Abstract
The resistance of advanced colorectal cancers to therapy is often related to mutations in the p53 tumor suppressor gene. Because somatostatin (SRIF) receptors (ssts) are present in colorectal carcinomas, the treatment with targeted cytotoxic SRIF analogue AN-238, consisting of 2-pyrrolinodoxorubicin (AN-201) linked to octapeptide SRIF carrier RC-121, may overcome this resistance by producing a higher concentration of the cytotoxic agent in the tumors. Four colon cancer cell lines, HCT-116 and LoVo expressing wild-type p53, and HCT-15 and HT-29 with mutated p53, were investigated. HCT-116, HCT-15, and HT-29, but not LoVo possess functional ssts. We analyzed changes in p53, p21, and proliferating cell nuclear antigen (PCNA) concentrations in these cells in vitro by immunoblotting after exposure to AN-238, its radical AN-201, or doxorubicin (DOX). Equitoxic doses of AN-238, AN-201, or DOX affected p53, p21, and PCNA differently. Analysis of the p21:p53 ratios revealed that DOX increased p53 levels, but most of p53 was mutated and inactive, whereas AN-238 produced smaller changes in p53 concentrations but enhanced its activity. In HCT-15 cells, PCNA:p21 ratios, which are indicators of proliferation and repair processes, remained unchanged after exposure to AN-238 but were increased by DOX. In vivo studies in nude mice demonstrated that AN-238, AN-201, and DOX were equally effective on HCT-116 tumors that express wild-type p53. However, AN-238 also inhibited the growth of HCT-15 and HT-29 cancers that express mutant p53, whereas AN-201 and DOX showed no effect. None of the compounds could suppress the proliferation of LoVo tumors that lack functional ssts. In conclusion, cytotoxic SRIF analogue AN-238 inhibits the growth of experimental colon cancers that express ssts, regardless of their p53 status.
INTRODUCTION
Colorectal cancer is the third most common cause of cancer-related deaths in the Western world in both men and women (1). Although an early diagnosis and adjuvant chemotherapy have slightly increased the survival of colorectal cancer patients (2), the response to treatment in advanced cases is generally temporary, and clinical trials with second-line therapeutic options have been disappointing (3). The resistance to chemotherapy may be the consequence of genetic alterations in tumors, and in colorectal cancer, a poor response to chemotherapy has been clearly associated with mutations in the p53 gene (2, 4). Thus, the development of new therapeutic agents that can overcome resistance caused by mutant p53 has become one of the most important goals in the management of this malignancy (5, 6).
Recently, we synthesized and evaluated a series of novel cytotoxic peptide conjugates based on luteinizing hormone-releasing hormone, SRIF,3 and bombesin that can be used for targeted therapy of tumors expressing specific receptors for these hormones (7, 8). Various studies have indicated that receptors for all three of these peptides may be present in colorectal cancers (9, 10, 11, 12, 13, 14, 15, 16). In this study, we investigated the effectiveness of our targeted cytotoxic SRIF analogue, AN-238, consisting of a highly active derivative of DOX, 2-pyrrolinodoxorubicin (AN-201) linked to SRIF octapeptide analogue RC-121 (17, 18). Although receptors for bombesin may be more prevalent in colorectal cancers (10) than receptors for SRIF (19) or luteinizing hormone-releasing hormone (9), we selected AN-238 because it has been shown to be very efficacious in a wide variety of cancers (7, 8, 20, 21), and it is in a more advanced stage of preclinical development than its cytotoxic bombesin counterpart, AN-215.
Thus we tested the efficacy of AN-238 in vivo using sst-positive HCT-116, HCT-15, HT-29, and sst-negative LoVo colon cancer lines. In addition, because HCT-15 differs from HCT-116 only in that it expresses a mutant p53 gene, we investigated whether targeting could overcome the chemoresistance associated with mutant p53 gene expression in colorectal cancer. AN-238, cytotoxic radical 2-pyrrolinodoxorubicin, and its parent compound DOX were also evaluated in vitro for their effects on p53, p21, and PCNA protein expression.
MATERIALS AND METHODS
Materials.
SRIF analog RC-121 (d-Phe-C⏞ys-Tyr-d-Trp-Lys-Val-Cy⏞s-Thr-NH2) was synthesized in our laboratory (18). RC-160 (vapreotide; d-Phe-C⏞ys-Tyr-d-Trp-Lys-Val-Cy⏞s-Trp-NH2), originally synthesized in our laboratory, was made by Novabiochem (Laufelfingen, Switzerland). The cytotoxic conjugate AN-238 was made by coupling one molecule of 2-pyrrolinodoxorubicin-14-O-hemiglutarate to the NH2 terminus of [Lys(Fmoc)5]RC-121 followed by deprotection and purification (10). 2-Pyrrolinodoxorubicin (AN-201) was synthesized as described (17). DOX × HCl salt was purchased from Chemex Export-Import GmbH (Vienna, Austria). Chemicals, unless stated otherwise, were obtained from Sigma Chemical Co. (St. Louis, MO).
In Vitro Studies.
The human colon cancer cell lines HT-29, HCT-116, HCT-15, and LoVo were purchased from the ATCC (Manassas, VA) and maintained in culture as recommended by the ATCC. Briefly, HT-29 and HCT-116 cells were grown in McCoy’s 5A medium + 10% fetal bovine serum (FBS), HCT-15 cells in RPMI 1640 medium + 10% FBS + 4.5 g/liter glucose + 1 mm pyruvate, and LoVo cells in F12 medium + 20% FBS. The cells were cultured in a humidified atmosphere of 5% CO2/95% air at 37°C. Cell growth was estimated by using the crystal violet assay (22). The absorbance of each well was measured at 600 nm using a Beckman (Palo Alto, CA) plate reader. Values of the percentage of T/C were calculated, where T = A600 nm of treated cultures, and C = A600 nm of control cultures. HCT-116 and LoVo lines express wild-type whereas HCT-15 and HT-29 cell lines express mutant p53 (23, 24).
