Manganese Porphyrin and Radiotherapy Improves Local Tumor Response and Overall Survival in Orthotopic Murine Mammary Carcinoma Models

Abstract

Novel synthetic compounds, known as manganese porphyrins (MnPs), have been designed to shift the redox status of both normal cells and cancer cells. When MnPs are coupled with cancer therapies, such as radiation, they have been shown to sensitize tumor cells to treatment and protect normal tissues from damage through the modulation of the redox status of various tissue types. Until now, our preclinical studies have focused on local effects of MnPs and radiation; however, we recognize that successful outcomes for cancer patients involve control of tumor cells throughout the body. In this study, using murine orthotopic mammary tumor models, we investigated how MnPs and radiation influence the development of distant metastasis. We hypothesized that the combination of MnP (MnP/RT), such as MnTnBuOE-2-PyP⁵⁺ and radiation treatment (RT) would increase local tumor control via a shift in the intratumoral redox environment, leading to subsequent downregulation of HIF1 in the primary tumor. Secondarily, we hypothesized that these primary tumor treatment effects would result in a reduction in pulmonary metastatic burden. Balb/c mice with orthotopic 4T1 mammary carcinomas were treated with saline, MnP, RT or MnP/RT. We found MnP/RT did extend local tumor growth delay and overall survival compared to controls and was associated with increased intratumoral oxidative stress. However, the primary tumor growth delay observed with MnP/RT was not associated with a reduced pulmonary metastatic burden. Future directions to investigate the effects of MnP/RT on the development of distant metastasis may include modifications to the radiation dose, the experimental timeline or using a murine mammary carcinoma cell line with a less aggressive metastatic behavior. Clinical trials are underway to investigate the clinical utility of MnTnBuOE-2-PyP⁵⁺ for patients undergoing radiotherapy for various tumor types. The promising preclinical data from this study, as well as others, provides support that MnP/RT has the potential to improve local tumor control for these patients.

INTRODUCTION

Radiation kills cells through the generation of highly reactive species such as superoxide, hydroxyl radical and hydrogen peroxide. These species damage critical cellular structures, such as DNA or cell membranes (1, 2). Advanced radiation therapy techniques allow for precise targeting of tumor tissue; however, normal regional tissue is commonly exposed and damaged by radiation as well, which can result in significant, and sometimes dose-limiting, acute and chronic side effects (3, 4). Attempts have been made using adjuvant agents, such as amifostine, to scavenge the reactive species in normal tissues to reduce radiation injury (5); however, it is possible that such compounds not only protect the normal tissues, but also protect the tumor cells.

With this in mind, ideal therapeutic approaches for treating cancer should balance the damaging oxidative stress to the cancer cells with antioxidant protection to normal tissues. An example of an endogenous antioxidant is superoxide dismutase (SOD). SOD is an enzyme that catalyzes the dismutation of the free radical, superoxide, into molecular oxygen and hydrogen peroxide (6). Hydrogen peroxide (H₂O₂), a product of this reaction, is also reactive and damaging to cells; however, under normal conditions, it can be removed quickly and efficiently by endogenous antioxidant defenses, such as catalase, glutathione, peroxidase and peroxiredoxins (6).

Novel synthetic compounds, known as manganese porphyrins (MnPs), have been designed to shift the redox status of both normal cells and cancer cells (7–10). MnPs are superoxide dismutase (SOD) mimics, which, like endogenous SOD, catalyze the dismutation of superoxide into molecular oxygen and hydrogen peroxide. However, it has been shown that they are also capable of undergoing a variety of additional reactions; among those, oxidation of protein cysteines appears to be a major mechanism by which MnPs affect the cellular signaling processes (9). Therefore, instead of classification as simply antioxidants, MnPs may be more appropriately described as modulators of the cellular redox environment (7, 9). When MnPs are coupled with cancer therapies, such as radiation, they have been shown to sensitize tumor cells to treatment and protect normal tissues from damage through the modulation of the redox status of these very different tissue types (3, 4, 11, 12).

In addition to the immediate free radical generation created by ionizing radiation, our group has shown that radiation results in tumor reoxygenation over several hours after treatment. The reoxygenation is tied to additional tumor microenvironmental oxidative stress (13). This oxidative stress leads to overexpression of the tumor promoting transcription factor, hypoxia-inducible factor-1 (HIF-1), via inhibition of the activity of prolyl hydroxylase, which leads to protection of the tumor vasculature, and ultimately causes radioresistance (13). MnPs have been shown to catalytically inactivate this radiation-induced oxidative stress, downregulate HIF-1 as prolyl hydroxylase activity is restored, sensitize the tumor vasculature to radiation, and improve radiation response in several tumor types (12–14). The addition of MnPs to radiotherapy has also been shown to induce high levels of oxidation in cancer cells while simultaneously creating a reducing environment in normal cells, leading to a wider therapeutic index (7, 8, 15).

