NOTE: End date changed to 9/30/2022 per HRP Space Radiation (Ed., 8/3/21)

NOTE: End date changed to 9/30/2021 per PI and NSSC information (Ed., 3/4/21)

NOTE: End date changed to 4/14/2021 per NSSC information (Ed., 1/16/2020)

GI-tumorigenesis study (Project 1, Georgetown University):

In the last eight years of NASA Specialized Center of Research (NSCOR) support, we have successfully applied mouse model approaches to acquire quantitative and qualitative gastrointestinal (GI) tumorigenesis data to address prevailing uncertainties in risk modeling. Our overall goal is to determine the relative biological effectiveness (RBE) of various parameters for space radiation compared to terrestrial radiation.

The risk estimates can then be extrapolated to space radiation using a RBE scaling factor. Pointwise details of progress made for GI tumorigenesis study are provided below:

1. Using Apc1638N/+ mice, we reported that the relative RBE for radiation-induced (IR)-induced gastrointestinal (GI) tumorigenesis is related to ion energy, linear energy transfer (LET), and gender. Male and female mice were exposed to sham or 5 to 200 cGy doses of ??-rays, 4He (250 MeV/n), 12C (290 MeV/n), 16O (325 MeV/n), 28Si (300 MeV/n), and 56Fe (1000 MeV/n). Calculation of RBE for intestinal and colon tumorigenesis showed the highest value with 28Si, and lower doses showed greater RBE relative to higher doses.

2. We reported downregulation of RXRa, an Apc-independent factor involved in targeting ß-catenin to the ubiquitin-proteasome pathway (UPP) for degradation after heavy-ion exposure. Decreased expression of RXRa in tumors as well as in adjacent normal epithelium was indicative of perturbations in ß-catenin proteasomal-targeting machinery, which suggests the plausibility of RXRa agonists as a potential countermeasure against heavy-ion-induced GI tumorigenesis.

3. We compared GI tumorigenesis in Apc1638N/+ mice after exposure to heavy ions at a high (50 cGy/min) and relatively low (0.33 cGy/min) dose rate. In both male and female mice, GI tumor number and size were not significantly different after high and low dose rate exposures. This study suggested that the carcinogenic potential of energetic heavy ions is independent of the dose rate.

4. Significantly higher intestinal tumor number and grade, along with decreased cell differentiation, were observed after acute radiation relative to fractionated radiation. Acute protons induced upregulation of ß-catenin and Akt pathways with increased proliferative marker phospho-histone H3. Increased DNA damage along with decreased DNA repair factors involved in mismatch repair and nonhomologous end joining were also observed after exposure to acute protons.

5. A comparison of tumor data from simplified-mixed-field (Smf)-galactic cosmic radiation (GCR) and equivalent doses of heavy ions revealed an association between higher GI tumorigenesis, where the dose received from heavy ions probably contributed to much of the total GI tumorigenic effect observed after Smf-GCR. This study provided the first experimental evidence that cancer risk after GCR exposure may largely depend on doses received from constituent heavy ions.

6. Aspirin (an inhibitor of COX), was evaluated as a potential countermeasure against 28Si-induced GI tumorigenesis in Apc1638N/+. Aspirin led to a significant reduction in PGE2 in a dose-dependent manner but did not reduce 28Si-induced GI tumorigenesis.

7. Metformin is a known suppressor of IGF1-mTOR signaling via activation of AMP-activated protein kinase (AMPK), we tested its potential GI-cancer preventive effects against heavy ion-induced GI tumorigenesis. We noted that long-term continuous intake of metformin confers significant protection against heavy-ion radiation-induced GI tumorigenesis.

8. We assessed radiation quality effects on colonic inflammation, colon cancer incidence, and associated signaling events in Il10-/- mice. IR-induced colitis and colitis-associated cancer (CAC) incidence in Il10-/- mice depend on radiation quality and display co-activation of ß-catenin and NF-?B signaling. Regardless of the Apc gene status of a mouse model, heavy-ion IR-induced GI tumorigenesis displays co-activation of oncogenic and pro-inflammatory signaling pathways.

9. Male Apc1638N/+ mice were exposed to 5, 10, and 50 cGy ?? rays or 28Si (69 keV/µm) radiation, and gastric tumorigenesis was assessed at 150 days post-exposure relative to the control group. Animals exposed to 28Si radiation had the most gastric tumors and carcinomas, followed by ?-ray-exposed and control groups. Both 28Si and ?-ray exposures showed a dose-dependent increase in gastric tumor and carcinoma development at the dose range of 5 to 50 cGy. At all doses, the radiation-quality-dependent increase in gastric tumorigenesis was evident. Our findings suggest that exposure to heavy ions is more likely to cause gastric cancer than low-LET radiation.

Dissection of the signaling events and consequences in GI cells of the persistent effects of space radiation (Project-2, Georgetown University):

Proliferative gastrointestinal (GI) tissue is radiation-sensitive, and heavy-ion space radiation with its high-linear energy transfer (high-LET) and higher damaging potential than low-LET ?-rays is predicted to compromise astronauts' GI function. However, much uncertainty remains in our understanding of how heavy ions affect coordinated epithelial cell migration and extrusion, which are essential for GI homeostasis. The overall goal of this project is to understand the long-term effects of space radiation on normal gastrointestinal (GI) cells including stem cells.

Pointwise details of progress made for persistent effects of space radiation on GI tissues are provided below:

1. We investigated the long-term effects of 56Fe radiation on adipokines and the insulin-like growth factor 1 (IGF1) signaling axis in the mouse intestine and colon. Irradiation increased leptin and IGF1 levels in serum and IGF1R and leptin receptor expression in tissues. Additionally, upregulated Jak2/Stat3 pathway and cell proliferation were also noted. Therefore, our data demonstrated chronic pathophysiologic and endocrine alterations in GI tissue after space radiation exposure.

