Human cervical carcinoma HeLa cells were provided by the Cell Center of the Union Medical College (Beijing, China). Cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) high glucose-complete medium at 37°C in a 5% CO₂ incubator. Logarithmic growth phase cells were collected, and the cell number was adjusted to 1×10⁷/ml.

Specific pathogen-free (SPF) female nude mice (BALB/c; 13–15 g, aged 4 weeks) were obtained from the Shanghai Silaike Experimental Animal Co., Ltd. (Shanghai, China). Mice were maintained in pathogen-free conditions and used for study in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. All surgical procedures and care administered to the animals were approved by the Institutional Ethics Committee.

All surgical procedures and care administered to the animals were approved by the Institutional Ethics Committee.

i) The experiment, in which the needs of the animals were fully considered, was approved by the ethics committee, including physiological (adequate food, water, temperature and illumination), environmental, psychological and social needs (socially raised, 4–6 animals per cage, avoid tiredness and overstimulation). The outcomes of the preliminary experiment and the literature were taken into consideration to make rational design of the sample size and operation standard. ii) A daily observation was preformed to prevent the animals from anger, comfortlessness, fear, nervousness, pain or damage and to keep them at normal status. Abuse, excessive or incorrect medication was avoided. For subcutaneous injection, which was easy to operate, narcotics were not applicable; for tail vein injection, which was not easy to operate, intraperitoneal anesthesia was given to alleviate the pain of the animals. iii) At the time of endpoint, the animals were sacrificed within 15 sec to avoid the nervousness of the other animals.

The xenograft tumor model was established by subcutaneous (s.c.) injection of ~1×10⁶ HeLa cells (100 μl suspension) into the left axillary lateral of each nude mouse. The model was deemed to be established successfully following the appearance of xenografted tumors after 7 days and with tumor diameters of ~4–5 mm after 14 days. Subsequently, successfully xenografted mice were randomly divided into six groups (n=6 per group): i) PBS buffer control group: 100 μl injected once per 3 days for a total of five times; ii) empty plasmid group: empty plasmid pDC316 (Benyuan Zhengyang Gene Technology Co., Beijing, China); 100 mg/l, 100 μl once per 3 days, for a total of five times) and liposome Lipofectamine 2000 (Invitrogen, Shanghai, China; 25 μl for transfection); iii) half-dose DDP (cisplatin) group: intraperitoneal (i.p.) injection of cisplatin (Qilu Pharmaceutical Co., Ltd., Jinan, China; 2.5 mg/kg, 100 μl once a day for 3 days); iv) IL-24 group: local multipoint intratumoral injection of pDC316-hIL-24 (Benyuan Zhengyang Gene Technology Co.); 100 mg/l, 100 μl once per 3 days for a total of five times) and liposome Lipofectamine 2000 (25 μl for transfection); v) full-dose DDP (cisplatin) group (DDP group): i.p. injection of cisplatin (5 mg/kg, 100 μl i.p. once a day for a total of 3 days; and vi) combined treatment group: cisplatin (2.5 mg/kg, 100 μl i.p. once a day for 3 days) followed by local multipoint intratumoral injection of pDC316-hIL-24 (100 mg/l, 100 μl once per 3 days for a total of five times). The tumor size (L) and short diameter (W) were measured every 3 days with a vernier caliper. Weight and the overall status of the mice were recorded at the same time. Nude mice were sacrificed by cervical dislocation at 2 weeks after the cessation of treatment. Tumor volume was measured and weighed; and the inhibition of tumor weight was calculated. Tumor size V (mm³) = L × W²/2.

