Anti-cancer effect of in vivo inhibition of nitric oxide synthase in a rat model of breast cancer

Increased expression of nitric oxide synthase (NOS) is associated with different cancers such as cervical, breast, lung, brain, and spinal cord. Inhibition of NOS activity has been suggested as potential tool to prevent breast cancer. The anti-tumor therapeutic effect of L-nitro arginine methyl ester (L-NAME), NOS inhibitor, using in vivo models is currently under investigation. We hypothesized that L-NAME will show an anti-tumor effect by delaying a progression of breast cancer via a modulation of cell death and proliferation, and angiogenesis. We used a novel model of anti-cancer treatment by the administration of L-NAME (30 mg/kg in a day, intraperitoneal) injected every third day for five weeks to rat model of 7,12-dimethylbenz[a]anthracene (DMBA)-induced breast tumor. Concentrations of nitrite anions, polyamines, malondialdehyde, NH4+ levels, and arginase activity in the blood were decreased in DMBA + L-NAME-treated rats compared with DMBA rats. The mortality rates, tumor number, weight, and volume, as well as the histopathological grade of breast cancer were also significantly reduced. In addition, L-NAME treatment showed a delay in tumor formation, and in body weight compared with rats administrated only with DMBA. In conclusion, our data show that L-NAME is a promising anti-cancer agent to treat breast cancer, which can lead to development of anti-tumor therapeutic tools in future.


Introduction
Breast cancer has a high cause of death among the female population worldwide. According to the World Health Organization, the age-standardized death rate due to breast cancer in Armenia is 24.44 (197.93 for all cancers) per 100,000 women; making it the 19th highest nation globally for breast cancer (2nd highest nation for all cancers) deaths (World Health Ranking 2020). Lung cancer is the most commonly diagnosed cancer (11.6% of the total cases) and the leading cause of cancer death (18.4% of the total cancer deaths), followed by female breast cancer (11.6%) and prostate cancer (7.1%) [1]. Therefore, considerable efforts are made to develop novel tools for the treatment [2]. The L-arginine amino acid is the substrate for cancerrelated enzymes, such as arginase, and nitric oxide synthase (NOS) [3]. The NOS isozymes, including neuronal, inducible, and endothelial NOS, turn L-arginine into nitric oxide (NO) and citrulline [4]. Previous data have shown that serum levels of NO in humans can be increased in direct correlation with breast cancer stage advancement and decreased post-chemotherapy [5]. Downregulation of NO production by NOS inhibition can be a key to therapeutic advantage in cancer prevention and treatment. NO is an important small signaling molecule that regulates cancer-related biological processes including angiogenesis, invasion, metastasis, immune response, and apoptosis [6,7]. At low levels, nitric oxide works as a mediator of muscle relaxation, neurotransmission, and cell proliferation [4,8,9]. The enhanced NO levels, induced by macrophages, can moderate anti-tumor activity [10], whereas chronic stimulation of NOS may promote several pathological processes, including cancer itself. A high quantity of NO synthesized by inducible NOS in macrophages has been shown to cause phosphorylation of p53 suggesting its stabilization [11].
Vascular endothelial growth factor (VEGF) generated by cancer cells activates the NO/cGMP (cyclic guanosine monophosphate) pathway to support vascular expansion [12]. Nitric oxide also induces invasion [13,14]. Amplified production of NO has been found in the blood of breast cancer patients and elevated NOS activity in invasive mammary tumors in contrast to healthy patients [7,15]. Literature data state that the production of nitrite/ nitrate and VEGF-C in MDA-MB-231 breast cancer cells was increased by treatment with the NO donor DETA NONO-ate. The NO synthase inhibitor NG-nitro-L-arginine methyl ester (L-NAME) prevented this increase. Nitro tyrosine levels were significantly correlated with VEGF-C immunoreactivity and lymph node metastasis. Survival curves showed that high nitrotyrosine levels were associated with reduced disease-free and overall survival [16].
The latter was the basis for studying the anti-cancer effect of L-NAME, considering that a decrease in NO levels could have a broad-spectrum effect, including an antiangiogenic effect with a quantitative decrease in VEGF. Abnormal upregulation of COX-2 and/or iNOS has been associated with the pathophysiology of certain types of human cancers as well as inflammatory disorders. Arginase expression and function in tumor-infiltrating myeloid cells are also controlled by COX-2 [17]. There is a strong correlation between COX-2 mRNA expression and VEGF-C expression or secretion levels in breast cancer cell lines and VEGF-C expression in breast cancer tissues [18]. Endogenous prostaglandin E2 (PGE2) resulting from cyclooxygenase (COX)-2 expression in a highly metastatic murine breast cancer cell line upregulates IFN-γ and LPSinduced NOS (iNOS) expression and NO production [19]. We can assume that the reduction of NO quantity and the regulation of arginase-NOS interaction allows influence the activity of COX-2, thus showing anti-inflammatory and anti-angiogenesis effects, inhibiting the activity of angiogenesis factors.
The L-NAME is one of the well-investigated inhibitors of NOS [20]. Till now, studies of the anti-cancer role of L-NAME were directed to promote the healthy functioning of either vascular endothelium or immune response in vitro [21]. For example, it was shown that L-NAME exhibited anti-cancer action in human colorectal cancer cell lines by suppressing the mRNA for matrix metalloproteinases 2 (MMP2) and upregulating mRNA for tissue inhibitors of MMP2 (TIMP2) expression [22]. In addition, it has been reported that L-NAME suppresses ovarian cancer (OVCA) cell proliferation [23]. A beneficial role for L-NAME in the treatment of pancreatic ductal adenocarcinoma has been also documented [24]. Collectively, these observations support the preclinical evaluation of L-NAME for the treatment of breast cancer.
Here, we explored regulatory interactions between arginase and NOS in the breast cancer experimental model. Earlier reports focused on the expression of arginase in murine or human primary cancer tissue as well as in malignant cell lines [14,25] and emphasized its potential role in the promotion of tumor growth via polyamine synthesis [10,26]. Recently rapid elevation of polyamines (spermine, spermidine, putrescine) quantity in blood serum and urine during malignant tumors in different organs is shown [27]. Polyamines are important partakers in cell proliferation and differentiation. High levels of polyamines in the blood and tumor microenvironment during cancer stimulate abnormal cell proliferation and inhibit the synthesis of adhesive molecules, which is observed in metastasis. Arginine metabolic pathway, as a source for the synthesis of polyamines, is considered a potential candidate for an effective target for anti-cancer therapeutic purposes. Reducing the level of polyamines by arginase activity inhibition can be effective in anti-breast cancer prevention and treatment.

