Background
Prostate cancer is the second most common cancer of men, affecting approximately 14% of patients [
1]. While the risk of developing prostate cancer might be beneficially influenced by proper diet and physical activity [
2], there are no confirmed pharmacological means for the prevention of these types of tumors. The majority of prostate cancer cases are diagnosed at the localized stage enabling effective treatment; however, a significant fraction of patients develops metastatic disease that often progresses into treatment-irresponsive, ultimately resulting in patient’s death [
3].
Initial treatment of prostate cancer usually comprises hormone therapy; however, when tumors are irresponsive to hormonal treatment (i.e., in case of castrate-resistant prostate cancer), the most common first-line treatment includes the simultaneous use of docetaxel and prednisone. Docetaxel is a semi-synthetic taxane that inhibits microtubular depolymerization and block
bcl-2 and
bcl-xl gene expression [
4,
5]. Prednisone, in turn, is a glucocorticoid that is used to improve symptoms such as pain [
6]. It was also shown to inhibit cell proliferation and induce apoptosis in prostate cancer cells [
7,
8], and thus decrease the level of prostate-specific antigen [
9]. Accordingly, in multiple studies, prednisone was shown to promote anticancer activity of docetaxel [
10‐
14]. However, the use of glucocorticosteroids in patients with prostate cancer is associated with the risk of adverse side effects (as reviewed, for example, by Dorff and Crowford [
15]), and it eventually leads to the development of resistance to chemotherapy [
16]. Therefore, there is still an urgent need for new treatment regimens that would enable efficient yet safe means for the therapy of patients suffering from prostate cancer.
1-methylnicotinamide (1-MNA) is an endogenous metabolite of nicotinamide (NA) that has recently gained attention due to its anti-inflammatory [
17] and anti-thrombotic [
18] activity driven by mechanisms dependent on prostacyclin (PGI
2) release [
18,
19]. Another compound that has been shown to modulate thrombus formation based on the PGI
2-related mechanisms is 1,4-dimethylpyridine (1,4-DMP) – a structural analog of 1-MNA that arises naturally in roasted coffee seeds [
20]. In addition, it has been recently shown that both 1-MNA and 1,4-DMP could inhibit metastases formation in the model of experimental and spontaneous metastasis of 4T1 murine mammary gland cancer [
21].
The present work is aimed to establish whether 1,4-DMP may have an anti-oncogenic effect in the prophylaxis and the treatment of prostate tumors.
Methods
Drugs
1,4-DMP and 1-MNA were used in the form of chlorides provided by the Institute of Applied Radiation Chemistry, Technical University of Lodz, Poland. Prior to use, both salts were diluted in drinking water such that mice received the predetermined dose of the drugs. Docetaxel (DTX) was purchased at Ak Scientific (USA). All drugs were administrated at the doses and according to the schedules presented in Table
1.
Table 1
Drugs, doses and therapeutic regimens applied in the presented studies
Spontaneous tumor formation (Fig. 1) | 1-MNA | per os in drinking water | 100 mg/kg/day | continuously from the age of 8 to 12 weeks to the day of the necropsy |
1,4-DMP | per os in drinking water | 100 mg/kg/day | continuously from the age of 8 to 12 weeks to the day of the necropsy |
PC-3 M-luc2 tumors (Fig. 2) | 1,4-DMP | per os in drinking water | 100 mg/kg/day | continuously from the day 1 to the end of the experiment |
DTX | intraperitoneally | 10 mg/kg | 2 doses at days 15 and 22 |
Mice
Eight- to twelve-weeks-old male C57BL/6-Tg(TRAMP) 8247 Ng/J (TRAMP) mice were purchased from the Jackson Laboratory (USA). Seven- to eight-weeks-old BALB/c Nude male mice were provided by Charles Rivers Laboratories (Germany) (Table
2). All experiments were performed according to the
Interdisciplinary Principles and Guidelines for the Use of Animals in Research, Marketing and Education issued by the New York Academy of Sciences’ Ad Hoc Committee on Animal Research and were approved by the 1st Local Committee for Experiments with the Use of Laboratory Animals, Wroclaw, Poland.
