50 mg/mL gentamycin, and 100 mg/mL streptomycin (Invitrogen, Carlsbad, Calif) and were maintained at 37°C in a humidified atmosphere with 5% CO2. All cells were transfected using Lipofectamine 2000 (Invitrogen).

RT-PCR Analysis

Total RNA was isolated from cervical cancer tumor specimens and from all cervical cancer cell lines (SiHa, CaSki, C-33 A and HeLa) by using TRI reagent. Subsequently, complementary DNA (cDNA) was prepared by using the FastLane Cell cDNA Kit (Qiagen, Hilden, Germany) as described previously.10HSP70-2 gene-specific primers were used for RT-PCR reactions (HSP70-2 forward primer, CCTACTCGGACAACCAGAG; HSP70-2 reverse primer, TCTCGTC TTCCACCGTCTG). PCR product was cloned into TOPO vector (Invitrogen) for confirming the HSP70-2 nucleotide sequence by DNA nucleotide sequencing.

Generation of Antibody Against HSP70-2

The primary antibodies that were used included murine monoclonal antibodies against HSP70 and Hsc70 (C92FBA-5 and N27F3-4; Calbiochem, San Diego, Calif). Antibody against HSP70-2 was generated by immunizing a rabbit with keyhole limpet hemocyanin and ovalbumin-conjugated NH2-SKLYQGGPGGGGSSGGPT peptide as described previously.2 Immunoglobulin G (IgG) from rabbit serum was isolated using a Nab protein G spin chromatography kit, and the concentration of purified IgG was determined using the bicinchoninic acid method as described previously.11 The purified IgG (1 mg/mL concentration) was used as an anti-HSP70-2 antibody for all experiments.

Indirect Immunofluorescence Assay

HSP70-2 protein expression was examined in all cervical cancer cell lines using anti-HSP70-2 antibody or a control IgG by immunofluorescence assay as described previously.12 Nuclear staining of cells was done using 4′6-diamidino-2-phenylindole (Sigma-Aldrich, St. Louis, Mo). The slides were mounted in antifade reagent (Molecular Probes/Invitrogen). The photomicrographs were captured using an ECLIPSE, E 400 Nikon microscope (Nikon, Fukok, Japan). Each experiment was repeated 3 times, and the fluorescence images were merged using Image-Pro Plus version 5.1 (Media Cybernetics, Inc., Bethesda, Md).

Flow Cytometric Analysis

To detect surface expression, flow-cytometric analyses were carried out in all cervical cancer cell lines as described previously.12 Briefly, cells were harvested and fixed with 0.4% paraformaldehyde. Next, the cells were incubated with anti-HSP70-2 antibody or control IgG followed by goat antirabbit IgG fluorescein isothiocyanate conjugate. Then, the cells were resuspended in phosphate-buffered saline and analyzed with a flow cytometer (BD-LSR model; Becton Dickinson, San Jose, Calif). Data acquisition and analysis was done using WinMDI software (version 2.8; Joseph Trotter, The Scripps Research Institute, Jupiter, Fla).

Immunohistochemistry

Endogenous expression of HSP70-2 protein was evaluated on 4-μm, paraffin-embedded, serial sections of cervical cancer tissues and ANCT by using anti-HSP70-2 antibody or control IgG as described previously.13 HSP70-2 immunostaining was examined by counting >500 cells from 5 random fields of each specimen under ×400 magnification in a stained tumor area from each section as described previously.14 Immunoreactivity was considered distinct and positive in specimens in which >10% of cancer cells were immunoreactive for HSP70-2.

HSP70-2–Specific Small Interfering RNA

The HSP70-2-specific shRNA (Sure Silencing shRNA plasmids) constructs were purchased from Super Array (Frederick, Md). The following target sequences were used: shRNA1, CAT AAC GGT CCC GGC CTA TT; shRNA2, GAG CGG TAC AAA TCG GAA GAT; shRNA3, CGG CGA CAA ATC AGA GAA TGT; and shRNA4, TTC GAC GCC AAG AGG CTG ATT. These sequences were designed from the Gene Bank accession number NM_021979 [gene symbol HASPA2], and the negative control (NC) sequence was GGAATCTCATTCGATGCATAC.