Reagents
7,12-Dimethylbenz[a]anthracene (Sigma-Aldrich, Cas number 57-97-6), L-NAME (Sigma-Aldrich, CAS Number: 51298-62-5) and chemicals for histopathological Fig. 1 Experimental design. Rats were separated into 5 groups (8 rats in each group, 10 rats in the DMBA group) with group 1 served as a Control; Group 2 served as vehicle controls (injected with 0.25 ml saline) and group 4 was assigned as L-NAME-injected group. Breast cancer was induced in the groups 3 and 5 by administration of DMBA, with group 3 remained untreated (0.25 mL saline), and group 5 being treated by L-NAME. Intraperitoneal administration of L-NAME started 10 days after the DMBA injection and continued for 5 weeks. The animals were euthanized within 145 days after DMBA administration examination, determination of arginase activity, malondialdehyde (MDA), and nitrite ions quantities, and for thinlayer chromatography (TLC) analysis were purchased from Sigma-Aldrich. And Carl Roth GmbH+ .

Experimental protocol
Overall, 42 adult female Wistar rats (52-day old) weighing 100 ± 20 g were utilized (Fig. 1.). Rats were separated into 5 groups. Group 1 (n = 8) served as Control. Group 2 (n = 8) rats served as vehicle control (injected with 0.25 ml saline) and group 4 was assigned as L-NAME-injected group (30 mg/kg L-NAME dissolved in 0.25 ml saline) at a concentration of 3 -3.6 mg from 2 to 7th weeks, respectively. Breast cancer was induced in groups 3 (n = 10) and 5 (n = 8) by the administration of DMBA by intragastrical gavage (20 mg/ml in a single dose, dispersed in the mix of 0.5 ml olive oil and 0.5 ml saline) as described before [29][30][31] with group 5 being treated by L-NAME and group 3 treated with the same amount of saline without L-NAME. The DMBA-induced breast cancer model in the rat is used for the study of mammary carcinogenesis because it closely mimics human breast disease [32]. Intraperitoneal injections of L-NAME/or saline started 10 days after administration of DMBA and continued for 5 weeks. Each group was housed in a cage (3500 cm 2 ) in the ventilated room at 25 °C. Animals were kept at similar conditions at 12 h light/ dark cycle and ad libitum feed and water. The L-NAME dosage was chosen based on previously published data [20,33,34]. Daily changes in the visible health status of rats were monitored after tumor hatching. Special conditions were created (the food was soaked, placed in a more accessible place, the litter and bedding were changed daily) for rats in the DMBA and DMBA + L-NAME groups to accurately assess the effect of L-NAME. Health monitoring and tumor measurements were performed daily by laboratory personnel. Body weights were continuously monitored throughout each experiment. Treatment options or euthanasia for cases of tumor ulceration or metastasis into other organ systems were determined under the guidance of the veterinarian. The animals were humanely euthanized within 145 days after DMBA administration [35,36].

