| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Division of Endocrine and Oncologic Surgery, Department of Surgery (S.K., I.P., R.C., C.M., D.F.), and Department of Pathology (K.M.-K.), University of Pennsylvania, Philadelphia, Pennsylvania 19104
Address all correspondence and requests for reprints to: Dr. Douglas Fraker, Division of Endocrine and Oncologic Surgery, HUP, 4 Silverstein Building, 3400 Spruce Street, Philadelphia, Pennsylvania 19104. E-mail: frakerd{at}uphs.upenn.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Objective: As a new approach to treatment, the efficacy of replication-selective, human thyroglobulin (TG) enhancer and promoter-driven, adenovirus (AdhTGEP)-mediated oncolysis was investigated using two well-differentiated thyroid cancer cell lines, XTC (TG positive) and FTC-133 (TG negative), and other control tumor and nontumor cell lines (all TG negative).
Design: A cohort study design was used.
Setting: The study setting was laboratory bench-top experiments.
Subjects/Participants: In vitro TG-expressing and nonexpressing thyroid cell culture lines, nonthyroid tumor cell lines, as well as preclinical thyroid tumor-bearing mice were studied.
Intervention: Adenoviral infection of cell lines was determined by immunohistochemistry, selective replication by one-step growth assays, and cytotoxicity by 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulphophenyl)-2H-tetrozolium (MTS) assay. In vivo tumor growth inhibition was determined by a single intratumoral injection of 1 x 109 plaque-forming units AdhTGEP, AdLacZ (control virus), or PBS to 50- to 75-mm3 tumors. XTC cells showed intense immunohistochemical staining, whereas FTC-133 and all other control cell lines showed minimal staining for viral infection with AdhTGEP.
Main Outcome Measures: Cell survival and tumor growth inhibition after adenoviral infection were the main outcome measures.
Results: One-step growth assays showed at least a more than 60-fold titer of AdhTGEP in XTC than in FTC-133 cells. Cytotoxicity assays showed approximately 68% cell kill in XTC and minimal cell kill in FTC-133 and all other control cell lines at a multiplicity of infection of 250. There was significant in vivo growth inhibition of AdhTGEP-treated XTC tumors (67 ± 49 mm3) compared with AdLacZ-treated XTC (228 ± 45 mm3; P < 0.01), PBS-treated XTC (372 ± 70 mm3; P < 0.001), or AdhTGEP-treated FTC-133 tumors (598 ± 168 mm3).
Conclusion: Replication-selective virus-mediated oncolysis is a potential therapy for recurrent, well-differentiated, TG-secreting thyroid cancer that is unresponsive to standard treatment.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Gene therapy approaches for the treatment of cancer are currently being investigated in numerous preclinical studies and clinical trials (8). Critical to the success of these strategies are the maintenance of tissue specificity and efficient transduction of target cells. Because the majority of gene delivery vectors are capable of infecting multiple tissue types, toxic effects may be observed in normal tissues if viral gene expression is not restricted to neoplastic cells. One strategy that has been used to maintain tissue specificity is the integration of tumor-specific promoters and/or enhancers into gene delivery vectors. Examples of tumor-specific promoters that have been successfully used in this manner include the 
-fetoprotein promoter (9), the prostate-specific antigen promoter (10), and the carcinoembryonic antigen promoter for prostate, colon, and hepatocellular carcinoma, respectively (11, 12, 13, 14).
The majority of gene therapy strategies use replication-incompetent viral vectors. A more recent approach, however, is the use of replication-competent viral vectors with highly regulated viral expression. In this strategy, the pathological replication of a virus within cells produces tumor destruction either from tumor cell lysis at the end of a viral replication cycle, from stimulation of the immune system by viral antigens, or from the production of a toxin during viral replication (15). This approach, known as replication-selective, viral-mediated oncolysis, may help to circumvent poor transduction efficiencies by providing in vivo viral amplification. The most widely studied oncolytic virus is the E1B-deleted adenovirus ONYX-015, which has been shown to preferentially replicate in p53-deficient tumor cells (16, 17). Tumor-specific promoters for prostate, breast, lung, and hepatocellular carcinomas have also been used to control oncolytic viral replication under the control of a prostate-specific antigen promoter (10), a DF3/MUC-1 promoter (18), a surfactant B promoter (19), and an -fetoprotein promoter (9), respectively. In most cases, the tumor-specific promoter was engineered to regulate the expression of the E1A and/or E1B region and, in turn, improved the selective replication of virus in the relevant tumor cells by 10- to 1000-fold compared with cells lacking the relevant tumor-specific antigen.