Until now, our preclinical studies have focused on local effects of MnPs and radiation; however, we recognize that successful outcomes for cancer patients involve control of tumor cells throughout the body. Shifts in the cellular redox environment and HIF-1 overexpression have been associated with increased tumor aggressiveness, leading to local tumor progression and the development of metastatic disease (16–19). Relevant to this project, we investigate the tumor redox environment, intratumoral HIF-1α expression, and the local and distant effects of orthotopic murine mammary carcinomas treated with MnPs and radiation. As prior studies investigating redox modulating MnPs have demonstrated both antioxidant and pro-oxidant activity within tissues, we aimed to define differences in the redox environment, the associated HIF-1α levels within, and the therapeutic outcome of mice treated in these experiments. Reactive oxygen species are known to stabilize HIF-1α via inhibition of prolyl hydroxylase enzyme activity (20–22). The heterodimer HIF-1, comprised of HIF-1α and HIF-1β subunits, is an established promoter of tumor aggressiveness and metastatic propensity (23). Clinically, HIF-1 is highly expressed in primary tumors and metastases in patients with various tumor types (24–27), and increased intratumoral HIF-1 expression is often linked with increased patient mortality [reviewed in (28)]. In experimental models, overexpression of HIF-1 in tumor cells promotes metastasis (29, 30), whereas inactivation of HIF-1 decreases the metastatic potential of tumor cells (19, 30–33). HIF-1 signaling regulates multiple steps within the metastatic cascade, including migration/invasion, intravasation/extravasation, establishing the premetastatic niche, and promoting distant growth at secondary site(s) [reviewed in (23, 34, 35)]. This metastatic cascade is complex and occurs over time. For these reasons, in our experiments, we evaluated the tumor redox environment, intratumoral HIF-1α expression, and the local and distant effects of MnPs and radiation in mice with orthotopic mammary carcinoma at various time points.

In this study, we aimed to investigate how MnPs and radiation influence the development of distant metastasis using murine orthotopic mammary tumor models. Ultimately, we hypothesized that the combination of radiation and the MnP, MnTnBuOE-2-PyP⁵⁺, would increase local tumor control via a decreased, antioxidant shift in the intratumoral redox environment, leading to subsequent downregulation of HIF-1 in the primary tumor. Secondarily, we hypothesized that these primary tumor treatment effects would result in a reduction in pulmonary metastatic burden.

MATERIALS AND METHODS

Cell Lines

4T1 cells were obtained from the American Type Culture Collection (ATCC®, Gaithersburg, MD) and cultured in RPMI supplemented with 10% fetal bovine serum (FBS), 10 mM HEPES, 1 mM sodium pyruvate, 100 I.U./ml penicillin and 100 μg/ml streptomycin, and incubated at 37°C in 5% CO₂. Cells were passed 2–3 times a week at a subcultivation ratio of 1:6 to 1:8, as recommended by ATCC. A modified 4T1 murine mammary carcinoma cell line was also used for experiments in this study (a gift from Dr. Michael Wendt, Case Western Reserve University, Cleveland, OH). This stably transfected cell line constitutively expresses firefly luciferase-2 under a CMV promoter with Zeocin resistance (500 μg/ml) for selection in vitro. These 4T1-luciferase-expressing cells (4T1-Luc) were cultured in Dulbecco’s modified eagle medium (DMEM, high glucose with sodium pyruvate and glutamine) with 10% FBS and Zeocin (500 μ/ml) and incubated at 37°C in 5% CO₂. Cells were passaged 1:10 when 75–90% confluent. These cell lines were routinely tested and confirmed to be free from Mycoplasma contamination. While we used this luciferase-expressing cell line for experiments with the intention of quantifying the metastatic burden via bioluminescent imaging, we recognized inconsistency in the luminescent signal produced by the cells in vivo. Because of this, we questioned the utility and reliability of the luminescence as a quantitative biomarker for our experiments, so the luciferase-expression property of the cell line was not utilized in this study.

Animal Husbandry

Six-to-eight-week-old female, Balb/c (Jackson Laboratory, Bar Harbor, ME) mice were used for all experiments. Five animals per cage were housed in standard mouse cages in the animal facilities at Duke University (Durham, NC) and at the University of Nebraska Medical Center (UNMC; Omaha, NE). Mice were exposed to a 12:12 h light-dark schedule, fed and watered ad libitum. All experimental protocols were reviewed and approved by the Duke University or UNMC Institutional Animal Care and Use Committees.

Tumor Implantation and Monitoring

Surgical tumor implantation.

4T1-Luc cells were surgically implanted into the dorsal mammary fat pads of mice. Mice were anesthetized using intraperitoneal injections of ketamine (85 mg/kg) and xylazine (8.5 mg/kg), fur overlying the dorsal thorax was removed using a chemical depilatory (Nair™, potassium thioglycolate), and the surgical field was prepared using three alternating applications of chlorhexidine gluconate scrub solution and 70% ethanol. A small skin incision was made overlying the dorsal mammary fat pad. The fat pad was visualized, and approximately 5 × 10⁵ 4T1-Luc cells suspended in 100 μl of serum-free DMEM were injected into the site. The incision was closed using 4–0 silk suture and a single cruciate suture pattern. Triple antibiotic was applied to the incision, and buprenorphine (0.1 mg/kg) was administered subcutaneously (s.c.) as analgesia for the recovering mice.

