Suppression of Colorectal Cancer Growth Using an Adenovirus Vector Expressing an Antisense K-ras RNA

doi:10.1006/mthe.2001.0302, available online at http://www.idealibrary.com on IDEAL

ARTICLE

Suppression of Colorectal Cancer Growth Using an Adenovirus Vector Expressing an Antisense K-ras RNA Masaru Nakano,*,† Kazunori Aoki,* Nobuyuki Matsumoto,* Shumpei Ohnami,* Kazuteru Hatanaka,* Toshihumi Hibi,† Masaaki Terada,* and Teruhiko Yoshida*,1 *Genetics Division, National Cancer Center Research Institute, 5-1-1 Tsukiji, Chuo-ku, Tokyo 104-0045, Japan † Department of Internal Medicine, School of Medicine, Keio University, 35 Shinanomachi, Shinjuku-ku, Tokyo 160-8582, Japan Received for publication August 9, 2000; accepted in revised form February 26, 2001; published online April 5, 2001

In human colorectal cancer, K-ras point mutations occur in approximately 40 –50% of the cases, a frequency second only to pancreatic cancer (80 –90%). Unlike pancreatic and lung cancers, however, the tumor-suppressive effect of antisense K-ras RNA expression has not been examined for colorectal cancers. A recombinant adenovirus vector expressing an antisense or sense K-ras gene fragment (AxCA-AS-K-ras or AxCA-S-K-ras) was first transduced into seven human colorectal cancer cell lines. Stable expression of antisense or sense K-ras RNA was detected by RNA blot analysis. Western blot analysis confirmed a reduction of up to 25% of K-ras-specific p21 protein in the antisense K-ras-transduced HCT-15 cells. In contrast to our previous findings on pancreatic cancer, the status of K-ras point mutations was not correlated with the growth-suppressive effect of the antisense K-ras vector: both the K-ras-mutation-positive and -negative colorectal cancer cell lines were suppressed for their growth in vitro. There was no growth-inhibitory effect on normal cells such as hepatocytes. Next, to test the efficacy in vivo, HCT-15 cells were inoculated subcutaneously into the left flank of SCID mice, and AxCA-AS-K-ras was injected intratumorally three times after the tumor mass was established. The infection of AxCA-AS-K-ras, but not the control AxCA-S-K-ras, significantly suppressed the growth of the HCT-15 subcutaneous tumor. This study shows that the adenovirus-mediated in vivo gene transfer of the antisense K-ras construct may be a useful therapeutic strategy for colorectal cancer. Key Words: colorectal cancer; antisense; K-ras; in vivo gene transfer; adenovirus vector.

INTRODUCTION In Japan, the incidence of colorectal cancer has been increasing rapidly over the past 20 years. This form of cancer is now the most common malignancy, after lung cancer, gastric cancer, and liver cancer, accounting for 34,397 deaths per year in 1998 (1). In the United States, colorectal cancer is the second leading cause of cancerrelated death and the third most common cancer, with 131,200 people diagnosed with cancer of the colon and rectum and 54,900 deaths from this disease during 1997 alone (2, 3). Eighty percent of the patients who die of colon cancer have metastases in the liver, and half have only liver metastases at the time of death (4). Liver me-

1 To whom correspondence and reprint requests should be addressed at Genetics Division, National Cancer Center Research Institute, 5-1-1 Tsukiji, Chuo-ku, Tokyo 104-0045, Japan. Fax: ⫹81-3-3541-2685. E-mail: [email protected].

MOLECULAR THERAPY Vol. 3, No. 4, April 2001 Copyright © The American Society of Gene Therapy 1525-0016/01 $35.00

tastasis is an important prognostic factor for survival of the patients with colorectal cancer (5), and the surgical removal of liver metastasis is indicated, if no other distant metastases are present (6). However, the results of the current nonsurgical treatments have been disappointing (7). Gene therapy could provide a new therapeutic alternative for the colorectal cancer, especially for the local control of liver metastasis. Colorectal cancer is associated with multiple genetic alterations, including activation of the K-ras oncogene and inactivation of the tumor-suppressor genes p53, DCC, MCC, and APC. Adenoma– carcinoma sequence accounts for carcinogenesis in nearly all sporadic colorectal cancers (8, 9). K-ras mutation appears to promote cell replication and progression from colorectal adenoma to adenocarcinoma in multistep carcinogenesis (10). In contrast to pancreatic cancer, the K-ras mutation appears to be an event at an intermediate stage of multistep carcinogenesis. In human colorectal cancer, K-ras point mutation