Recent studies have demonstrated that for adenoviral gene therapy to be effective, the adenoviral receptors, coxsackievirus receptor (CAR) and adenovirus receptor, and integrins, vß3 and vß5, must be expressed on the cell surface (20, 21). These adenoviral receptors are especially important when using replication-selective viral-mediated oncolysis as a gene therapy strategy. The potential therapeutic advantage offered by in vivo viral amplification would be significantly hindered by poor intratumoral spread of viral progeny due to the inability to infect neighboring tumor cells lacking adenoviral receptors. Other strategies are to develop adenoviral vectors that target other surface receptors to facilitate efficient transduction in tumor cells that lack the CAR (22, 23).
In the case of well-differentiated thyroid cancer, transcriptional targeting using the thyroid-specific thyroglobulin (TG) promoter and/or enhancer has been evaluated in several preclinical models using replication incompetent viral vectors for gene delivery (24, 25). Braiden et al. (24) and Zhang et al. (25, 26) demonstrated the feasibility of this approach for suicide gene therapy. In in vivo animal models, Nagayama et al. (27) demonstrated enhancement of the transcriptional activity of the rat TG promoter using the Cre-LoxP system, and Barzon et al. (28) combined suicide gene therapy and immunopotentiation using two transcriptionally targeted retroviral vectors containing the human TG enhancer sequence.
Thyroid cancers are good targets for gene therapy treatments because they express certain tissue-specific markers, such as the gene promoter, TG, TSH receptor, and calcitonin, which are very specific to the tumor and the thyroid. Because TG synthesis and radioiodine uptake reflect different functions of the thyroid cancer cell (29), we constructed a replication-selective adenovirus under the control of a TG promoter that would cause oncolysis of cancer cells that expressed TG and had minimal toxic effect on cells that lacked TG expression. Several transcriptional regulatory elements, such as TATA and capsite of the E1A promoter region of the adenovirus, were deleted so the TG promoter could drive the replication of the virus in TG-expressing cells, leading to viral oncolysis. The replication selectivity, infectivity, and selective oncolytic activity in vitro and in vivo of an adenoviral vector driven by the human TG enhancer and promoter regions (AdhTGEP; Fig. 1A
) were examined in several thyroid cancer and other cell lines, including the well-differentiated, TG-expressing XTC cell line and the TG-nonexpressing FTC-133 cell line.
|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
XTC (Hurthle cell line expressing TG) and FTC-133 (follicular cell line lacking TG expression) thyroid cancer cell lines (30, 31) were gifts from Dr. Orlo Clark (University of California, San Francisco, CA). Both cell lines were cultured according to previous protocols (29, 30). Experiments were performed with XTC cells and FTC-133 cells between passages 27–29 and 17–21, respectively. The RmcB hybridoma cell line and 293 cells (transformed human embryonic kidney cells) were obtained from American Type Culture Collection (ATCC; Manassas, VA) and cultured according to ATCC guidelines. ARO (anaplastic thyroid cancer cell line, non-TG expressing), WRO (follicular thyroid cancer cell line, non-TG expressing), and NPA (papillary thyroid cancer cell line, non-TG expressing) were gifts from Dr. Matthew Ringel (Ohio State University, Columbus, OH). The latter three cell lines were cultured in RPMI supplemented with 10% fetal calf serum, 2 mM L-glutamine, 100 U/ml streptomycin/penicillin, 0.25 µg/ml amphotericin B (Invitrogen Life Technologies, Inc., Gaithersburg, MD), 10 µg/ml insulin, and 10 mIU/ml TSH (Sigma-Aldrich Corp., St. Louis, MO). WI-38, a fibroblast cell line, and IMR-32, a neuroblastoma-derived cell line, were purchased from the ATCC and cultured according to ATCC guidelines. NIH1286, a human melanoma cell line, was a gift from Dr. Steven Rosenberg (National Cancer Institute, Bethesda, MD). The NIH1286 cells were cultured in RPMI supplemented with 10% fetal calf serum, 2 mM L-glutamine, 100 U/ml streptomycin/penicillin, and 0.25 µg/ml amphotericin B (Invitrogen Life Technologies, Inc.).