Table 2
Strains and number of mice used in the experiments
Spontaneous tumor formation (Fig. 1) | TRAMP | 15 | 45 |
PC-3M-luc2 tumors (Figs. 2, 3 and 4) | C57BL/6 | 9 | 36 |
Cell culture and transplantation
Human prostate cancer PC-3M-luc2 cell line stably expressing the firefly luciferase gene (luc) was obtained from Caliper Life Sciences Inc. (USA). Cells were cultured in RPMI 1640 + Gluta-MAX™ medium (Life Technologies, USA) supplemented with 10% fetal bovine serum (Sigma-Aldrich, Germany) and antibiotics (penicillin and streptomycin—Polfa Tarchomin, Poland). Cell line cultures were maintained at 37 °C in a humidified atmosphere with 5% CO2.
Prior to the transplantation, cells were trypsinized (IIET, Poland), centrifuged (200 × g, 4 °C, 5 min) and counted. Then, cells were resuspended in Hank’s Balanced Salt Solution (HBSS; IIET, Poland).
Male BALB/c Nude mice were intraperitoneally injected with ketamine at a dose of 50 mg/kg (VET-AGRO Sp. z o.o., Poland) and anesthetized with the mixture of air and isoflurane (3% v/v). 1.0 cm wide abdominal wall incision was made just above the bladder, in the lower part of abdomen, and the prostate gland was exposed for the injection. Then, 5 × 106 PC-3M-luc2 cells in 0.05 ml of HBSS were inoculated into the dorsal prostate lobes of mice. Immediately after the transplantation, incised abdominal wall and skin were sewed with soluble surgical suture.
Estimation of the antitumor activity
The development of prostate tumors in TRAMP mice was monitored weekly by physical examination. Adenocarcinoma formation was confirmed by histological examination of the tumors isolated from mice during necropsy carried out in animals with clear physiological (e.g., body weight or body temperature decrease, body posture, ruffled fur) and behavioral symptoms (e.g., decreased movement) of an advanced disease. Briefly, prostate tumors were isolated and fixed in buffered formalin, and then cut into 4-μm-thick sections that were subsequently dewaxed with xylene. Following rehydration in a gradient of ethanol, the sections were washed in distilled water, cytoplasm was stained with eosin while nuclei were counterstained in hematoxylin. Finally, the preparations were dehydrated in an alcohol gradient and coverslip mounted. The histological appearance of the tissue was examined at 50× or 100× magnitude.
Using an In vivo MS FX PRO system (Carestream Health INC., USA), in vivo visualizations of PC-3 M-luc2 tumors growing in prostate gland of BALB/c Nude mice were performed no more often than every 4 days starting from the 15th day of the experiment. In brief, about 10 min before imaging, D-luciferin potassium salt (Synchem INC., Germany) was administered to each mouse intraperitoneally at a dose of 150 mg/kg. Then, animals were anesthetized with a 3–5% (v/v) mixture of isoflurane (Forane, Abbott Laboratories, USA) in synthetic air (200 ml/min). Anesthesia was maintained with 1.5–2% (v/v) mixture of isoflurane and synthetic air delivered via individual masks. Visualization was carried out using the following settings: for X-ray — t = 2 min, f-stop = 5.57, FOV = 198.6; for luminescence capture — t = 3 min, binning 2 × 2, f-stop = 5.57, FOV = 198.6. Images were analyzed with Carestream MI SE software (Carestream Health INC., USA). The intensity of the luminescent signal is presented as the sum intensity of the region of interest and expressed in arbitrary units (a.u.). Tumor tissue was also excised and weighted on the last day of the experiment (day 46).
