Clinical Implications of Expression of Interleukin 8 Related to Angiogenesis in Uterine Cervical Cancers

Angiogenesis is essential for development, growth, and advancement of solid tumors (1). Although angiogenic factors for tumor angiogenesis have been considered to be produced by tumor cells, it has been demonstrated that they are also produced by interstitial cells within and around the tumor. In uterine cervical cancers, especially adenocarcinomas of the cervix, VEGF² (particularly the VEGF₁₆₅ and VEGF₁₂₁ isoforms) was dominantly expressed in cancer cells (2). Furthermore, VEGF levels correlated with microvessel density (2). However, bFGF and PD-ECGF were expressed in both cancer and interstitial cells of uterine cervical cancers, and PD-ECGF was dominantly expressed in interstitial cells of squamous cell carcinomas of the cervix (3, 4). Furthermore, PD-ECGF levels correlated with microvessel density and patient prognosis in metastatic and primary lesions of uterine cervical cancers (4, 5).

It is well known that infiltration of inflammatory cells is activated in malignant tumors, which contributes to angiogenesis. The tumor-associated macrophage has been recognized as a candidate among inflammatory cells for tumor angiogenesis (6). The macrophages that infiltrate tumors of the liver and the gastrointestinal tract supply bFGF (7), and those that infiltrate breast and ovarian cancers supply VEGF (8) and TNF-α (9). Although TNF-α possesses weak angiogenic activity in itself, the angiogenic potential of TNF-α appears to be modulated through induction of the strong angiogenic factors IL-8,VEGF, and bFGF, and this pathway is regulated through paracrine and/or autocrine mechanisms (10). IL-8 is expressed in macrophages and fibroblasts derived from the interstitium(7) and was recognized as a macrophage-derived mediator of angiogenesis (11). This prompted us to study the clinical implications of macrophage-derived angiogenesis in uterine cervical cancers.

Consent for the following studies was obtained from all patients and from the Research Committee for Human Subjects, Gifu University School of Medicine. Eighty patients ranging from 37–74 years of age underwent curative resection for cervical cancer. Patient prognosis was analyzed with a 24-month survival rate after curative resection at the Department of Obstetrics and Gynecology, Gifu University School of Medicine between June 1994 and April 1997. None of the patients had received any preoperative therapy. A part of each uterine cervical cancer tissue was obtained immediately after hysterectomy and snap-frozen in liquid nitrogen to determine the levels of IL-1α,IL-1β, TNF-α, IL-8, bFGF, VEGF, and PD-ECGF, and a neighboring part of the tissue was submitted for histopathological study. The clinical stage of uterine cervical cancers was determined by the FIGO classification (12).

Four-μm sections of formalin-fixed paraffin-embedded tissue of uterine cervical cancers were cut with a microtome and dried overnight at 37°C on a silanized slide (DAKO, Carpinteria, CA). Samples were deparaffinized in xylene at room temperature for 80 min, washed with a graded ethanol/water mixture, and then with distilled water. The samples for IL-8 and CD68 antigens were soaked in a citrate buffer and then microwaved at 100°C for 10 min, and those for factor VIII-related antigen were treated with 0.3 μg/ml trypsin in PBS at room temperature for 20 min. The protocol for DAKO LSAB2 Kit Peroxidase(DAKO) was followed for each sample. In the described procedures,rabbit antihuman IL-8 (Biosource, CA), mouse antihuman macrophage CD68 (DAKO), and rabbit anti-factor VIII-related antigen(Zymed, San Francisco, CA) were used at dilutions of 1:50, 1:50, and 1:2, respectively, as the first antibodies. The addition of the first antibody, rabbit antihuman IL-8, mouse antihuman macrophage CD68, or rabbit anti-factor VIII-related antigen, was omitted in the protocols for negative controls of IL-8, CD68, or factor VIII-related antigen,respectively.

Vessels and macrophages were counted in the five highest density areas at ×200 magnification (using a combination of a ×20 objective and a×10 ocular; 0.785 mm²/field). MVC and IMC were expressed as the mean numbers of vessels and macrophages in these areas,respectively (13, 14). Microvessel density and macrophage infiltration were evaluated by counting microvessels and macrophages,respectively.

All steps were carried out at 4°C. Tissues of uterine cervical cancers (wet weight, 10–20 mg) were homogenized in HG buffer [5 mm Tris-HCl (pH 7.4), 5 mm NaCl, 1 mm CaCl₂, 2 mmethyleneglycol-bis-[β-aminoethyl ether]-N,N,N′,N′-tetraacetic acid, 1 mmMgCl₂, 2 mm DTT, 25 μg/ml aprotinin, and 25μg/ml leupeptin] with a Polytron homogenizer (Kinematics, Luzern,Switzerland). This suspension was centrifuged in a microfuge at 12,000 rpm for 3 min to obtain the supernatant. The protein concentration of samples was measured by the method of Bradford (15) to standardize IL-1α, IL-1β, TNF-α, IL-8, bFGF, VEGF, and PD-ECGF antigen levels.