Construction of recombinant adenovirus
Human thyroid tissue that was snap-frozen in liquid nitrogen obtained from the Cooperative Human Tissue Network (University of Pennsylvania, Philadelphia, PA) was used to provide genomic DNA template for the TG promoter and enhancer. Primers for the human TG promoter and enhancer were designed from known genebank sequences. For the TG promoter fragment, BglI and XhoI sites were introduced upstream of the promoter, and SpeI and NotI sites were introduced downstream of the promoter. The upstream and downstream TG promoter primers were 5'-CGA TCC CTC GAG GCC ACC ACG GCG GAT CCA GCA ATA TGG TGG CAG GCT G-3' and 5'-ACT AGT GCG GCC GCG CCC TAA AAT GCA TGC ATA AAG CTT GCA C-3', respectively. For the TG enhancer fragment, SalI and EcoRI sites were introduced upstream of the enhancer, and BglI and SalI sites were introduced downstream of the enhancer. The upstream and downstream TG enhancer primers were 5'-GAA TTC GTC GAC GAG CTC AGT GGA GAA GAG GTA AAA GTA G-3' and 5'-GAA TCC GCG GCC GCG CCA CCA CGG CGC ATG CAT AAA GCT TGC ACA GGT TG-3', respectively. Primers were purchased from Invitrogen Life Technologies, Inc. (Carlsbad, CA). PCR to generate the TG promoter and enhancer fragments was carried out in a 100-µl volume containing 200 µM deoxy-NTPs, 1 U Taq polymerase, 1x PCR buffer (Roche, Indianapolis, IN), 0.25 µM each of upstream and downstream primers, and 2 µg genomic DNA. The reaction was allowed to run for 35 cycles at the following settings: 94 C for 1 min and 72 C for 2 min. The PCR fragments were then run on a 1% agarose gel, quantitated, and cloned into the TA Vector pCR II (Invitrogen Life Technologies, Inc.). The fragments were sequenced using the upstream primer, with evidence of 100% homology in the enhancer sequence and up to 1 bp mutation in the promoter sequence. The TG enhancer fragment from the TATGE plasmid was then cloned into the TA vector containing the TG promoter (TATGP) using the EcoRI and BglI sites. The hTGETGP fragment from the TA vector was then excised using the KpnI and XhoI sites in the TA vector, blunted, and cloned into an adenoviral shuttle vector pAd460 using the PacI site (pAd460hTGEP). The pAd460 is the pXCI vector from Microbix (Toronto, Canada) with deletion of several transcriptional regulatory elements (TATA and Capsite) of the E1A promoter region (Fig. 1A
). The pAd460hTGEP was then transfected along with the linearized adenovirus backbone into 293 cells with the CellPhect transfection kit (Amersham Biosciences, Piscataway, NJ). The adenovirus generated as a result of the homologous recombination was screened for the TG promoter and enhancer by PCR. This adenovirus, AdhTGEP, then underwent three rounds of plaque purification and high titer viral stocks were generated using 293 cells (32).
RT-PCR for endogenous TG mRNA expression in thyroid cancer cell lines
RNA was extracted from cells using the Stratagene micro-RNA extraction kit (La Jolla, CA) and was quantitated using a DU-640 (Beckman Coulter, Inc., Fullerton, CA) at 260 nm. For the RT reaction, 3.5 µg RNA were reverse transcribed in a 20-µl volume using 0.5 mg oligo(deoxythymidine) (Promega Corp., Madison, WI), 0.5 mM dexoy-NTPs, 10 mM dithiothreitol, 4 µl first-strand buffer, and 10 U/µl Moloney murine leukemia virus reverse transcriptase (Invitrogen Life Technologies, Inc., Gaithersburg, MD) for 1 h at 37 C. After heat inactivation of the Moloney murine leukemia virus enzyme, 2 µl reaction mixture were PCR amplified with human TG-specific primers. The TG upstream and downstream primers were 5'-CCC CAT GTC ATC CAG CCA AGA AGT and 5'-CCC GGG AGA AGG AGG CAT AGT CAT, respectively. PCR for a 490-bp TG fragment was carried out in a 100-µl volume containing 200 µM deoxy-NTPs, 1 U Taq polymerase, 1x PCR buffer (Roche), and 0.25 µM each of upstream and downstream primers. The reaction was allowed to run for 35 cycles at the following settings: 94 C for 1 min, 58 C for 1 min, and 72 C for 1 min. PCR products were then electrophoresed on a 1% agarose gel in 1x Tris-acetate-EDTA buffer and stained with ethidium bromide.
ELISA for endogenous TG protein secretion by cell lines
Cells were plated in 48-well plates at a density of 1 x 105 cells in 1 ml complete medium and incubated at 37 C in 5% CO2. Supernatant was harvested 48 h later and stored at –20 C until ELISA was performed. The MESACUP TG ELISA kit (MBL Medical and Biological Laboratories Co., Nagoya, Japan) was used to determine the TG concentration in the harvested supernatants. To determine whether increasing levels of TSH induced an increase in TG concentrations in cell supernatants, XTC and FTC-133 cells were plated in complete medium with either 10 or 100 mIU/ml TSH. Supernatant was harvested 48 h after plating the cells, and TG concentration was determined as described above.