Transfection and Western Immunoblot Analysis

Cervical cancer cells were transfected with all 4 HSP70-2–specific shRNA targets and 1 NC using lipofectamine (Invitrogen). After 48 hours of transfection, the cells were harvested and lysed in lysis buffer containing protease inhibitor cocktail (Sigma-Aldrich), and protein concentrations were determined. For further analysis, 40 μg of each cell line lysate were resolved on 10% sodium dodecyl sulfate-polyacrylamide electrophoresis gels and transferred to polyvinylidene difluoride membranes (Hybond-P; Amersham Biosciences, Cowley, United Kingdom) as described previously.10 After protein transfer, the membranes were blocked and probed for HSP70-2, HSP70, and HSC70 using anti-HSP70-2 antibodies or murine monoclonal antibodies against HSP70 and heat shock cognate 70 (HSC70) for 2 hours followed by incubation with horseradish peroxidase-conjugated secondary antibody. Immunoreactive bands were observed using a chromogen (0.05% 3,3′-diaminobenzidine; Sigma-Aldrich).

Cell Invasion, Migration, and Wound-Healing Assay

Cell invasion, migration, and wound-healing assays were performed to validate the role of HSP70-2 in invasiveness and cell motility of SiHa, CaSki, C-33 A, and HeLa cells in vitro as described previously.12 Briefly, in the invasion assay, 24-hour post-transfection with HSP70-2 shRNA3 or control NCshRNA, cells were trypsinized, and 2 × 104 cells per filter were added to 8-μm pore inserts that were coated with 15 μg Matrigel (Becton Dickinson Labware, Bedford, Mass) in triplicate wells. After 24 hours, the cells that passed through the filter into the lower chamber were fixed with 5% glutaraldehyde, stained with 0.5% toluidine blue, and quantified in 5 areas under light microscopy. In the migration assay, 8-μm pore inserts were used without Matrigel, and each experiment was done in triplicate as described previously.12

The wound-healing assay was performed as described previously.12 Briefly, 24 hours after transfection with HSP70-2 shRNA3 or control NCshRNA, cells were seeded at a density of 2 × 106 cells per 35-mm Petri dish. After 12 hours of incubation, a wound was created using an aerosol P200 pipette tip, and photomicrographs were taken immediately (time 0 hours) and thereafter at 12 hours, 24 hours, and 48 hours to observed the effects on wound healing.

Cell Proliferation Assay

SiHa, CaSki, C-33 A, and HeLa cells that were transfected with HSP70-2 shRNA3 or control NCshRNA were seeded (2 × 104 cells) in 35-mm Petri dishes. After 24 hours, 48 hours, and 72 hours of transfection, the cells were trypsinized and counted with 0.4% Trypan blue using a hemocytometer.12

Colony-Formation Assay

Transfected SiHa, CaSki, C-33 A, and HeLa cells with HSP70-2 shRNA3 or control NCshRNA were seeded into 35-mm Petri dishes with 400 to 1200 cells per plate in triplicate experiments and were stored at 37°C in a humidified incubator (5% CO2) for 7 days. After 7 days, cell colonies were fixed with glutaraldehyde and stained with toluidine blue, and, after staining, we counted the number of cells in 5 random fields.12