491

ARTICLE occurs in about 40 –50% of the cases, an incidence second only to pancreatic cancer, which shows a frequency of 80 –90% for the activated mutation of the K-ras gene (11). Since the frequency of K-ras mutation occurs in less than 20% of other cancers, oncogenic K-ras might have a specific role in carcinogenesis of the colon, rectum, and pancreas. Therefore, K-ras may be a good target in gene therapy of these cancers, and the suppression of K-ras activation may provide some benefits for half of patients with advanced colorectal cancer. Specific suppression of the ras family oncogene expression has been examined in other cancers. Suppression of H-ras expression by antisense oligonucleotide, antisense RNA, or ribozyme led to inhibition of the neoplastic phenotype of bladder carcinoma cells and NIH3T3 cells transformed by H-ras oncogene (12–16). The antisense K-ras retroviral vector infection was useful in suppressing the tumorigenicity of lung cancer in a nude mouse orthotopic transplantation model (17, 18). Adenovirus-mediated gene transfer of antisense K-ras produced an inhibition of monolayer growth and colony formation of lung cancer cells (18). We showed that the liposome-mediated in vivo gene transfer of antisense K-ras construct inhibits pancreatic tumor dissemination in the murine peritoneal cavity (20). However, the fact that antisense K-ras RNA expression vector was effective in the pancreatic and lung cancers does not automatically guarantee that a similar vector is also effective in other types of cancers. A previous study examined the effect of an antisense K-ras oligonucleotide on colorectal cancer cell growth in vitro (21). We believe that the antisense RNA strategy coupled with the highly efficient adenovirus vector expression system is better suited to in vivo gene therapy for metastatic colorectal cancers. This article is the first report evaluating the tumor-suppressive effect and safety of an adenovirus vector expressing antisense K-ras RNA in an in vivo model of colorectal cancer. In this preclinical study, we employed a subcutaneous tumor model to develop a gene therapy for liver metastasis based on the intratumoral injection of the vector; the most important factor of the direct injection gene transfer is the intratumoral vector spread, and the subcutaneous model allows a well-controlled repeated vector injection and accurate follow-up of the change in the tumor size. Although we previously used a liposome DOGS as a gene transfer reagent in murine peritoneal dissemination models, we here selected an adenovirus vector to transfer a gene, because we found that the efficiency of the intratumoral gene delivery by the direct injection of DNA:liposome complexes was too low [(22) and our unpublished data]. Antisense K-ras RNA expression could effectively suppress the growth of colorectal cancer cells in vitro and in vivo. Our data suggest that the direct intratumoral injection of this construct is a safe and efficient way of exerting a tumor-suppressive effect in vivo. This study will be an essential preclinical step toward considering a future clinical trial on colorectal cancer.

492

FIG. 1. Structure of adenovirus vectors used in this study. A 347-bp K-ras cDNA fragment containing exons 1 and 2 and part of exon 3 of wild-type K-ras (nucleotides 171–517) was inserted in sense or antisense orientation downstream of the CAG promoter in the adenovirus vector AxCAwt.

MATERIALS

AND

METHODS

Cell lines and culture conditions. HT-29, WiDr, SW1116, and HCT-15 were obtained from the American Type Culture Collection (ATCC; Rockville, MD). NCCCO33, NCCCO34, and NCCCO31 were established by Drs. Maruyama and Ochiai at the Pathology Division, National Cancer Center Research Institute, Tokyo, Japan (23). All cell lines were maintained in RPMI 1640 medium with 10% fetal bovine serum. Primary cultures of normal human cells, human umbilical endothelial cells (HUVEC), lung microvascular endothelial cells, hepatocytes, smooth muscle cells, and mesangial cells were purchased from Dainippon Pharmaceutical Co. (Osaka, Japan) and cultured according to the supplier’s instructions. Construction of recombinant adenovirus vectors. K-ras cDNA fragments of 347 bp (20) corresponding to exons 1 and 2 and part of exon 3 of wild-type K-ras (nucleotides 171–517) were cloned into the cosmid pAxCAwt (24) downstream of the CAG promoter (25, 26), which is a ␤-actin promoter fused to a cytomegalovirus enhancer (Fig. 1). The EcoT22I-digested adenoviral DNA-terminal protein complex (DNA-TPC) was prepared from the parent Ad5-dlX adenovirus (27). The recombinant adenovirus was produced by cotransfecting 293 cells with the cosmid pAxCAwt containing K-ras and the DNA-TPC by the calcium-phosphate method using a CellPhect Transfection Kit (Amersham Pharmacia Biotech, Buckinghamshire, England). The resulting adenovirus vectors, AxCA-S-K-ras and AxCA-ASK-ras, express the K-ras cDNA fragments in the sense and antisense orientation, respectively. The viruses were purified through CsCl gradients followed by dialysis (28), and the purified viruses were stored at ⫺80°C in PBS containing 10% glycerol. Virus titer was determined by plaque forming assays. The adenovirus vector expressing lacZ, AxCA-LacZ (29), was generously provided by Dr. Izumu Saito (Institute of Medical Science, University of Tokyo, Tokyo, Japan). Analysis of gene transduction efficiency in vitro. Three days following infection with AxCA-LacZ, the seven human colorectal cancer cell lines were washed with PBS, fixed with 0.25% glutaraldehyde in PBS for 5 min at room temperature, and developed in a substrate solution [5 mM K3Fe(CN)6, 5 mM K4(CN)6, 2 mM MgCl2, and 1 mg/ml X-gal] at 37°C for 3 h. The unstained and X-gal-stained cells from four representative highpower fields were counted in each section, and the percentage of X-galstained cells was calculated. RNA blot analysis. Poly(A)⫹ RNA was isolated from HCT-15 cells transduced by AxCA-S-K-ras and AxCA-AS-K-ras. Two micrograms of poly(A)⫹ RNA was size-fractionated on a 1.0% denaturing agarose gel, transferred onto a nitrocellulose membrane (NitroPlus; MSI, Westboro, MA), and hybridized with a strand-specific RNA probe. A 372-bp K-ras cDNA fragment containing exons 1 and 2 and part of exon 3 sequences were amplified by PCR and subcloned into a Bluescript plasmid vector. Sense and MOLECULAR THERAPY Vol. 3, No. 4, April 2001 Copyright © The American Society of Gene Therapy