Immunoblotting.
HT-29, HCT-116, HCT-15, and LoVo cancer cells, 5 × 106 each, were placed in six-well plates and allowed to adhere. The cells were incubated for 90 min in the respective culture medium described earlier, containing 10−8 m AN-238 or AN-201, or 10−5 m DOX. The cells were then washed and further incubated in fresh medium. At 1.5, 4, 24, and 48 h after the start of the treatment, the cells were collected, rinsed with PBS (pH 7.2), and lysed for 1 h in extraction buffer containing 50 mm Tris-HCl, 150 mm NaCl, 1% NP40, 1 mm EGTA, 1 mm sodium orthovanadate in the presence of freshly added protease inhibitors (aprotinin, leupeptin, pepstatin, and phenylmethylsulfonyl fluoride). The lysates were centrifuged for 10 min at 10,000 × g at 4°C. The protein content of the total cell lysates was quantified using Bio-Rad (Hercules, CA) Bradford Assay according to the manufacturer’s instructions. Samples containing equivalent amounts of protein were separated using 7.5% or 10% SDS-polyacrylamide gel and transferred to a 0.45-μm nitrocellulose membrane. The blots were probed with goat polyclonal antibodies for p53 (FL-393)-G and p21 (C-19)-G proteins for 1 h at 1:1,500 dilution in Tris-buffered saline, followed by horseradish peroxidase-conjugated secondary antibody at 1:10,000 dilution (all antisera from Santa Cruz Biotechnology, Inc, Santa Cruz, CA). The bands were visualized by the enhanced chemiluminescence method (Amersham Pharmacia Biotech, Piscataway, NJ), according to the manufacturer’s instructions. To confirm that equal amounts of proteins were loaded onto the gels, the blots were reprobed with actin as a control (anti-actin I-19, in a dilution of 1:100; Santa Cruz Biotechnology). For the detection of PCNA in blots, anti-PCNA (Ab-I) mouse monoclonal antibody (Calbiochem, Cambridge, MA) was used at 1:1,500 dilution, and horseradish peroxidase-conjugated antimouse IgG was used as secondary antibody at 1:10,000 dilution.
Animals and Tumors.
Male athymic nude mice (Ncr nu/nu), approximately 6 weeks old on arrival, were obtained from the National Cancer Institute (Frederick Cancer Research and Development Center, Frederick, MD) and maintained under pathogen-limited conditions.
To obtain tumor donor animals, HCT-116, HCT-15, HT-29, or LoVo cells (106/mouse) were injected s.c. into groups of three nude mice. The well developed tumors were dissected and mechanically minced, and 2-mm3 pieces of tumor tissue were transplanted s.c. to both flanks of the experimental animals. When tumors became measurable, groups were formed with about equal average tumor sizes, and the control and treatment groups were selected at random. Body weights and tumor sizes were measured weekly. All animal experiments were carried out in accordance with institutional ethical guidelines for animal care.
Experimental Protocol.
In experiment I, the treatment of HCT-116 cancers was started 26 days after transplantation when tumor volume reached 56 ± 5 mm3. The treatment groups were as follows: group 1, control, vehicle only; group 2, AN-238 at 150 nmol/kg; group 3, AN-201 at 150 nmol/kg; and group 4, DOX at 13.8 μmol/kg (8 mg/kg). Each group contained 12 mice. Injections were given on days 1, 15, 35, and 50. The animals were sacrificed on day 55. In experiment II on HCT-15 cancers, the groups and treatment schedules were the same as in experiment I, the treatment being started 26 days after transplantation when the average tumor volume was 50 ± 2.5 mm3. In experiment III, the treatment of HT-29 cancers was initiated 19 days after transplantation when the tumors measured 33 ± 2 mm3. The treatment groups were: group 1, control, injection vehicle only; group 2, AN-238 at 200 nmol/kg; group 3, AN-201 at 200 nmol/kg; group 4, AN-238 at 150 nmol/kg; and group 5, AN-201 at 150 nmol/kg. Groups 2 and 3 were treated on days 1 and 38, and groups 4 and 5 on days 1, 8, and 38. Each group consisted of 10 mice. The experiment was terminated on day 56. In experiment IV, the treatment of LoVo cancers started 35 days after transplantation when the tumor volume was 64 ± 3 mm3. The treatment groups were as follows: group 1, control, vehicle only; group 2, AN-238 at 150 nmol/kg; and group 3, AN-201 at 150 nmol/kg given on days 1, 15, 29, and 43. Each group consisted of 7 animals. The mice were sacrificed on day 62.
The cytotoxic compounds were dissolved in 20 μl of 0.01 n acetic acid, diluted with 6% (w/v) aqueous d-mannitol, and 0.2 ml/20 g body weight were injected into the jugular vein. The doses and treatment schedules for the cytotoxic compounds were based on previous studies in our laboratory (7, 8). Tumor growth reduction (TGR) was calculated with the formula: TGR% = 100 − 100 × (T − t)/(C − c), where t = the volume of a treated tumor at the beginning of therapy, T = volume of the same tumor at the end of the experiment, c = volume of a control tumor at the start of the treatment, and C = volume of the same tumor at the end of the experiment. At the termination of the studies, the mice were sacrificed by decapitation under Metofane (methoxyflurane; Schering-Plough, Union, NJ) anesthesia.