Immunocytochemistry for endogenous TG expression
Cells plated in Permanox eight-well chamber slides (Nalge Nunc, Rochester, NY) at 2.5 x 104 cells/well were incubated for 24 h at 37 C in 5% CO2 and were fixed in ice-cold 100% methanol/100% acetone/37% formaldehyde (10:10:1) for 90 sec. The slides were then incubated in 15 mM sodium azide and 0.03% hydrogen peroxide for 5 min to quench endogenous peroxidases. Slides were rinsed twice in water, twice in 1x PBS for 5 min each time, blocked for 45 min with 5% rabbit serum (Vector Laboratories, Inc., Burlingame, CA) and 4% BSA (Sigma-Aldrich Corp.) in PBS, and incubated with rabbit anti-TG antibody (DakoCytomation, Carpinteria, CA) at a 1:10,000 dilution for 30 min at room temperature. A nonspecific rabbit Ig antibody from DakoCytomation was used as a negative control. The slides were incubated with secondary antibody, horse antirabbit biotin (Vector Laboratories, Inc.), at a concentration of 1:200 for 1 h at room temperature, followed by a 15-min wash in 1x PBS. The AB-ELITE kit (Vector Laboratories, Inc.) was used to develop the slides. Photomicrographs were taken using a camera attached to an Olympus IX70 microscope (New Hyde Park, NY) at a magnification of x400.
FACS analysis of cell lines for adenoviral receptors
Cells were evaluated for the expression of CAR and integrins, vß3 and vß5, by flow cytometry. The CAR antibody was purified from supernatant of the RmcB hybridoma cell line (ATCC) using the Hi-Trap r-Protein kit (Pharmacia Biotech AB, Uppsala, Sweden). The vß3 receptor and vß5 receptor antibodies were purchased from Chemicon International (Temecula, CA). The secondary antibody, goat antimouse fluorescein isothiocyanate (FITC) and IgG1 FITC control antibody were purchased from Roche (Mannheim, Germany), and the streptavidin-fluorescein secondary antibody was obtained from NEN Life Science Products (Boston, MA). Cells (2 x 105) were incubated with 1 µg each of primary antibody or control antibody (IgG1 FITC) for 30 min at 4 C in FACS buffer made of 2% BSA (Sigma-Aldrich Corp.) in 1x PBS (Invitrogen Life Technologies, Inc.). Cells were rinsed with FACS buffer and incubated with 0.5 µg secondary antibody for 30 min at 4 C. The cells were rinsed again, resuspended in FACS buffer, and analyzed on a FACScan (BD Biosciences, Mountain View, CA) flow cytometer. Propidium iodide (0.01 µg/sample) was used to distinguish live from dead cells.
Luciferase assays to determine TG promoter activity
To determine TG promoter activity, the TGETGP fragment from the TA vector was cloned into a luciferase expression vector, PGL-2 basic (Promega Corp.) using the KpnI and XhoI sites. PGL-2 basic and PGL-2 control plasmids (Promega Corp.) were used in these assays as negative and positive controls, respectively. All cell lines were plated in a six-well plate at 2.5 x 105 cells/well and incubated overnight. For the luciferase activity assays, 2 µg each of PGL-2TGEP plasmid, PGL-2 basic, and PGL-2 control plasmids were transfected into the cell lines using FuGene (Roche). A Renilla luciferase plasmid pRL-thymidine kinase (pRL-TK; 100 ng; Promega Corp.) was cotransfected into the cell lines as an internal transfection control. Forty-eight hours after transfection, luciferase and Renilla luciferase activities in the cell lines were determined using the Dual Assay Reporter Kit (Promega Corp.) and measured using a Lumat LB 9507 luminometer (Berthold Technologies, Oak Ridge, TN).
Immunocytochemistry for adenovirus infection
XTC and FTC-133 cells plated on Permanox eight-well chamber slides at 1 x 105 cells/well were incubated for 24 h at 37 C in 5% CO2. Cells were infected with AdhTGEP at a multiplicity of infection (MOI) of 10 or 100 or AdWT (Institute for Human Gene Therapy, University of Pennsylvania) at an MOI of 10. One well was left uninfected and served as a control. The cells were immunostained for adenovirus 24, 48, 72, 96, and 120 h after infection. Cells were fixed in cold acetone for 5 min. Endogenous peroxidases were quenched by incubation in 1% hydrogen peroxide for 10 min, followed by a rinse in tap water and a rinse in 1x PBS. Slides were blocked with 5% horse serum and 4% BSA in 1x PBS for 45 min at room temperature. The slides were incubated with primary antibody to adenovirus, goat anti-Ad2 (Chemicon International) at a concentration of 1:2000 for 1 h at room temperature, followed by a 30-min rinse in 1x PBS. The slides were incubated with secondary antibody, horse antigoat biotin (Vector Laboratories, Inc.), at a concentration of 1:200 for 1 h at room temperature, followed by a 15-min rinse in 1x PBS. The AB-ELITE kit (Vector Laboratories, Inc.) was used to develop the slides. Photomicrographs were taken using a camera attached to an Olympus IX70 microscope at a magnification of x400.
