Relaxin Expression From Tumor-Targeting Adenoviruses and Its Intratumoral Spread, Apoptosis Induction, and Efficacy

Joo-Hang Kim 0 1 Young-Sook Lee 0 1 Hoguen Kim 0 1 Jing-Hua Huang 0 1 A-Rum Yoon 0 1 Chae-Ok Yun 0 1 0 Affiliations of authors: Brain Korea 21 Project for Medical Sciences, Institute for Cancer Research, Yonsei Cancer Center (JHK , YSL, JHH, ARY, COY) and Department of Pathology (HK), Yonsei University College of Medicine , Seoul, Korea . College of Medicine , 134 Shinchon-Dong, Seodaemun-Gu, Seoul, Korea ( 1 Journal of the National Cancer Institute , Vol. 98, No. 20, October 18, 2006 - Although the selective replication and spread of oncolytic adenoviruses within cancer cells and tissues has had some success as an anticancer treatment, its promise has been limited because of the uneven penetration and distribution of these viruses in tumor tissues. To date, ONYX-015, an oncolytic adenovirus in which the E1B 55-kDa gene has been deleted, has been administered to more than 300 cancer patients with various tumor types (1,2) but has had only limited success (i.e., local tumor regression rates of 0%14%). Sauthoff et al. (3) demonstrated that high levels of adenoviruses persisted in xenograft tumors for at least 8 weeks after intratumoral injection of wild-type adenoviruses, but the viral distribution pattern in the tumor tissue was uneven and patchy. Harrison et al. (4) detected a high level of viruses in persistent viable tumors up to 100 days after the initial viral injection. Thus, despite long-term viral persistence in tumors, the limited spread of virus to cells throughout the tumor could explain the low response rates observed. Connective tissue and the extracellular matrix appear to play a role in inhibiting viral spread in tumors. Kuriyama et al. (5) demonstrated that treatment of human U87, U251, or SF767 glioblastoma multiformederived brain tumor xenografts with collagenase, dispase, or trypsin, before the intratumoral injection of adenovirus enhanced virus-mediated gene transduction, and Maillard et al. (6) reported that elastase pretreatment before delivery of adenoviral vectors into rabbit iliac arteries increased viral transduction efficiency. Relaxin is a peptide hormone that is structurally related to insulin and insulin-like growth factors (7). Treatment of human lung fibroblasts and bleomycin-induced mouse lung fibrosis (i.e., the common end stage of many pneumopathies) tumors with relaxin decreases the synthesis and secretion of interstitial collagens and increases the expression of matrix metalloproteinase and procollagenase (8). Relaxin is a potent inhibitor of collagen expression when collagen is overexpressed, but it does not markedly alter basal levels of collagen expression, in contrast to other collagen-modulatory cytokines, such as interferon gamma (9). To further explore the barrier role of the extracellular matrix and connective tissue in inhibiting viral spread and penetration within tumor masses, we determined whether the expression of relaxin by adenoviruses increases their spread in tumor tissues. MATERIALS AND METHODS Cell Lines and Cell Culture The human embryonic kidney cell line 293 that expresses the adenoviral E1 region, the brain cancer cell lines U343 and U87MG, the cervical cancer cell line C33A, the liver cancer cell line Hep3B, and the nonsmall lung cancer cell line A549 were purchased from the American Type Culture Collection (ATCC, Manassas, VA). Murine B16BL6 melanoma cells (a metastatic variant of B16 melanoma cells) were obtained from Dr Y. S. Park (Yonsei University, Wonju, South Korea). All cell lines were cultured in Dulbeccos modified Eagle medium (DMEM; Gibco BRL, Grand Island, NY) supplemented with 10% fetal bovine serum (Gibco BRL), 2 mM l-glutamine, penicillin (100 IU/mL), and streptomycin (50 g/mL), except for Hep3B cells, which were cultured in modified Eagle medium (MEM; Gibco BRL), and B16BL6 cells, which were cultured in MEM supplemented with 5% fetal bovine serum, MEM vitamin solution (1 mM; Gibco BRL, product 11120-052), 100 mM sodium pyruvate, 10 mM of MEM nonessential amino acids solution (10 mM; Gibco BRL, product 11140-050), penicillin (500 IU/mL), and streptomycin (50 g/mL). All cell lines were maintained at 37 C in a humidified atmosphere at 5% CO2 and 95% air. Escherichia coli was propagated in Luria Bertani medium at 37 C. Generation of Relaxin-Expressing Adenoviruses To generate adenoviruses that express relaxin at the E3 region, we first excised the relaxin gene from the vector pDNR-LIB-RLX (ATCC) with the endonucleases SalI and HindIII and subcloned into the vector pCA14 (Microbix, Ontario, Canada) to generate pCA14-RLX. CMV-relaxin-polA expression cassette was then excised from pCA14-RLX and cloned into the adenovirus E3 shuttle vector pSP72-E3 (10) that had been predigested with the endonuclease BamHI to generate pSP72-E3/CMV-RLX. The newly constructed pSP72-E3/CMV-RLX shuttle vector was then linearized with PvuI digestion. The replication-incompetent adenoviral vector pdl-lacZ that expresses lacZ (i.e., -galactosidase) at E1 region of adenovirus and E1B 19-kDa and E1B 55-kDa deleted replication-competent adenoviral vector pAd-ElB were linearized with SpeI digestion. The linearized pSP72-E3/CMVRLX shuttle vector was then cotransformed into E. coli BJ5183 with the SpeI-digested pdl-lacZ or pAd-ElB for homologous recombination (11) to generate pdl-lacZ-RLX and pAd-E1BRLX adenoviral vector, respectively (Fig. 1, A). E1-deleted replication-incompetent adenovirus (dl-lacZ) and E1B 19-kDa and E1B 55-kDadeleted replication-competent adenovirus (Ad-ElB) were also prepared as previously described (12). All viruses were propagated in 293 cells, and the purification, titration, and quality analysis of all adenoviruses used were performed as previously described (13). The number of viral particles was calculated from measurements of optical density at 260 nm, where 1 absorbency unit is equivalent to 1012 viral particles per milliliter, and infectious titers (plaque-forming units per milliliter) were determined by limiting dilution assay on 293 cells; the plaque-forming unit was calculated from infectious titers as follows: T = 101+d(S0.5), where d and S were the log10 of the dilution and the sum of ratios, respectively. The viral particle/plaqueforming unit ratios for dl-lacZ, dl-lacZ-RLX, Ad-ElB, and Ad-ElB-RLX were 57 : 1, 70 : 1, 22 : 1, and 43 : 1, respectively. Enzyme-Linked Immunosorbent Assay for Relaxin Expression We infected 5 105 U343 or C33A cells with dl-lacZ, dl-lacZRLX, Ad-E1B, or Ad-E1B-RLX at various multiplicities of infection (MOIs) in 25-T culture flasks. Forty-eight hours later, the supernatant was collected by centrifugation at 15 000g for 10 minutes at 4 C, and the level of relaxin protein was assessed by enzyme-linked immunosorbent assay (Endogen, Woburn, MA). Serial dilutions of a purified recombinant human relaxin preparation with a known concentration were used to generate a standard curve. Animal Studies Male athymic nu/nu and C57BL/6 mice, weighing approximately 20 g (...truncated)