ARTICLE antisense RNA probes were synthesized using either a T7 or a T3 bacteriophage DNA-dependent RNA polymerase. The hybridization was performed in 50% formamide, 5⫻ Denhardt’s solution, 0.1% SDS, 5⫻ SSPE, and 100 ␮g/ml salmon testis DNA at 42°C for 16 h. The filters were then washed in 2⫻ SSPE and 0.1% SDS at 65°C. Western blot analysis. The colorectal cancer cells transduced by AxCA-SK-ras and AxCA-AS-K-ras were lysed in RIPA buffer (10 mM Tris–HCl, pH 7.4, 1% deoxycholate, 1% Nonidet P-40, 150 mM NaCl, 0.1% SDS, 0.2 mM phenylmethylsulfonyl fluoride, 1 ␮g/ml aprotinin, 1 ␮g/ml leupeptin). Eighty micrograms of the parental or vector-transduced HCT-15 cell lysates was heated at 90°C for 5 min, size-fractionated by 8 –16% SDS– polyacrylamide gel (TEFCO, Tokyo, Japan), and electroblotted onto polyvinylidene difluoride membranes (Millipore, Bedford, MA). The K-ras protein was detected by a K-ras-specific p21 monoclonal antibody (Calbiochem, Darmstadt, Germany) (30) using the enhanced chemiluminescence system (Amersham Pharmacia Biotech). In vitro growth analysis. The seven colorectal cancer cell lines (HT-29, WiDr, SW1116, HCT-15, NCCCO33, NCCCO34, NCCCO31) were seeded at 1 ⫻ 105 in 100-mm plates. Twenty-four hours later, cell monolayers were incubated for 1 h with the virus at a m.o.i. of 30 pfu/cell in a minimal amount of culture medium (0.3 ml/100-mm plate, 37°C in 5% CO2 incubator, rocking the plates every 15 min to avoid drying). More culture medium was then added and the plates were incubated for an additional 7 days in triplicate. Cells were harvested and counted by trypan blue exclusion at days 1, 3, 5, and 7 following infection. The five normal cell lines were seeded at 2 ⫻ 103 per well in 96-well microtiter plates and infected with AxCA-S-K-ras or AxCA-AS-K-ras at m.o.i. of 10, 30, and 100. The cell numbers were assayed by cell proliferation assay (Tetracolor One; Seikagaku Corp., Tokyo) 5 days after infection. Gene transduction efficiency into the subcutaneous tumor. Since HCT-15 could rapidly form a subcutaneous tumor, we used this cell line in an in vivo gene transfer model. Five-week-old male SCID mice were obtained from Charles River Japan (Kanagawa, Japan) and kept in a specific-pathogen-free environment. HCT-15 cells were harvested with trypsin and resuspended in Hanks’ balanced salt solution. The HCT-15 cell suspensions (5 ⫻ 106 cells) were injected subcutaneously into the left flank. Adenovirus-mediated gene transfer in the HCT-15 subcutaneous tumor was first examined using AxCA-LacZ. When the HCT-15 subcutaneous tumor nodule reached the size of ⬃4 mm in diameter, 0.5 ⫻ 109 pfu of AxCA-LacZ was injected intratumorally three times every 24 h. One day after the last administration of AxCA-LacZ, the subcutaneous tumor was removed and fixed with 0.25% glutaraldehyde in PBS for 5 min at room temperature and developed as above. Then the X-gal-stained tissues were counterstained with hematoxylin– eosin. Effect of AxCA-AS-K-ras on the tumor growth in vivo. The HCT-15-derived subcutaneous tumors ⬃4 mm in diameter were established as described for the AxCA-LacZ transduction experiment. Fifty microliters of viral solution (AxCA-S-K-ras or AxCA-AS-K-ras; 1 ⫻ 1010 pfu/ml) was injected into the tumor with a 27-gauge hypodermic needle at 24-h intervals a total of three times. Ten mice were injected with PBS alone as a control group. Mice were injected with AxCA-S-K-ras or AxCA-AS-K-ras in groups of 12. Two mice of each group were sacrificed for the histological examination of the subcutaneous tumor, and the remaining animals were observed for tumor growth. The short (r) and long (l ) diameters of the tumors were measured every day for 21 days and the tumor volume was calculated as r 2 l/ 2.

RESULTS

TABLE 1 Status of K-ras Gene in Colorectal Cancer Cell Lines Cell line

Histologya

Codon 12

Codon 13

Codon 61

SW1116 HT-29 WiDr HCT-15 NCCCO33 NCCCO34 NCCCO31

Well Moderate Moderate Moderate Moderate Moderate Poor

GGT 3 GCTb WT WTc WT WT WT WT

WTb WT WTb GGC 3 GAC GGC 3 GAC GGC 3 GAC WT

WT WT WT WT WT WT WT

a Histology of the tumor produced by intramuscular injection of each cell line to SCID mice: well, moderate, and poor refer to well-, moderately, and poorly differentiated adenocarcinomas, respectively (23). b Our analysis of K-ras sequences showed the same results as reported in Ref. 21. WT, wild type. c In Ref. 21, WiDr cells were reported to contain a GGT to GCT point mutation at codon 12.

line from ATCC, and the wild-type K-ras sequence was confirmed again.