Tumor inhibition
The effect of L-NAME on DMBA-induced tumors was determined at 5, 8, 13, 16, and 20 weeks after DMBA administration. The occurrence of tumors and their numbers per rat were quantified at those time points. Nitrite ions, polyamines, NH 4 + , and MDA quantities, as well as arginase activity in the blood were also measured at week 5 (tumor development), 8 (end of the treatment), 13 (tumor stabilization), 16 (after treatment), and 20 (end of the study) post-DMBA injection. To determine the long-term efficacy of L-NAME, rats were monitored for 2 months after tumor formation and stabilization. The baseline time was established based on tumor formation in all rats in the DMBA group, which usually occurred around between week 12 and 13 after DMBA injection. In the study of the various components, based on the non-significant values between some weeks, not all the key weeks were presented in the charts for simplicity. In particular, in the case of arginase activity and nitrite ion determination, it was non-significant between the 13th and 16th weeks (16th-week values are not presented); in the case of polyamines, it was non-significant between the 8th and 13th weeks (not presented 13th-week values) and it was non-significant between the 16th and 20th weeks (16th week not shown), and the 16th-week values were not included in the charts when assessing rat weight change, MDA, and ammonia levels.

Blood sampling
Blood samples were collected from the lateral tail vein as described [38]. All biochemical parameters were measured in 1 ml of blood plasma.

Histology
About 20 weeks after the DMBA administration, palpable mammary tumors were removed and fixed in 10% buffered formalin for 24 h. Tumors were then embedded in paraffin blocks, sectioned at approximately 5 μm thickness, and stained with hematoxylin and eosin (H&E) (Sigma-Aldrich, Taufkirchen, Germany) [39,40]. Stained sections were visualized by light microscopy (BM-190/T/SP Trinocular, Boeco, Germany), and representative images were captured using a CCD camera (B-CAM10, Boeco, Germany). Tissue sections were examined by a pathologist at the YSU Department of Human & Animal Physiology who was blinded to the experimental design. Histological scores of breast cancer were evaluated quantitatively based on the Nottingham Histologic Score system [41] by investigators blinded to experimental groups. The score has given based on the amount of gland formation, the nuclear features, and the mitotic activity. Each of these features was scored from 1 to 3, and then the scores were added to create a final score ranging from 3 to 9. The final scores were used to determine the grade in the following order: Grade I-tumors have a total score of 3-5, Grade II-tumors have a total score of 6-7, and Grade III-tumors have a total score of 8-9. As stated in this section, a high histological score indicates profound pathological changes, while a low value indicates tissue morphology relatively more akin to normal.

Measurement of NO content
NO quantity was determined by nitrite anions, which were measured as shown before [42]. Briefly, 100μL Griess reactive was added to 100μL of each sample. The supernatants were added to the Cadmium pellet containing tubes that convert nitrate to nitrite and incubated at room temperature for 12 h. The samples were measured at 550 nm and normalized to the standard curve of NaNO 2 .

Measurements of blood urea content
Arginase activity (kat) in blood was determined by the diacetyl monoxime colorimetric method with modifications for blood [5].

Analysis of polyamines (putrescine, spermidine, and spermine) by TLC
The assay was performed with some modifications for blood [43]. Briefly, 50μL of dansylated extract (Dansylpolyamines (DP)) was added to the pre-absorbing part of silica gel TLCplates and developed for 2 h with chloroform/triethylamine (25:2 v/v) mobile phase. DP lanes were scraped and collected, dissolved in 2 mL ethyl acetate, and evaluated under 505 nm.

MDA assay
Evaluation of lipid peroxidation was assayed by MDA colorimetric measurement using the thiobarbituric acidmalondialdehyde protocol with modifications as described by Ohkawa et al. [44].

Determination of ammonia
Ammonia in blood plasma was measured with the indophenol direct method as described by Huizenga et al. [45].

Statistics
Results are presented as mean ± SEM. A p-value of less than 0.05 was considered significant. Data were analyzed using GraphPad Prism 8 software. The sample size in the DMBA group was n = 10 at week 5, n = 8 at week 8, n = 6 at week 13, and n = 4 at week 20. The DMBA + L-NAME group included n = 8 at weeks 5, 8, and 13, and n = 6 at week 20. Statistical significance was determined by (1) two-way ANOVA and Tukey's multiple comparisons test for nitrite anions, MDA, polyamines, NH 4 + quantity, arginase activity, and animals' weight, (2) by Log-rank (Mantel-Cox), and Gehan-Breslow-Wilcoxon tests for Kaplan-Meier survival curve, and (3) by a two-tailed unpaired T-test for tumor incidence, multiplicity, tumor volume (cm 3 ), tumor weight (g), and histopathological scoring.