One-step growth curves to determine selective replication of AdhTGEP
XTC and FTC-133 cells plated in 3 ml complete medium in six-well plates at a density of 2.5 x 105 cells/well were incubated for 24 h, then infected with either the AdhTGEP or AdWT virus at an MOI of 0.3 in 1 ml serum-free medium. Cells were incubated at 37 C for 1 h, and then the virus was removed. One well was immediately harvested into 1 ml complete medium by scraping the cells for a zero hour point. The remaining wells were incubated in 1 ml complete medium at 37 C and harvested at 4, 19, 24, 48, 72, and 96 h after infection. The viral titer of the harvested cells at each time point was determined by performing plaque assays on 293 cells (32). The assays were repeated at least three times for each time point.
AdhTGEP cytotoxicity assays
Cells plated at a density of 1 x 104 or 5 x 103 cells/well in a 96-well culture plate were incubated for 24 h at 37 C in 5% CO2. The cells were infected for 1 h in serum-free medium with AdhTGEP, AdWT, or AdLacZ (IHGT, University of Pennsylvania) in replicates of six at MOIs of 500, 250, 100, 10, 1, 0.1, and 0.001. A set of six wells was left uninfected and used as the control. Cell survival was determined using the 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulphophenyl)-2H-tetrozolium (MTS) reagent in the one-step growth proliferation kit (Promega Corp.). The percentage of dead cells in each group was calculated relative to the uninfected control wells. Photomicrographs of cells in each group were taken with a camera attached to an Olympus IX70 microscope at a magnification of x400.
In vivo tumor growth inhibition by AdhTGEP
Six- to 8-wk-old athymic nude mice were tumored on their right flank with 2 x 106 XTC or FTC-133 cells in a 1:1 ratio of Matrigel to serum-free medium in a 100-µl volume. When tumor sizes were between 50 and 75 mm3 (based on the ellipsoid formula, 1.33 x width x height x length), tumors were injected with PBS, or 1 x 109 plaque-forming units AdLacZ, AdhTGEP, or AdWT in a 50-µl volume. Mice were then followed for tumor growth by measuring tumors twice a week. These in vivo experiments were performed using a protocol approved by the institutional animal care and use committee of University of Pennsylvania.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Experiments to evaluate the transcription and translation of TG in various thyroid cancer cell lines were a preliminary step in evaluating the use of replication selective viral-mediated oncolysis. To determine TG mRNA expression, RT-PCR was performed on the thyroid and other cell lines. A 490-bp TG PCR fragment was seen only in the lane corresponding to the XTC cell line, whereas no band was observed in the lanes corresponding to FTC-133, ARO, WRO, NPA, IMR-32, WI-38, or NIH1286 cell lines (Fig. 1B).
Only XTC cells secrete TG protein
TG concentrations in cell culture supernatants of were determined by ELISA. TG levels were observed in XTC cell supernatants as early as 24 h after plating cells (data not shown). Forty-eight hours after plating, the concentration of TG in the XTC cell line was 84 ± 9 ng/ml when cells were cultured in 10 mIU/ml TSH and were at a low passage number (p27). When cells were cultured in 100 mIU/ml TSH, there was a slight increase in the amount of secreted TG (90.4 ± 8 ng/ml). With serial passaging of XTC cells, TG levels decreased. At passage 32, the TG concentration in cell supernatants was 12 ng/ml, and by passage 37, TG was not detectable by ELISA. There were no detectable levels of TG in the FTC-133, ARO, WRO, NPA, IMR-32, WI-38, and NIH1286 cell lines. Endogenous TG protein expression in these cell lines was also analyzed by immunocytochemistry. Relatively intense cytoplasmic staining for TG was observed in the XTC cells at passage 27 (Fig. 2B), whereas there was no staining seen in the other cell lines (Fig. 2). Minimal background staining was observed with the rabbit control antibody in all cell lines.
|
To examine differential expression of cell surface adenoviral receptors that could impact adenoviral entry into the cell lines, cells were analyzed for the expression of CAR and integrin receptors, vß3 and vß5, by flow cytometry. XTC and FTC-133 cells expressed all three adenoviral receptors (Table 1). Both cell lines showed similar levels of expression of CAR receptor, whereas XTC cells expressed the vß3 receptor at a higher level than FTC-133 cells. The FTC-133 cells, however, expressed higher levels of the vß5 receptor than XTC cells. The ARO, WRO, and NPA cell lines showed minimal expression of CAR compared with XTC and FTC-133 cell lines. Although there was relatively slightly higher expression of vß3 and vß5 expression compared with CAR in these cell lines, the expression levels were lower than those in XTC and FTC-133 (Table 1). IMR-32 had some CAR expression, but minimal vß3 and vß5 receptor expression, whereas the WI-38 and NIH1286 cell lines had minimal expression of the three adenoviral receptors (Table 1). Because the XTC and FTC-133 cell lines had similar CAR expressions, we chose these cell lines to determine the selective replication efficiency of AdhTGEP.