Xenograft Experiments

Athymic nude mice (National Institute of Immunology [NII], National Institutes of Health, [S] nu/nu) were inoculated subcutaneously on the lower back with 5 × 106 SiHa cells suspended in Matrigel collagen basement membrane (BD Biosciences, Bedford, Mass). These nude mice were maintained at an NII animal facility in a pathogen-free atmosphere. Fifteen days after injection, the tumors averaged in size from 50 mm3 to 100 mm3. Mice that had developed subcutaneous xenograft tumors of that measured ∼100 mm3 were divided into 2 treatment groups with 8 mice per group: 1) HSP70-2 shRNA3 and 2) control NCshRNA. Plasmids (50 μg) suspended in 300 μL of phosphate-buffered saline were injected into the mouse tail vein followed by a booster injection of 25 μg plasmid twice weekly for 7 weeks. The in vivo response was evaluated based on the changes in tumor volume over time. Tumor volume (V) was calculated according to the following formula: V = π/6 × larger greatest dimension × (smaller greatest dimension)2 by measuring tumor dimensions before each injection using digital calipers.12 Tumors were excised for immunohistochemical analyses 1 week after the last treatment.

Statistical Analysis

The Pearson chi-square test and the Student t test for unpaired data were used in the SPSS 16.0 statistical software package (SPSS Inc, Chicago, Ill). The results are expressed as the mean ± standard error. All P values <.05 were considered statistically significant.

RESULTS

Analysis of HSP70-2 Messenger RNA Expression

HSP70-2 messenger RNA (mRNA) expression in 4 cervical cancer cell lines, in 76 specimens of cervical SCC, and in 60 matching ANCT specimens was investigated using RT-PCR (Fig. 1a). HSP70-2 mRNA expression was observed in all cervical cancer cell lines, in tumor specimens, and in human testis specimens (positive control), but not in ANCT specimens. Figure 1a indicates that the PCR product was same size in cervical cancer cell lines, in cervical SCC tumor specimens, and in testis specimens. Furthermore, the PCR-amplified product was cloned and confirmed by nucleotide sequencing. No mutation was observed in the HSP70-2 nucleotide sequence. HSP70-2 mRNA expression was observed in 3 of 5 stage IA specimens (60%), in 24 of 31 stage IB specimens (77%), 18 of 20 stage IIA specimens (90%), and in 20 of 20 stage IIB specimens (100%) from patients with cervical SCC, as shown in Table 1. Furthermore, HSP70-2 mRNA expression was detected in 12 of 19 grade 1 tumors (63%), in 41 of 44 grade 2 tumors (93%), and in 12 of 13 grade 3 tumors (92%) from patients with cervical SCC (Table 1). A significant association with HSP70-2 gene expression was observed among the various tumor stages (P = .046) and grades (P = .006) using the Pearson chi-square test.

Figure 1.

Heat shock protein 70-2 (HSP70-2) expression was analyzed in cervical carcinoma specimens and cervical cancer cell lines. (a) HSP70-2 messenger RNA was detected by reverse transcriptase-polymerase chain reaction analysis in representative samples of cervical carcinoma from patients with (Lane 1) stage IA, (Lane 2) stage IB, (Lane 3) stage IIA, and (Lane 4) stage IIB disease; in the cervical carcinoma cell lines (Lane 5) SiHa, (Lane 6) CaSki, (Lane 7) C-33 A, and (Lane 8) and HeLa; and (Lane 9) in testis (positive control). HSP70-2 gene expression was not observed in a representative adjacent noncancerous tissue sample (Lane 10). β-Actin gene amplification was used as an internal control. Lane M is a molecular size marker. (b) A predominant cytoplasmic localization of HSP70-2 protein in fixed and permeabilized SiHa, CaSki, C-33 A, and HeLa cells is shown. Nuclei were stained blue by 4′6-diamidino-2-phenylindole. (c) No staining was detected with control immunoglobulin G (IgG). (d) Flow cytometric analysis demonstrated distinct displacement of fluorescence on the x-axis in cervical cancer cell that were treated with anti-HSP70-2 antibodies, indicating the surface localization of HSP70-2 protein in live SiHa, CaSki, C-33 A, and HeLa cells (green histogram) compared with cells that were stained with control IgG (black histogram) or secondary antibody only (red histogram). Results from 1 of 3 representative experiments are shown.