2. We demonstrated that heavy-ion exposure resulted in persistently delayed intestinal epithelial cell (IEC) migration involving chronic sublethal genotoxic and oncogenic stress-induced altered cytoskeletal dynamics, which were seen even a year later after IR exposure.

3. 56Fe radiation triggered a time-dependent increase in ?H2AX foci and senescent cells but without a noticeable increase in apoptosis. Some senescent cells acquired the senescence-associated secretory phenotype (SASP), and this was accompanied by increased intestinal epithelial cell (IEC) proliferation. Additionally, using the Lgr5-EGFP-IRES-creERT mice model, we observed increased ROS and ongoing DNA damage by staining for ?H2AX, and 53BP1, along with the accumulation of senescence markers in intestinal stem cells. Results also showed increased acquisition of SASP in senescent intestinal stem cells (ISCs) after heavy-ion exposure relative to ?-rays. Collectively, our data indicate that heavy-ion-induced chronic stress and ongoing DNA damage promoted SASP in a fraction of the ISCs.

4. Exposure to 56Fe resulted in significantly higher levels of DNA base lesions in the GI tissue, relative to control and ?-radiation. Both base excision repair (BER) and nucleotide excision repair (NER) pathways were downregulated. Moreover, DNA damage response (DDR) alterations could result in both somatic gene mutations as well as activation of the pro-inflammatory, pro-oncogenic SASP signaling known to accelerate adenoma-to-carcinoma progression during radiation-induced GI cancer development.

5. In addition to GI tissues, differential accumulation of senescence and SASP cells was observed as part of the persistent IR-induced alterations in mouse bone marrow (BM). A remarkable increase in p16-positive cells indicated a persistent increase in cell senescence; whereas, an increased RANKL/OPG ratio, reductions in the number of osteoblast progenitor cells, and osteocalcin provided clear evidence that exposure to both proton and 56Fe-ions promotes pro-osteoclastogenic activity in BM.

6. Senescent cells can be removed intrinsically by immune cells. 28Si exposure decreased the NK cell population in the LP, which suggests inefficient immune surveillance of senescent cells that could contribute to the persistent accumulations of senescent and SASP cells after heavy ion exposure.

7. Exposure to a low dose of ?-rays were associated with upregulation of nicotinamide adenine dinucleotide phosphate (NADPH) oxidase and its modifiers, along with increased Ras/p38/Gata6 signaling in the colon. When considered along with oxidative damage and proliferative markers, our observations suggest that the NADPH oxidase pathway could be playing a critical role in propagating long-term oxidative stress after radiation.

Lung tumorigenesis and countermeasure studies (UTSW) (Project-3):

1. Developed an isogenic series of human colon and lung normal, premalignant and cancer cell lines to investigate carcinogenesis after low and high LET irradiation. 2. Determined that 3D culture of human normal colonocytes can be used to investigate radiation-induced transformation.

3. Quantitatively assessed lung and colon tumorigenesis in mouse models of radiation-associated cancer susceptibility. Determined that the low threshold for polyp formation in the colon cancer susceptible mouse model was between 5 and 10cGy of HZE silicon particle irradiation.

4. To simulate the total dose expected by astronauts on a round trip to Mars, we utilized a simplified 3-beam GCRsim fast-switching approach in mice using a total of 30 cGy or less. 30 cGy single-dose of protons and a single dose of 5 cGy silicon did not increase carcinogenesis.

5. This 3-beam GCR simulation used 20cGy proton, 5cGy helium, and 5cGy silicon which resulted in an increase in plasma lipid peroxidation (a biomarker of increased persistent inflammation). We also observed an increase in premalignant lesions and advanced cancers (~2-fold). However, if we reduced the amount of silicon to 2 cGy the increased cancer risk was decreased back to unirradiated control mice. If we reversed the order of ions so that silicon was provided first followed by helium and then protons, we did not observe the increases in cancer risk. We concluded that the order of GCR irradiation matters.

6. Determined that aspirin does not protect against space radiation-associated cancers.

7. Demonstrated that Nrf2 activation by the synthetic triterpenoid, CDDO (an FDA approved oral drug to treat patients with Friedreich’s Ataxia), protects colonic epithelial cells against IR-induced transformation in part by enhancing repair of DNA damage. CDDO prevents human cell transformation using both low and high LET irradiation. CDDO protects specific brain functions after GCR simulations and protects normal lung, colon, and breast epithelial cells but not cancer cells from irradiation. CDDO increases mouse survival after 10Gy total body ionizing radiation.

8. Metformin pre-administration acts as a radioprotector through AMPK phosphorylation, increasing OGG1 and decreasing cleaved PARP expression. Altogether, we interpret these results to suggest that metformin is a potential oral available and safe GCR radioprotector and has the potential to lower the risk of cancer initiation/promotion in astronauts.