Gene Transduction Efficiency in Vitro The AxCA-LacZ vector was used to estimate gene transduction efficiency in the colorectal cancer cell lines. AxCA-LacZ experiment showed that the transduction efficiency at m.o.i. of 30 was 28.2 ⫾ 3.9% in HT29, 74.5 ⫾ 5.9% in WiDr, 43.9 ⫾ 9.4% in SW1116, 100% in HCT-15, 67.8 ⫾ 4.0% in NCCCO33, 39.0 ⫾ 9.1% in NCCCO34, and 26.8 ⫾ 9.5% in NCCCO31.

Expression of Antisense or Sense K-ras RNA in Transduced Cells Poly(A)⫹ RNA was extracted from the parental and vector-transduced HCT-15 cells and analyzed by RNA blot hybridization with sense or antisense K-ras RNA probe. About 1 kb antisense or sense K-ras RNA, which is the expected size of the transcript driven by the CAG promoter, was detected (Fig. 2A). The autoradiogram exposure time was 30 min, and the endogenous K-ras mRNA [major 5.5 kb and minor 3.8 kb in size (31)] was not detectable in HCT-15 cells. The endogenous K-ras messages were clearly visible after 48 h exposure, suggesting an excess expression of the transgene compared to the relatively low level of endogenous K-ras expression.

K-ras Gene Status of the Colorectal Cancer Cells

Downregulation of K-ras p21 Protein by AxCA-AS-K-ras

First, we sequenced the K-ras gene for all the cell lines examined in this study, and the results are shown in Table 1. The K-ras codons 12 and 13 were previously sequenced for SW1116 and WiDr cells, but our WiDr cells showed the wild-type sequence and not the GGT to GCT mutation as reported before (21). We reobtained the WiDr cell

Western blot analysis using a K-ras-specific p21 monoclonal antibody showed a reduction of the K-ras p21 protein in the colorectal cancer cells transduced by AxCA-ASK-ras (Fig. 2B). As a parallel, AxCA-LacZ experiment showed that the transduction efficiency at m.o.i. of 30 was higher than that at m.o.i. of 10 (data not shown), and

MOLECULAR THERAPY Vol. 3, No. 4, April 2001 Copyright © The American Society of Gene Therapy

493

ARTICLE

FIG. 2. (A) RNA blot analysis of K-ras RNA expression in HCT-15 cells after adenovirus vector infection. Poly(A)⫹ RNA was extracted 2 days after infection at m.o.i. of 30 pfu/cell. The filter was hybridized with a strand-specific RNA probe (antisense, left; sense, right). P, parent (uninfected) cells; S, cells infected by the sense K-ras vector; AS, cells infected by the antisense K-ras vector. (B) Western blot analysis of K-ras p21 protein expression in colorectal cancer cells. Protein was extracted 2 days after infection at m.o.i. of 30. Each lane was loaded with 80 ␮g of the protein and probed with a K-ras-specific p21 monoclonal antibody. P, parent (uninfected) cells; S10, cells infected by the control sense K-ras vector at m.o.i. of 10; AS10, cells infected by the antisense K-ras vector at m.o.i. of 10; S30, cells infected by the sense K-ras vector at m.o.i. of 30; AS30, cells infected by the antisense K-ras vector at m.o.i. of 30.

494

MOLECULAR THERAPY Vol. 3, No. 4, April 2001 Copyright © The American Society of Gene Therapy

ARTICLE the level of p21 suppression in HCT-15 cells was greater at m.o.i. of 30 than at m.o.i. of 10 (Fig. 2B).

Growth Suppression of AxCA-AS-K-ras-Transduced Cells All seven colorectal cancer cells expressing antisense K-ras RNA showed remarkable growth retardation compared to those expressing sense K-ras RNA, irrespective of the status of K-ras point mutation (Fig. 3A). AxCA-S-K-ras induced growth suppression in HCT-15, NCCCO33, and NCCCO34 cells, suggesting that these cells are sensitive to adenoviral toxicity even at the low m.o.i. of 30. However, the antisense K-ras-specific effect was still significant in these cells. To study the effect of antisense K-ras expression on normal cells, five primary cultures of human normal cells, HUVEC, lung microvascular endothelial cells, hepatocytes, smooth muscle cells, and mesangial cells, were infected with AxCA-AS-K-ras. The growth-inhibitory effect of antisense K-ras was not observed in those normal cells (Fig. 3B).

Adenovirus-Mediated Transduction into Subcutaneous Tumor Adenovirus-mediated gene transfer and expression in the subcutaneous tumor was analyzed using transduction of the lacZ gene. Intratumoral administration of 1.5 ⫻ 109 pfu of AxCA-LacZ resulted in transgene expression in 80 –90% of established HCT-15 subcutaneous tumor cells (Fig. 4A, left), which was congruous with our previous report (32). Importantly, the lacZ gene expression was not detected in the peritumoral normal tissues, showing a good confinement of the vector distribution within the tumor mass. AxCA-AS-K-ras-injected tumor did not show any blue staining (Fig. 4A, right).