Livers, lungs, kidneys, bones and axillary as well as inguinal lymph nodes were isolated and fixed in buffered formalin on the day of the necropsy, in order to detect metastases in the mice bearing prostate tumors. Then, tissue samples were cut into 4-μm-thick sections and stained as described hereinabove. The number of metastases in isolated tissues was counted at 50× or 400× magnitude.
Platelet activation status
Blood samples were collected on days 87, 122, 213 and during animal’s necropsy in the model of the spontaneously formed prostate tumors or on the last day of the experiment (day 46) in case of mice bearing PC-3M-luc2 tumors. Samples were collected in tubes containing 0.05 ml of 5% ethylenediaminetetraacetic acid (EDTA) solution (Sigma-Aldrich, Germany). Platelet-related morphology analyzes were performed using Mythic 18 analyzer (C2 Diagnostics, France). Then, blood plasma was obtained by centrifugation (2000 × g, 15 min, 4 °C) and stored at −80 °C until further analyzes. Prostacyclin generation in the treated mice was determined by the quantification of plasma 6-keto-prostaglandin F1α (6-keto-PGF1α) levels. Based on thromboxane B2 (TXB2), von Willebrand factor (vWF) and soluble P-selectin plasma concentrations, platelet activation status was estimated. Using commercial kits available from Cusabio Biotech Co. Ltd. (Wuhan, China), all analyzes were conducted via the ELISA technique. In addition, plasma concentration of transforming growth factor β1 (TGF-β1) was determined with ELISA kit from Boster Biological Technology (USA). All ELISA-based analyzes were conducted according to the manufacturer’s instructions.
Protein expression in tumor tissue
Protein expression in prostate tumor tissue was analyzed according to the standard Western blot procedure [
22]. In brief, using a FastPrep®-24 MP Bio device (Mp Biomedicals LLC., USA), samples of tumor tissue that were collected and immediately frozen on the last day of the experiments were homogenized in RIPA Buffer (Sigma-Aldrich, Germany) with the following settings: CP 24 × 2, 6 m/s, 40 s. According to the manufacturer’s protocol, protein content in all samples was analyzed using a Bio-Rad Protein Assay (Bio-Rad Laboratories Inc., USA). Samples containing 100 μg of protein were separated on the pre-cast 4–20% gradient gels (Bio-Rad Laboratories, Inc., USA) and transferred onto 0.45 μm polyvinylidene fluoride (PVDF) membranes (Merck Millipore, USA). Next, the membranes were probed with primary rabbit polyclonal anti-E-cadherin (1:1000), anti-N-cadherin (1:1000), anti-VEGFR-1 (1:200) antibodies (all from Proteintech Group, USA) or mouse anti-β-actin (1:1000, Sigma-Aldrich, Germany) antibody. Finally, according to the manufacturer’s instruction, the analyzed proteins were detected with IRDye® 800CW Goat anti-Rabbit IgG or IRDye® 680RD Donkey anti-Mouse IgG (both from LI-COR, USA). Blots were visualized in ODDYSEY® CLx Imager (LI-COR, USA) and analyzed with ImageJ Software as follows. The total E-cadherin cellular content comprising truncated and unprocessed E-cadherin (with a molecular weight of approximately 100 and 130 kDa, respectively) was calculated. Similarly, total N-cadherin cellular content comprising mature and unprocessed N-cadherin (with a molecular weight of approximately 70 and 100 kDa, respectively) was determined. Then, E-cadherin and N-cadherin contents were normalized to β-actin. Finally, E-cadherin to N-cadherin ratios in individual samples were calculated and presented as mean ± SD values.
Toxicity of the anticancer treatment
The toxicity of the proposed anticancer treatment strategy and its influence on the overall health condition were estimated based on body weight changes as well as morphological and biochemical blood analyzes. The body weight of experimental animals was measured thrice each week throughout the course of all studies.
Blood morphology was performed with Mythic 18 analyzer (C2 Diagnostics, France). Using reagents and procedures provided by the manufacturer, biochemical analyzes were performed in Cobas C 111 analyzer (Roche Diagnostics, Switzerland).