Cancer risk modeling (Columbia University) (Project-4):

Long-duration space exploration missions, such as a journey to Mars, pose significant health risks to astronauts due to exposure to a unique mixture of sparsely ionizing and densely ionizing radiations present in space. The biological effects of this mixture can be divided into two categories: targeted effects (TE) and non-targeted effects (NTE). The combination of both TE and NTE can result in non-linear dose-response shapes for space radiations, particularly in the dose range relevant for space exploration missions. To predict the likelihood of developing cancer and other diseases following space travel, it is important to use radiation dose-response models that incorporate both TE and NTE components. In this project, such a model was developed and applied to data on intestinal and gastric carcinogenesis in Apc1638N/+ tumor-prone mice and are described below:

1. Dose-response and dose rate effect analyses: Modeling of our intestinal carcinogenesis data using the TE+NTE formalism suggested considerable NTE involvement at low doses that are relevant for space travel. The NTE contribution was predicted to be particularly strong for the most densely ionizing tested ions Si and Fe. Model-based estimates for the dose-rate effect on heavy ion-induced carcinogenesis at doses/dose rates expected during a Mars mission show only a small and not statistically-significant dose-rate effect. Consequently, modeling analysis suggests that heavy ion carcinogenesis estimates from moderate/high dose-rate experimental data may be applicable to doses/dose rates relevant for space exploration.

2. Metrics for radiation quality comparisons: Reliable methods are needed for scaling low-LET to high-LET radiation risks for humans, based on animal or in vitro studies comparing these radiations. A proposed new metric, radiation effects ratio (RER), compares the effects of two radiations at the same dose of interest (vertical scaling). In contrast, the traditional RBE metric relies on horizontal comparisons of dose responses for the compared radiations, for which more extensive information at several doses is generally needed. RER can be used without the need for detailed information about dose-response shapes for compared radiations and allows animal carcinogenesis experiments to be simplified by reducing the number of tested radiation doses. For simple linear dose-effect relationships, RBE = RER. However, for more complex dose-effect relationships, such as those with NTE, RER can be different from RBE.

3. Analyses of radiation mixture effects: Based on the modeled responses for each radiation type, the tumor yield for a Mars-mission-relevant mixture of these radiations was predicted using the recently developed incremental effect additivity (IEA) synergy theory. The proposed modeling approach can enhance current knowledge about the quantification of space radiation quality effects, dose-response shapes, and ultimately the health risks for astronauts.

4. Microdosimetric quality factor calculations: The microdosimetric approach, using measured or calculated distributions of microdosimetric energy deposition together with empirical biological weighting functions, is conceptually and practically simpler than the alternative fluence-based risk cross-section approach. The microdosimetric approach is practical and can allow continuous readout of biologically effective doses during spaceflight in conjunction with in-situ measurements of microdosimetric spectra.

NOTE: End date changed to 9/30/2022 per HRP Space Radiation (Ed., 8/3/21)

NOTE: End date changed to 9/30/2021 per PI and NSSC information (Ed., 3/4/21)

NOTE: End date changed to 4/14/2021 per NSSC information (Ed., 1/16/2020)

GI-tumorigenesis study (Project 1, Georgetown University):

In the last six years of NASA Specialized Center of Research (NSCOR) support, we have successfully applied mouse model approaches to acquire quantitative and qualitative gastrointestinal (GI) tumorigenesis data to address prevailing uncertainties in risk modeling. Our overall goal is to determine the relative biological effectiveness (RBE) of various parameters for space radiation compared to terrestrial radiation. The risk estimates can then be extrapolated to space radiation using an RBE scaling factor. Our strategy is to: i] Quantitatively analyze tumor incidence and grade following space type radiation exposure. ii] Assess potential dose rate and gender variation. iii] Delineate cancer progression-specific biological signatures associated with space radiation exposure. Since low-linear energy transfer (LET) gamma radiation risk estimates are available for human populations from studies in atom bomb cohorts and the like, we will then be able to combine our data to model the relative risk for space radiation.

Pointwise details of progress made for GI tumorigenesis study are provided below: 1. We participated in the NASA Space Radiation Laboratory / NSRL-21B (summer) and NSRL-22B (summer) beam campaigns. During these beam runs, we conducted a pilot GI-tumorigenesis experiment with a limited number of male and female Apc1638N/+ mice exposed to acute (25 to 75 cGy) and chronic full-spectrum galactic cosmic radiation (GCR) beams. 2. Sequentially delivered proton (H) to Helium (He) to Oxygen (O) to Silicon (Si) beams that were designed to simulate simplified-mixed-field GCR. A comparison of tumor data from simplified-mixed-field GCR and equivalent doses of heavy ions revealed an association between higher GI-tumorigenesis where the dose received from heavy ions contributed to >95% of the total GI-tumorigenic effect observed after mixed-field GCR. This study provides the first experimental evidence that cancer risk after GCR exposure could largely depend on doses received from constituent heavy-ions. 3. Aspirin, a well-known inhibitor of the cyclooxygenase-2/prostaglandin E2 (COX/PGE2) pathway, was evaluated as a potential countermeasure against 28Si-induced PGE2 and GI-tumorigenesis in Apc1638N/+. Aspirin led to a significant reduction in PGE2 in a dose-dependent manner but did not reduce 28Si-induced GI tumorigenesis even at the highest (300 mg/day) dose. In summary, this study suggests that aspirin has no preventive effect against high-LET radiation-induced GI tumorigenesis in Apc1638N/+ mice. 4. Metformin is known to suppress insulin-like growth factor 1 (IGF1) signaling, via activation of AMP-activated protein kinase (AMPK). Here we tested metformin for its potential GI-cancer preventive effects against heavy-ion radiation-induced GI-tumorigenesis. Male Apc1638N/+ mice were continuously fed either standard or metformin supplemented diet (0.1% W/W) starting at the age of 4 weeks, and one month later mice were irradiated with 10 cGy of 28Si at the NSRL. At 150-day post-exposure intestinal tumor number, cluster, size, and tumor grade were assessed. The systemic impact of metformin on serum IGF1 was analyzed using an antibody array. We have reported that long-term continuous intake of metformin confers significant protection against heavy-ion radiation-induced GI-tumorigenesis in Apc1638N/+ mice. Here we report that metformin has a preventive effect against heavy ion-induced serum level IGF1, a pro-carcinogenic growth factor with an established role in GI-cancer promotion.