Inhibition of Subcutaneous Tumor Growth by AxCA-AS-K-ras To evaluate the usefulness of the adenovirus-mediated antisense K-ras RNA expression in the local control of the colorectal cancer growth in vivo, HCT-15 cells were inoculated subcutaneously into the left flank of SCID mice. After nodules reached ⬃4 mm in diameter, the tumors were injected with AxCA-S-K-ras and AxCA-AS-K-ras three times at 24-h intervals. The infection of AxCA-AS-K-ras suppressed the growth of HCT-15 subcutaneous tumor, while AxCA-S-K-ras did not (Fig. 5). The subcutaneous tumor growth was suppressed for 7 days from the last AxCA-AS-K-ras injection and then the tumor began to grow again. Histological analysis of the tumors after the injection of AxCA-AS-K-ras revealed massive cell death and mild peritumoral infiltration of mononuclear cells and lymphocytes (Fig. 4B, right). Injection of the AxCAS-K-ras at the same dose had no effect on tumor growth of HCT-15 cells, and no significant histological changes were observed in the tumor tissue (Fig. 4B, left). There was no difference in mouse survival between antisense group and sense group. MOLECULAR THERAPY Vol. 3, No. 4, April 2001 Copyright © The American Society of Gene Therapy

DISCUSSION Adenovirus vectors expressing antisense or sense K-ras RNA were transduced into the seven human colorectal cancer cell lines and normal cell lines. RNA blot analysis showed overexpression of the transgene RNA over the endogenous K-ras mRNA, and Western blot analysis confirmed a reduction of K-ras-specific p21 protein in colorectal cancer cells transduced with AxCA-AS-K-ras. The growth of all colorectal cancer cells was significantly suppressed following transduction of AxCA-AS-K-ras irrespective of the status of K-ras point mutation. This lack of mutation dependency was not observed in pancreatic cancer cells, in which the growth inhibitory effect of the antisense K-ras RNA was more pronounced in the cells with K-ras point mutation than in those with wild-type K-ras gene (20). On the other hand, AxCA-AS-K-ras did not show growth-suppressive effect in any of the five normal cell lines, including the hepatocytes. It is conceivable that the role of K-ras mutation in the carcinogenic sequences is different between the colorectal and pancreatic cancers; in colorectal cancer development, the K-ras mutation occurs mostly in cancers and also in adenomas larger than 1 cm in diameter (33), while in pancreatic cancers, K-ras mutation has been detected in earlier stages, including ductal hyperplasia (34 –36). Difference in the mutation dependency of the antisense Kras RNA-mediated growth suppression observed in this study may also suggest a presence of a distinct abnormality of K-ras signaling between the colorectal and the pancreatic cancers. From the standpoint of gene therapy, the lack of mutation dependency of the antisense K-ras strategy in colorectal cancers, as well as the apparent resistance of normal primary cells to the antisense K-ras RNA expression, may allow the use of this therapy for a wide range of patients. A previous study using the K-ras antisense oligonucleotides claimed a K-ras mutation-specific growth suppression in vitro; they examined four colorectal cancer cell lines with K-ras mutation, including WiDr, and all of the four cell lines showed the antisense oligonucleotide-induced growth inhibition. In contrast, only one colorectal cancer cell line with wild-type K-ras gene, COLO 201, was analyzed and found to be resistant to the antisense oligonucleotides (21). Although we did not include the COLO 201 cells in our study, we also observed that AxCA-AS-Kras suppressed the growth of WiDr. However, our sequence analysis revealed only the wild-type K-ras gene in the WiDr cells. Thus, the discrepancy between the previous report and the current study could be due to the use of different cell lines or sublines, but it is also possible that high gene transduction efficiency and strong expression of antisense RNA by adenovirus vector generate a quite different condition from addition of antisense oligonucleotides to the cell culture medium. Adenovirus-mediated gene transfer and expression in the subcutaneous tumor was analyzed by transduction of the lacZ gene. LacZ expression was observed in 80 –90% of the cells in the subcutaneous tumor following the intra-

495

ARTICLE

FIG. 3. In vitro growth curve of seven colorectal cancer cell lines and normal cells. (A) Colorectal cancer cells were infected by AxCA-S-K-ras (open circles) or AxCA-AS-K-ras (closed squares) at m.o.i. of 30 pfu/cell and counted by trypan blue exclusion 1, 3, 5, and 7 days after infection and compared with uninfected control cells (closed circles). K-ras mutation status of each cell line is shown in Table 1. (B) Normal cell lines. The relative cell numbers of AxCA-AS-K-ras-infected cells compared with those of AxCA-S-K-ras-infected cells are shown. The gene transduction efficiency at m.o.i. of 30 was more than 89% by X-gal staining in all of the five normal cell lines.

tumoral injection of a total 1.5 ⫻ 109 pfu AxCA-LacZ, but was not detected in the peripheral mouse normal tissues. Direct intratumoral injection of the adenovirus vector

496

showed good transduction efficiency and safety in clinical trials (37– 40). In patients with advanced non-small-cell lung cancer, wild-type p53 gene therapy by intratumoral MOLECULAR THERAPY Vol. 3, No. 4, April 2001 Copyright © The American Society of Gene Therapy

ARTICLE

FIG. 4. (A) Adenovirus-mediated lacZ gene expression in subcutaneous tumor of HCT-15. After intratumoral administration of AxCA-LacZ in mice, expression of the lacZ gene product was demonstrated in 80 –90% of established HCT-15 subcutaneous tumor cells (left). AxCA-AS-K-ras was injected as a negative control for X-gal staining (right). (B) Cell death induced by adenovirus vector expressing an antisense K-ras RNA in vivo. Subcutaneous tumors were stained by hematoxylin– eosin 3 days after intratumoral administration of AxCA-S-K-ras (left) and AxCA-AS-K-ras (right). N, necrotic tissue; V, viable tissue.