|
To determine the efficiency of the TG promoter in the cell lines, luciferase activity assays were performed in XTC (passage 28), FTC-133 (passages 17–21), ARO (passages 40–45), WRO (passages 36–41), NPA (passages 45–50), IMR-32 (passage unknown), WI-38 (passage unknown), and NIH1286 (passage unknown) cell lines using the PGL-2TGEP plasmid. Luciferase activity in the XTC cells transfected with PGL2TGEP was 7- to 20-fold higher than that in the other cell lines transfected with the same vector relative to the PGL-2 basic control vector (Table 2). All luciferase values were corrected for differential transfection efficiencies in the cell lines using an internal Renilla luciferase plasmid pRL-TK. Results are the average of three separate experiments.
|
To determine selective infection in TG-expressing and nonexpressing cell lines by AdhTGEP and AdWT, XTC and FTC-133 cells were immunostained for adenovirus with a goat anti-adenovirus antibody at 24, 48, 72, 96, and 120 h after infection. In the AdhTGEP-infected XTC cells, a few cells stained positively for adenovirus at 24 h after infection, a large number stained positively at 48 h after infection, and an even larger number stained positively at 72 h after infection (Fig. 3). In the AdhTGEP-infected FTC-133 cells, no staining was observed for adenovirus at 24 h after infection, with minimal staining at 48 and 72 h after infection (Fig. 3). In the AdWT-infected cells, adenoviral immunostaining was observed as early as 24 h after infection in both cell lines (data not shown). The AdWT-infected XTC and FTC-133 cells showed a good proportion of the cells staining positively for adenovirus at 48 h after infection, and almost all cells stained positively for adenovirus at 72 h after infection (data not shown). A cytopathic effect was observed in the AdWT-infected cell lines at 96 h after infection and in AdhTGEP-infected XTC cells at 120 h after infection.
|
To establish selectivity of replication of AdhTGEP in XTC cells, one-step growth curves were generated by determining virus titer after infecting XTC and one of the control cell lines (FTC-133) with either AdhTGEP or AdWT and by harvesting cells at 0, 4, 19, 24, 48, and 72 h after infection (Table 3). The titer of virus was determined by plaque assays that were performed using 293 cells. Initial virus burst in both cell lines for AdhTGEP was observed at 19 h after infection. AdhTGEP infection of XTC cells produced a 68.5 ± 8.5-fold higher virus titer at 72 h after infection compared with virus generated by infection of FTC-133 cells at the same time point. Table 3 represents the average of several experiments that were performed. Although the actual titer values of virus differed in each experiment, the degree of increase in virus titer of XTC-generated AdhTGEP virus compared with that in titer of FTC-133-generated virus remained similar. AdWT infection of both cell lines, however, produced similar viral titers, and the virus burst for AdWT in both cell lines was also at 19 h after infection (data not shown). These results suggest that the AdhTGEP virus replicated in a cell-specific manner in the XTC cell line, but not in the FTC-133 cell line.
|
The oncolytic activity of AdhTGEP in the thyroid and other cell lines was determined at various MOIs ranging from 0.001–500. XTC cells cultured in 10 mIU/ml TSH, infected with AdhTGEP at MOIs of 500, 250, 100, and 10 showed 22, 32, 57, and 82% cell survival, respectively, relative to uninfected controls (Table 4). Minimal toxicity (93–100% cell survival) was observed in FTC-133 cells infected at MOIs up to 250. At the lower ranges of MOI, 0.1–0.001, both XTC and FTC-133 cells showed no toxicity to AdhTGEP. Both cell lines, however, were equally sensitive to AdWT infection, with greater than 95% cell kill at MOIs of 500, 250, and 100, whereas minimal toxicity was observed at MOIs of 1 and lower. Microscopically, widespread viral oncolysis was seen in both XTC (Fig. 4C) and FTC-133 cells (Fig. 4F) infected with AdWT at an MOI of 100 and in XTC cells infected with AdhTGEP (Fig. 4B) at an MOI of 250, whereas minimal oncolysis was seen in FTC-133 cells infected with AdhTGEP at the same MOI (Fig. 4E). No oncolytic effect was observed in uninfected XTC or FTC-133 cell lines (Fig. 4, A and D, respectively). Infection of XTC and FTC-133 cells with a replication-deficient AdLacZ virus resulted in minimal oncolytic effect (Fig. 4) even though the cells stained positively for the lacZ gene. In contrast, minimal cytotoxicity was observed in NPA cells after infection with AdhTGEP or AdWT even at higher MOIs of 250 and 500. IMR-32, WI-38, and ARO cells were more susceptible to AdWT infection than AdhTGEP. Minimal oncolytic cell death was observed in NPA, WRO, or NIH1286 cells infected with AdWT (Table 4). These results probably reflect the lack of adenoviral receptor expression in these cell lines.