Dissection of the signaling events and consequences in GI cells of the persistent effects of space radiation (Project 2, Georgetown University):

Proliferative gastrointestinal (GI) tissue is radiation-sensitive; and heavy-ion space radiation, with its high-linear energy transfer (high-LET) and higher damaging potential than low-LET gamma-rays, is predicted to compromise astronauts' GI function. However, much uncertainty remains in our understanding of how heavy ions affect coordinated epithelial cell migration and extrusion, which are essential for GI homeostasis. The overall goal of this project is to understand the long-term effects of space radiation on normal gastrointestinal (GI) cells, including stem cells. Coordinated epithelial cell migration is key to maintaining functional integrity and preventing pathological processes in gastrointestinal (GI) tissue and is essential for astronauts’ health and space mission success.

Pointwise details of progress made for persistent effects of space radiation on GI-tissues are provided below: 1. Heavy ion radiation exposure induces senescence, and triggers accelerated senescence-associated secretory phenotype (SASP) in mouse intestinal Lgr5+ stem cells. Higher levels of senescence and SASP were observed at 1 year compared to the 60d timepoint. 2. ß-gal, lysozyme, IL1R, or IL1ß showed the highest staining in 28Si irradiated compared to control intestinal stem cells (ISCs) samples. Phospho-Histone H3 co-staining showed a mutually exclusive staining pattern with a higher level of proliferating cells in 28Si compared to control or gamma-rays irradiated groups. g-H2AX staining showed greater foci numbers in 28Si relative to gamma-rays or the control group. 3. The higher expression of secretory cytokines IL8 and IL15 was observed in 28Si irradiated ISCs relative to gamma-rays or the control group at 60d and 1y time points. Enhanced expression of Dickkopf-related protein (DKK2) and activation of Nuclear factor kappa B and p38 mitogen-activated protein kinase (MAPK) were visualized in ß-gal positive cells of 28Si irradiated ISCs relative to gamma-rays or control group at these time points. 4. Senescent cells can be removed intrinsically by immune cells; however, we know very little about the effects of heavy ion radiation on immune cells. Flow cytometry analysis showed a significant decrease in NKp46 (natural killer/NK cell marker) expression in 28Si relative to the gamma-ray or control group. Our results demonstrate that 28Si exposure decreased NK cell infiltration into the Lamina Propria, and inefficient senescent surveillance may contribute to the persistent accumulations of senescent cells, and increased SASP signaling.

Lung tumorigenesis and countermeasure studies (Project 3, University of Texas Southwestern Medical Center / UTSW):

This project involves lung tissue from normal wildtype (WT) mice, and a mouse model of lung cancer (LA-1). The rationale for these models is that whole body exposure to space radiation will affect organ/tissues and cells in different ways. The cells in the lung only turn over with damage, while the colon turns over in humans every 7 days irrespective of damage. Thus, to model increased risks of more lethal cancers from space radiation exposure, it is important to examine tissues with both a high rate of cellular turnover (colon, Georgetown plan) and tissues with a low rate of cell turnover (lung, UTSW plan). The inclusion of a lung model reflects the major solid tumor risk for astronauts traveling in space. Results from these studies were designed to evaluate the effects of both acute and chronic doses of the GCR Simulator on radiation-induced health risks using a radiation-induced lung cancer in male and female wildtype and LA1 mutant (a lung cancer susceptible mouse) mice on a 129 background. Each study included an acutely exposed group, a fractionated exposed group with consecutive doses of 20.8 mGy/day delivered over a four-week period, and sham controls. The simulator was baselined to 50 cGy for the first series of experiments. Acute exposures were delivered halfway through the fractionated exposure to support age-matching post irradiation. Recorded exposure times for either the acute or fractionated exposure were ~70 to 75 minutes, with the largest portion of time devoted to beam switching. During this initial experimental campaign, approximately 300 mice received daily irradiations and ~200 mice treated as sham controls. This required the additional daily loading and unloading of all control animals from the cage array over the four-week period as well. Based on camera footage from the target room, the mice adapted nicely to the exposure boxes and remained active especially with the nestlet/bedding material provided. This was the case even after being loaded into and out of the cages over the course of four weeks and alleviated concerns over the possibility of mice being shielded by cage mates while sleeping in the corners and not receiving a more uniform dose in all directions. We then initiated a series of GCRsim with acute and chronic 75cGy but the results are not complete yet. Even with 75cGy total dose, there is only 6.3cGy of HZE >3. Finally, the Brookhaven National Laboratory (BNL) team irradiated our mice with a 25cGy and 100cGy acute GCRsim during the NSRL 20B, and results are still in progress. Thus, there are experiments that need to be completed. The survival studies and overall incidence of invasive, more lethal cancers require a year post irradiation. At NSRL-22A, we irradiated an additional 100 mice with acute 75cGy GCRsim, With sufficient mice for 22B, we also irradiated mice each with acute 25, 50, and 100cGy 33 beam GCR to increase our numbers of mice from previous experiments to show reproducibility and to determine if there are sex differences in cancer initiation and progression.