Arginase inhibition, decrease of NO and polyamines quantity by L-NAME prevent tumor growth
Upregulation of nitrite ions levels in the blood of the rats in the DMBA group was detected at all key weeks compared to Control (p < 0.0001, Fig. 2a). At week 5, levels of nitrite ion decreased as result of L-NAME administration in the L-NAME and DMBA + L-NAME groups compared to the control group and the DMBA group (p < 0.0001, Fig. 2a). In the DMBA group, nitrite ions levels were increased by 31.2% (p < 0.0001) and 16.3% (p < 0.001) on the week 13 and 20, respectively, compared with week 8. Treatment with L-NAME in the DMBA + L-NAME group decreased blood nitrite ions at weeks 8 (32.65%), 13 (57.2%), and 20 (51.1%) post-DMBA injection compared with group treated only with DMBA (p < 0.0001, Fig. 2a). In the L-NAME control group, a 35% reduction in nitrite ions levels (p < 0.0001) was observed at week 8, at the end of the treatment phase (Fig. 2a). Interestingly, in the DMBA + L-NAME group, nitrite ion concentrations were resumed to levels observed in Control groups (Fig. 2a). At week 5, arginase activity was increased as a result of L-NAME treatment in both the L-NAME and DMBA + L-NAME groups vs. control groups. This increase was maintained between weeks 8 and 13 and decreased to normal levels at week 20 (p < 0.0001, Fig. 2b). In the DMBA group, blood arginase activity was doubled at 8, 13, and 20 weeks vs. Control group (p < 0.0001, Fig. 2b). A significant decrease in arginase activity was found in the DMBA + L-NAME group when compared to DMBA group by 32.1% and 45.9% at weeks 13 and 20, respectively (p < 0.0001) (Fig. 2b). We also found that L-NAME administration increases arginase activity in the blood of the L-NAME group vs. Control at week 8 (p < 0.0001) and 13 (p = 0.013). However, at week 20 of treatment, the arginase activity in both L-NAME and DMBA + L-NAME groups was resumed to control levels (Fig. 2b). We found that L-NAME induced arginase activity in rats in L-NAME and DMBA + L-NAME groups at weeks 8 and 13 vs. control group and decreased the enzyme activity compared to DMBA group in weeks 13 (p < 0.0001) and 20 (p < 0.0001).
Quantitative changes in putrescine and spermidine levels in the blood as result of L-NAME did not occur at week 5 in all study groups. An increase in spermine level in the rats from the DMBA group was observed, whereas DMBA + L-NAME group (p < 0.0001) showed a decrease in spermine (Fig. 3c). Our results showed that blood polyamines levels (nmol/ml blood) were increased in the DMBA group at weeks 8 and 20 post-DMBA injection vs. Control group (p < 0.0001) (Fig. 3). In the DMBA + L-NAME vs. DMBA group, blood putrescine levels were reduced by 6.1%, and 16.2% at weeks 8 (p < 0.001) and 20 (p < 0.0001) respectively (Fig. 3a). Spermidine content was reduced by 16.3% after 20 weeks (p < 0.0001), compared to the DMBA group (Fig. 3b). However, in the week 20, the level of spermidine in the DMBA + L-NAME group was decreased significantly and was only slightly different from the Control group (Fig. 3b). Decrease in levels of Spermine was 7.7% and 16.1% at weeks 8 and 20 after tumor induction in the DMBA + L-NAME group compared with the DMBA group (p < 0.0001). In the week 20, the quantity of polyamines in the DMBA + L-NAME group was decreased and reached the values of the Control group. We found that the treatment of rats with a 30 mg/kg/day dosage of L-NAME only did not change the concentration of polyamine in the blood of L-NAME-treated rats vs. Control and Saline-treated groups during throughout the entire experiment. We then measured the effect of L-NAME on the mortality rates, tumor number, and volume, as well as on histopathological changes.