injection of a replication-defective adenovirus expression vector was shown to be safe and effective (39). Intratumoral injection of the AxCA-AS-K-ras prevented HCT-15 tumor growth. In histological analysis 3 days after the last injection of the vectors, a more massive cell death was apparent in the tumor tissues injected with AxCA-AS-K-ras compared to the AxCA-S-K-ras- or PBSinjected control tumors. In this in vivo experiment, m.o.i. is estimated to be approximately 30, a figure comparable to that in the in vitro study. Moreover, it was reported that the suppression of K-ras inhibits tumor angiogenesis through the down-regulation of VEGF (41, 42). The combination of antiproliferative and antiangiogenic effect may result in growth inhibition of the subcutaneous tuMOLECULAR THERAPY Vol. 3, No. 4, April 2001 Copyright © The American Society of Gene Therapy

mor in vivo. No mouse died following subcutaneous intratumoral injection of the AxCA-AS-K-ras, and no sign of general toxicity was seen in the AxCA-AS-K-ras-infected mice. The expression of exogenous gene mediated by an adenovirus vector is transient. It begins 4 h after injection, peaks on day 3, and lasts for 2–3 weeks (43). Our data suggest that repeated intratumoral injection of the AxCAAS-K-ras may be necessary to achieve a sustained tumorsuppressive effect. In fact, treatment with up to six intratumoral injections of replication-defective adenovirus vector expressing wild-type p53 at monthly intervals was well tolerated in patients with advanced non-small-cell lung cancer (39).

497

ARTICLE

FIG. 5. Growth-inhibitory effect of AxCA-AS-K-ras in vivo. The HCT-15 cells were injected subcutaneously into the left flank of SCID mice. 0.5 ⫻ 109 pfu of the recombinant adenovirus vector was injected into the tumor three times every 24 h after the tumor nodule reached the size of ⬃4 mm in diameter. The short (r) and long (l ) diameters of the tumors were measured every day for 21 days and the tumor volume was calculated as r2l/2.

In summary, in vitro delivery of the antisense K-ras into seven colorectal cancer cell lines suppressed the cell growth irrespective of the status of K-ras point mutation, and direct intratumoral antisense K-ras delivery led to growth suppression of established tumor without any specific toxicity in animal experiments, which is a very important piece of information for the future development of clinical protocols. Further research, including a clinical study, is expected to show the usefulness of the intratumoral injection of the antisense K-ras RNA-expressing adenovirus vector as a viable option for the local control of advanced colorectal cancer, such as treatment of unresectable liver metastases and improvement of ileus with advanced primary lesion. ACKNOWLEDGMENTS We thank Dr. Izumu Saito for providing adenovirus vector expressing lacZ, Drs. Maruyama and Ochiai for their valuable information and suggestions on colorectal cancer cell lines. This work was supported in part by a grant-in-aid from the 2nd-Term Comprehensive 10-Year Strategy for Cancer Control and Health Sciences Research Grants from the Ministry of Health and Welfare of Japan. Masaru Nakano and Kazuteru Hatanaka are awardees of a Research Resident Fellowship from the Foundation for Promotion of Cancer Research.

REFERENCES 1 Vital Statistics of Japan 1998. (2000). Statistics and Information Department, Minister’s Secretariat, Ministry of Health and Welfare, Tokyo. 2 Wingo, P. A., Tong, T., and Bolden, S. (1995). Cancer statistics, 1995. CA Cancer J. Clin. 45: 8 –30. 3 Alberts, D. S., Slattery, M. L., Giovannucci, E., Shike, M., and Garcia, D. J. (1998). Primary prevention of colon cancer with dietary and micronutrient interventions. Cancer 83: 1734 –1739. 4 Drebin, J. A., and Niederhuber, J. E. (1993). In Current Therapy in Oncology: Colon Cancer (J. E. Niederhuber, Ed.), pp. 426 – 431. Dekker, St. Louis. 5 Fong, Y., Kemeny, N., Paty, P., Blumgart, L. H., and Cohen, A. M. (1996). Treatment of colorectal cancer: Hepatic metastasis. Semin. Surg. Oncol. 12: 219 –252. 6 Hughes, K., Scheele, J., and Sugarbaker, P. H. (1989). Surgery for colorectal cancer metastatic to the liver: Optimizing the results of treatment. Surg. Clin. North Am. 69: 339 –359. 7 Lebovic, G. S., and Niederhuber, J. E. (1993). In Current Therapy in Oncology: Colorectal Cancer Metastatic to the Liver: Hepatic Arterial Infusion (J. E. Niederhuber, Ed.), pp. 389 –396. Dekker, St. Louis. 8 Morson, B. C., Dawson, I. M. P., Day, D. W., Jass, J. R., Price, A. B., and Williams, G. T. (1990). Morson and Dawson’s Gastrointestinal Pathology, 3rd ed. Blackwell Sci. Oxford. 9 Fearon, E. R., and Vogelstein, B. (1990). A genetic model for colorectal tumorigenesis. Cell 61: 759 –767.