Cancer risk modeling (Project 4, Columbia University):

A trip to Mars and back, or another lengthy deep space exploration mission, will result in exposure of astronauts to a multi-component mixture of ionizing radiations which damage cells in multiple ways. Such damage mechanisms can be roughly classified into two categories: (1) Targeted effects (TE), involving the consequences of direct traversals of cells by ionizing tracks leading to DNA double strand breaks and other lesions. (2) Non-targeted effects (NTE, or “bystander” effects) caused by release of signals from cells directly hit by the tracks and the effects of these signals on other cells. The presence of both types of effects generates complex dose response shapes for space radiations, particularly in the low dose region relevant for space missions.

Here we used our mechanistically-motivated model formalism that includes both TE and NTE to analyze updated mouse tumorigenesis data from the current NSCOR. The updated data set now includes more detailed information on space-relevant doses of multiple radiation types. This new information and improved modeling approach allowed better estimation of dose response shapes and radiation effectiveness metrics in the space-relevant dose region. Specifically, we predicted the tumor yield for a Mars-mission-relevant mixture of radiation types, using the new Incremental Effect Additivity (IEA) synergy theory, which addresses many of the weaknesses of the simple effects additivity (SEA) theory that is commonly used. These improvements can provide enhanced estimation of cancer risks in astronauts.

NOTE: End date changed to 9/30/2021 per PI and NSSC information (Ed., 3/4/21)

NOTE: End date changed to 4/14/2021 per NSSC information (Ed., 1/16/2020)

Aim 1. Quantitatively assess GI tumorigenesis in mouse models of GI cancer and collect samples for qualitative analysis.

Aim 2. Dissection of the signaling events and consequences in GI cells of the persistent effects of space radiation.

Aim 3. Quantitatively assess lung tumorigenesis and test potential countermeasures in lung cancer susceptible mouse model.

Aim 4. Development of mathematical modeling for cancer risk assessment.

GI-tumorigenesis study (Project-1, Georgetown University): In the last six years of NSCOR funding, we have successfully applied mouse model approaches to acquire quantitative and qualitative GI tumorigenesis data to address prevailing uncertainties in risk modeling. Our overall goal is to determine the relative biological effectiveness (RBE) of various parameters for space radiation compared to terrestrial radiation. The risk estimates can then be extrapolated to space radiation using a RBE scaling factor. Our strategy is to i] quantitatively analyze tumor incidence and grade following space type radiation exposure. ii] Assess potential dose rate and gender variation. iii] Delineate cancer progression-specific biological signatures associated with space radiation exposure. Since low-linear energy transfer (LET) gamma radiation risk estimates are available for human populations from studies in atom bomb cohorts and like, we will then be able to combine our data to model the relative risk for space radiation. Pointwise details of progress made for GI tumorigenesis study are provided below:

1. Beam run participation in 2020: Georgetown University team participated in two-beam runs (NASA Space Radiation Laboratory (NSRL) 20B and 20C)), while planned experiments in the Spring-run (20A) were canceled due to COVID-19 restrictions at Brookhaven National Laboratory (BNL)/NSRL.

2. Dose-rate effect analysis using acute and chronic full GCR sim: In order to understand dose rate effects on GI-tumorigenesis after full-spectrum GCR sim exposures, we completed our initial study using 50 cGy dose of acute and chronic full-spectrum (33-ion) GCR beams during the NSRL 20B run. This initial study using a limited number of mice revealed a good signal after both acute and chronic full-spectrum GCR sim exposures, relative to controls. However, additional data from more mice are required; this initial study provided crucial insight and validated our mouse model for future experiments using lower doses of chronic full spectrum GCR beams.

3. Role of HZE in GCR-induced GI-tumorigenesis: Our previous study comparing quantitative tumor data obtained from the equivalent doses of a single HZE beam to the 4-ion mixed-field GCR-induced tumorigenesis revealed an association between higher intestinal tumor number and dose received from the HZE fraction of mixed-field GCR. Therefore, in order to understand the relative contributions of HZEs components in a full GCRsim induced tumorigenesis, we also exposed male Apc1638N/+ to p+He (45.7 cGy) and HZE only (4.3 cGy) during NSRL 20C run and quantitative tumorigenesis data will be obtained in early 2021. Once quantitative data from p+He (45.7 cGy) and HZE only (4.3 cGy) exposures are available, a comparison with 50 cGy full-GCRsim is planned to understand and model the relative role of HZE in GCR-induced tumorigenesis.

4. Role of non-targeted effects (NTEs) in HZE-induced tumorigenesis: In order to understand the role of non-targeted effects (NTEs) in HZE-induced tumorigenesis, male Apc1638N/+ mice (6-8 weeks) were whole-body exposed to sham-radiation, gamma-rays, and 28Si-ion at 10 cGy dose and mice were sacrificed at 60 d and 150 d after radiation exposure and intestinal tumor frequency were scored and analyzed. Our analysis of tumorigenesis rate data clearly indicated a remarkable acceleration in tumorigenesis rate at the late time point in 28Si exposed mice, relative to gamma-rays and control groups.

5. Addition of acute neutron beam to our range of single beam GI tumorigenesis data: Male Apc1638N/+ mice (6-8 weeks) were whole-body exposed to sham-radiation, gamma-rays, 4He, 12C, 16O, 28Si, and 56Fe-ion at 10 cGy dose at NSRL, while neutron (10 cGy) exposure was conducted at Radiological Research Accelerator Facility (RARAF) (Columbia University). Mice were sacrificed at 150 d after radiation exposure and intestinal tumor frequency was scored and analyzed. The highest number of tumors was observed after 28Si followed by 56Fe, neutron, 16O, 12C, and 4He radiation. No statistical difference in tumorigenesis count was evident between neutron, 16O, and 12C at 10 cGy dose.