L-NAME prolongs survival of DMBA-treated animals
During the experiments, 0 death was registered in the Control, Saline, and L-NAME groups. 6 deaths (60%) happened in the DMBA group, and 2 deaths (25%) happened in the DMBA + L-NAME group (Fig. 4a). As shown on the Kaplan-Meier survival curve, in DMBA + L-NAME group, the mortality rates were reduced by 50% (p < 0.05). It is noteworthy to mention that the first death in the DMBA + L-NAME group was registered much later than in the DMBA group. Deaths in the DMBA group were observed at 6, 8, 12, 13, 17, and 19 weeks, whereas in the DMBA + L-NAME group, death had occurred at weeks 13 and 19 after DMBA administration.
The mean weights of rats in the Control, Saline, and L-NAME groups were comparable and kept tumor-free throughout the study. Therefore, only the Control group was shown in the weight change assessment chart. At week 5, no change in rat weight was observed in all study groups. Rats in these groups had no mortality rate and ending weights at 20 weeks reached 240 g. In the DMBA group, the weights of the animals during weeks 8, 13, and 20 were lower vs. control groups by 17.4%, 15.2%, and 23.1%, respectively (p < 0.0001, Fig. 4b). In the DMBA + L-NAME group at the end of the treatment (week 8), the body weights of the animals were lower by 10% compared with the Control group (p < 0.0001). Importantly by weeks 13 and 20, it resumed to the Control group rats' body weights as a result of L-NAME treatment. In addition, treatment with L-NAME significantly decreased tumor incidence and multiplicity ( Fig. 5a and b). Animals were monitored daily for appearance of tumor. The earliest tumor was detected after 51 days (approximately at 7 weeks) in DMBA-treated group and after 79 days (~ 11 weeks) in the DMBA + L-NAME group (p = 0.039, Fig. 5a). The tumor incidence achieved 100% in the DMBAtreated group after 12 weeks compared to week 14 in the Fig. 2 The change in nitrite anions quantity (a) and arginase activity (b) in the rat blood of experimental groups at different key weeks after DMBA administration. In the DMBA group, n = 10 at 5th, n = 8 at 8th week, n = 6 at 13th, and n = 4 at 20th week, in the DMBA + L-NAME group n = 8 at 5th, 8th, and 13th weeks, n = 6 at 20th week. Control and Saline (as vehicle control) showed no differences; therefore, only Control group is shown. Significance (p < 0.05) was determined by two-way ANOVA and Tukey's multiple comparisons test. Data are represented as mean ± SEM. * p < 0.05, ** p < 0.01 , *** p < 0.001, **** p < 0.0001 1 3 DMBA + L-NAME-treated group (Fig. 5a). Our results show that tumor multiplicity was greater in the DMBA group vs. DMBA + L-NAME group (Fig. 5b). The quantitative examination of tumors during week 20 after DMBA administration indicated that in the DMBA + L-NAME group the number of tumors was reduced by 31% compared with the DMBA group (p = 0.024, Fig. 5b). Tumor volumes were threefold smaller in the DMBA + L-NAME group compared to the DMBA group (p < 0.0001, Fig. 5c), and tumor weights were also decreased by 42% in the DMBA + L-NAME group compared with the DMBA group (p < 0.0001, Fig. 5d).

NOS activity inhibition by L-NAME reduces oxidative stress and prevents the development of hyperammonemia
The risk of hyperammonemia due to inhibition of arginase activity and potential cell membrane damage by L-NAME in all experimental groups was evaluated in blood samples in all key weeks after cancer induction by DMBA. DMBA markedly increased blood MDA and NH 4 + quantity at all time points after DMBA injection in comparison with the Control group (p < 0.0001, Fig. 6a and b). No changes in ammonia levels were observed between L-NAME and DMBA + L-NAME groups at week 5. L-NAME administration brought to increase of ammonia quantity by 19% at weeks 8 and 13 in the L-NAME control group compared to the Control group (p < 0.0001), which withdraws at the 20th week. In the DMBA group, there was an acute increase in the quantity of ammonia in the blood, which decreased during the week 8, but gradually decreased in the following weeks. Blood ammonia levels were reduced by 13.3%, 31.2%, and 40.3% in the DMBA + L-NAME group vs. DMBA group at weeks 8 (p < 0.01), 13 (p < 0.0001) and 20 (p < 0.0001), respectively (Fig. 6a). Importantly, L-NAME stopped this growth in the DMBA + L-NAME group, reversing it to values similar to the Control group at weeks 13 and 20 even though the amount of ammonia was elevated in the DMBA group at the 20th week compared to the DMBA group at week 8 (p < 0.01). In the 5th week, as result of L-NAME treatment, there was an increase in the quantity of MDA compared to the control group, both in the L-NAME group and in the DMBA + L-NAME group. In the following weeks (post-treatment period), there was a decrease in the quantity of MDA in the DMBA + L-NAME group compared to the Fig. 3 The change in polyamine content (PUT (a), SPD (b), and SPM (c)) in the blood samples from all experimental groups at different time points post-DMBA administration. PUT-putrescine, SPDspermidine, and SPM -spermine. The levels of polyamines were unchanged in the Control, Saline (as vehicle control), and L-NAME groups; thus, for simplicity, only Control group is shown (b and c). Data are shown as mean ± SEM. * p < 0.05. ** p < 0.01, *** p < 0.001, * *** p < 0.0001 ▸ DMBA group. This may be due to a dysregulation in the NO biosynthesis during treatment, also a result of the effect of L-NAME, which was stabilized in the coming weeks. L-NAME did not bring to increase of MDA quantity during all 3 key weeks compared to the Control group, which was confirmed by the values of comparative analysis. Compared to the DMBA group, MDA levels in the DMBA + L-NAME group were reduced by 32.9%, 46.2%, and 55.7% at weeks 8, 13, and 20, respectively (p < 0.0001). During the periods of all 3 key weeks, MDA quantity in the DMBA + L-NAME group decreased compared to the DMBA group (p < 0.0001) and approached the levels of the Control group at weeks 13 and 20. Thus, L-NAME did not cause hyperammonemia and lipid peroxidation in the post-treatment period (Fig. 6).