Direct injection tumor implantation.

4T1 breast cancer cells (3.65 × 10⁵ in 100 μl of Hank’s Balanced Salt Solution) were injected s.c. into the left 4th mammary fad pads of mice while mice were anesthetized with isoflurane.

Tumor monitoring.

Mice were examined daily after tumor cell implantation. Once mice developed palpable tumors, the tumors were measured daily with calipers, and volumes were calculated using the formula V = (A² xB x π)/6, where A is the shortest diameter and B is the longest diameter.

Manganese Porphyrin Preparation and Administration

The chemical structure of MnTnBuOE-2-PyP⁵⁺ is shown in Fig. 1. For consistency in discussions, MnTnBuOE-2-PyP⁵⁺ will be referred to as MnP, the general abbreviation for manganese porphyrin compounds. MnP was synthesized and purified according to procedures described elsewhere (36, 37). A loading dose of 0.2 mg/kg MnP was administered s.c. in the left hind legs 24 h prior to irradiation. Then, a dose of 0.1 mg/kg MnP was administered in the same manner three times a week beginning at least 48 h after the loading dose until sacrifice. This dosing protocol was selected as it was determined to yield pharmacologically-active concentrations in tissues (3, 38).

FIG. 1.

Chemical structure of the studied compound.

Irradiations

Mice with tumors directly injected into the 4th mammary fat pad were anesthetized with ketamine 100 mg/kg:xylazine 5 mg/kg and then placed under lead shielding so that their tumor received 15 Gy X-ray irradiation at ~1 Gy/min, using an X-ray box irradiator (Rad Source RS-2000; Buford, GA).

Mice with surgically implanted tumors in the dorsal mammary fat pad were anesthetized with 1.5% isoflurane gas mixed with oxygen and placed in an X-RAD 225Cx (Precision X-ray Inc., North Branford, CT) small animal micro-CT irradiator. A collimating cone that produced a 10 × 10 mm radiation field was used to target the radiation beam to the dorsal mammary tumors. The field was rotated 45 degrees to create a diamond shape over the irradiated area (Supplementary Fig. S1; https://doi.org/10.1667/RADE-20-00109.1.S1). The cranial and caudal aspects of the tumors were identified using small barium markers, which had been applied to the tumors of anesthetized mice. Fluoroscopy at 40 kVp and 2.5 mA with a 2-mm aluminum filter allowed proper alignment of the radiation field. Mice were treated with one fraction of 15 Gy administered as two exposures of 7.5 Gy from parallel-opposed fields with a dose rate of 257 cGy/min at target depth with 225 keV and 13 mA and a 0.3-mm copper filter. Control mice were anesthetized, but not irradiated.

Experimental End Points

Treatment results were evaluated using three different end points (Fig. 2). For the acute effects end point, mice were euthanized 24 h postirradiation. For the tumor progression extended end point, mice were euthanized when the tumor reached 1,500 mm³, mice developed neurologic deficits due to tumor invasion of the spine, or mice lost greater than 20% of starting body weight. Finally, a fixed end point was used to directly compare treatment effects with a controlled duration of time with respect to tumor growth and the development of pulmonary metastasis. At each time point, mice were injected with Euthasol® (pentobarbital sodium and phenytoin sodium, 0.5 μl per mouse) to euthanize the mice.

FIG. 2.

Experimental design for orthotopic mammary tumor experiments. Panel A shows the acute and delayed effects of MnP/RT on local tumor response and distant metastasis were investigated. Mice received a loading dose of MnP (0.2 mg/kg) or saline, followed by primary tumor irradiation 24 h later. MnP or saline was administered at a maintenance dose (0.1mg/kg) every Monday, Wednesday and Friday until completion of the study. For acute effects, mice were euthanized and tumors harvested 24 h after irradiation. For the extended study, mice reached their end point when tumors were >1,500 mm³, they developed neurologic deficits or when they demonstrated >20% body weight loss. Panel B shows the design to reduce the influence of time for the evaluation of treatment effects for local and distant metastatic disease. Irradiation was performed 11 days after tumor implantation. All mice were euthanized and tissues harvested 11 days after the start of treatment.

Tissue Collection

After injection of euthanasia agent, but prior to death, 50 μl of fresh blood was collected from an incision made in the region of the large vessels in the lateral necks of the mice. The blood was added to GSH-stabilization buffer for GSH/GSSG analysis [20 mM N-ethylmaleimide (NEM; Sigma-Aldrich® LLC, St. Louis, MO), 2% 5-sulfosalicilic acid (SSA) and 2 mM EDTA in 15% ethanol]. Blood was mixed gently and kept at room temperature for 45 min. Samples were centrifuged at 2,000g for 2 to 5 min, and the supernatants and blood cells were stored separately at −80°C. Tumors were resected after euthanasia, flash frozen in liquid nitrogen and stored at −80°C. Lungs were resected and gently perfused with 1 ml of phosphate buffered saline (PBS) followed by 1 ml of 4% paraformaldehyde (PFA) in PBS via the trachea. The lungs were placed in 5 ml of 4% PFA and stored at 4°C for 48 h. The lungs were transferred to 5 ml of PBS and stored at 4°C for 48 h, and stored in 5 ml of 70% ethanol at 4°C until processed for histology.