Histological and Immunohistochemical Analysis.
The methods of histological examination were described (22). For the detection of p53 in tumor tissue with the immunohistochemical method, after a standard deparaffination procedure, the slides were heated in a microwave oven for two times for 5 min in 0.01 m sodium citrate buffer at pH 6.0. Incubation with anti-p53 (FL-393-G; Santa Cruz Biotechnology), at a dilution of 1:200, was followed by incubation with biotin-conjugated rabbit antigoat IgG at a dilution of 1:300, and Extravidin-peroxidase at a dilution of 1:100. All incubations were carried out at room temperature for 60 min. The reaction products were visualized with SigmaFast 3,3′-diaminobenzidine (Sigma). The nuclei with positive reaction were counted in areas showing the highest density, and their number per 1000 cells was used as the p53 index. It is generally accepted that a tumor with an index higher than 250 expresses the mutated gene (25).
Receptor Assays and mRNA for ssts.
Statistical Analysis.
The SigmaStat software (Jandel Scientific, San Rafael, CA) was used for statistical evaluation of the data. Tumor volume changes were evaluated by two-way repeated measures ANOVA, and the groups were compared by Dunnett’s comparison method. The results of sequential tests in the experiments in vitro were compared by one-way repeated measures ANOVA and Tukey’s procedure. All other data were evaluated by one-way ANOVA and Dunnett’s method.
RESULTS
In Vitro Studies
Cell Proliferation Assay
Preliminary experiments using various concentrations and exposure times of 30, 60, and 90 min revealed that AN-238 and AN-201 at 10−8 m, and DOX at 10−5 m show comparable inhibitory effects on cell growth after exposure for 90 min. Cell proliferation was assayed 24, 48, and 72 h after addition of the drugs, and the results are shown in Fig. 1. Under the conditions described in “Materials and Methods,” AN-238 and AN-201 had similar cytotoxic effects on all cell lines, independently of their sst or p53 status.
Immunoblotting
The levels of p53, p21, and PCNA were determined in various colon cancer cells 1.5, 4, 24, and 48 h after the addition of 10−8 m AN-238 or AN-201, or 10−5 m DOX to the medium for 90 min. The cytotoxic compounds affected the cellular regulatory proteins in vitro, depending on the p53 status of the cells. The effects are described below.
Effects on p53 Levels.
In HCT-116 cells, the immunologically detectable p53 levels were elevated by each compound, DOX causing the biggest and AN-238 the smallest increase. AN-238 slightly raised p53 concentrations in LoVo cells at 48 h only. In HCT-15 and HT-29 cells with mutant p53, the level of p53 was practically unchanged by the compounds, only DOX causing a moderate increase at 24 and 48 h in HCT-15 cells (Figs. 2 and 3).
Effects on p21 Levels.
The p21 levels were slightly elevated by each compound in HCT-116 cells but remained unchanged in LoVo cells (Figs. 3 and 4). In these two cell lines with wild-type p53, the p21:p53 ratios were significantly higher after exposure to AN-238 than after treatment with DOX (P = 0.036 for HCT-116 and P = 0.047 for LoVo cells). Very low concentrations of p21 were detected in control HCT-15 cells with mutant p53. In these cells, exposure to AN-238 slightly increased concentrations of p21, but AN-201 and DOX produced no changes. The p21:p53 ratios were significantly lower in HCT-15 cells exposed to AN-201 (P = 0.005) or DOX (P = 0.001) than in those treated with AN-238 (Fig. 3). p21 was not detectable in HT-29 cells.
Effects on PCNA Levels.
In the HCT-116 cell line, PCNA levels were elevated by all three compounds, but with a different time course (Figs. 2 and 5). In HCT-15 and HT-29 cells, changes in PCNA content depended on the compounds. AN-238 caused a slight increase in HCT-15 cells and a marginal decrease in HT-29 cells. DOX increased PCNA levels in both cell lines. Interestingly, AN-201 acted similarly to AN-238 on HCT-15 cells but like DOX on HT-29 cells (Fig. 5). PCNA:p21 ratios were low in both HCT-116 and HCT-15 cells exposed to AN-238. However, these ratios were much higher than control values in HCT-15 cells after addition of either AN-201 or DOX (Fig. 6). There was a significant difference in PCNA:p21 ratios for HCT-15 cells treated with AN-238 versus DOX (P = 0.002). A similar difference was observed for AN-201 versus DOX (P = 0.027).
Effect of Treatment with Cytotoxic Compounds on Tumor Growth in Vivo
In experiment I, repeated administration of AN-238 and AN-201 at 150 nmol/kg doses and DOX at 13.8 μmol/kg resulted in a similar inhibition of the growth of HCT-116 colon cancers (Fig. 7,A; Table 1). One mouse died in the group treated with AN-238 and one in the group given AN-201. Two mice were dead in the group receiving DOX. There were no deaths among controls. The mean heart weight was significantly decreased and the average weight of testicles was lowered by DOX, but not by AN-238 or AN-201. In contrast to HCT-116 cancers, only AN-238 inhibited significantly the growth of HCT-15 tumors in experiment II (P < 0.05 compared with controls; Fig. 7,B; Table 1). One mouse died in the control group and in that treated with AN-201. Two animals died in the groups receiving AN-238 or DOX. Body, liver, and spleen weights were lowered only by DOX. Weights of hearts were the smallest in the group receiving DOX, but the differences from control were statistically not significant.