|
|
Statistically significant tumor growth inhibition was observed in both AdWT- and AdhTGEP-treated XTC tumors (Fig. 5). The volume of AdhTGEP-treated XTC tumors at 41 d after treatment was 68 ± 49 mm3, whereas the volume of AdWT tumors was 40 ± 20 mm3. In contrast, PBS- and AdLacZ-treated tumors showed no growth inhibition, and volumes of tumors were 372 ± 70 mm3 (P < 0.001) and 228 ± 45 mm3 (P < 0.01), respectively. AdhTGEP-treated FTC-133 tumors showed no growth inhibition, and the volume of tumors at 21 d after treatment was 598 ± 168 mm3. Complete tumor regression was observed in one mouse in each of the AdhTGEP- and AdWT-treated XTC tumors.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Several investigators have demonstrated that the success of adenoviral gene therapy is contingent on the presence of specific cell surface adenoviral receptors (20, 21, 22, 35). In the present study, the adenoviral receptor profiles of thyroid cancer and other tumor and nontumor cell lines were investigated before initiating experiments for viral replication and oncolysis. By FACS analysis, similar CAR expression was observed in two thyroid cancer cell lines, XTC and FTC-133, whereas minimal CAR expression was seen in the remaining six cell lines. These receptor profiles help explain why adenoviral transduction, especially when using wild-type adenovirus, was significantly more successful in the XTC and FTC-133 cell lines. In addition, these data confirm that the cell-specific oncolysis observed in XTC cells compared with FTC-133 cells is not solely a consequence of the differential expression of adenoviral receptors in these two cell lines.
Another key to the success of replication-selective viral oncolysis is maintenance of tissue specificity, viral titers, and cytotoxicity. Significantly greater immunocytochemical staining for adenovirus was observed in XTC cells than in FTC-133 cells as early as 48 h after infection with AdhTGEP, and AdhTGEP viral titers were more than 60-fold higher in XTC cells than in FTC-133 cells by 72 h after infection. In addition, cytotoxicity assays demonstrated increasing cell death in XTC cells after infection with AdhTGEP at increasing MOIs, whereas minimal cytotoxicity was observed in FTC-133 cells after infection at MOIs up to 250. These results confirm that the AdhTGEP virus selectively replicates and leads to oncolysis in thyroid cancer cells that express TG (XTC), whereas minimal replication and cytotoxicity are observed in thyroid cancer cells that lack TG expression (FTC-133). Conversely, when both cell lines were infected with wild-type adenovirus, equivalent viral titers and adenoviral immunohistochemical staining were observed (data not shown). This suggests that the differences in AdhTGEP viral replication and cytotoxicity in the XTC and FTC-133 cell lines are not simply the result of inherent differences in adenoviral infection.
Results from the present study demonstrate selective viral replication and oncolysis in the XTC cell line, but 100% cytotoxicity in XTC cells was never obtained in vitro after infection with AdhTGEP. There are two potential explanations for this lack of 100% efficacy. First, the TG promoter is a weak promoter, and although it is capable of driving replication, it is not vigorous enough to kill 100% of the cells. Second, the transcription factor levels that regulate TG expression and adenoviral replication in the XTC cell line may not be sufficient even at lower passages of the cell lines. One method that has been shown to increase promoter activity and thereby to potentially increase viral efficacy is the insertion of more than one enhancer sequence (36). In the current study, an adenoviral vector with a single TG enhancer sequence was used. In a separate set of experiments, an attempt was made to further enhance the activity of the TG promoter through the insertion of an additional TG enhancer sequence upstream of the first enhancer sequence. However, promoter assays demonstrated that this additional enhancer actually decreased TG promoter activity (data not shown). It appears that the position of the second enhancer sequence relative to the first could influence overall promoter activity. Additional experiments along these lines are warranted. A second approach that may improve efficacy is increasing the MOI. However, this could lead to a loss of viral specificity. We observed increasing cytotoxicity in the XTC cell line with increasing MOI, and minimal cytotoxicity in FTC-133 up to an MOI of 250. At MOIs greater than 250, however, a significant decrease in specificity was observed. Another drawback of increasing the MOI is that it exposes animals to higher viral loads, which may increase morbidity and mortality.