Lung Metastasis Quantification

Gross metastatic lesions were visualized and counted manually using a Leica StereoZoom® 4 (Lincolnshire, IL) microsurgery magnification system. For histologic quantification of metastasis, paraffin embedded inflated whole lung tissues were sectioned all the way through the block and 5 sections (7 microns in thickness) were taken for each lung. Sections were then stained with hematoxylin and eosin, and each lung section was blindly evaluated for lung metastatic lesions by a pathologist (GT). The total number of lesions per mouse lungs was enumerated for each animal.

8-Hydroxydeoxyguanosine (8-OHdG) and 4-Hydroxynonenal (4-HNE) Immunostaining

Fixed tumor tissues were paraffin embedded and sectioned by the Tissue Science Facility at UNMC. Sections were immunostained for markers of oxidative stress, 4-hydroxynonenal (4-HNE) and 8-hydroxydeoxyguanosine (8-OHdG). Tissues were de-paraffinized in xylenes and rehydrated through graded alcohols. For antigen retrieval, slides were heated to 95°C in 10 mM sodium citrate buffer (pH 6.0) with 0.05% Tween® 20. Slides were then allowed to cool in phosphate buffer (pH 7.0) for 30 min. For blocking, an M.O.M.™ kit (Vector® Laboratories, Burlingame, CA) was used according to the manufacturer’s directions. After blocking, tissue sections were incubated with a primary antibody (4-HNE, 1:50; R&D Systems™, Minneapolis, MN; and 8-OHdG, 1:100, Abcam®, Cambridge, MA) overnight at 4°C in a humidified chamber. The following day, slides were washed in Super Sensitive Wash Buffer (BioGenex Laboratories, Fremont, CA) and stained for 1 h with a secondary antibody conjugated to Alexa Fluor® 555 (1:200, goat anti-mouse; Invitrogen™, Carlsbad, CA). Slides were mounted under coverslips with ProLong™ Gold Antifade with DAPI (Invitrogen). Sections were imaged using a Leica DM 4000B LED fluorescent microscope, followed by analysis using ImageJ software (National Institutes of Health, Bethesda, MD). The background staining was removed with thresholding and positively staining pixels were selected as regions of interest. Mean fluorescent intensity in 6–10 images per animal were averaged.

GSH/GSSG in Whole Blood and Tumor Tissue by LC-MS/MS

The ratio of reduced to oxidized glutathione (GSH/GSSG) was used as a measurement of the cellular redox environment within the blood and treated tumors. Mice were euthanized and samples were obtained either at 24 h postirradiation or sham treatment to evaluate acute effects or 24 h after the last dose of MnP in the extended end point experiment. The assay for tissue samples was based on the assay for analysis of GSH and GSSG in blood published elsewhere (9, 39).

Partially-thawed blood (50 μl) or tumor tissue (~20 mg) was homogenized immediately in 500 μl conical polypropylene vials containing the following: 4 parts (g/vol) of GSH-trapping solution [20 mM N-ethylmaleimide, 2% sulphosalicylic acid (SSA) and 2 mM EDTA in 15% methanol], 10 parts (g/vol) of 1% formic acid (Fluka) and two 2.5 mm Zr-silica beads (BioSpec Products Inc., Bartlesville, OK) in a FastPrep™ (Thermo-Savant™, Waltham, MA) homogenizer at speed 5 for 40 s. After 30 min incubation at room temperature, 150 μl of the homogenate was transferred into a 2-ml conical polypropylene vial, 500 μl of chloroform (B&D) was added, agitated in the FastPrep at speed 5 for 40 s, followed by centrifugation at 16,000g for 5 min. For the GSH analysis, 5 μl of supernatant, 1 ml of deionized water, and 100 μl of 1 μM GSH-NEM-d3 (internal standard) were combined and placed into the autosampler at 4°C. For the GSSG analysis, 20 μl of supernatant and 20 μl of 1 μM GSSG-d6 (internal standard) in mobile phase A (see below) were combined and placed into the autosampler at 4°C.

The LC-MS/MS analysis was performed on Agilent 1100/1200 series liquid chromatography (LC) (Santa Clara, CA) and Applied Biosystems®/SCIEX™ API-5500 QTRAP® tandem-mass spectrometer (MS/MS) (Carlsbad, CA) at Duke Cancer Institute PK/PD Core Laboratory. LC conditions: Column Agilent ZORBAX Eclipse Plus, C18 4.6 × 50 mm 1.8 μm particle size (P/N 959941–902) analytical column and Phenomenex, C18 4 × 3 mm guard cartridge (P/N AJ0–4287) at 45°C. Mobile phase solvents (all MS-grade) were: A. Formic acid (0.1%) in water, 2% acetonitrile; and B. acetonitrile. Elution gradient at 1 ml/min: 0–3 min 0–50% B, 3–3.1 min 50–95% B, 3.1–3.5 min 95% B, 3.5–3.6 min 95–0% B. Run time: 7 min. MS/MS conditions: MRM transitions (m/z) followed for GSH-NEM, GSH-NEM-d3, GSSG and GSSG-d6 were 433–304, 436.2–307.1, 613.1–355 and 619.1–361, respectively. Calibration samples (n = 6) in mobile phase A were prepared in the following appropriate ranges: 0.375–6.00 mM for GSH and 1.88–30 μM for GSSG analysis and analyzed alongside the study samples. Quantification was performed using Analyst® 1.6.2 software (SCIEX).