Table 1. Detailed Clinicopathologic Characteristics of Patients and Heat Shock Protein 70-2 Expression: Reverse Transcriptase-Polymerase Chain Reaction and Immunohistochemical Analyses
Clinicopathologic CharacteristicNo. RT-PCR/IHC- Positive/No. Tested (%)Pa
  • RT-PCR indicates reverse transcriptase-polymerase chain reaction; IHC, immunohistochemistry; ANCT, adjacent noncancerous tissue.

  • a

    Pearson chi-square test.

All tumors65/76 (86) 
ANCT0/60 (0) 
Tumor stage  
 I [IA+IB]27/36 (75) 
 II [IIA+IIB]38/40 (95) 
 IA03/05 (60) 
 IB24/31 (77) 
 IIA18/20 (90) 
 IIB20/20 (100) 
Lymph node involvement  
 Negative52/61 (85) 
 Positive13/15 (87) 
Histopathologic grade  
 112/19 (63) 
 241/44 (93) 
 312/13 (92) 
Statistical analysis  
 Stages I and II .013
 Stages IA, IB, IIA, and IIB .046
 Grades 1, 2, and 3 .006

HSP70-2 Protein Expression in Cervical Cancer Cells and Cervical Cancer Specimens

HSP70-2 protein expression was investigated in SiHa, CaSki, C-33 A, and HeLa cells by using indirect immunofluorescence and flow-cytometric analysis with antibodies against HSP70-2. HSP70-2 protein was localized predominantly in the cytoplasm of fixed and permeablized cervical cancer cells (Fig. 1b). No staining was detected with control IgG, as indicated on Figure 1c. Furthermore, surface expression of HSP70-2 protein was confirmed using flow cytometry. Flow-cytometric analysis revealed the distinct displacement of fluorescence on the x-axis in cervical cancer cell lines that were treated with anti-HSP70-2 antibodies, indicating the surface localization of HSP70-2 protein in all cancer cell lines (Fig. 1d, green histogram). Cells that were probed with control IgG (Fig. 1d, black histogram) or secondary antibody (Fig. 1d, red histogram) revealed no or very low surface distribution.

Furthermore, HSP70-2 protein expression was analyzed in cervical cancer specimens, ANCT specimens, and human testis specimens by using immunohistochemistry. The human testis specimens had distinct and intense immunoreactivity against HSP70-2 in spermatocytes and spermatids (data not shown). In cervical cancer specimens (but not in ANCT specimens), HSP70-2 protein expression was observed in 65 of 76 patients (86%) irrespective of disease stage or tumor grade (Fig. 2). Most reactivity was heterogeneous and was localized in the cytoplasm of tumor cells in the tissue sections. However, no HSP70-2 reactivity was observed with control IgG in serial sections. HSP70-2 protein expression was detected in 60% of stage IA specimens (3 of 5 patients), in 77% of stage IB specimens (24 of 31 patients), in 90% of IIA specimens (18 of 20 patients), and in 100% of stage IIB specimens (20 of 20) of cervical SCC, as indicated in Table 1. For different tumor grades, HSP70-2 protein expression was detected in 12 of 19 grade 1 tumors (63%), in 41 of 44 of grade 2 tumors (93%), and in 12 of 13 grade 3 tumors (92%) (Table 1). A significant association with HSP70-2 protein expression was observed among the various tumor stages (P = .046) and grades (P = .006) using the Pearson chi-square test.

Figure 2.

Immunohistochemical staining of heat shock protein 70-2 (HSP70-2) protein is seen in specimens from patients with cervical cancer and in adjacent noncancerous tissue (ANCT) specimens. Representative hematoxylin and eosin (H&E)-stained sections of (a) stage IA, (e) stage IB, (i) stage IIA, and (m) stage IIB tissues are shown. HSP70-2 cytoplasmic localization was observed in (b) stage IA, (f) stage IB, (j) stage IIA, and (n) stage IIB tissues using anti-HSP70-2 antibody. (c,g,k,o) No immunoreactivity was observed in serial tissue sections from these patients when they were probed with control immunoglobulin G (IgG). (d,h,l,p) ANCT specimens failed to demonstrated any reactivity when they were probed with anti-HSP70-2 antibody (original magnification, ×400).