Well-differentiated thyroid cancer appears to be an excellent model for studying transcriptional targeting and replication-selective viral oncolysis because of the several thyroid-specific proteins that are available. However, limitations have been observed in preclinical models. Although various investigators have successfully isolated TG-expressing thyroid cancer cell lines in culture from primary tumors, these cell lines appear to dedifferentiate in vitro with serial passaging and to lose TG expression. Similar observations were made in this study as well. At low passages (p27), XTC cells expressed 84 ng/ml TG, and widespread viral oncolysis was observed after infection with AdhTGEP. However, at later passages (greater than p36), there was considerable loss of TG expression (3 ng/ml), and this resulted in a significant decrease in the oncolytic effect of AdhTGEP in the XTC cell line (20% cell death at an MOI of 500; data not shown). Xing et al. (37) demonstrated that aberrant methylation of the human TSH receptor promoter is a likely molecular pathway that results in silencing of the TG gene in thyroid cancers, leading to a loss of TG expression. Chun et al. (38) have also demonstrated that in cell lines that do not express TG, overexpression of the TG transcription factors thyroid transcription factor-1 (TTF-1) and Pax-8 restores TG expression.
Currently, we are investigating mechanisms by which TG expression is lost in XTC cells. In preliminary experiments to determine the TG transcription factor levels in XTC and FTC-133 cells, TTF-1 expression was observed in both cell lines; however, both cell lines lacked Pax-8 expression (data not shown). A decrease in TTF-1 expression with serial passaging of the XTC cell line was also observed. Because several investigators have shown that Pax-8 transcription factor is important for TG expression (38, 39), we are presently looking at methods to restore Pax-8 expression and possibly TG expression in the XTC cell line. In addition, we are investigating the effect of demethylating agents on TG expression in the XTC cell line, because previous studies have indicated that the use of histone-deacetylating agents restores sodium iodide symporter and Pendred syndrome gene expression in this cell line (40).
Kitazono et al. (41) demonstrated that an adenoviral vector expressing TK that was driven by a TG enhancer and promoter showed preferential ganciclovir sensitivity in thyroid cancer cells. In these studies, the FTC-133 cell line was used to demonstrate this effect. In our studies we found that FTC-133 cells did not express TG and as a result did not show specific viral oncolysis when infected with AdhTGEP. FTC-133 cells were first cultured by Goretzki et al. (30) and demonstrated TG expression at passage 4. Our experiments were conducted when the FTC-133 cells were at passages 17–21. It appears that the FTC-133 cells used in the present study had TG expression at an earlier point in time and lost TG expression over repeated passaging, as was seen with the XTC cell line. The FTC-133 cell line and the higher passages of XTC cells were exposed to sodium butyrate and 5-azacytidine, both histone-deacetylating reagents. Some expression of TG was restored in the cell lines (data not shown). Additional experiments have to be conducted to determine whether treating the cells with both histone deacetylases and AdhTGEP will enhance the oncolytic effect of AdhTGEP.
Although gene therapy remains a promising novel approach for the treatment of cancer, many obstacles remain in realizing its full clinical potential. Replication-selective viral oncolysis may help to overcome the problem of low transduction efficiency, but iv injection of adenovirus has been shown to lead to rapid accumulation of virus in the liver, and lack of efficacy in several preclinical gene therapy models has been attributed to antiadenoviral immune responses (8). Another limitation that exists in translating this model of replication-selective viral oncolysis to the clinical setting is that it fails to target those patients with anaplastic thyroid cancers. Although these patients comprise a small percentage of the population that develop thyroid cancer each year (<5%), these tumors are very aggressive and respond poorly to standard surgical, chemotherapeutic, and radiation interventions (42).
In conclusion, this study has shown that recombinant adenovirus, AdhTGEP, selectively replicates in a thyroid cancer cell line that expresses TG and leads to tumor cell oncolysis in vitro. In addition, in a single in vivo experiment, significant growth inhibition of XTC tumors treated with AdhTGEP was obtained compared with PBS- and AdLacZ-treated tumors. Replication-selective viral oncolysis, therefore, is potentially a novel approach to the treatment of recurrent well-differentiated thyroid cancer that is unresponsive to standard treatment modalities.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 S.K. and I.P. contributed equally to this study. 
Abbreviations: CAR, Coxsackie adenoviral receptor; FITC, fluorescein isothiocyanate; MOI, multiplicity of infection; TG, thyroglobulin; TK, thymidine kinase; TTF-1, thyroid transcription factor-1.
This work was supported in part by the Georgene S. Harmelin Endowment Fund.
Received September 9, 2004.
Accepted March 17, 2005.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
vß3 and vß5 promote adenovirus internalization but not virus attachment. Cell 73:309–319[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  A. S. Rao, P. E. Goretzki, J. Kohrle, and G. Brabant Letter re: Id1 Gene Expression in Hyperplastic and Neoplastic Thyroid Tissues J. Clin. Endocrinol. Metab., October 1, 2005; 90(10): 5906 - 5906. [Full Text] [PDF]
|
|
, AdLacZ; 
, AdhTGEP; 
, AdWT.