Progression of breast cancer is abolished by L-NAME therapy
The histological alterations seen in the mammary gland samples at week 20 in DMBA and DMBA + L-NAME groups are summarized in Fig. 7. The histological score of breast cancer was calculated for every rat in the DMBA and in all groups at 8, 13, and 20 weeks after DMBA administration. We observed no differences between Control, Saline (vehicle control), and L-NAME groups, that is why only Control group is presented. Significance (p < 0.05) determined by Logrank (Mantel-Cox) (p = 0.049) and Gehan-Breslow-Wilcoxon (0.021) tests in a two-way ANOVA and Tukey's multiple comparisons test in b. Data are shown as mean ± SEM. * p < 0.0 5, ** p < 0.01, *** p < 0.001, **** p < 0.0001 DMBA + L-NAME groups, and then, the mean value was calculated for each group. The histological mean value in the DMBA group was 8-9, which is equivalent to Grade III (poor prognosis). In the DMBA + L-NAME group, the mean value was 6, which is equivalent to Grade II. As shown in Fig. 7, the treatment with L-NAME shifted the histological abnormalities from Grade III to Grade II (p = 0.001) (Fig. 7h). In spindle cell carcinoma (Fig. 7a), round spindle cells with pleomorphic nuclei were forming an uncertain pattern (black arrows). In invasive lobular carcinoma, the malignant cells formed single-file lines and showed vacuoles within the cytoplasm (Fig. 7b, black  arrows). In first-grade papillary carcinoma (Fig. 7c), many papillary ledges sustained by thin connective tissue kernels were shown (black arrows). In sarcomatoid carcinoma spindle cell of the malignant tumor is without visible epithelial separation (Fig. 7d). Invasive ductal carcinoma (IDC) is based on the appearance of determining the growth of malignant epithelial cells (intraductal proliferation, black arrows) into the neighbor stroma (Fig. 7e). After treatment by L-NAME, the following histopathological alterations in the DMBA + L-NAME group were detected: tubular carcinoma in 2 rats (33.3% in DMBA + L-NAME group), defined as Grade I good prognosis (Fig. 7f); pleomorphic lobular carcinoma (in 2 rats, Grade III, poor prognosis, (Fig. 7g); and invasive ductal carcinoma (in 2 rats, Grade II, poor prognosis, Fig. 7e). In tubular carcinoma, tumor cells had a low nuclear grade (black arrows), and the tumor cells form tubular structures with open lumens (Fig. 7f). Pleomorphic lobular breast cancer emerged in milk-generating glands (c), and tumor weight (g) (d) in the DMBA and DMBA + L-NAME groups. Tumor volume and quantity have been counted equivalent per one rat. The tumor incidence achieved 100% in the DMBAtreated group at 12 weeks in contrast to the DMBA + L-NAMEtreated group in which the tumor incidence achieved 100% in the week 14 (p = 0.039). Results have shown tumor multiplicity values were greater by 44.7% in the DMBA group compared with the DMBA + L-NAME treatment group (p = 0.024). Significance (p < 0.05) was determined by a two-tailed unpaired T-test. Data are represented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 (lobules) of the breast (Fig. 7g). It resembles the classical tumor in growth pattern but the tumor cells showed a more abundant cytoplasm, marked nuclear pleomorphism, prominent nucleoli, and increased mitotic rate (black arrows).