In experiment III, AN-238 at three 150 nmol/kg doses significantly inhibited the growth of HT-29 cancers as shown by tumor volume and weight data (Fig. 7,C, Table 1). Equimolar amounts of AN-201 had no significant effect. The mice tolerated the treatment well, and only one mouse died in each of the treated groups, whereas there were no deaths among controls. AN-238 and AN-201 at two doses of 200 nmol/kg had an inhibitory effect on tumor growth similar to that of three doses of 150 nmol/kg, but the two-times, 200 nmol/kg regimen was toxic, resulting in the deaths of five mice in each of the treated groups. Mean body weights and testes weights were also significantly decreased in these groups.
Histological Analysis
In HCT-116 cancers, AgNOR numbers were decreased and apoptosis was increased by all four compounds, but the mitotic index was reduced only by AN-238. In HCT-15 cancers, AN-238 caused a decrease in mitoses and AgNORs and an increase in apoptosis, whereas treatments with AN-201 and DOX had no effect on these parameters (Table 1). Administration of AN-238 resulted in a decrease in AgNOR counts in HT-29 cancers, whereas AN-201 had no significant effect on this marker of cell proliferation rate. Mitotic and apoptotic indices remained unchanged in HT-29 tumors after treatment with both compounds (Table 1). In LoVo cancers, AgNOR counts were reduced by AN-238.
Detection of p53 with Immunohistochemistry
The p53-positive nuclei were counted on slides of control tumors, and their numbers per 1000 cells were used as p53 indices. The mean p53 indices for the control tumors are shown in Table 2. Our data confirm the results of other investigators (23, 24) that HCT-116 and LoVo cancers express wild-type p53, and HCT-15 and HT-29 have mutated p53 (Table 2).
SRIF Binding Sites and Expression of mRNA for ssts
Radiolabeled SRIF octapeptide RC-160 was bound to a single class of high-affinity, low-capacity binding sites on membrane fractions of control HCT-116, HCT-15, and HT-29 cancers (Table 2). LoVo cancers showed no binding. mRNA for sst5 was present in samples of all four tumors. mRNA for sst2A was expressed in HCT-116, HCT-15 and LoVo cancers, but could not be detected in HT-29 tumors (Fig. 8, Table 2). mRNA for sst3 was present in HCT-116 and LoVo cancers, but not in p53-mutated HCT-15 and HT-29 tumors (Fig. 8; Table 2).
DISCUSSION
The presence of ssts on various tumors can be used for imaging and targeted therapy based on radionuclide or cytotoxic analogues of SRIF octapeptides such as octreotide, lanreotide, RC-160, and RC-121 (8, 13, 28). It is well established that neuroendocrine tumors express high concentrations of sst2 and sst5 that bind SRIF octapeptide analogues with high affinity (11, 29). However, cancers of non-neuroendocrine origin such as ductal pancreatic carcinomas or colorectal cancers appear to be entirely different from endocrine tumors with regard to sst expression (8). Various research groups reported diverse results on the occurrence and binding characteristics of receptors for SRIF in colorectal cancers. Radulovic et al. (15) showed high-affinity binding sites for radiolabeled [Tyr11]SRIF-14 in 8 of 15 human colon cancer specimens examined and found no correlations between the presence of the receptors and the tumor stage or differentiation. RC-160 and RC-121 showed binding affinities similar to SRIF receptors on these colon tumors (15). Virgolini et al. (16) demonstrated in vivo the accumulation of 111In-DOTA-lanreotide in liver or lung metastases of colorectal cancers in all three patients examined. In contrast, in one moderately differentiated colon adenocarcinoma sample, Reubi et al. (30) could not detect binding of radiolabeled octreotide or [Leu8,d-Trp22,Tyr25]SRIF-28 and, in a subsequent study, showed the presence of binding sites for SRIF in only 2 of 25 colon cancer specimens (19). The available findings on the presence of mRNAs for the five known receptor subtypes (sst1–5) in colorectal cancers are also somewhat controversial (11, 12, 14). Laws et al. (12) and Pinzani et al. (14) detected mRNA for sst2 in 90–100% of colon tumors, and the frequency of sst5 expression was found to be lower with the progression of the disease (12). In contrast, Buscail et al. (11) demonstrated mRNA for sst5 in most colon cancer samples, but found a much lower incidence of mRNA for sst2 in advanced cases and metastatic lesions. More recently, a study on a relatively large number of samples revealed that there was an increased expression of mRNA for sst5 in colon cancers, and this expression was higher in tumors than in normal mucosa. The expression of sst4, sst3, and sst2 was low (31). There was no difference in receptor subtype expression among tumors of different stages, but the presence of sst5 depended on the localization of the tumor (32). Three of the four colon cancers investigated in this study, HCT-116, HCT-15, and HT-29, possess functional ssts and show binding affinities for octapeptide analogue RC-160 in the nanomolar range and concentrations between 400 and 700 fmol/mg membrane protein. mRNA analysis revealed the expression of sst5 in all cancers tested, and sst2 was absent only in HT-29 tumors. Interestingly, only the two cell lines HCT-116 and LoVo with wild-type p53 express mRNA for sst3. However, in LoVo cancers, no functional SRIF receptors are found, despite the expression of mRNA for ssts 2A, 3, and 5. The presence of subtype 2, 3, or 5 on colorectal cancers might allow a treatment with targeted cytotoxic SRIF analogue AN-238. The feasibility of this approach was recently demonstrated in various experimental tumors including exocrine pancreatic and prostate cancers, which express low concentrations of sst2 and/or 5 (8).