Dissection of the signaling events and consequences in GI cells of the persistent effects of space radiation (Project-2, Georgetown University): Proliferative gastrointestinal (GI) tissue is radiation-sensitive, and heavy-ion space radiation with its high-linear energy transfer (high-LET) and higher damaging potential than low-LET gamma-rays is predicted to compromise astronauts' GI function. However, much uncertainty remains in our understanding of how heavy ions affect coordinated epithelial cell migration and extrusion, which are essential for GI homeostasis. The overall goal of this project is to understand the long-term effects of space radiation on normal gastrointestinal (GI) cells including stem cells. Coordinated epithelial cell migration is key to maintaining functional integrity and preventing pathological processes in gastrointestinal (GI) tissue and is essential for astronauts’ health and space mission success. Pointwise details of progress made for persistent effects of space radiation on GI-tissues are provided below:

1. We studied space radiation effects on intestinal stem cells (ISC) in male Lgr5 (Lgr5-EGFP-IRES-creERT2) mice in C57BL/6J background using 0.5 Gy of gamma-ray or 28Si ions. Lgr5 mice were exposed to 0.5 Gy of 28Si-ion and GFP+ Lgr5 cells were sorted from the isolated intestinal cells using flow-cytometry. Sorted cells were analyzed on flow cytometer for mitochondrial and cellular reactive oxygen species (ROS) using Mitosox or CellRox staining. Higher intracellular and mitochondrial ROS was observed in the 28Si exposed group, relative to gamma-ray and control GI stem cells.

2. We also demonstrated an increased level of beta-gal (senescence marker) positive intestinal stem cells in the 28Si exposed group, relative to gamma-ray. This suggests a higher number of stem cells were undergoing senescence after 28Si relative to gamma-ray or control group. This also correlated well with an increased number of gamma-H2AX foci in GI stem cells after 28Si compared to gamma ray or control group.

3. Emerging evidence indicates that senescent cells acquire a secretory phenotype (SASP) and production of inflammatory cytokines that lead to altered intestinal tissue homeostasis. A key pro-inflammatory secretory cytokine IL8 (SASP marker) was co-stained with p21 (senescent marker) and the data shows an increased level of IL8 expression in senescent cells (high p21 marker) indicating acquisition of the senescence-associated secretory phenotype (SASP) in a subset of senescent ISCs after 28Si exposure compared to gamma ray or control group.

4. A time-dependent analysis of persistent stress was conducted on samples collected at 2, 5, and 12-months post-exposure that indicated a progressive increase in DNA damage (gamma -H2AX), senescence (p16 and p21), and SASP (IL8) cells after 28Si exposure compared to gamma ray or control group.

5. Characterization of heavy ion-induced persistent oxidative modifications to DNA: We detected and quantified the spectrum of oxidative DNA base damage species using gas chromatography (GC)/mass spectrometry (MS) techniques two months after radiation exposure. Levels of damaged DNA base were significantly higher after 56Fe irradiation relative to sham-irradiated control as well as gamma-irradiated samples.

6. We also assessed the status of the repair pathways involved in repairing damaged DNA bases. Expression of BER (base excision repair) and NER (nucleotide excision repair) genes involved in the repair of identified DNA damage species was assessed in irradiated as well as in control samples. 56Fe exposure caused significant downregulation of many BER (Ogg1, Nth1, Neil1, Neil2, Neil3, Ape2, Xrcc1, and Lig3) and NER genes (Rbx1, ddb1, Ddb2, Xpc, Ercc8, Xpa, and Xpf). Additionally, immunoblot analysis of both BER (OGG1, NEIL1) and NER (DDB1, RBX1, CSB, NEIL1, and ERCC8) proteins showed significantly lower expression after 56Fe, relative to gamma radiation. We demonstrated a quantitatively higher accumulation of base damages that was associated with greater downregulation of BER and NER pathways after 56Fe, relative to gamma radiation.

Quantitatively and qualitatively compare effects of GCR-type irradiations on lung cancer initiation and progression in lung cancer susceptible mice. (Lead: Jerry W Shay) (Project-3, UTSW): The UTSW (University of Texas Southwestern) component of this project involves lung tissue from normal WT mice, and a mouse model of lung cancer (LA-1). The rationale for these models is that whole body exposure to space radiation will affect organ/tissues and cells in different ways. The lung only turns over cells with damage, while the colon turns over in humans every 7 days irrespective of damage. Thus, to model increased risks of more lethal cancers from space radiation exposure it is important to examine tissues with both a high rate of cellular turnover and tissues with a low rate of cell turnover. The inclusion of a lung model reflects the major solid tumor risk for astronauts traveling in space where the radiation-induced lung cancer incidence risk was estimated to be approximately 1% and 2.5% for male and female astronauts on a Mars mission, respectively, with upper 95% CIs extending well above 3% in both cases.

The overall hypothesis for this project is that space radiation induces molecular manifestations of a pro-inflammatory response as a function of radiation type, radiation doses, doses rates, LET value, and time that leads to increased risk of cancer initiation and progression. There remain uncertainties about cancer risks with single beam experiments and even simplified GCR simulations, but most experiments at the NSRL suggest late cancer effects are likely. It can be argued that we cannot predict what would happen in a mixed beam of fixed doses based on current data since most experiments were done in the acute setting with higher dose rates. While we recognize that there are issues of scaling from a mouse to human, and actually phenocopying the continuous low dose rates in the space environment, we have a large amount of legacy radiation carcinogenesis data and molecular baselines with mouse models using single beam experiments, a simplified 3 beam fast switching series of experiments, and now experiments with the 33-beam official full spectrum GCR simulation. We observed several hallmarks of cancer (even using a total dose of 30 cGy) in both our lung and colon cancer susceptible mouse models. Since our original NSCOR was focused on colon cancer, we had several mouse models of susceptibility to colon cancer and several papers have been published. However, the former space radiation element scientist recommended we add a lung cancer model for the GCR experiments in 2018 as part of our NSCOR.