Discussion
The study aimed to show the influence of the regulation of unique interaction between NOS and arginase activity, thereby affecting the course of cancer and leading to its regression. Based on obtained data, the changes in nitrite ions quantity (hence NOS activity) and arginase activity were found in the blood plasma of rats of all experimental groups. We assumed that the interaction of these enzymes could affect the regulation of polyamine levels, which in turn should affect suppressing cancer progression. The increasing risks of cancer lead to the search for more effective sources to battle the disease. It is now clear that nitric oxide plays significant roles at different stages of tumor progression such as regulation of oncogenes and tumor suppressor genes, apoptosis, angiogenesis, and metastasis. However, the exact role of NO in tumor progression or prevention is not clear; therefore, the present research aimed to better characterize the effects of NOS inhibition by L-NAME on tumor multiplicity, growth and weight, animals' survival and weight, blood arginase activity, nitrite anions, MDA, NH4 + , polyamines quantity, and histological scores in a DMBA-induced rat model of breast cancer.
Based on the published evidence, inhibition of NO synthase with L-NAME induces an anti-tumor effect either by limiting the access of nutrients and oxygen to the tumor tissue or by affecting the vascular growth [46]. Inhibition of NOS has been previously reported to show anti-metastatic and anti-angiogenic effects in short-term treatments in the settings of laboratory manipulatable cancers [33,47,48]. The mechanisms acting during long-term treatment in chemical-induced cancer remained unknown and were investigated in our study. In addition, many reports are focused on the effects of L-NAME in various in vitro models or in vivo cell culture-induced tumors [33,46,[48][49][50]. In contrast, in our studies, we used in vivo rat tumor model induced by carcinogen. 7,12-Dimethylbenz[a]anthracene (DMBA) is a highly potent carcinogen that is activated by microsomal enzymes to a diol epoxide metabolite that binds covalently to DNA in mammalian cells, ultimately leading to tumor induction. DMBA is a hazardous component present in tobacco smoke and polluted environments [51,52]. In contrast, other study showed the induction of tumor by subcutaneous implants of growth factor-reduced C3L5 cells [47]. In other report, 9-week-old female BALB/c mice were inoculated in the fourth mammary gland using 2.5 × 10 5 LM3, LMM3, or LM2 cancer cells [47,53]. Using an injection of MDA-MB-231 or SUM159 cells (3 × 10 6 ) into the right mammary fat pad [33], Granados-Principal et al. (2015) induced tumors in female severe combined immunodeficiency (SCID) Beige mice, a model of immunodeficiency that affects B and T lymphocytes, and natural killer cells. It has been previously reported that C3L5 cells migration was reduced in the presence of L-NAME in a concentration-dependent manner, and restored in the presence of excess L-arginine, confirming a migration promoting role of endogenous NO [50]. The study showed that high endogenous iNOS expression was associated with a worse prognosis in patients with TNBC [33]. Selective iNOS (1400 W) and pan-NOS (L-NMMA and L-NAME) inhibitors diminished cell proliferation, cancer stem cell self-renewal, and cell migration in vitro, together with the inhibition of EMT transcription factors [33]. Inhibition of iNOS significantly reduced tumor growth, the number of lung metastases, tumor initiation, and self-renewal [33,49]. De Wilt et al. (2000) explored the anti-tumor effect of L-NAME after systemic administration in a renal subcapsular CC531 adenocarcinoma model in rats. In isolated limb perfusion, reduced tumor growth was observed when L-NAME was used alone [46]. A synergistic anti-tumor effect of L-NAME is observed in combination with melphalan (chemotherapeutical agent) and/or TNF using isolated limb perfusion. Inhibition of NO synthase reduced tumor growth after both systemic and regional treatment [46]. In another paper [48], the effects of NO deficiency on cancer development were also studied, where 4T1 cancer cells were administered orthotopically or intravenously to Balb/c mice. In a 4T1 murine metastatic breast cancer model, NO played a major role in primary tumor development, while NO was not the key mediator of cancer cell extravasation to the lungs. Furthermore, NO deficiency activated a PGI2dependent compensatory mechanism only in the intravenous model of 4T1 breast cancer. In terms of the novel, pharmacologically actionable targets, nitric oxide synthases have been Fig. 7 The histopathological alteration (a-g), and scoring (h) of mammary glands of the DMBA and DMBA + L-NAME groups' animals (at the 20th week after DMBA administration, 28-week-old rats, H&E × 400). Histopathological alteration in DMBA group rats has revealed spindle cell carcinoma (a), invasive lobular carcinoma (b), papillary carcinoma grade 1 (c), sarcomatoid carcinoma (d), and invasive ductal carcinoma (e). In the DMBA + L-NAME group, the histopathological examination has revealed tubular carcinoma (f), pleomorphic lobular breast cancer (g), and invasive ductal carcinoma (e) which has numerous intraductal (IDP) proliferations. His-tological scores of breast cancer were evaluated quantitatively based on the Nottingham Histologic Score system (h). The score has given based on the amount of gland formation, the nuclear features, and the mitotic activity. Each of these features has scored from 1-3, and then the scores have been added to give a final total score ranging from 3-9. The final total score has used to determine the grade in the following way: Grade I-tumors have a total score of 3-5, Grade IItumors have a total score of 6-7, and Grade III-tumors have a total score of 8-9. Significance (p < 0.05) was determined by a two-tailed unpaired T-test (h). Data are represented as mean ± SEM. **p < 0.01 implicated in the etiology of KRAS-driven cancers, including lung cancer; therefore, small molecular weight NOS inhibitors have been developed for the treatment of these diseases [4,24]. Various studies evaluated the anti-neoplastic activity of the oral NOS inhibitor L-NAME in a randomized preclinical trial using a genetically engineered mouse model of Kras and p53 mutation-positive non-small-cell lung cancer (NSCLC). Based on assessment of sequential radiological imaging, it was shown that L-NAME decreased lung tumor growth in vivo and provided a survival advantage, perhaps the most difficult clinical parameter to improve upon [4]. These results indicate a promising role of L-NAME in the treatment of solid tumors in a systemic setting, which was also performed in our study. These observations served as the basis for the investigation of the anti-cancer effect of L-NAME in DMBA-induced breast cancer. It is important to point out that in our rat model, carcinogenesis has been induced with DMBA, a chemical that is abundant in the environment due to improper human activity and can be one of the main reasons for the high prevalence of mammary cancer worldwide. Therefore, the use of our model can be quite indicative as it mimics the development of many clinical cases.
