Background
Renal cell carcinoma (RCC) incidence has increased in recent years with approximately 38,890 new cases each year in the United States of America [
1]. The disease is responsible for an estimated 12,840 deaths each year [
1]. This increased RCC incidence may be linked to certain risk factors including smoking, obesity, high protein diets and hypertension [
2]. Nearly half of the patients present only with localized disease that can be treated by surgical removal [
2‐
4]. However, one third of the patients also present with metastatic disease and half of the patients treated for localized carcinomas subsequently develop metastatic disease [
2‐
4]. The median survival of patients with metastases is only eight months, with a five-year survival rate of less than 10% [
2‐
4]. Patients with metastatic RCC frequently present with pulmonary metastases that are poorly responsive to conventional treatment including most chemotherapeutic drugs, hormones and radiation therapy [
2‐
5]. The treatment of metastatic disease has been and remains a difficult clinical challenge.
To develop new and alternative therapeutic modalities for metastatic disease and to investigate the metastatic progression and the molecular genetics of RCC, various pre-clinical animal models were established (reviewed in ref 6). Among others, tumor xenograft models established in immunodeficient mice by implantation of RCC cells, isolated from a human tumor specimen, have been valuable to assess responsiveness to therapy [
6‐
9]. To investigate the combination of radiotherapy with other treatment modalities, we have established a xenograft metastatic RCC tumor model by orthotopic renal implantation of a new KCI-18 human RCC cell line in athymic nude mice. The KCI-18 RCC model was used to study,
in vivo, the responsiveness of human RCC primary tumors and metastases to the soy isoflavone genistein and radiation.
Genistein, the most bioactive isoflavone of soybeans, was extensively used in cancer studies and demonstrated inhibition of tumor cell growth
in vitro by affecting the cell cycle and inducing apoptosis [
10]. We further showed that genistein potentiated radiation-induced tumor cell killing [
11,
12]. This was demonstrated in various human tumor cell lines including RCC cell lines and in particular the KCI-18 line [
12]. Genistein significantly increased tumor cell death when given prior to radiation in KCI-18 cells similar to the effect observed with the human PC-3 prostate carcinoma cell line [
12].
In vivo, we previously showed that genistein potentiated inhibition of prostate tumor growth by radiation and controlled spontaneous metastasis to regional para-aortic lymph nodes using two different orthotopic metastatic prostate tumor models [
13,
14]. Paradoxically, we discovered that pure genistein, administered as a single treatment modality, promoted increased metastasis to lymph nodes [
13]. This intriguing observation was reproduced in two independent orthotopic prostate tumor models, the human PC-3 xenograft model in nude mice [
13] and the mouse RM-9 model syngeneic in C57BL/6 mice [
14] raising concerns regarding soy-based clinical trials for cancer patients. The goals of the current study were to investigate whether treatment with genistein alone also promotes metastasis in the KCI-18 orthotopic RCC model, and whether genistein combined with primary tumor irradiation is an effective treatment approach for RCC treatment. We found that genistein treatment demonstrated a tendency to stimulate the growth of the primary kidney tumor and increase the incidence of metastasis to the mesentery lining the bowel. In contrast when given in conjunction with kidney tumor irradiation, genistein significantly inhibited the growth and progression of established kidney tumors.
Discussion
We have previously reported the potentiation of radiotherapy by the soy isoflavone genistein. This phenomenon was demonstrated for prostate cancer using prostate tumor cells
in vitro [
11,
12] and orthotopic prostate tumor models
in vivo [
13,
14]. Genistein combined with primary tumor irradiation caused a greater inhibition of prostate tumors and control of lymph node metastasis than each modality alone. These findings were reproduced in two independent orthotopic tumor models [
13,
14], however, genistein used as single therapy caused increased metastasis to regional para-aortic lymph nodes [
13,
14]. This effect was observed with a wide range of genistein doses, from 20 μg/day (Hillman, personal communication) up to 1–5 mg/day [
13,
14]. To clarify whether these intriguing adverse effects of genistein are intrinsic to the orthotopic prostate tumor model, these studies were repeated in the orthotopic metastatic KCI-18 RCC model established in our laboratory.
The KCI-18 RCC model was established from a cell line generated by culture of cells isolated from a human papillary RCC tumor specimen. Following serial renal implantation of KCI-18 cells in the kidney of nude mice, we confirmed the establishment of a reliable orthotopic RCC experimental model. This model showed the properties of an ideal RCC tumor model including histologically proven carcinoma, predictable growth rate and ability to metastasize similarly to human RCC in a reasonable time frame [
6]. The karyotype and histological characteristics of KCI-18 cell line and kidney tumors were comparable to those of the human primary tumor specimen from which the cell line was produced. Like human RCC, these tumors were highly vascularized and stained positively for cytokeratin and vimentin [
5,
20]. We showed that renal implantation of the KCI-18 cell line in the kidney of nude mice caused the formation of a primary kidney tumor that metastasizes to the lungs, liver and mesentery; mimicking the development and progression of metastatic RCC in human [
2,
3,
21]. Our studies and others demonstrate that orthotopic models of heterotransplanted human RCC cells are representative of human RCC progression and preserve the karyotypic and histological characteristics of the human primary tumor [
6‐
9,
20]. These models can be used to study human RCC progression and metastasis and responsiveness of human RCC tumors to treatment
in vivo.
