HeLa (human cervical cancer), 293 (human embryonic kidney) and HT1080 (human fibrosarcoma) cells were purchased from Bioresource Collection and Research Center (BCRC, Taiwan) and cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum, 2 mM L-glutamine, 100 IU/ml penicillin/streptomycin, 1 mM sodium pyruvate, and 0.1 mM non-essential amino acids. HT2 (murine IL2/IL15 dependent) cells were purchased from BCRC and cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum, 2 mM L-glutamine, 100 IU/ml penicillin/streptomycin, 1 mM sodium pyruvate, and 0.1 mM non-essential amino acids. In addition, HT2 cells were supplemented with 1 μg/ml human recombinant IL15 protein (Santa Cruz Biotechnology, Santa Cruz, California, USA). DMEM, RPMI, fetal bovine serum, L-glutamine, penicillin/streptomycin, sodium pyruvate and non-essential amino acids were bought from Invitrogen (Carlsbad, California, USA).

The rAAV2 helper-free system, containing packaging plasmid pAAV-MCS plasmid, pAAV-RC plasmid, pHelper, pAAV-hrGFP plasmid was purchased from Strategene (La Jolla, California, USA). The plasmid containing human IL15 gene (hIL15) was provided thru the kindness of Dr L. KW (National Chiao Tung University, Taiwan). The hIL15 containing EcoR I and BamH I sites was amplified for cloning using the polymerase chain reaction. Oligonucleotide primers containing EcoR I and BamH I sites were synthesized as follows: sense primer: 5'GAATTC AAA GAA TTC ATG TAC AGG ATG CAA CTC CT, and anti-sense primer, 3'GGATCC AAA GGA TCC TTA AGA AGT GTT GAT GAA CAT TTG G. The PCR conditions used a thermal profile of denaturing at 95°C for 30 sec, annealing at 95°C for 30 sec, and extending at 72°C for 30 sec for 30 cycles, followed by 72°C for 10 min. The amplified hIL15 cDNA was inserted between the EcoR I and BamH I sites of pAAV-MCS to yield pAAV-hIL15 plasmid.

Production of rAAV2-hIL15 and rAAV2-hrGFP was done in the helper-free system as described previously 25. For rAAV2-hIL15 production, 293 cells were cultured on fifty 15-cm-dishes and transfected with calcium chloride with 2 mg pAAV-hIL15 plasmid, 2 mg pAAV-RC plasmid and 2 mg pAAV-Helper plasmid. For rAAV2-hrGFP production, 293 cells were cultured on fifty 15-cm-dishes and transfected with calcium chloride with 2 mg pAAV-hrGFP plasmid, 2 mg pAAV-RC plasmid and 2 mg pAAV-Helper plasmid. After 65 hours of transfection, rAAV2-hIL15 and rAAV2-hrGFP were produced in 293 cells.

Purification of rAAV2-hIL15 and rAAV2-hrGFP was done using a single-step column purification (SSCP) method 26. The rAAV2 was concentrated to about 1 ml with an Amicon Ultra-15 centrifugal filter (Millipore, Billerica, MA) and stored at -80°C. Finally, the viral titers of rAAV2 were determined by Real-time PCR method as described previously 24. A total of 2–5 × 10¹² genomic particles of rAAV2 can be obtained from fifty 15-cm-dishes. Purified rAAV2 (rAAV2-hIL15 and rAAV2-hrGFP) was determined by loading 10¹⁰ viral particles on 10% SDS gel 26 and viral proteins (VP1, VP2 and VP3) were detected by silver stain

HT1080 cells can be infected with AAV2 easily for the production of AAV2 expressing genes. In this study, HT1080 (1.5 × 10⁵) cells were divided into three groups and cultured on 6-well-plates, then each group was treated with 100 μl of rAAV2-hIL15 (10¹³ viral particles/ml), rAAV2-hrGFP (10¹³ viral particles/ml) and PBS, respectively. After 2 days, the media were collected for quantification of IL15. IL15 titers were determined using an ELISA kit (Biosource, Camarillo, California, USA).