HSP70-2 Down-Regulation Affects Cellular Proliferation and Colony Formation in SiHa Cells

To examine the potential role of endogenous HSP70-2 expression on various malignant properties of cervical cancer cell lines, we used shRNA methodology. Four HSP70-2–specific shRNA targets and 1 control NCshRNA target were transfected with SiHa, CaSki, C-33 A, and HeLa cells. The suppression of HSP70-2 expression was confirmed by Western blot analyses in transfected cells (Fig. 3a). HSP70-2 shRNA3 caused approximately >80% suppression in SiHa cells, 79% suppression in CaSki cells, 77% suppression in C-33 A cells, and 78% suppression in HeLa cells. HSP70-2 shRNA4 caused approximately >80% suppression in SiHa cells, 79% suppression in CaSki cells, 66% suppression in C-33 A cells, and 75% suppression in HeLa cells (Fig. 3b). However, cells that were transfected with control NCshRNA had no effect on HSP70-2 expression (Fig. 3a). Furthermore, the specificity of these targets was investigated in 2 other members of the HSP70 protein family. HSP70-2 shRNA-specific targets did not produce the ablation of HSP70 and Hsc70 proteins, indicating that the HSP70-2 shRNA targets were specific to HSP70-2 testis-specific protein only (Fig. 3a). Hence, the subsequent experiments were restricted to HSP70-2 shRNA3. Figure 4a shows that all cervical cancer cell lines transfected with HSP70-2 shRNA3 had a reduction in cellular growth after 72 hours (P = .001; SiHa cells, 66%-70%; CaSki cells, 68%-72%; C-33 A cells, 67%-70%; and HeLa cells, 64%-70%). Similar results were obtained with the colony-formation assay, which revealed significant suppression in HSP70-2 shRNA3-transfected cells but not in control NCshRNA-transfected cells (P < .0001). Cells that were treated with HSP70-2 shRNA3 revealed a significant reduction in colony-formation ability for various numbers of seeded SiHa cells (range, 68%-70% for 400-1000 cells) (Fig. 4b), CaSki cells (range, 70%-75% for 400-1000 cells) (Fig. 4b), C-33 A cells (range, 66%-70% for 400-1000 cells) (Fig. 4b), and HeLa cells (range, 68%-71% for 400-1000 cells) (Fig. 4b). These results collectively support the hypothesis that HSP70-2 plays a role in cellular growth and colony-forming ability in cervical cancer cells.

Figure 3.

The silencing of heat shock protein 70-2 (HSP70-2) expression in SiHa, CaSki, C-33 A, and HeLa cells is shown. (a) HSP70-2 short-hairpin RNA3 (shRNA3) and HSP70-2 shRNA4 demonstrated ablation of HSP70-2 protein expression compared with negative control (NC) shRNA. The levels of actin as detected by anti-β-actin antibodies were used as loading controls. No effect of silencing of HSP70-2 was observed on HSP70 or heat shock cognate 70 (HSC70) protein expression. (b) A representative histograph shows the HSP70-2 protein level expressed as the ratio of HSP70-2 to β-actin. Densitometric analysis revealed a reduction in HSP70-2 protein expression in HSP70-2 shRNA3-transfected and HSP70-2 shRNA4-transfected cells compared with control NCshRNA.

Figure 4.