498

10 Bos, J. L. (1988). The ras gene family and human carcinogenesis. Mutat. Res. 195: 255–271. 11 Bos, J. L. (1989). Ras oncogenes in human cancer: A review. Cancer Res. 49: 4682– 4689. 12 Lu, G. R., Wu, C. J., Ke, Y., Xiong, X., Li, Y. J., Liu, J. M., Xu, Y., Xu, W., Li, J. Y., Deng, G. R., and E, Z. (1991). Inhibited neoplastic phenotype by the c-Ha-ras antisense RNA. Sci. China Series B Chem. Life Sci. Earth Sci. 34: 1485–1489. 13 Daaka, Y., and Wickstrom, E. (1990). Target dependence of antisense oligodeoxynucleotide inhibition of c-Ha-ras p21 expression and focus formation in T24-transformed NIH3T3 cells. Oncogene Res. 5: 267–275. 14 Saison-Behmoaras, T., Tocque, B., Rey, I., Chassignol, M., Thuong, N. T., and Helene, C. (1991). Short modified antisense oligonucleotides directed against Ha-ras point mutation induce selective cleavage of the mRNA and inhibit T24 cells proliferation. EMBO J. 10: 1111–1118. 15 Kashani-Sabet, M., Funato, T., Florenes, V. A., Fodstad, O., and Scanlon, K. J. (1994). Suppression of the neoplastic phenotype in vivo by an anti-ras ribozyme. Cancer Res. 54: 900 –902. 16 Monia, B. P., Johnston, J. F., Ecker, D. J., Zounes, M. A., Lima, W. F., and Freier, S. M. (1992). Selective inhibition of mutant Ha-ras mRNA expression by antisense oligonucleotides. J. Biol. Chem. 267: 19954 –19962. 17 Georges, R. N., Mukhopadhyay, T., Zhang, Y., Yen, N., and Roth, J. A. (1993). Prevention of orthotopic human lung cancer growth by intratracheal instillation of a retroviral antisense K-ras construct. Cancer Res. 53: 1743–1746. 18 Zhang, Y., Mukhopadhyay, T., Donehower, L. A., Georges, R. N., and Roth, J. A. (1993). Retroviral vector-mediated transduction of K-ras antisense RNA into human lung cancer cells inhibits expression of the malignant phenotype. Hum. Gene Ther. 4: 451– 460. 19 Alemany, R., Ruan, S., Kataoka, M., Koch, P. E., Mukhopadhyay, T., Cristiano, R. J., Roth, J. A., and Zhang, W.-W. (1996). Growth inhibitory effect of anti-K-ras adenovirus on lung cancer cells. Cancer Gene Ther. 3: 296 –301. 20 Aoki, K., Yoshida, T., Sugimura, T., and Terada, M. (1995). Liposome-mediated in vivo gene transfer of antisense K-ras construct inhibits pancreatic tumor dissemination in the murine peritoneal cavity. Cancer Res. 55: 3810 –3816. 21 Sakakura, C., Hagiwara, A., Tujimoto, H., Ozaki, K., Sakakibara, T., Oyama, T., Ogaki, M., Imanishi, T., Yamazaki, J., and Takahashi, T. (1995). Inhibition of colon cancer cell proliferation by antisense oligonucleotides targeting the messenger RNA of the Ki-ras gene. Anti-Cancer Drugs 6: 553–561. 22 Coll, J. L., Chollet, P., Brambilla, E., Desplanques, D., Behr, J. P., and Favrot, M. (1999). In vivo delivery to tumors of DNA complexed with linear polyethylenimine. Hum. Gene Ther. 10: 1659 –1666. 23 Yasui, N., Sakamoto, M., Ochiai, A., Ino, Y., Akimoto, S., Orikasa, A., Kitajima, M., and Hirohashi, S. (1997). Tumor growth and metastasis of human colorectal cancer cell lines in SCID mice resemble clinical metastatic behaviors. Invasion Metastasis 17: 259 –269. 24 Kanegae, Y., Lee, G., Sato, Y., Tanaka, M., Nakai, M., Sakaki, T., Sugano, S., and Saito, I. (1995). Efficient gene activation in mammalian cells by using recombinant adenovirus expressing site-specific Cre recombinase. Nucleic Acids Res. 23: 3816 –3821. 25 Miyazaki, J., Takaki, S., Araki, K., Tashiro, F., Tominaga, A., Takatsu, K., and Yamamura, K. (1989). Expression vector system based on the chicken ␤-actin promoter directs efficient production of interleukin-5. Gene 79: 269 –277. 26 Niwa, H., Yamamura, K., and Miyazaki, J. (1991). Efficient selection for high-expression transfectants with a novel eukaryotic vector. Gene 108: 193–200. 27 Saito, I., Oya, Y., Yamamoto, K., Yuasa, T., and Shimojo, H. (1985). Construction of nondefective adenovirus type 5 bearing a 2.8-kilobase hepatitis B virus DNA near the right end of its genome. J. Virol. 54: 711–719.