Statistical analysis
Data normality was estimated using the Shapiro-Wilk test with a predetermined value of p < 0.05. The Tukey-Kramer multiple comparison test for parametric data or the Kruskal–Wallis Test for non-parametric data was applied; p values lower than 0.05 were considered significant. All calculations were performed using GraphPad Prism 7 (GraphPad Software, Inc., USA) software.
Unless stated otherwise, all data presented on graphs correspond to mean ± SD values.
Discussion
1-MNA is an endogenous metabolite of NA that was previously shown to possess significant anti-inflammatory and anti-thrombotic activity [
17,
18]. 1-MNA is synthetized by nicotinamide N-methyltransferase (NMMT), an enzyme expressed primarily in liver cells where it participates in methylation of NA and other pyridine compounds [
24]. Concurrently, the expression of NMMT was reported in multiple types of cancer in which it was associated with tumor-promoting activity [
25‐
27] that could be further attributed to 1-MNA [
28]. On the contrary, some of the published reports show the beneficial correlation between NMMT expression and cancer survival [
29,
30]. Along the lines with such data in a recently published study, we have shown that exogenous 1-MNA does not enhance the growth of cancer cells neither in vitro nor in vivo but, in contrary, may possess antimetastatic activity, most likely resulting from its PGI
2-releasing capacity. We have also shown that 1,4-DMP, a structural analog of 1-MNA, possesses similar antimetastatic activity; however, both compounds seemed to have different mechanisms of action that ultimately resulted in platelet-dependent metastasis inhibition. Importantly, pyridine compounds, but particularly 1,4-DMP, when given in a combination with cyclophosphamide contributed to its anticancer activity enhancing both antitumor and antimetastatic activity of cytostatic drug [
21]. Referred studies were, however, carried out exclusively in the mouse model of breast cancer, and the report did not mention any activity of 1-MNA or 1,4-DMP in other types of malignant tumors or with different anticancer agents.
In the present work, we investigated the activity of both compounds in the model of TRAMP mice that spontaneously develop prostate tumors. Similar to our previously reported studies [
21], both compounds, when administrated alone to mice developing prostate tumors, revealed no significant anticancer activity; however, to some extent these compounds delayed the disease onset (Fig.
1a–c). Such a delayed disease onset might be attributed to the prostacyclin-dependent activity of the compounds, as prostacyclin was shown to inhibit lung tumor development in PPARγ-dependent mechanism [
31] that was also shown to be involved in tumor growth arrest in prostate tumors [
32,
33]. Lack of the significant tumor preventing activity of both pyridine compounds while being somehow disappointing in terms of the possible application of 1-MNA, and its analog in the prevention of prostate cancer, is important for their implementation in anticancer treatments, in general, as once again we demonstrated that neither 1-MNA nor 1,4-DMP promoted the growth of solid tumors. On the contrary, we have not observed any antimetastatic activity of neither of the studied compounds. In contrast, among 1-MNA-treated animals, we even observed a slight increase of metastases frequency (Fig.
1c and d). Such a surprising result might be the consequence of prostacyclin-related inhibition of natural killer cells [
34] that, in turn, was shown to stimulate prostate tumor metastasis [
35]. Another explanation of the observed limited antimetastatic activity of both 1-MNA and 1,4-DMP might be associated with previously reported relationship between thrombin generation and the growth and metastasis of prostate tumors in TRAMP mice [
36]. Possibly, thrombin as a potent coagulation and platelet activator that was proven to facilitate metastasis [
37] counteracts a possible antiplatelet activity of pyridine compounds in this model. Importantly, in this study, we noted that during the prostate tumor development, TGF-β1 plasma concentration in TRAMP mice increased (Fig.
1p), which remains consistent with the previous reports indicating the usefulness of this molecule as a prognostic factor in prostate tumors [
38].