Our results suggest unique relationships between arginase and NOS in the L-arginine pathway during breast carcinogenesis and treatment (Figs. 2 and 3). This intercorrelation is related to the NOHA (N G -hydroxy-L-Arginine), the intermediate compound in nitric oxide biosynthesis, which is a strong inhibitor for arginase isoenzymes [7,13]. Based on our results, it should be noted that the increase of arginase activity in the L-NAME group after the injection of NOS inhibitor at weeks 5 and 8 could be a reason for low NOHA quantity. The reason that treatment with the L-NAME inhibited arginase activity in the DMBA + L-NAME group at weeks 13 and 20 may be an improvement in the pathological condition, which in turn affects the change in the activity of these enzymes (Fig. 2a). The results suggest that the decrease in polyamines is not due to the direct effect of L-NAME, but rather the result of the decrease in arginase activity (Fig. 2b, and Fig. 3c). Thus, after the 8 weeks, L-NAME had an effective inhibitory effect on the levels of NO and polyamines. Our results once again confirm the potential of spermine for diagnostic purposes and its potential influences at different stages of cancer development. The reasons for maintaining the levels of spermidine during the week 8 in the DMBA group remain to be elucidated. However, we can assume that in the PUT/SPD/SPM pathway, spermidine has intermediate importance and at the end of the treatment, there are still active mechanisms that elevate polyamines during cancer. Particularly, spermine oxidase, polyamine oxidase, and spermidine synthase enzymes are still active and allow to maintain enough quantity of polyamine influence on the lipid metabolism, cell signaling, cell cycle, and growth regulation processes [54].
Our studies show long-term anti-tumor effects of L-NAME. As we mentioned in the Methods section, cotreatment by L-NAME was performed in parallel to cancer progression. Both the appearance of tumors and animal survival were delayed in L-NAME-treated group (Fig. 4a  and Fig. 5a). These results are highlighting the preventive effect of L-NAME as well. Although the samples from the DMBA + L-NAME group have histological mean values of Grade II, two out of six survived rats (20 weeks after DMBA injection) in that group still have Grade III value. The latter forces the therapeutic model to be modified in terms of dosage and combination with natural synthetic anti-cancer compounds to obtain a more effective therapeutic model in future. In any case, our data allow us to conclude that the inhibition of NOS activity affects the L-arginine pathway, and particularly nitric oxide and polyamines biosynthesis, a phenomenon that has an inhibiting effect on cancer progression.
Several published studies examined the properties of various cancer-related components and their quantitative changes in the tissue, mainly in the microenvironment of the tumor. Meanwhile, in our studies, we explored the changes of metabolites in the blood that can have diagnostic value. Taking this into account, we performed the treatment with L-NAME in parallel to the development of breast cancer. We studied the inhibitory effect of L-NAME on mortality, tumor incidence, multiplicity, tumor volume (cm 3 ), and tumor weight as well (Figs. 4 and 5). Moreover, in this study, we observed the crosstalk between NOS and arginase and their influence on the quantitative changes of polyamines, which are important factors during metastasis and progression of cancer. We assume that regulation of the polyamine's levels via L-NAME administration could represent one of the potential mechanisms of its anti-cancer activity. Most of the studies investigating the anti-cancer effect of L-NAME are designed for a short period, lasting from a few hours to 2 weeks [53]. Long-term, about 5-week period, studies are a few and served as a basis for this work [33]. This study illustrates the cancer-preventive role of NOS activity inhibition and its long-term therapeutic effect between weeks 5 and 20th weeks. In other similar studies, L-NAME was administered either orally or by intratumorally injection [4,53]. In contrast, we observed long-termed treatment (5 weeks) conducted through intraperitoneal injection which allowed every animal to receive a precise dose of the inhibitor. We also evaluated the possible side effects of L-NAME during longterm treatment by lipid peroxidation and hyperammonemia assessment that is not performed in many similar studies. This can be used to evaluate the differences in the effect of L-NAME in different animal species, which requires further meta-analysis. Our research work includes detailed histopathological analysis, which is missing in most other studies.

Conclusions
Thus, we showed that inhibition of NOS activity by L-NAME influences the L-arginine pathway, specifically nitric oxide and polyamines biosynthesis, which have an inhibiting effect on cancer progression. We measured several parameters of potential side effects of L-NAME treatment including its effect on lipid peroxidation and hyperammonemia and observed no significant changes. Perspectives of targeting NOS in cancer diseases can serve as a base to reveal new inhibiting models using experimental and computer modeling approaches. Collectively, these results and analysis support the clinical evaluation of L-NAME for the prevention and treatment of early-stage breast cancer.

Limitations of the study
Our study has several limitations which will be addressed in our further research: • Only one dose of L-NAME was used in the study, • The effect of L -NAME on angiogenesis and during the metastasis has not been studied (VEGF, MMP2 and histological examinations of other tissues).