MOLECULAR THERAPY Vol. 3, No. 4, April 2001 Copyright © The American Society of Gene Therapy

ARTICLE 28 Kanegae, Y., Makimura, M., and Saito, I. (1994). A simple and efficient method for purification of infectious recombinant adenovirus. Jpn. J. Med. Sci. Biol. 47: 157–166. 29 Miyake, S., Makimura, M., Kanegae, Y., Harada, S., Sato, Y., Takamori, K., Tokuda, C., and Saito, I. (1996). Efficient generation of recombinant adenoviruses using adenovirus DNA–terminal protein complex and a cosmid bearing the full-length virus genome. Proc. Natl. Acad. Sci. USA 93: 1320 –1324. 30 Mukhopadhyay, T., Tainsky, M., Cavender, A. C., and Roth, J. A. (1991). Specific inhibition of K-ras expression and tumorigenicity of lung cancer cells by antisense RNA. Cancer Res. 51: 1744 –1748. 31 Capon, D. J., Seeburg, P. H., McGrath, J. P., Hayflick, J. S., Edman, U., Levinson, A. D., and Goeddel, D. V. (1983). Activation of Ki-ras 2 gene in human colon and lung carcinomas by two different point mutations. Nature 304: 507–513. 32 Aoki, K., Akyurek, L. M., San, H., Leung, K., Parmacek, M. S., Nabel, E. G., and Nabel, G. J. (2000). Restricted expression of an adenoviral vector encoding Fas ligand (CD95L) enhances safety for cancer gene therapy. Mol. Ther. 1: 555–565. 33 Smith, A. J., Stern, H. S., Penner, M., Hay, K., Mitri, A., Bapat, B. V., and Gallinger, S. (1994). Somatic APC and K-ras codon 12 mutations in aberrant crypt foci from human colons. Cancer Res. 54: 5527–5530. 34 Yanagisawa, A., Ohtake, K., Ohashi, K., Hori, M., Kitagawa, T., Sugano, H., and Kato, Y. (1993). Frequent c-Ki-ras oncogene activation in mucous cell hyperplasias of pancreas suffering from chronic inflammation. Cancer Res. 53: 953–956. 35 Cerny, W. L., Mangold, K. A., and Scarpelli, D. G. (1992). K-ras mutation is an early event in pancreatic duct carcinogenesis in the Syrian golden hamster. Cancer Res. 52: 4507– 4513. 36 DiGiuseppe, J. A., Hruban, R. H., Offerhaus, G. J., Clement, M. J., van den Berg, F. M., Cameron, J. L., and van Mansfeld, A. D. (1994). Detection of K-ras mutations in mucinous pancreatic duct hyperplasia from a patient with a family history of pancreatic carcinoma. Am. J. Pathol. 144: 889 – 895.

MOLECULAR THERAPY Vol. 3, No. 4, April 2001 Copyright © The American Society of Gene Therapy

37 Zhang, W.-W., Alemany, R., Wang, J., Koch, P. E., Ordonez, N. G., and Roth, J. A. (1995). Safety evaluation of Ad5CMV-p53 in vitro and in vivo. Hum. Gene Ther. 6: 155–164. 38 Roth, J. A., Nguyen, D., Lawrence, D. D., Kemp, B. L., Carrasco, C. H., Ferson, D. Z., Hong, W. K., Komaki, R., Lee, J. J., Nesbitt, J. C., Pisters, K. M. W., Putnam, J. B., Schea, R., Shin, D. M., Walsh, G. L., Dolormente, M. M., Han, C.-I., Martin, F. D., Yen, N., Xu, K., Stephens, L. C., Mcdonnell, T. J., Mukhopadhyay, T., and Cai, D. (1996). Retrovirusmediated wild-type p53 gene transfer to tumors of patients with lung cancer. Nat. Med. 2: 985–991. 39 Schuler, M., Rochlitz, C., Horowitz, J. A., Schlegel, J., Perruchoud, A. P., Kommoss, F., Bolliger, C. T., Kauczor, H.-U., Dalquen, P., Fritz, M. A., Swanson, S., Herrmann, R., and Huber, C. (1998). A phase I study of adenovirus-mediated wild-type p53 gene transfer in patients with advanced non-small cell lung cancer. Hum. Gene Ther. 9: 2075–2082. 40 Roth, J. A., Swisher, S. G., Merritt, J. A., Lawrence, D. D., Kemp, B. L., Carrasco, C. H., El-Naggar, A. K., Fossella, F. V., Glisson, B. S., Hong, W. K., Khurl, F. R., Kurie, J. M., Nesbitt, J. C., Pisters, K., Putnam, J. B., Schrump, D. S., Shin, D. M., and Walsh, G. L. (1998). Gene therapy for non-small cell lung cancer: A preliminary report of a phase I trial of adenoviral p53 gene replacement. Semin. Oncol 25(Suppl. 8): 33–37. 41 Rak, J., Mitsuhashi, Y., Bayko, L., Filmus, J., Shirasawa, S., Sasazuki, T., and Kerbel, R. S. (1995). Mutant ras oncogenes upregulate VEGF/VPF expression: Implications for induction and inhibition of tumor angiogenesis. Cancer Res. 55: 4575– 4580. 42 Okada, F., Rak, J. W., Croix, B. S., Lieubeau, B., Kaya, M., Roncari, L., Shirasawa, S., Sasazuki, T., and Kerbel, R. S. (1998). Impact of oncogenes in tumor angiogenesis: Mutant K-ras up-regulation of vascular endothelial growth factor/vascular permeability factor is necessary, but not sufficient for tumorigenicity of human colorectal carcinoma cells. Proc. Natl. Acad. Sci. USA 95: 3609 –3614. 43 Zhang, W.-W., Fang, X., Mazur, W., French, B. A., Georges, R. N., and Roth, J. A. (1994). High-efficiency gene transfer and high-level expression of wild-type p53 in human lung cancer cells mediated by recombinant adenovirus. Cancer Gene Ther. 1: 5–13.

499