Because mutations in tumor suppressor gene p53, which can affect therapeutic responses, were detected in 50–80% of colorectal cancers (2, 4, 5), we studied two colon cancer cell lines, HCT-116 and LoVo, which express wild-type, and two lines, HCT-15 and HT-29, with mutant p53. Tumor suppressor p53 protein is implicated in multiple functions including cell cycle control and apoptosis (4, 33), and direct correlations were found between p53 mutations and aggressiveness of growth (4). Wild-type p53 protein is rapidly degraded, hence usually it is not detectable by immunohistochemistry. In contrast, mutated p53 proteins have a much longer half-life, and their presence can be revealed by immunological methods (34). Another approach to determine the p53 status of cells is based on the quantitative measurement of its downstream effector p21 (p21WAF1/Cip1), which can arrest cell cycle progression by inhibiting cyclin-dependent kinases (4, 35, 36, 37). The p53 gene mutation and the nuclear accumulation of its p53 protein product are associated with down-regulation of p21 (36). Thus, we also evaluated changes in the levels of p21 after treatment with the cytotoxic compounds. A very low concentration or absence of p21 protein in HT-29 and HCT-15 cells found in our studies is consistent with the mutant p53 status of these lines. PCNA is likewise a regulatory protein that has an essential role in DNA replication and in various forms of DNA repair (38). p21 inhibits DNA replication by a PCNA-dependent pathway. The molecular ratio of p21:PCNA is 1:1 in normal fibroblasts, and p21 is absent during the S phase of the cell cycle (38).
The aim of the in vitro segment of the present studies on colon cancer cell lines was the analysis of early cellular events caused by the compounds, and the elucidation of possible differences in the effects of targeted analogue AN-238, radical AN-201, and its parent compound DOX on the specific proteins cited above that regulate cell cycle checkpoints, and DNA proliferation and repair. Some of these proteins such as p53 and p21 can also influence the cellular responses to therapeutic agents (39). The results of our in vitro studies demonstrate that despite their similar short-term effects on cell proliferation, AN-238, AN-201, and DOX affect differently the expression of various regulatory proteins in the colon cancer cells tested. This may explain, at least in part, the different long-term effects of these agents on tumor growth in vivo. In the presence of DOX in the medium, there is a strong increase in concentrations of p53 in HCT-116, HCT-15, and LoVo cells, but this is accompanied by an elevation of p21 only in HCT-116 cells. Similar alterations in p53 and p21 levels in DOX-treated breast cancer cells with wild-type or mutant p53 have been found by others (40). This shows that in response to DOX, an inactive p53 is produced by the cells or a selection process is started that results in a progressive dominance of cells with mutant p53. Such a selection of the mutant cell population by cisplatin has been reported (33), and these changes can lead to increased resistance to therapy. In contrast to DOX, AN-238, and to a lesser degree AN-201, exhibited no such discriminating effect between cells expressing wild-type or mutant p53 and slightly increased p21 levels while causing only small changes in detectable p53 concentrations. This means that the cytotoxic SRIF analogue either activates p53 or raises the levels of p21 through a pathway independent of p53. It has to be noted, however, that p21 was not induced and activated by AN-238 in HT-29 cells. The three compounds had a similar effect on p21:p53 ratios in LoVo cells, which do not express ssts. The amount of PCNA was increased strongly by DOX, moderately by AN-201, and only slightly by AN-238. Thus, PCNA:p21 ratios remained unchanged after treatment with AN-238. In contrast, this ratio was increased by AN-201 and DOX, which also reflected increased DNA replication in the cells.
In the studies in vivo, we compared the tumor-inhibitory effect of the targeted cytotoxic SRIF analogue AN-238 with that of its radical AN-201 in all four colon cancer models. DOX was also tested in HCT-116 and its counterpart with mutant p53, HCT-15. The results show that AN-238 has a stronger tumor-inhibitory effect than cytotoxic radical AN-201 on sst-positive HT-29 and HCT-15 cancers. The two compounds were about equally effective on HCT-116 tumors that also express ssts, but were ineffective in LoVo cancers that lack functional ssts.
Some investigators detected mRNA for ssts in specimens of human colon cancers (11, 12, 14, 31, 32), but mRNA expression does not necessarily mean that functional receptors would be present on tumor cells, as shown also by our findings on LoVo cancers. Various studies demonstrated that clinically available radiolabeled SRIF analogues such as Octreoscan (28) and DOTALAN (13) can bind to a broad range of human tumors, including some colon cancers, that express sst2 or sst5 (8, 13, 28). However, because at this time the data are inconclusive, more clinical investigations are needed to ascertain the percentage and types of colorectal cancers that have functional receptors and that could eventually be targeted by radiolabeled or cytotoxic SRIF analogues. Such studies are in progress in our laboratory.
The present study demonstrates that AN-238 has a strong inhibitory effect on three sst-positive colon cancer lines, independently of their p53 status. In contrast, AN-201 and DOX had no effect on tumors expressing the mutant p53 gene. This indicates that AN-238 may have a higher therapeutic potential on colorectal cancers than do the nontargeted cytotoxic compounds AN-201 or DOX. Our results suggest that patients with colorectal cancers expressing sst subtype 2 and/or 5 and possibly 3, detectable with radiolabeled SRIF octapeptides, might benefit from treatment with cytotoxic SRIF analogue AN-238, independently of the p53 status.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Supported in part by the Medical Research Service of the Department of Veterans Affairs (to A. V. S.), and by a grant from ASTA Medica, Frankfurt am Main, Germany to Tulane University School of Medicine (to A. V. S.).
The abbreviations used are: SRIF, somatostatin; sst1–5, somatostatin receptors; DOX, doxorubicin; FBS, fetal bovine serum; AgNOR, argyrophilic nucleolar organizer region; PCNA, proliferating cell nuclear antigen; RT-PCR, reverse transcription-polymerase chain reaction.