HIF-1α ELISA

An HIF-1α enzyme linked immunosorbent assay (ELISA; R&D Systems) was used to detect HIF-1α levels in tumor samples. Tumor lysates were prepared following the protocol recommended by the manufacturer. Briefly, approximately 50 mg of tumor tissue was solubilized in Lysis Buffer no. 11 and incubated on ice for 15 min. Samples were centrifuged at 2,000g for 5 min, and supernatants transferred to clean tubes for use in the ELISA kit. Plates were scanned at 450 nm with wavelength correction at 540 nm using a SpectraMax® M3 plate reader (Molecular Devices, Sunnyvale, CA).

Statistical Analysis

Statistical analyses were performed using GraphPad Prism 6 Software version 6.0.5 or 8.0.1 for Windows (La Jolla, CA). Five to eight animals were included in each experimental group for all experiments. Data are expressed as the mean ± standard error of the mean (SEM). The statistical significance between different groups was evaluated using one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test for multiple comparisons and a P value ≤ 0.05 was considered statistically significant. Additionally, statistical evaluation of results was conducted using two-way ANOVA, which verified the one-way analysis (data not shown).

RESULTS

Influence of MnP and Radiation on Local and Distant Tumor Control

The variation in local and distant tumor control was evaluated across the treatment groups with a tumor growth delay study in mice with surgically implanted tumors. After implantation of tumor cells, tumors reached treatment size (minimum 100 mm³, longest diameter <12 mm) in nine days. In this experiment, the combination of MnP and radiation (MnP/RT) slowed local tumor progression and increased survival time twofold compared to untreated controls (P = 0.032) (Fig. 3). As mice treated with MnP or radiation lived significantly longer than mice in the untreated control group, this provided additional time for pulmonary metastasis to develop. When considering the metastatic burden measured per mouse as a function of time postirradiation (Fig. 4B), a multiple regression model revealed that there was no difference across the treatment groups in the ability of the model to predict number of metastases from days postirradiation (P = 0.15). Based on this analysis, there is no evidence that MnP or radiation effects the relationship of the development of pulmonary metastasis beyond the influence of time to euthanasia.

FIG. 3.

MnP/RT increases tumor growth delay and extends survival time in the extended end point experiment in mice with orthotopic dorsal mammary 4T1 tumors. Mice treated with MnP/RT lived significantly longer than control mice (P = 0.032).

FIG. 4.

Metastasis associated with MnP/RT correlates with survival time in the extended end point experiment in mice with orthotopic dorsal mammary 4T1 tumors. Metastatic burden measured per mouse was considered as a function of time. There was no significant difference across the treatment groups in development of pulmonary metastasis over the influence of time (P = 0.15).

To verify that metastases were not being enhanced by MnP/RT, we performed a second experiment in which all mice were euthanized at the same time after treatment. We postulated that if MnP/RT was enhancing metastases, we would observe more lung metastases in the MnP/RT group at this fixed time point. After the implantation of tumor cells, tumors reached treatment size (minimum 100 mm³, longest diameter <12 mm) in 10 days, and were then treated with MnP and/or radiation. Mice with directly injected mammary fat pad tumors were monitored for 11 days after the start of treatment and then euthanized. In this experiment, MnP alone did not affect local primary tumor growth. Radiation reduced primary tumor growth twofold (P = 0.0004) as a monotherapy compared to control. The combination of MnP/RT resulted in a threefold reduction in tumor volume compared to the untreated controls (P = 0.0001) (Fig. 5A). Tumors were weighed at the time of euthanasia as another approach to compare local tumor effects. There was a significant (25%) reduction in primary tumor weight in the irradiated group compared to controls (P = 0.0047), and the MnP/RT treatment group (P = 0.0001) showed one half the size compared to control and MnP treatment groups (Fig. 5B). While local tumor growth was reduced by the treatment, there was no difference in the number of pulmonary metastases between the treatment groups (Fig. 5C).

FIG. 5.

MnP/RT reduces tumor growth, but does not affect metastasis, in the fixed end point experiment in mice with orthotopic ventral mammary 4T1 tumors. Panel A: Radiation reduced primary tumor volume (P = 0.0004) compared to control volume at the defined end point of eight days. MnP/RT further reduced tumor growth (P = 0.0001) relative to controls. Panel B: Tumors were weighed at the time of euthanasia. MnP/RT (P = 0.0001) and irradiation (P = 0.0047) resulted in significant reduction in tumor weight at end point compared to control and MnP treatment groups. Panel C: No significant difference in the magnitude of pulmonary metastatic burden was found between treatment groups.