Knockdown of heat shock protein 70-2 (HSP70-2) inhibits in vitro tumorigenicity of cervical cancer cell lines. (a) Decreased HSP70-2 protein by HSP70-2 short hairpin RNA3 (shRNA3) reduces the cellular growth of SiHa, CaSki, C-33 A, and HeLa cells compared with negative control (NC) shRNA. (b) Inhibition of HSP70-2 protein expression decreases the colony formation ability of SiHa, CaSki, C-33 A, and HeLa cells transfected with HSP70-2 shRNA3 compared with NCshRNA. (c) A migration assay representing cells that have migrated through the membrane is shown. SiHa, CaSki, C-33 A, and HeLa cells transfected with HSP70-2 shRNA3 demonstrate that significantly fewer cells migrated compared with NCshRNA-transfected cells. (d) HSP70-2 shRNA inhibits tumor cell invasion through Matrigel-coated Transwell filters. SiHa, CaSki, C-33 A, and HeLa cells transfected with HSP70-2 shRNA3 or control NCshRNA were seeded onto Transwell filters coated with Matrigel. Subsequently, the number of cells that invaded through the Matrigel were stained with toluidine blue and counted. The number of invading cells was significantly lower in HSP70-2 shRNA3-transfected SiHa cells compared with control NCshRNA-transfected cells. The differences in growth ratio, colony formation, migration, and invasion between control NCshRNA-transfected and HSP70-2 shRNA3-transfected SiHa, CaSki, C-33 A, and HeLa cells achieved statistical significance. A single asterisk indicates P < .001; double asterisks, P < .0001 (Student t test). Representative results from 3 experiments are shown. Columns represent the mean of 3 independent experiments; bars, standard error.

HSP70-2 Is Essential for Cellular Motility and Invasion

The hallmark of cancer metastasis is the cellular migration and invasion of cancer cells from the primary site to distant organs. It is believed that the migration and invasion of cancer cells through basement membranes is an important event in the cascade of metastasis. In this context, we investigated the effect of inhibiting HSP70-2 expression on the migration and invasion of cervical cancer cells in vitro. In the migration assay, all cervical cancer cell lines that were transfected with HSP70-2 shRNA3 had significantly reduced cell motility (P < .0001; SiHa cells, 71% reduction; CaSki cells, 72% reduction; C-33 A cells, 70% reduction; HeLa cells, 70% reduction) compared with control NCshRNA-transfected cells (Fig. 4c). Similar results were obtained in the invasion assay. All cervical cancer cell lines that were transfected with HSP70-2 shRNA3 displayed a significantly reduced ability to invade (P < .0001; SiHa cells, 70% reduction; CaSki cells, 70% reduction; C-33 A cells, 68% reduction; HeLa cells, 70% reduction) compared with control NCshRNA-transfected cells (Fig. 4d). Furthermore, cell mobility was assessed by using a wound-healing assay, and we observed that cell motility was greatly reduced in all cervical cancer cell lines transfected with HSP70-2 shRNA3 and failed to close the wound even after 48 hours. Representative images in Figure 5a illustrate the reduction in the motility for SiHa cells only. In contrast, the cells that were transfected with control NCshRNA produced complete wound closing within 24 hours. These results collectively suggest that HSP70-2 may be involved in the cellular motility and invasiveness of cervical cancer cells that are key features of the early spread of cancer.

Figure 5.

(a) A wound-healing migration assay is shown. The healing of wounds by migrated SiHa cells that were transfected with heat shock protein 70-2 (HSP70-2) short-hairpin RNA 3 (shRNA3) or with negative control (NC) shRNA (NCshRNA) at 0 hours, 12 hours, 24 hours, and 48 hours was imaged. SiHa cells transfected with HSP70-2 shRNA3 did not show the closing of the wound, even after 48 hours. In contrast, SiHa cells transfected with NCshRNA closed the wound after 24 hours. Representative results from 3 experiments are shown. (b) The effect of HSP70-2 knockdown on an SiHa cell tumor xenograft is shown. The macroscopic appearance of the tumors on Day 49 after the subcutaneous inoculation of SiHa cells is shown. Note the formation of a large, subcutaneous tumor mass in control NCshRNA-treated mice compared with HSP70-2 shRNA3-treated mice. (c) Tumor growth curves in nude mouse xenograft models are shown. The volume of subcutaneous tumors was significantly greater in control NCshRNA-treated mice than in HSP70-shRNA3-treated mice on Day 49 (P < .0001). (d) This representative histograph of tumor weight reveals a significant reduction in HSP70-2 shRNA3-treated mice (P < .0001) compared with control NCshRNA-treated mice, consistent with tumor volume data. Points indicate mean; bars, standard error. Double asterisks indicate P < .0001 (Student t test). (e) Knockdown of HSP70-2 expression is shown. A representative photomicrograph reveals reduced immunoreactivity (indicated by brown areas) in HSP70-2 shRNA3-treated tumors compared with control NCshRNA-treated tumors (original magnification, ×400).