To investigate the therapeutic effect of genistein and radiation in the KCI-18 model, we first assessed KCI-18 cell responsiveness to these modalities
in vitro. KCI-18 cells were pre-treated with genistein for 24 hr, then irradiated and assayed in a clonogenic assay in the presence of genistein [
12]. KCI-18 RCC cells showed a comparable dose-dependent inhibition of cell growth when treated with genistein alone, but a lower response to radiation compared to PC-3 cells. The combination of 15 μM genistein with 3 Gy radiation, for KCI-18 cells, was more effective at inhibition of colony formation (80%) than genistein alone (45%) or radiation alone (50%) [
12]. These data confirm that RCC cell killing by radiation is also enhanced by genistein pre-treatment as demonstrated for PC-3 prostate cancer cells [
11,
12]. Based on these findings, the treatment of established KCI-18 kidney tumors with genistein alone or combined with radiation was tested.
In the KCI-18 RCC model, treatment of kidney tumor-bearing mice with genistein alone demonstrated a tendency to augment the growth of kidney tumors compared to control mice whereas radiation slowed the growth of kidney tumors. In contrast, genistein given in conjunction with tumor irradiation caused a significant inhibition in the growth of kidney tumors. The anti-tumor activity mediated by genistein and radiation on kidney tumors correlates with the levels of genistein measured in the mice blood. We have previously reported that mice treated with 5 mg/day genistein reached serum levels of about 30 μM compared to 0.2 μM in control mice without apparent toxicity [
13]. These levels are comparable to the 15 μM range measured in human volunteers consuming 50 mg isoflavone consisting of 40 mg of genistein and daidzein and in other studies [
22,
23].
The effects of genistein combined with radiation observed in KCI-18 kidney tumors in vivo are in agreement with our in vitro data showing a greater inhibition of KCI-18 cell division by the combined treatment. However, the in vivo genistein effect of stimulating kidney tumor growth data does not corroborate the findings of KCI-18 tumor cell inhibition in vitro suggesting the influence of tumor microenvironment factors in vivo.
Interestingly, when used as a single treatment modality, genistein also showed a trend in increasing the number and frequency of tumor nodules in the mesentery lining the bowel but no increase in lung metastasis was noted. Radiation alone and combined with genistein caused limited mesentery metastasis and no lung metastases in the majority of the mice.
These findings were confirmed by histological analysis of the kidney tumors and tissue sections. Genistein-treated kidney tumors showed histological changes comparable to those observed in PC-3 prostate tumors [
13], including necrosis, apoptosis and giant tumor cells. As a result of genistein anti-angiogenic effects [
24] and the high vascularity of kidney tumors, extensive hemorrhages were observed in genistein-treated kidney tumors. In contrast to genistein or radiation alone, tumors treated with the combined therapy provided histological evidence for significant inhibition of tumor growth and progression in the kidney, with residual focal abnormal tumor cells covering only 40–50% of the kidney. These findings confirm that genistein combined with radiation inhibit the growth and invasion of established kidney tumors and cause marked aberrations in tumor cells that are reminiscent of those observed in prostate tumors [
13]. Atypical giant cells are induced by either genistein or radiation alone but are more prominent in tumors treated with both modalities and represent a slow death due to alterations in cell division at the level of cytokinesis [
13,
17].
Genistein treatment showed a trend in increasing the number and frequency of metastatic mesentery metastases but not that of lung metastases. The mesentery tumor nodules, located in the fat tissue lining the bowel, could have developed from tumor cell migration through adipose tissue blood vessels due to the high vascularization of the kidney tumor. Combining genistein with tumor irradiation controlled the genistein-induced increase of mesentery metastases.
These data indicate that genistein treatment alone has tendency to stimulate the growth of the primary kidney tumor and to promote metastasis to proximal organs but not to distant organs. These findings in the orthotopic RCC model suggest that the effect of genistein on primary tumors located in different organs may differ, as prostate tumor growth was not increased by genistein treatment in orthotopic prostate cancer models. However, genistein promoted metastasis in both RCC and prostate tumor models.
Previous animal studies have emphasized the role of genistein in the prevention of prostate cancer and mammary tumors [
25‐
30]. In contrast, other studies showed increase in tumor growth by genistein in experimental colon cancer in rats, orthotopic mouse mammary tumors and rat prostate tumors [
31‐
33]. Genistein induced stimulation of kidney tumor growth in our orthotopic KCI-12 RCC is in agreement with these studies. Our study emphasizes the adverse effects of pure genistein when treatment is initiated on established tumors.
Clarification of the mechanism by which genistein causes increased metastasis in proximal organs including mesentery in the kidney tumor model or regional lymph nodes in the prostate tumor model is relevant for the clinical application of soy for cancer prevention and cancer treatment. Whether additional isoflavone compounds from the soybean (e.g., daidzein and glycitein) will protect against increased metastasis mediated by purified genistein is currently under investigation.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
G.G.H. designed and supervised the study and prepared the manuscript. Y.W. performed the animal experiments. M.C. is a pathologist who analyzed all histology slides with G.G.H. J.J.R. assisted with histology and data analysis. M.Y. is a medical physicist in charge of calibrating the X-ray machine and shielding of the mice. O.K. and F.H.S. participated in the design of the study. All authors read and approved the manuscript.