Cellular Redox Environment within Tumor and Blood

The cellular redox environment was characterized according to GSH/GSSG within the primary tumors and whole blood at two time points: 1. at 24 h postirradiation to study acute effects; and 2. at the time of euthanasia to study effects throughout the extended growth delay. There were no significant differences across the treatment groups in GSH/GSSG at either time point in tumor or whole blood samples (Fig. 6). Additionally, there were no significant differences in the absolute levels of GSH or GSSH across the treatment groups at either time point (data not shown).

FIG. 6.

MnP/RT does not alter the cellular redox environment in tumor or whole blood at 24 h postirradiation or at the extended end point. GSH/GSSG was measured within the primary tumors (panels A and B) and whole blood (panels C and D) at two time points: 1. 24 h postirradiation; and 2. end point of time of euthanasia. There were no significant differences across the treatment groups in GSH/GSSG at either time point in tumor or whole blood samples.

However, the combination of MnP and radiation generated a significant increase in intratumoral oxidative stress at 11 days after treatment, as measured by immunofluorescent staining of 4-HNE for lipid peroxidation (P < 0.05) and 8-OHdG for RNA/DNA oxidation (P < 0.05) (Fig. 7).

FIG. 7.

MnP/RT increased intratumoral oxidative stress at the fixed end point. Panel A: 4-Hydroxynonenal (4-HNE) staining for lipid peroxidation. Red = 4HNE, blue = DAPI/nuclei. ^#Significant increase (P ≤ 0.05) compared to all other groups. Panel B: 8-Hydroydeoxyguanosine (8-OHdG) staining for RNA/DNA oxidation. Red = 8-OHdG, blue = DAPI/nuclei. *Significant increase (P ≤ 0.05) compared to control-PBS. Control-PBS, BuOE, and RAD-PBS groups contain n = 5 animals and RAD + BuOE contains n = 8 animals. Results are expressed as mean fluorescent intensity and SEM. White bars =100 μm.

Intratumoral HIF-1α Expression

HIF-1α levels of treated tumors were quantified via ELISA at the same time points as the cellular redox environment analyses: 1. at 24 h postirradiation to study acute effects; and 2. at the time of euthanasia in the tumor growth delay study. There was a significant difference in HIF-1α levels across the treatment groups at 24 h (P = 0.0309); the HIF-1α levels of the tumors in the MnP/RT group increased fivefold compared to the irradiation only group (P = 0.0346) (Fig. 8A). At the extended end point, there were no significant differences in HIF-1α levels across the treatment groups (P = 0.0686) (Fig. 8B).

FIG. 8.

MnP/RT increases intratumoral HIF-1α levels at 24 h postirradiation, but not at the extended end point. HIF-1α levels of treated tumors were quantified using ELISA at 24 h postirradiation (panel A) and at the end point time of euthanasia (panel B). There was a significant difference in HIF-1α levels across the treatment groups at 24 h (P = 0.0309), with the HIF-1α levels of the tumors in the MnP/RT group increased compared to irradiated mice (P = 0.0346). No significant differences were seen when the levels in the MnP/RT group were compared to the other treatment groups (MnP/RT vs. control, P = 0.0899; MnP/RT vs. MnP, P = 0.1129). There were no significant differences in HIF-1α levels across the treatment groups at the extended end point (P = 0.0686).

DISCUSSION

In this study, the therapeutic effects of radiation and adjuvant MnPs for primary and pulmonary metastatic tumor growth were investigated in orthotopic mammary tumor models. Until now, our preclinical studies have focused on local effects of MnPs and radiation; however, we recognize that successful therapeutic outcomes for patients involve control of tumor cells throughout the body. Therefore, we investigated whether MnPs and radiation influence the development of distant metastasis using murine orthotopic mammary tumor models. We hypothesized that adjuvant MnP would shift the cellular redox environment within the irradiated tumor and downregulate HIF-1α in tumor cells. Secondarily, we hypothesized that these local tumor effects would lead to abrogation or inhibition of the development of distant metastases. We found MnP/RT treatment did extend local tumor growth delay compared to controls and was associated with increased intratumoral oxidative stress. However, the primary tumor growth delay seen with MnP/RT was not associated with a reduced pulmonary metastatic burden.

We performed these experiments with orthotopic murine mammary tumors that spontaneously metastasize, in order to study the development of distant metastases in the context of MnP/RT treatment. The development of metastatic disease is a serious clinical scenario correlating with advanced stages of disease for breast cancer patients. The 4T1 mammary tumor model is aggressive and progresses rapidly at the site of tumor cell implantation. For the tumor growth delay study, we utilized the dorsal mammary tumor model in anticipation that this would allow for the longest duration of survival with primary tumor growth to evaluate metastatic disease burden uniformly across treatment groups. In this experiment, however, we found that a subset of mice required humane euthanasia before tumors reached the defined end point due to the development of neurologic deficits or paralysis secondary to spinal invasion by the tumor. Thus, a proportion of mice in the study were euthanized before their primary tumors reached the end point of 1,500 mm³. Because of this variability in time of euthanasia in the growth delay study, we designed a second study, wherein all mice were euthanized at a fixed time point after treatment. We performed this fixed end point experiment using the ventral mammary tumor model to minimize the number of mice lost for analysis due to spinal invasion of the tumors.