HT2 cells are IL2/IL15 dependent cells and can not survive without IL2 or IL15 supply. Therefore growth of HT2 cells is determined by the IL15 bioactivity. HT2 cells were divided into three groups and their survival rates were determined by using MTS assay (Promega, Madison, Wisconsin, USA) under an optical density 490 nm. The first group of HT2 cells was cultured with the media obtained from 100 μl of rAAV2-hIL15 (10¹³ viral particles/ml)-infected HT1080 cells. The second group was cultured with the media obtained from 100 μl of rAAV2-hrGFP (10¹³ viral particles/ml)-infected HT1080 cells. The third group of HT2 cells was cultured with the media obtained from purified 1 μg/ml human recombinant IL15-incubated HT1080 cells (This group was used as the positive control). These were then observed under the microscope.

Nude mice are T-cell defective in which NK cell activity cannot be determined. Previous research showed that the increase in the immunoglobulin titer can be an indication of an increase in the IL15 activity 10. In addition, it has been reported that this immunoglobulin is related to antitumor activity on nude mice 27. Therefore immunoglobulin level was determined in this study. Twenty four (24)-four-week-old female nude mice were purchased from National Laboratory Animal Center (Taiwan). All mice were divided into three groups (8 mice/each group). The first group of mice was injected with 500 μl of rAAV2-hIL15 (10¹³ viral particles/ml) on one site of the quadriceps muscles of their hind limbs, the second group with 500 μl of rAAV2-hrGFP (10¹³ viral particles/ml) and the third group with 500 μl of PBS (the steps and the amount injected were the same as the first group). After 28 days, the mice sera were collected for immunoglobulin measurement. Quantification of immunoglobulin levels was performed using immunoglobulin ELISA kit (Becton, Dickinson and Company) measured at an optical density 450 nm. The animal studies have been approved by the animal committee of the Tzu-Chi General Hospital.

Twenty four, four-week-old female nude mice were purchased from National Laboratory Animal Center (Taiwan). All mice were divided into three groups (8 mice/group). The first group of mice was injected with 500 μl of rAAV2-hIL15 (10¹³ viral particles/ml) on one site of their quadriceps muscles of the hind limbs, the second group with 500 μl of rAAV2-hrGFP (10¹³ viral particles/ml) and the third group with 500 μl of PBS (the steps and amount injected were the same as the first group). Every week, the weight of the treated-mice was determined using electronic weighing-scale. The animal studies have been approved by the animal committee of the Tzu-Chi General Hospital.

Twenty four, four-week-old female nude mice were purchased from National Laboratory Animal Center (Taiwan). All mice were divided into three groups (8 mice/group). The first group of mice was injected with 500 μl of rAAV2-hIL15 (10¹³ viral particles/ml) on one site of quadriceps muscles of their hind limbs, the second group with 500 μl of rAAV2-hrGFP (10¹³ viral particles/ml) and the third group with 500 μl of PBS (steps and amount injected were the same as the first group). After 28 days, 10⁷ HeLa cells were injected subcutaneously on the lower abdominal region of all mice. The day when mice began treatment with HeLa cells corresponded to day zero. After that, tumor size was measured every week according to the following formula: (A × B²)/2, where A is the greater and B is the smaller measurement of the two dimensions 2829. The animal studies have been approved by the animal committee of the Tzu-Chi General Hospital.

Twenty four, four-week-old female nude mice were purchased from National Laboratory Animal Center (Taiwan). All mice were divided into three groups (8 mice/group) and were injected with 10⁷ HeLa cells on the lower abdominal region. The next day, the first group of mice was injected with 500 μl of rAAV2-hIL15 (10¹³ viral particles/ml) on one site of quadriceps muscles of the hind limbs, the second group with 500 μl of rAAV2-hrGFP (10¹³ viral particles/ml) and the third group with 500 μl of PBS (steps and amount injected were the same as the first group). The day when mice began treatment with HeLa cells corresponded to day zero. The, tumor size was then measured according to the following formula: (A × B²)/2, where A is the greater and B is the smaller measurement of the two dimensions 2829. The animal studies have been approved by the animal committee of the Tzu-Chi General Hospital.

The Student t-test was utilized in the analysis of data presented as the mean ± SE. A value of P < 0.05 was considered statistically significant.