We were the first team testing the NSRL-GCR 33 beam simulation with our lung cancer susceptible mouse model compared to WT mice starting in late 2018. Results from these studies were designed to evaluate the effects of both acute and chronic doses of the GCR simulator on radiation-induced health risks using a radiation-induced lung cancer in male and female wildtype and LA1 mutant (a lung cancer susceptible mouse) mice on a 129 background.

Each study included an acutely exposed group, a fractionated exposed group with consecutive doses of 20.8 mGy/day delivered over a four-week period, and sham controls. The simulator was baselined to 50 cGy for the first series of experiments. Acute exposures were delivered half-way through the fractionated exposure to support age-matching post irradiation During this initial experimental campaign approximately 300 mice received irradiations and ~200 mice treated as sham controls. The NSRL 18C (Fall 2018) run was a highly valuable experience in establishing concept of operations for future studies and providing a realistic determination of set-up and run times for future schedule planning.

Cancer risk modeling (Columbia University) (Project-4, CU): A trip to Mars and back, or another lengthy space exploration mission, will result in exposure of astronauts to a multi-component mixture of ionizing radiations which damage cells in multiple ways. Such damage mechanisms can be roughly classified into two categories: (1) Targeted effects (TE), involving the consequences of direct traversals of cells by ionizing tracks leading to DNA double strand breaks and other lesions. (2) Non-targeted effects (NTE, or “bystander” effects) caused by release of signals from cells directly hit by the tracks and the effects of these signals on other cells. The presence of both types of effects generates complex dose response shapes for space radiations, particularly in the low dose region relevant for space missions.

Here we used our mechanistically-motivated model formalism that includes both TE and NTE to analyze updated mouse tumorigenesis data from the current NSCOR. The updated data set now includes more detailed information on space-relevant doses of multiple radiation types. This new information and improved modeling approach allowed better estimation of dose response shapes and radiation effectiveness metrics in the space-relevant dose region. Specifically, relative biological effectiveness (RBE) and radiation effects ratio (RER) values at such doses of heavy ions and protons were estimated based on the most detailed data set version and improved methodology for describing variability in tumor numbers per mouse. These improvements provided updated and enhanced estimation of cancer risks in astronauts.

We continue to employ two models of GI cancers: the APC1638N/+ model, which was generated by targeted modification of the Apc gene at a position corresponding to aa1638. These mice develop very few adenomas in the small as well as in the large intestine but post-radiation tumor numbers show marked increases. Radiation also induced substantial gastric tumorigenesis in this model. The second model is the CDX2P Apc flox/+ that has one APC allele inactivated almost exclusively in the colon leading to adenomas and subsequent low rate of adenocarcinomas in the colon that is increased post-irradiation. Data being collected from these two relevant mouse models will be used to calculate risk estimates for NASA. Since low-LET risk estimates are available for human populations, we will then be able to model the relative risk for space radiation. In particular, this ratio can then be extrapolated to human exposure to low-LET radiation to then make an estimate of risk for colorectal cancer (CRC) in humans exposed to space radiation compared to the standard risks based on atomic bomb survivors and other exposure populations.

In order to answer the critical question of risk for colonic tumorigenesis after acute and chronic space radiation exposures, in vitro human cell culture and animal models for CRC initiation and progression are needed. We are investigating the effects of high-LET particle and high-energy protons in colonic epithelial cells with respect to the nature of damage and consequent tumor induction using biological cancer-related endpoints and systems biology (genomics, proteomics, and metabolomics) approaches. In addition, we are comparing these results to the biological consequences of low dose rates and mixed fields of radiation which astronauts are likely to be continuously exposed. The studies of low- and high-LET radiation on colonocytes gain more relevance with increasing age when premalignant adenomas and other colonic lesions such as flat lesions and microscopic aberrant crypt foci become more prevalent. Pre-existing undetectable genetic or chromosomal aberrations in astronauts in addition to de novo mutations resulting from long-term irradiation may increase the chances of colonic stem cell malignant transformation.

Plans for this proposal build on our past accomplishments with GI tumorigenesis models and are focused on studies at low doses (5 to 50 cGy) with an expanded range of NASA’s high priority beams. This proposal is also focused on assessing the effects of gender, dose-rate, age, and genetic heterogeneity in mouse models. Molecular events triggered by space radiation will be investigated in-depth and the mechanistic basis for persistent aberrant cell signaling after HZE radiation will be dissected. Genome-wide NGS combined with other omics approaches will contribute to a systems biology understanding of the effects of space radiation in the intestines and stomach. Both quantitative murine tumor data and results for underlying molecular events in mouse and human GI cells will contribute to the modeling of human risks for GI cancer by space travel. The overall aims of this NSCOR are summarized below:

Aim 1 (Project 1). Quantitatively assess GI tumorigenesis in mouse models of GI cancer and collect samples for qualitative analysis.

Aim 2 (Project 2). Dissection of the signaling events and consequences in GI cells of the persistent effects of space radiation.

Aim 3 (Project 3): Characterization of IR-induced neoplastic events in normal diploid human colonocytes.

Aim 4 (Project 4). Development of systems biology (Project 4A) and mathematical modeling (Project 4B) approaches for GI cancer risk assessment.