The therapeutic effects of MnP/RT with respect to local tumor control were evaluated in these experiments. In the tumor growth delay experiment, the combination of MnP/RT slowed tumor progression and increased survival time compared to control mice. In the fixed end point experiment, radiation reduced primary tumor growth compared to untreated controls, and MnP/RT reduced tumor growth even further. These differences in primary tumor burden, as measured by size, were verified by tumor weights from mice treated in the radiation and MnP/RT treatment groups compared to control and MnP treatment groups.

The addition of MnPs to irradiation has been shown to induce high levels of oxidation in cancer cells while simultaneously creating a reducing environment in normal cells, leading to a wider therapeutic index (7, 8, 15). We have previously shown in other tumor models, including head and neck cancer, prostate cancer and melanoma (3, 15, 40), that the treatment combination of MnP and radiation inhibits primary tumor growth. The results of our study reinforced these findings utilizing the orthotopic 4T1 tumor model and recapitulated the results obtained in the 4T1 mouse flank model (39). The MnP/RT-treated tumors had significantly increased levels of oxidative stress compared to other treatment groups at a fixed end point; however, analyses of the cellular redox environment via GSH/GSSG did not reveal significant differences with respect to tumor or whole blood samples at 24 h postirradiation or at extended tumor growth end points. These GSH/GSSG results may be associated with the time points of tissue collection and analysis. In a published experiment with 4T1 flank tumors treated with both MnTE-2-PyP⁵⁺ and MnTnBuOE-2-PyP⁵⁺, it was noted that the inhibitory effect of MnTnBuOE-2-PyP⁵⁺ on tumor growth, with concurrent treatment with radiation and/or ascorbate, was strongly correlated to the levels of H₂O₂ within the tumor, irrespective of the amount of MnP administered (being either 2 mg/kg or 0.2 mg/kg) (8, 38, 39).² Batinic-Haberle and Spasojevic attribute this finding to the major role of H₂O₂ in the actions of MnP (9). We also know that MnTnBuOE-2-PyP⁵⁺ accumulates more in tumor than in normal cells (39). Under such conditions of high MnP and H₂O₂ concentrations, MnP catalyzes cysteine oxidation/S-glutathionylation of numerous critical proteins; massive oxidation of proteins enhances apoptotic processes and reduces tumor growth (8, 9). The increased levels of protein S-glutathionylation and reduced levels of GSH both in 4T1 cells and flank tumors was previously reported by Tovmasyan et al. (39). Increased oxidative stress was demonstrated in this study in the increased levels of 4-HNE and 8-OHdG, demonstrating increased lipid peroxidation and DNA damage, but was not seen with the GSH/GSSG level in this study. Comprehensive study of oxidative damage and redox status at different time points during tumor growth may be needed. Furthermore, we recognize that GSH/GSSG gives a partial indication of the redox environment. Future directions will incorporate additional assays to measure the redox environment, such as NAD⁺/NADH and NADP⁺/NADPH ratios, as well as measurement of thiol oxidation.

As the mechanism of the effect of MnP/RT for improving local tumor control is associated with an increased intratumoral concentration of H₂O₂ and oxidative stress, we consider an additional level of explanation for why MnP/RT treatment did not influence distant metastasis. In our model, mice treated with MnP/RT were treated systemically with MnP and then received targeted irradiation administered to the orthotopic mammary tumor. The tumor microenvironment would have undergone an increased accumulation of MnP, then an enhancement of H₂O₂ when the tumor was irradiated. Alternatively, the normal lung tissue would have maintained a normal concentration of H₂O₂ as it was spared from irradiation and MnP does not distribute well in normal lung. Therefore, the sensitizing effect of MnP/RT on local tumor control would not be expected to translate to a secondary treatment effect of MnP/RT in the lung microenvironment. Future directions to explore treatment combinations to improve both local and distant tumor control with MnP/RT could include treatment with an additional source of systemic H₂O₂, such as ascorbate, to slow or inhibit the formation of distant metastasis.

We also consider that MnP/RT did not affect the development of distant metastasis in our experiment due simply to the high metastatic rate of the 4T1 model. It is possible that by the time the primary orthotopic mammary tumors had reached treatment size, the mice could have already developed micrometastatic disease in the lungs. In a published study investigating doxorubicin-conjugated polypeptide nanoparticles to treat 4T1 orthotopic mammary tumors, the day of surgical resection of the primary tumor affected the development of metastasis; specifically, metastasis-forming cells spread between days 8 and 12 after tumor implantation for most mice (41). In our experiments, after implantation of tumor cells, tumors reached treatment size in 9 to 10 days, which would fall within the metastasis time frame reported for orthotopic 4T1 cells. Future experiments with 4T1 tumors will involve a modified timeline from implantation to primary tumor treatment to address this challenge in the experimental design. Alternatively, a less metastatic variant of the 4T1 mammary carcinoma cell line could be considered for future studies. 67NR, 66c14, 4T07 and 4T1 are mammary carcinoma cell lines which were derived from a single spontaneous arising mammary tumor from a Balb/cfC3H mouse (42). Although they share a common origin, these lines are phenotypically heterogeneous in their metastatic behavior. Like the 4T1 cell line, 66c14 cells will also spontaneously metastasize to lungs of mice, but not nearly as rapidly as the 4T1 cells (42–44). Experiments of this project could be repeated in 66c14 cells to evaluate local and distant cancer effects of Mn/RT in orthotopic models of mammary carcinoma. Another consideration for future experiments investigating the distant cancer effects of MnP/RT is to evaluate metastasis of 4T1 cells to the lungs, as well as other tissues, including the liver and spleen. Furthermore, quantification of metastatic burden in the mice could be performed according to counts of grossly visible counts, as done in this study, and then these results could be enhanced by evaluating the size of metastatic lesions, in addition to quantity.