Determination of pDC316-hIL-24 expression in the xenografted tumors was performed using PrimeScript™ First Strand cDNA synthesis kit (Takara Biotechnology, Dalian, China). Primers of IL-24 were: forward, 5′-GCCAAGCTTATGAATTTTCAACAGA GG-3′ and reverse, 5′-GCCGTCGACCTAGAGCTTGTAGAATTT-3′. Expression of GADPH was analyzed as a control using the following primers: forward, 5′-TGAACGGGAAGCTC ACTGG-3′ and reverse, 5′-TCCACCACCCTGTTGCTGGA-3′. The amplification was performed using the Applied Biosystems GeneAmp PCR system 2700 with the following reaction conditions: a 94°C pre-denaturation for 2 min, followed by 32 cycles of a 94°C denaturation for 20 sec, a 58°C annealing for 20 sec and a 72°C extension for 20 sec, with a final extension at 72°C for 8 min. PCR products were separated by 1.5% agarose gel electrophoresis.

After dewaxing, hydration and antigen retrieval, VEGF expression was detected by immunohistochemical staining with primary and secondary detection antibodies (Boster, Wuhan, China). Immunoreactivity was visualized by incubation with DAB chromogenic agent DAB kit (Zhongshan Jinqiao Biological Engineering Co., Beijing, China). Staining was evaluated in 10 randomly selected fields per section, with intensity divided into four levels as follows: 0, no staining; 1, pale yellow; 2, brown; and 3, tan. Scores were assigned according to the percentage of positive cells in each field of vision as follows: 10%, 0 points; 11–20%, 1 point; 21–60%, 2 points; and 61–100%, 3 points. The average staining intensity score (IS) was calculated for each section based on Bresalier’s semiquantitative method (8). IS represents the average multiplied product of the staining intensity score and the positive cell score per 10 randomly selected fields of vision in each section.

Solid tumor growth and metastasis require sustained angiogenesis, the extent of which is indicated by MVD. Microvessel enumeration can be achieved by IHC labeling of CD34, which is expressed stably and specifically by tumor capillaries and endothelial cells (9). Immunohistochemical detection of CD34, specimen preparation and immunohistochemical staining procedures were performed using an SP kit according to the manufacturer’s instructions. Subsequently, areas rich in vasculature tissue (known as hot spots) were identified by visualization of the highest density of yellow-brown areas in whole sections under a light microscope (magnification, ×100). MVD was calculated at higher magnification microscopy (×400) using the following criteria (10): any brown-stained endothelial cells or cell clusters were recorded as a separate vessel, with clear demarcation between all the vessels. MVD was calculated as the average number of microvessels in five randomly selected fields of vision.

Analysis of PDGF-B protein levels in tumor tissue was performed using WesternBreeze^® Chemiluminescent kit (Life Technologies, Grand Island, NY, USA) according to the manufacturer’s instructions. Tumor tissue protein was extracted, and the concentrations were determined. After SDS-PAGE electrophoresis, it was transferred to a membrane, and immune response and exposure were determined (anti-mouse PDGF-B primary antibody; BioWorld Products, Visalia, CA, USA). The analysis of experimental results was performed using Quantity One image analysis software.

SPSS 16.0 statistical software was used for statistical analysis. Multi-group comparisons were performed using ANOVA after the test of normality. t-tests were used for comparisons between two groups. Bonferroni corrections were used if necessary. Non-normal data were compared using SNK tests. Data are presented as means ± SD. A P-value <0.05 was considered to indicate a statistical significant result.

Mice in the cisplatin groups (full-dose and half-dose) exhibited poor mental state and reduced feeding. The overall status of mice in the IL-24 and half-dose DDP+IL-24 groups was better than that in the DDP full-dose group. There were no adverse changes in the mental status and feeding habits of nude mice in the IL-24 group (data not shown).

The changes in tumor volumes after the different interventions are shown in Table I. The tumor volumes were subjected to two-factor analysis of variance. Results of Mauchly’s test of sphericity did not support the ‘spherical symmetry’ hypothesis (P<0.001); therefore, the coefficient was corrected using Greenhouse-Geisser, Huynh-Feldt and Lower-bound calibrations, and the test results showed statistical significant P<0.01 after calibration. Compared to the control groups (PBS and empty vector), tumor growth was significantly reduced following treatment with cisplatin and/or recombinant IL-24 plasmid. These effects were more marked following treatment with the IL-24 recombinant plasmid and the combined therapy, which mediated almost complete inhibition of tumor growth. The difference was statistically significant at different time-points within the same group (P<0.001) as well as among different groups (P<0.001), and there were interaction effects between time and treatment. There were significant differences between groups (all P-values <0.05) except between the control group and empty vector group, between the IL-24 group and half-dose DDP+IL-24, and between the full-dose and half-dose DDP groups (P>0.05).