When investigating the anticancer activity of therapeutic regimen including simultaneous use of docetaxel and 1,4-DMP (that seemed to be more potent in the model of prostate tumor as compared with 1-MNA) in the therapy of metastatic human prostate cancer PC-3M-luc2, we observed 60% enhancement of the antitumor activity of docetaxel given alone (Fig.
2c) and complete abolition of metastases formation (Table
3) in mice treated with docetaxel administrated with 1,4-DMP. These results confirm that 1,4-DMP may promote anticancer activity of various cytotoxic drugs. Beneficial therapy outcome was also reflected in the decreased plasma level of TGFβ-1, a molecule often acknowledged as a prognostic marker in prostate cancer (Fig.
2i).
TGF-β1 is commonly recognized as a molecule inducing epithelial-to-mesenchymal transition (EMT) in tumor-forming cells. EMT is a phenomenon in result of which non-invasive tumor cells of epithelial phenotype acquire mesenchymal properties and become able to migrate and invade distant tissues [
39]. Therefore, in our study, lower plasma concentration of TGF-β1, and by implication lower metastatic capacity, was associated with higher expression ratio of E-cadherin to N-cadherin (Fig.
2f), cell adhesion molecules commonly accepted as important markers of EMT in cancer cells, including those of prostate origin [
40]. Lower metastatic capacity of tumor-forming cells was additionally accompanied by the lower level of short E-cadherin fragments (40 kDa) observed in the tumor mass of mice lacking metastases. Such short intracellular protein fragments arise because of full-length E-cadherin cleavage resulting in the release into extracellular matrix and next to bloodstream of 80 kDa E-cadherin extracellular domain [
41]. Indeed, 80 kDa fragments identified in metastatic sites or serum were previously discussed as potential prostate cancer progression markers [
42,
43]. Accordingly, in our study, we observed that 40 kDa intracellular domain of E-cadherin was abundant in tumor mass isolated from mice bearing PC-3M-luc2 tumors diagnosed with metastases (Fig.
2g). Finally, the prominent efficacy of the combined treatment comprising the use of docetaxel and 1,4-DMP is additionally confirmed by the decreased expression of VEGFR-1 (Fig.
2h), another prognostic marker that has been previously linked to enhanced metastatic potential of prostate tumors [
44].
We have previously shown that observed enhanced antitumor and antimetastatic activity of cytotoxic drugs when in a combination with 1,4-DMP might be a result of anti-platelet activity of the latter compound [
21]. Platelets, in turn, contribute to metastases formation by several mechanisms as comprehensively reviewed in the literature [
45,
46]. To confirm that increased anticancer efficacy of docetaxel observed when cytotoxic drug was given with 1,4-DMP was associated with diminished platelets activity, we have analyzed morphological and biochemical parameters reflecting platelet activation status. In this regard, in mice receiving the studied combined treatment, we observed lowered values of mean platelet volume and PDW (Fig.
3a–d) that may suggest decreased platelet activity [
47,
48]. Additionally, in mice treated with docetaxel and 1,4-DMP, we also noted a marked reduction in plasma concentrations of TXB
2, vWF and soluble P-selectin (Fig.
3e–h), constituting biochemical markers, further confirming diminished platelet activity.
Interestingly, increased antitumor activity of docetaxel when administrated simultaneously with 1,4-DMP was accompanied by its reduced toxicity manifested in the decreased incidence of treatment-related deaths and improved liver function (Fig.
4). Although we are currently investigating these phenomena, our initial results indicate that the observed protective activity of 1,4-DMP may involve acetylcholinesterase and consequently histamine-dependent pathways. It seems possible that in response to the treatment with 1,4-DMP, the level of histamine is increased that, in turn, may prevent liver injury [
49]. This novel and unexpected feature of the 1,4-DMP treatment might not only be of a great value for possible improvement of side effects in patients undergoing chemotherapy, but may also allow to increase dosages in patients with drug-resistant tumors to induce desired response while maintaining acceptable treatment toxicity.