In a previously published study utilizing the dorsal skin fold window chamber model, 4T1 cells displayed a significant increase in HIF-1 activity at 24–48 h postirradiation (13). Furthermore, the combination of MnP and radiation-sensitized 4T1 tumors via reduction in radiation-induced oxidative stress, leading to the downregulation of HIF-1 activity at 48 h postirradiation (13). In these experiments, these tumor microenvironmental effects were not observed during 12–24-h time points. An interesting finding to emerge from this study was the significant difference in HIF-1α protein levels across the treatment groups at the 24-h time point, with the HIF-1α levels of the tumors in the irradiated group significantly reduced compared to the increased HIF-1α levels measured in the MnP/RT group. However, at the extended end point, there was no significant difference in HIF-1α levels across the treatment groups. A potential reason for the reduced HIF-1α levels at the 24-h time point in the irradiated group is the substantial cell killing from the 15 Gy dose. Additionally, the irradiated tumors may have been experiencing a phase of reduced oxygen consumption whereby there was an overall reduction in HIF-1α protein levels, as a result of there being less hypoxia. The increased HIF-1α levels measured in the MnP/RT-treated tumors at 24 h is an effect that has not been previously described to our knowledge. Therefore, the relevance of this result will require additional investigation. We recognize that the quantification of HIF-1α protein levels does not directly translate to HIF-1α activity, since HIF-1α is required to translocate along with HIF-1α to the nucleus to function as a transcription factor. HIF-1 activity has been measured in previous studies via cell lines that were transfected with a HIF-1 activity reporter gene. The 4T1 cells used in this study did not have the same reporter gene, so HIF-1 activity could not be measured in that way. In future experiments, we may investigate key downstream genes, such as vascular endothelial growth factor (VEGF) and glucose transporter 1 (GLUT-1), to determine if any of these downstream genes experience differential upregulation.

Clinical trials are underway to investigate the clinical utility of MnTnBuOE-2-PyP⁵⁺ for patients undergoing radiotherapy for high grade glioma (NCT02655601), multiple brain metastases (NCT03608020), locally advanced head and neck cancer (NCT02990468) or anal cancer (NCT03386500). The promising preclinical data from this study, as well as others, provides support that MnP/RT has the potential to improve local tumor control for these patients. With respect to breast cancer therapy, it is recognized that effective local control is associated with improved overall survival, particularly for women with early-stage cancer (45). We demonstrated that MnP/RT treatment for mice with orthotopic mammary carcinoma resulted in increased local tumor control and overall survival. A clinical trial investigating this treatment combination for women with breast cancer could be considered as a future direction of our work.

CONCLUSIONS

We investigated the therapeutic effect of MnP/RT with respect to the development of distant pulmonary metastasis. We found that the mice in the MnP/RT treatment group lived significantly longer than mice in the other treatment groups and metastatic burden was correlated with an increased survival time. To further investigate this, we performed an experiment with a fixed end point. There was no significant difference in the number of pulmonary metastatic nodules across the treatment groups when evaluated at the same time. In a recently published study, mice with orthotopic prostate tumors were treated with radiation and MnTE-2-PyP, and, in addition to investigating the influence of MnP/RT treatment on primary tumor control, the number of metastatic nodules within the peritoneal cavity was measured. It was found that MnP did not affect metastatic progression of the orthotopic prostatic tumors (15). Based on the results of that study and ours, we recognize that MnP/RT provides local tumor control but does not reduce or affect the development of metastatic disease. We acknowledge that we did not treat the mice in our study with a curative dose of radiation. We did not observe any acute skin toxicity or other radiation injury in the population of mice in our study; therefore, it seems possible to increase the radiation dose in subsequent experiments. Additionally, utilizing a higher dose of radiation to an extent where evaluable skin toxicity is induced in control mice would allow for another level of evaluation of the radioprotective properties of MnPs in mice treated with MnP/RT. Future directions may involve additional experiments to define the dose to control 50% of the orthotopic 4T1 tumors (TCD₅₀) and continue investigations into the role of MnP/RT in enhancing local and metastatic treatment effects.

Supplementary Material

Supplementary file

Fig. S1. Orthotopic dorsal mammary tumor irradiation technique. A diamond-shaped radiation field was used to target the radiation beam to the dorsal mammary tumor while avoiding radiation exposure to the lungs.