Effect of 90 min exposure of colon cancer cells to 10−8 m AN-238 or AN-201, or 10−5 m DOX on cell proliferation in vitro. HCT-116, HCT-15, HT-29, and LoVo cells were grown in culture media recommended by the ATCC and described in “Materials and Methods.” Cell growth was estimated by the crystal violet assay 24, 48, and 72 h after the addition of the compounds. The % T/C was determined by measuring T = A600 nm of treated cultures and C = A600 nm of control cultures. Vertical bars, SE.
Effect of 90 min exposure of colon cancer cells to 10−8 m AN-238 or AN-201, or 10−5 m DOX on cell proliferation in vitro. HCT-116, HCT-15, HT-29, and LoVo cells were grown in culture media recommended by the ATCC and described in “Materials and Methods.” Cell growth was estimated by the crystal violet assay 24, 48, and 72 h after the addition of the compounds. The % T/C was determined by measuring T = A600 nm of treated cultures and C = A600 nm of control cultures. Vertical bars, SE.
The p53 and PCNA levels in HCT-116, HCT-15, and HT-29, and p53 levels in LoVo human colon cancer cells in vitro after treatment with 10−8 m AN-238 or AN-201, or 10−5 m DOX for 90 min. Cell homogenates were assayed by immunoblotting 1.5, 4, 24, and 48 h after the addition of the compounds to the media as described in “Materials and Methods.” The controls (C) were tested at 1.5 h. Representative blots of two experiments are shown. In HT-29 tumors, the appearance of double bands may be the result of phosphorylation of p53. Actin was used as an internal control.
The p53 and PCNA levels in HCT-116, HCT-15, and HT-29, and p53 levels in LoVo human colon cancer cells in vitro after treatment with 10−8 m AN-238 or AN-201, or 10−5 m DOX for 90 min. Cell homogenates were assayed by immunoblotting 1.5, 4, 24, and 48 h after the addition of the compounds to the media as described in “Materials and Methods.” The controls (C) were tested at 1.5 h. Representative blots of two experiments are shown. In HT-29 tumors, the appearance of double bands may be the result of phosphorylation of p53. Actin was used as an internal control.
Changes in p53 and p21 levels in HCT-116, HCT-15, HT-29, and LoVo human colon cancer cells proliferating in vitro after treatment with 10−8 m AN-238 or AN-201, or 10−5 m DOX for 90 min. p53 and p21 were analyzed by immunoblotting assays and measured with a densitometer. The results were normalized versus or adjusted according to the actin content of the samples. □, mean p53 ▪, p21 levels measured in two experiments and compared with control values. p21 was not found in HT-29 cells. Bars, SE. For p53 findings: a, P < 0.05 versus DOX; b, P < 0.05 versus AN-201. For p21 data: c, P < 0.05 versus DOX; d, P < 0.05 versus AN-201.
Changes in p53 and p21 levels in HCT-116, HCT-15, HT-29, and LoVo human colon cancer cells proliferating in vitro after treatment with 10−8 m AN-238 or AN-201, or 10−5 m DOX for 90 min. p53 and p21 were analyzed by immunoblotting assays and measured with a densitometer. The results were normalized versus or adjusted according to the actin content of the samples. □, mean p53 ▪, p21 levels measured in two experiments and compared with control values. p21 was not found in HT-29 cells. Bars, SE. For p53 findings: a, P < 0.05 versus DOX; b, P < 0.05 versus AN-201. For p21 data: c, P < 0.05 versus DOX; d, P < 0.05 versus AN-201.
Changes in p21 levels in HCT-116, HCT-15, and LoVo human colon cancer cells in vitro after treatment with 10−8 m AN-238 or AN-201, or 10−5 m DOX for 90 min. The cells were homogenized and specific proteins were detected by immunoblotting assays as described in “Materials and Methods” 1.5, 4, 24, and 48 h after the addition of the compounds to the media. The controls (C) were tested at 1.5 h. The figure shows representative results of two experiments.
Changes in p21 levels in HCT-116, HCT-15, and LoVo human colon cancer cells in vitro after treatment with 10−8 m AN-238 or AN-201, or 10−5 m DOX for 90 min. The cells were homogenized and specific proteins were detected by immunoblotting assays as described in “Materials and Methods” 1.5, 4, 24, and 48 h after the addition of the compounds to the media. The controls (C) were tested at 1.5 h. The figure shows representative results of two experiments.
Changes in PCNA levels in HCT-116, HCT-15, and HT-29 human colon cancer cells in vitro after treatment with 10−8 m AN-238 or AN-201, or 10−5 m DOX for 90 min. PCNA was analyzed by immunoblotting assays and measured with a densitometer. The results were normalized by the actin content of the samples. ▨, mean PCNA levels determined in two experiments and compared with control values. Bars, SE. a, P < 0.05 versus DOX; b, P < 0.05 versus AN-201.
Changes in PCNA levels in HCT-116, HCT-15, and HT-29 human colon cancer cells in vitro after treatment with 10−8 m AN-238 or AN-201, or 10−5 m DOX for 90 min. PCNA was analyzed by immunoblotting assays and measured with a densitometer. The results were normalized by the actin content of the samples. ▨, mean PCNA levels determined in two experiments and compared with control values. Bars, SE. a, P < 0.05 versus DOX; b, P < 0.05 versus AN-201.
Changes in PCNA:p21 ratios in HCT-116 and HCT-15 human colon cancer cells in vitro after treatment with 10−8 m AN-238 or AN-201, or 10−5 m DOX for 90 min. PCNA and p21 concentrations are normalized versus control values.