IL-1α, IL-1β, TNF-α, IL-8, bFGF, and VEGF antigen levels in the samples were determined by a sandwich enzyme immunoassay using human IL-1α quantikine (R&D Systems, Minneapolis, MN), human IL-1βquantikine (R&D Systems), human TNF-α quantikine (R&D Systems), human IL-8 quantikine (R&D Systems), human bFGF quantikine (R&D Systems), and human VEGF assay kit-IBL (Immuno Biological Laboratories, Gunma,Japan), respectively, and PD-ECGF antigen levels were determined by the method described by Nishida et al. (16). The levels of IL-1α, IL-1β, TNF-α, IL-8, bFGF, VEGF, and PD-ECGF were standardized with the corresponding cellular protein concentrations.

Survival curves were calculated using the Kaplan-Meier method and analyzed by the log-rank test. IL-1α, IL-1β, TNF-α, IL-8, bFGF,VEGF, and PD-ECGF levels were measured from three parts of the same tissue in triplicate. Statistical analysis was performed with Student’s t test. Differences were considered significant when P was less than 0.05.

There was a significant correlation between MVCs and IL-8(P < 0.01, as shown in Fig. 1), bFGF (P < 0.01), VEGF(P < 0.01), and PD-ECGF levels(P < 0.01) in uterine cervical cancers, but not between MVCs and IL-1α, IL-1β, or TNF-α.

There was a significant correlation between IMCs and IL-8 levels in uterine cervical cancers, as shown in Fig. 2 (P < 0.01), but not between IMCs and IL-1α, IL-1β, TNF-α, bFGF, VEGF, or PD-ECGF (data not shown).

Immunohistochemical staining for IL-8 and CD68 in a representative case of adenocarcinoma is shown in Fig. 3. IL-8 was distributed in the interstitium within and around the tumor. CD68 was found in a localization and manner similar to IL-8 in all cases.

There was no significant correlation between histopathological type and IL-8 levels, as shown in Fig. 4. There was a significant difference (P < 0.05) in IL-8 levels between early-stage (stage I) and late-stage(stages II–IV) uterine cervical cancers, as shown in Fig. 5.

The prognosis of the 20 patients with high IL-8 (>1000 pg/mg protein)in uterine cervical cancers was extremely poor, whereas the 24-month survival rate of the other 60 patients with low IL-8 (<1000 pg/mg protein) was 67% (Fig. 6).

Infiltrated macrophages in tumors are activated, which leads to the induction of angiogenic potential; in fact, macrophage infiltration correlates with microvessel density in breast cancers (6). In the present study, there was a significant correlation between IMC and MVC (P < 0.05) in uterine cervical cancers (data not shown). Therefore, we studied which angiogenic factors derived from tumor-associated macrophages promote tumor angiogenesis in uterine cervical cancers. Generally, bFGF, VEGF,PD-ECGF, TNF-α, and IL-8 have been recognized as tumor-associated macrophage-derived angiogenic factors (7, 8, 9, 10, 11).

In the present study, the positive correlation between MVCs and IL-8,bFGF, VEGF, and PD-ECGF demonstrates that these factors work as angiogenic factors. In uterine cervical cancers, VEGF from cancer cells is related to patient prognosis, but only in early invasion (2, 17, 18). PD-ECGF from the interstitial cells correlates with patient prognosis, especially with lymph node metastasis (4, 5). bFGF from both the cancer and interstitial cells relates to clinical stage and advancement (3). Although PD-ECGF and bFGF were provided in part from tumor-associated macrophages,macrophage infiltration was positively correlated with only IL-8 among IL-1α, IL-1β, TNF-α, bFGF, VEGF, PD-ECGF, and IL-8, which demonstrates that IL-8 might be dominantly supplied from tumor-associated macrophages. Furthermore, localization of IL-8 in the tumors was similar to that of macrophage CD68. Although there is no distinct cytokine network among IL-1α, IL-1β, TNF-α, and IL-8 in tumor angiogenesis derived from infiltrated macrophages, it is strongly speculated that IL-8 works as an angiogenic factor in uterine cervical cancers. In other angiogenesis-dependent diseases, IL-8 contributes to growth related to angiogenesis in bronchogenic carcinoma(19), glioblastoma (20), melanoma(21), ovarian carcinoma (22, 23), and other cancers. Furthermore, IL-8 activates metastatic potential(21).

In the present study, IL-8 levels were significantly increased during the advancement from stage I to stages II, III, and IV. Therefore, IL-8 might be the angiogenic switch in advancement to later stages of uterine cervical cancers. Furthermore, IL-8, as an angiogenic factor supplied from infiltrated macrophages within and around the tumor,might be a prognostic indicator.