With the use of a helper-free system and SSCP method, rAAV2-hIL15 and rAAV2-hrGFP were produced and purified successfully. As shown in Figure 1, purified rAAV2-hIL15 and rAAV2-hrGFP have VP1, VP2 and VP3 of AAV2 capsid proteins. This result is similar to previous study 26. HT1080 cells can be infected with AAV2 easily for production of AAV2 expressing genes. Purified rAAV2-hIL15, rAAV2-hrGFP and PBS were added to HT1080 cells, respectively. After 2 days, the culture media of HT1080 cells were collected. For quantification of IL15, our data showed that the media obtained from rAAV2-hIL15 treated-HT1080 cells have IL15 of about 500 ng/ml (Figure 2). However the media obtained from rAAV2-hrGFP or PBS treated-HT1080 cells have less and less IL15 than rAAV2-hIL15 treated-cells (Figure 2). The data indicated that rAAV2-hIL15 can express IL15. HT2 cells are IL2/IL15 dependent cells and they cannot survive without IL15 (or IL2) supply. Therefore the survival rate of HT2 cells can be used as an indicator for IL15 bioactivities. The media obtained from rAAV2-hIL15, rAAV2-hrGFP-treated HT1080 cells were collected and added to HT2 cells respectively. The purified human IL15 added to HT2 cells was used as a positive control. Observation on HT2 survival rate, the data showed that HT2 cells can grow with the media containing purified human IL15 and with the media obtained from rAAV2-hIL15-treated HT1080 cells (Figure 3). However, media obtained from rAAV2-hrGFP-treated HT1080 cells resulted in survival inhibition on HT2 cells (Figure 3). The morphologies of HT2 cells were also observed under the microscope (Figure 4). The morphologic data indicated that HT2 cells were able to survive under the media containing purified human IL15 and the media obtained from rAAV2-hIL15-treated HT1080 cells. However, HT2 cells were not able to survive under the media obtained from rAAV2-hrGFP-treated HT1080 cells (this result has been demonstrated in figure 3). Taken together, we successfully produced and purified rAAV2-hIL15 which has bioactivity in vitro.

Sera were collected from mice that were treated by intramuscular injection of rAAV2-hIL15, rAAV2-hrGFP and PBS for 28 days. The sera were analyzed for the production of immunoglobulins (IgG1, IgG2a, IgG2b, IgA, and IgM). The data showed that sera obtained from rAAV2-hIL15-treated mice had higher titers of IgG1 and IgM than sera obtained from rAAV2-hrGFP-treated and PBS-treated mice (Figure 5). However, production of IgG2a, IgG2b and IgA has no significant difference in rAAV2-hIL15-treated, rAAV2-hrGFP-treated and PBS-treated mice. The data suggested that rAAV2-hIL15 can induce production of IgG1 and IgM in nude mice. Furthermore, we examined whether rAAV2-hIL15 and rAAV2-hrGFP interfered with mice growth. Our data indicated that during the 18 weeks, there was no difference in growth rate among mice treated by rAAV2-hIL15, rAAV2-hrGFP, and PBS (Figure 6).

Previous study had demonstrated that AAV, a single-stranded DNA virus, must convert single-stranded DNA into double stranded templates for transcription 3031. This rate-limiting step is achieved in about 3 weeks. Therefore, we first prescribed rAAV2-hIL15 before tumor implantation in the mice in this study. After rAAV2-hIL15, rAAV2-hrGFP and PBS were injected into the muscles of mice for 28 days, these mice received HeLa tumor cells implantation by subcutaneous injection. After which, tumor size was determined every week. Our data showed that tumor growth rate in rAAV2-hIL15-treated mice was slower than that in rAAV2-hrGFP- and PBS-treated mice (Figure 7). Meanwhile, tumor growth rates are similar in rAAV2-hrGFP-treated and PBS-treated mice. Comparing the tumor size on rAAV2-hIL15-treated mice with rAAV2-hrGFP- and PBS-treated mice, the difference in the size of the tumor becomes greater after the ninth week. These data indicated that pre-treated rAAV2-hIL15 could suppress tumor growth efficiently.

We examined next whether post-treated rAAV2-hIL15 has an anti-tumor effect. HeLa cells were first injected subcutaneously. in the lower abdominal region of the mice. The next day, these mice were given intramuscular injection of rAAV2-hIL15, rAAV2-hrGFP and PBS respectively. Before the 12th week, tumor sizes had no significant differences, but we found out that the tumor size was smaller in rAAV2-hIL15 treated-mice than in rAAV2-hrGFP-treated-mice and PBS-treated mice after 12 weeks (Figure 8). The result indicated that post-treated rAAV2-hIL15 could delay tumor growth. Altogether, our studies suggested pre-treated and post-treated rAAV2-hIL15 are a good approach for cancer gene therapy, yet pre-treated rAAV2-hIL15 is more efficient in inhibiting tumor growth.