Changes in PCNA:p21 ratios in HCT-116 and HCT-15 human colon cancer cells in vitro after treatment with 10−8 m AN-238 or AN-201, or 10−5 m DOX for 90 min. PCNA and p21 concentrations are normalized versus control values.
Effect of treatment with cytotoxic SRIF analogue AN-238, and cytotoxic radical AN-201 and its parent compound DOX on the growth of human colon cancer xenografts in nude mice. Arrows, days of treatment; vertical bars, SE. *, P < 0.05; **, P < 0.01 versus control.
Effect of treatment with cytotoxic SRIF analogue AN-238, and cytotoxic radical AN-201 and its parent compound DOX on the growth of human colon cancer xenografts in nude mice. Arrows, days of treatment; vertical bars, SE. *, P < 0.05; **, P < 0.01 versus control.
RT-PCR analysis of mRNA expression for human (h) sst (SSTR) subtypes 2A, 3, and 5 in untreated human colon cancers grown in nude mice. PCR products were separated using 1.8% agarose gel electrophoresis and stained with ethidium bromide. The sizes of the expected PCR products were 1104 bp for hSSTR2A, 183 bp for hSSTR3, and 222 bp for hSSTR5. Lanes: M, molecular weight marker; +, positive control (cDNA plasmids); 1–2, HCT-116; 3–4, HCT-15; 5–6, LoVo; 7–8, HT-29.
RT-PCR analysis of mRNA expression for human (h) sst (SSTR) subtypes 2A, 3, and 5 in untreated human colon cancers grown in nude mice. PCR products were separated using 1.8% agarose gel electrophoresis and stained with ethidium bromide. The sizes of the expected PCR products were 1104 bp for hSSTR2A, 183 bp for hSSTR3, and 222 bp for hSSTR5. Lanes: M, molecular weight marker; +, positive control (cDNA plasmids); 1–2, HCT-116; 3–4, HCT-15; 5–6, LoVo; 7–8, HT-29.
Effect of treatment with cytotoxic SRIF analog AN-238 and cytotoxic radical AN-201, and its parent compound DOX on the growth of human colon cancers in nude mice
Groups . | Tumor growth reduction (%) . | Tumor weights (g) . | Histological characteristics of tumors . | . | . | . | |||
---|---|---|---|---|---|---|---|---|---|
. | . | . | Area of necrosis (%) . | Mitotic index . | Apoptotic index . | AgNOR count . | |||
Experiment I: HCT-116 | |||||||||
1. Control | 9.59 ± 3.29 | 61 ± 6 | 7.5 ± 0.8 | 4.4 ± 0.7 | 6.33 ± 0.11 | ||||
2. AN-238 | 74a | 2.47 ± 0.69a | 67 ± 6 | 4.6 ± 0.5a | 9.3 ± 1.7a | 5.42 ± 0.10a | |||
3. AN-201 | 73a | 2.58 ± 0.83a | 49 ± 7 | 5.9 ± 0.7 | 9.9 ± 1.5a | 5.81 ± 0.10a | |||
4. DOX | 77a | 2.15 ± 0.67a | 63 ± 8 | 6.6 ± 0.8 | 9.2 ± 0.9a | 5.81 ± 0.06a | |||
Experiment II: HCT-15 | |||||||||
1. Control | 4.89 ± 1.35 | 47 ± 7 | 9.4 ± 1.2 | 4.9 ± 0.6 | 6.45 ± 0.17 | ||||
2. AN-238 | 78a | 1.37 ± 0.62a | 73 ± 5a | 5.7 ± 0.9 | 8.8 ± 1.2a | 5.33 ± 0.33a | |||
3. AN-201 | 51 | 2.63 ± 0.88 | 54 ± 6 | 8.3 ± 1.4 | 6.6 ± 0.6 | 5.90 ± 0.26 | |||
4. DOX | 48 | 2.74 ± 0.90 | 60 ± 11 | 10.5 ± 2.31 | 6.0 ± 0.6 | 6.20 ± 0.21 | |||
Experiment III: HT-29 | |||||||||
1. Control | 3.88 ± 1.03 | 54 ± 5 | 9.9 ± 0.7 | 7.3 ± 0.7 | 6.99 ± 0.15 | ||||
2. AN-238, 200 nmol/kg, 2 times | 74b | 1.18 ± 0.49a | 62 ± 9 | 7.3 ± 0.9 | 8.4 ± 1.0 | 5.22 ± 0.07a | |||
3. AN-201, 200 nmol/kg, 2 times | 49 | 2.01 ± 0.67 | 56 ± 8 | 9.1 ± 1.1 | 11.8 ± 1.4 | 5.78 ± 0.13 | |||
4. AN-238, 150 nmol/kg, 3 times | 79b | 1.06 ± 0.34b | 71 ± 6 | 10.4 ± 1.2 | 10.4 ± 1.1 | 5.15 ± 0.05a | |||
5. AN-201, 150 nmol/kg, 3 times | 43 | 2.12 ± 0.61 | 45 ± 8 | 9.0 ± 0.7 | 9.3 ± 1.0 | 5.69 ± 0.11 | |||
Experiment IV: LoVo | |||||||||