shRNA-Mediated Knockdown of HSP70-2 Reduces Tumorigenic Potential of SiHa Cervical Cells In Vivo

Next, we attempted to determine whether the results derived from in vitro systems could be translated into an in vivo model system. Cervical cancer (SiHa) cells were injected subcutaneously into mice, and tumorigenic potential was monitored for 7 weeks. Athymic mice that bore tumor volumes from 50 mm3 to 100 mm3 were treated with HSP70-2 shRNA3 plasmid or with control NCshRNA plasmid twice weekly and were observed regularly for 49 days. HSP70-2 shRNA3 administered to mice produced a significant reduction in tumor volume (at Day 49; P < .0001) (Fig. 5b), whereas tumors grew linearly with time in mice that were injected with control NCshRNA (Fig. 5c). Neither HSP70-2 shRNA3 nor control NCshRNA had a toxic effect, and none of the mice died during the treatment period. Consistent with tumor volume data, the average wet weight of the tumor was reduced significantly (P < .0001) in HSP70-2 shRNA3-treated mice compared with control NCshRNA-treated mice (Fig. 5d). We also confirmed the effect of HSP70-2 shRNA3 by immunohistochemical analysis in serial tumor sections from both groups. In addition, HSP70-2 protein expression was reduced in HSP70-2 shRNA3-treated mice compared with control NCshRNA-treated mice, as shown in Figure 5e. These results suggest that HSP70-2 plays an essential role in the growth of cervical tumor cells in vivo.

DISCUSSION

Cancer prevention is an important field of medicine. Despite significant developments in advanced surgical practices and screening procedures for cervical cancer, there is no effective screening program or preventive strategy in developing countries to lower its burden.15 In less developed countries like India, cytology-based screening programs have not been as successful as they are in developed countries, because no medical infrastructure is available.15 Alternative and novel biomarkers for detecting cervical cancer need to be investigated to develop more efficient and affordable screening methods that can detect cervical cancer early and facilitate the timely treatment of this morbidity causing cancer. In addition, new diagnostic tools with novel therapeutic strategies, which may involve the development of molecular-targeted agents, antibody therapy, and cancer vaccines for proper treatment modalities, should be explored. Therefore, we initiated the current study of HSP70-2 expression in cervical cancer. To best of our knowledge, this is the first report of HSP70-2 expression in patients with early stages and various grades of cervical SCC and its involvement in malignant properties of cervical carcinoma cells.

Several cervical cancer biomarkers, including SCC antigen, serum cytokeratin fragment 21.1, and 2 mucins (carbohydrate antigen [CA] 15-3 and CA 125), have been discovered9; however, the clinical application and practical use in clinical routine of these tumor-associated antigens remain controversial.16 Thus, we carried out the current study to examine HSP70-2 expression in patients and to investigate the association between HSP70-2 expression and clinicopathologic characteristics in patients with cervical SCC. Our results suggest that the expression of HSP70-2 plays a potential role in the development of cervical cancer, and it may be a novel diagnostic marker for cervical SCC. Previous studies have reported the overexpression of HSP70-2 in a cervical carcinoma cell line (HeLa) and its association with cell survival and cellular growth in various other cancer cell lines.6 HSP70-2 was expressed in considerably greater numbers of patients with SCC (86%), especially in early stages of cancer (stage I, 75%; stage II, 95%). We observed that HSP70-2 was expressed in cervical cancer cells of different origin and in clinical cervical SCC specimens at both the transcriptional and translational levels. Furthermore, statistical analyses revealed a significant association between HSP70-2 expression and the various disease stages and tumor grades in patients with cervical cancer, suggesting that, as disease progression is initiated, HSP70-2 expression increases with increased stage and grade. Therefore, HSP70-2 may serve as a biomarker for early detection and progression of cervical cancer.