Tumor growth was faster in the two control groups, leading to overall weight gain. Cisplatin is associated with serious side-effects. The weight changes of nude mice after intervention are also shown in Table I. Mauchly’s test of sphericity was performed and weight changes in the nude mice also refused the‘spherical symmetry’ hypothesis. The difference was statistically significant at different time-points within the same group (P<0.001) as well as among different groups (P<0.001), and there were interaction effects between time and treatment. Results of t-test with Bonferroni corrections revealed that, there were significant differences between groups (all P-values <0.05) except between the control group and empty vector group, and between the half-dose DDP and half-dose DDP+IL-24 groups (P>0.05).

Fig. 1A shows expression of IL-24 using RT-PCR in xenograft tumors of the control group, empty plasmid group, half-dose DDP group, IL-24 group, full-dose DDP group, and half-dose DDP+IL-24 group. IL-24 was expressed in the IL-24 group and half-dose DDP+IL-24 group. IHC assays were performed to detect VEGF and CD34 expression in xenograft tumors of the control, empty plasmid and IL-24 groups (Fig. 1B and C). Fig. 1D revealed that there were significant differences in VEGF and CD34 expression between the control group and IL-24 group (P<0.001 and P=0.001, respectively).

Visible tumor metastatic lymph node was found near the axilla, beside abdominal aortas and in the neck and groin (Fig. 2A). Lymph node metastasis results in the IL-24, cisplatin, combined therapy, empty plasmid and control group mice were 1/6, 4/6, 1/6, 5/6 and 5/6, respectively. Lymph node metastasis was significantly different between the five groups (χ²=11.25, P=0.024). Lymph node metastasis was significantly reduced in the IL-24 and half-dose DDP+IL-24 groups compared with the other groups (P<0.05).

As a marker of tumor cell infiltration, P-CK expression was detected using IHC as brown or tan granules in the cytoplasm of cancer cells (Fig. 2B). Cancer cells in the control and empty plasmid groups were typically observed as hyperchromatic cells with a large nucleus. Tumor cells were arranged in clusters or cords; the volume of tumor cells was large with visible mitotic phase, while the lymphocytes and neutrophils were less. HeLa cell invasion was significantly lower in the IL-24 group compared with that of the control groups. The majority of the infiltrate in the half-dose DDP+IL-24 group comprised lymphocytes and neutrophils, while only minor infiltration and necrosis, larger cells, loose cytoplasm and marked nuclear staining were observed. HeLa cell infiltration was less marked in the cisplatin group, although large nuclei were still observed.

Expression of VEGF-C and VEGFR-3 was predominantly detected in the cytoplasm of the cancer cells as brownish yellow granules (Fig. 3A and B). Univariate analysis of variance indicated statistically significant differences in the average grayscale values of VEGF-C-positive cells (F=17.49, P<0.001) and of VEGFR-3-positive cells (F=19.23, P<0.001) between the five groups. The average grayscale values of VEGF-C and VEGFR-3 in the IL-24 and half-dose DDP+IL-24 groups were significantly higher than those in the full-dose DDP, control and empty plasmid groups (P<0.001), while there were no statistically significant differences between the latter three groups (P>0.05) (Fig. 3C).

PDGF-B protein was detected in the xenografts of each group by western blotting. Immunohistochemical analysis showed high expression of PDGF-B in the tumor cells in the PBS group, empty plasmid group, DDP group and half-dose DDP group, with no significant differences between the four groups (P>0.05). The levels were significantly reduced in the IL-24 and half-dose DDP+IL-24 groups, compared with that in the other four groups (P<0.001) (Fig. 4).