1. Control | 1.27 ± 0.34 | 66 ± 6 | 26.3 ± 3.1 | 14.8 ± 1.9 | 4.70 ± 0.08 | ||||
2. AN-238 | 38 | 0.67 ± 0.14 | 82 ± 7 | 19.3 ± 1.8 | 15.6 ± 2.0 | 4.29 ± 0.09a | |||
3. AN-201 | 41 | 1.05 ± 0.41 | 73 ± 7 | 24.0 ± 3.2 | 18.4 ± 2.4 | 4.44 ± 0.09 |
Groups . | Tumor growth reduction (%) . | Tumor weights (g) . | Histological characteristics of tumors . | . | . | . | |||
---|---|---|---|---|---|---|---|---|---|
. | . | . | Area of necrosis (%) . | Mitotic index . | Apoptotic index . | AgNOR count . | |||
Experiment I: HCT-116 | |||||||||
1. Control | 9.59 ± 3.29 | 61 ± 6 | 7.5 ± 0.8 | 4.4 ± 0.7 | 6.33 ± 0.11 | ||||
2. AN-238 | 74a | 2.47 ± 0.69a | 67 ± 6 | 4.6 ± 0.5a | 9.3 ± 1.7a | 5.42 ± 0.10a | |||
3. AN-201 | 73a | 2.58 ± 0.83a | 49 ± 7 | 5.9 ± 0.7 | 9.9 ± 1.5a | 5.81 ± 0.10a | |||
4. DOX | 77a | 2.15 ± 0.67a | 63 ± 8 | 6.6 ± 0.8 | 9.2 ± 0.9a | 5.81 ± 0.06a | |||
Experiment II: HCT-15 | |||||||||
1. Control | 4.89 ± 1.35 | 47 ± 7 | 9.4 ± 1.2 | 4.9 ± 0.6 | 6.45 ± 0.17 | ||||
2. AN-238 | 78a | 1.37 ± 0.62a | 73 ± 5a | 5.7 ± 0.9 | 8.8 ± 1.2a | 5.33 ± 0.33a | |||
3. AN-201 | 51 | 2.63 ± 0.88 | 54 ± 6 | 8.3 ± 1.4 | 6.6 ± 0.6 | 5.90 ± 0.26 | |||
4. DOX | 48 | 2.74 ± 0.90 | 60 ± 11 | 10.5 ± 2.31 | 6.0 ± 0.6 | 6.20 ± 0.21 | |||
Experiment III: HT-29 | |||||||||
1. Control | 3.88 ± 1.03 | 54 ± 5 | 9.9 ± 0.7 | 7.3 ± 0.7 | 6.99 ± 0.15 | ||||
2. AN-238, 200 nmol/kg, 2 times | 74b | 1.18 ± 0.49a | 62 ± 9 | 7.3 ± 0.9 | 8.4 ± 1.0 | 5.22 ± 0.07a | |||
3. AN-201, 200 nmol/kg, 2 times | 49 | 2.01 ± 0.67 | 56 ± 8 | 9.1 ± 1.1 | 11.8 ± 1.4 | 5.78 ± 0.13 | |||
4. AN-238, 150 nmol/kg, 3 times | 79b | 1.06 ± 0.34b | 71 ± 6 | 10.4 ± 1.2 | 10.4 ± 1.1 | 5.15 ± 0.05a | |||
5. AN-201, 150 nmol/kg, 3 times | 43 | 2.12 ± 0.61 | 45 ± 8 | 9.0 ± 0.7 | 9.3 ± 1.0 | 5.69 ± 0.11 | |||
Experiment IV: LoVo | |||||||||
1. Control | 1.27 ± 0.34 | 66 ± 6 | 26.3 ± 3.1 | 14.8 ± 1.9 | 4.70 ± 0.08 | ||||
2. AN-238 | 38 | 0.67 ± 0.14 | 82 ± 7 | 19.3 ± 1.8 | 15.6 ± 2.0 | 4.29 ± 0.09a | |||
3. AN-201 | 41 | 1.05 ± 0.41 | 73 ± 7 | 24.0 ± 3.2 | 18.4 ± 2.4 | 4.44 ± 0.09 |
Values are mean ± SE.
P < 0.05 vs. control.
P < 0.01 vs. control.
Characteristics of ssts and p53 content in untreated human colon cancers in nude mice
p53 index was determined by immunohistochemistry as described in “Materials and Methods.” Binding characteristics were obtained from two to three independent 12-point displacement experiments. [125I]RC-160 was used as a radioligand. mRNA expression for subtypes of ssts was analyzed by RT-PCR as described in “Materials and Methods.”
Tumor . | ssts . | . | mRNA for . | . | . | p53 index . | |||
---|---|---|---|---|---|---|---|---|---|
. | Bmax (fmol/mg protein) . | Kd (nm) . | sst2A . | sst3 . | sst5 . | . | |||
HCT-116 | 677.8 ± 59.4 | 8.24 ± 0.67 | + | + | + | 64 ± 11 | |||
HCT-15 | 398.3 ± 92.4 | 5.05 ± 0.18 | + | ND | + | 768 ± 61a | |||
HT-29 | 432.0 ± 21.1 | 3.4 ± 0.12 | ND | ND | + | 911 ± 13a | |||
LoVo | ND | ND | + | + | + | 108 ± 10 |
Tumor . | ssts . | . | mRNA for . | . | . | p53 index . | |||
---|---|---|---|---|---|---|---|---|---|
. | Bmax (fmol/mg protein) . | Kd (nm) . | sst2A . | sst3 . | sst5 . | . | |||
HCT-116 | 677.8 ± 59.4 | 8.24 ± 0.67 | + | + | + | 64 ± 11 | |||
HCT-15 | 398.3 ± 92.4 | 5.05 ± 0.18 | + | ND | + | 768 ± 61a | |||
HT-29 | 432.0 ± 21.1 | 3.4 ± 0.12 | ND | ND | + | 911 ± 13a | |||
LoVo | ND | ND | + | + | + | 108 ± 10 |
Values are mean ± SE,
P < 0.001 vs. HCT-116 or LoVo. ND, not detectable.