In addition to establishing HSP70-2 expression in cervical cancer tissues and cell lines, next, we evaluated the outcome of HSP70-2 knockdown in cervical cancer cells in various steps of carcinogenesis. We demonstrated that HSP70-2 protein was involved with the various malignant properties of cervical cancer cells. We observed its potential contribution in the cellular proliferation and colony-formation ability of cervical carcinoma cells and confirmed its role as a therapeutic target in the treatment of cervical cancer. Plasmid-based shRNA uses short RNA duplexes of defined sequences to silence a targeted gene and is an important technique in knocking down gene expression and studying gene function. For our study, we used specific shRNA against HSP70-2 gene and demonstrated decreased HSP70-2 protein expression levels in SiHa, CaSki, C-33 A, and HeLa cells. Our data suggest that a reduction in HSP70-2 expression by shRNA greatly decreases growth and the ability of cells to form colonies. The results from our in vitro studies suggested that reduced HSP70-2 expression is not merely cytotoxic but exclusively affects tumorigenic cells. This makes HSP70-2 a very attractive target for developing cancer therapies. Therapies that are designed to impede tumor cell migration and invasion are expected to be advantageous in the management of cervical cancer. Because the invasiveness of cancer cells depends on increased migratory and invasive properties, we investigated the effect of HSP70-2 knockdown on the migration and invasiveness of cervical cancer cells. For the first time to our knowledge, we have demonstrated HSP70-2 shRNA-mediated inhibition of the migration and invasion of cancer cells. Our results suggest that HSP70-2 expression is critical for the migration and invasive properties of cervical cancer cells.

Although the progression from early to advanced stage cervical cancer is a critical determinant of survival, little is known about the molecules that contribute to the progression and metastasis of cervical tumors. The role of molecules involved in tumorigenesis is an emerging area in the field of cancer research and, thus, represents a potential area for therapeutic targeting. The current study represents a proof-of-principle concept that HSP70-2 expression is involved in cervical carcinogenesis. In vivo studies using a xenograft mice model revealed that the inhibition of HSP70-2 significantly reduced the tumorigenic potential of cervical cancer cells. HSP70-2 down-regulation inhibited the tumor growth rate by 75% in nude mice. This result indicates that HSP70-2 expression is important for cervical tumor cell survival. Given the involvement of HSP70-2 gene in cervical carcinoma progression and the pronounced inhibitory effects of HSP70-2 shRNA on growth, colony formation, invasion, migration, and tumorigenesis reported in this study, we believe that the HSP70-2 molecule requires more attention for further investigation as a new strategy for the treatment of human cervical carcinoma and warrants further studies.

In summary, the current findings indicate that HSP70-2 is expressed in various cervical cancer cell lines of different origin and in patients with cervical SCC. Our data also support the role of HSP70-2 in cell migration, invasion, and tumorigenesis. We present HSP70-2 as a prospective diagnostic biomarker in cervical cancer associated with early disease stage and different tumor grades. Furthermore, HSP70-2 may be proposed or recommended as a novel and innovative target for the treatment of cervical cancer. Additional studies are warranted to explore its potential role as a biomarker for early detection and for implementing the accurate, precise, early, and aggressive cancer treatment that may produce a cure and higher survival rates in patients with cervical cancer.

CONFLICT OF INTEREST DISCLOSURES

Supported by grants from the Cancer Research Program, Associated Cancer Center for Immunotherapy, Department of Biotechnology, Government of India.

Ancillary