Task book accomplishments using APC1638N/+ and Lgr5 (Lgr5-EGFP-IRES-creERT2) mice:

1. Proton tumorigenesis study: In our earlier proton beam studies we have compared results of acute vs. fractionated exposure at high doses (1.2 Gy vs 0.4 Gy x3 fractions). Here, we exposed male and female APC1638N/+ to 50 cGy of proton radiation and tumors in the GI tract were counted after 150 days post-exposure. Further, proton tumorigenesis data was compared to the already acquired GI-tumorigenesis data after 50 cGy exposure of 4He, 12C, 16O, 28Si, and 56Fe –ions to understand radiation quality effects on GI-tumorigenesis.

2. Qualitative analysis of HZE-induced tumors: HZE-ion exposure led to higher carcinoma relative to gamma-rays. The highest % of carcinoma was observed after 56Fe radiation (25% were carcinoma) at 10 cGy dose. A similar trend of highest percentage pf carcinoma after 56Fe radiation was also observed after 50 cGy as well as after 2 Gy equitoxic doses of heavy-ion radiation.

3. Gastric tumorigenesis study: Male APC1638N/+ mice were whole-body exposed to sham-radiation, gamma-rays, and 28Si (300 MeV/n) in summer-2019 and mice were sacrificed150 days after irradiation, and gastric tumor-frequency was scored. A significant increase in gastric tumorigenesis after 10 cGy of 28Si (~3 fold higher) was observed compared to the same dose of gamma-rays.

4. Mixed-field GI tumorigenesis study in APC1638N/+ mice: Male and female APC1638N/+ mice were whole-body exposed to sham-radiation and mixed-field GCR type beam in fall of 2018 and mice were sacrificed 150 days after irradiation, and gastrointestinal tumor-frequency was scored. We compared quantitative tumor data obtained from the equivalent doses of single HZE beam to the mixed-field GCR-induced tumorigenesis that revealed a higher relative contributions of the dose received from HZE.

5. Persistent effects study: Lgr5+ mice (6 to 8 weeks) were exposed to 50 cGy of 28Si-ions and intestinal stem cells were isolated through flow-sorting and markers of persistent stress, senescence, and SASP are being analyzed. Additionally, these samples are also being used for next-gen transcriptomics (RNAseq) and epigenomic studies.

6. Metformin risk mitigation study: Tested risk mitigation efficacy of metformin (0.1% in diet). The male APC1638/+ mice were put on metformin diet at 4 weeks of age and continued on metformin to the end of the study. Mice were exposed to 28Si-ion (10 cGy) at 8 week of age and GI-tumorigenesis was estimated at 150 days post-exposure. Metformin administration led to a statistically significant reduction in HZE-induced GI-tumorigenesis.

In addition to our primary focus on GI-tumorigenesis using APC-based mouse models (APC1638N/+ and CDX2P Apc flox/+), in consultation with NASA, we also added a lung cancer mouse model (LA1) to study space radiation-induced lung tumorigenesis and mitigation study using CDDO.

Task book accomplishments using CDX2P Apc flox/+ and LA1 mice:

1. Simulated solar particle events (sSPE) radiation induces prolonged DNA damage, and senescence-associated inflammation in colon tissues.

2. sSPE increases colon cancer initiation and progression in CPC;APC mice. Increased p53 mutations in colon cancer after sSPE exposure was also observed.

3. SPE-irradiation results in a ~3-fold increased incidence of colon carcinoma in the CPC;APC mouse model while administration of CDDO diet during irradiation exhibit fewer polyps in comparison to mice on control diet.

4. CDDO provided 3 days prior to radiation exposure protected mice from sSPE induced colon cancer initiation and progression. The CPC;APC mice on CDDO diet during solar particle irradiation simulations exhibit a decreased incidence in invasive carcinoma in comparison to mice on control diet.

5. 5cGy (300 MeV/n) of a single dose exposure to Silicon does not increase cancer initiation or overall survival but 10 cGy results in a >2-fold cancer initiation and an overall decrease in lifespan.

6. LA1 mice on CDDO diet during fractionated 56Fe- irradiation exhibit a decreased incidence in invasive carcinoma in comparison to mice on control diet.

7. LA1 mice on CDDO diet during solar particle irradiation simulations exhibit a decreased incidence in invasive carcinoma in comparison to mice on control diet.

8. LA1 mice fed CDDO without irradiation show reduced tumor progression.

9. Mixed beam studies were also carried out in CDX2P Apc flox/+ mice using both acute and protracted GCR sim beans.

Task book accomplishments on genetic and epigenetic analysis of space radiation tumor and normal tissues: Using next-gen sequencing approaches, we defined the features of the tumor genomes arising in our murine models after gamma, proton, and HZE exposure. During 2019, efforts have focused on two primary areas described in detail below:

1. DNA sequence analysis of mouse colon tumor samples arising after radiation exposure: Using our well established next-generation sequencing pipeline, we first used a targeted approach to sequence the Apc and Trp53 regions. From targeted sequencing, there is evidence that the loss of the Apc region is the primary mechanism of tumor initiation. However, inactivating mutations in Apc without LOH have been identified in some tumors from Si irradiated mice. These results have been validated with the sensitive digital droplet PCR assay. At a whole-genome scale, we will have the potential to detect differences in the mutation signature between these groups including the presence of structural rearrangements as well as small mutations. We have now obtained whole-genome sequencing (WGS) on some samples and are in the process of obtaining WGS on a collection of tumors from control mice and those exposed to gamma or 28Si.

2. Development of a targeted bisulfite-seq platform for the epigenetic analysis of irradiated mouse tissue: It has now been firmly established that HZE radiation leads to a persistent disturbance in the architecture of the gut epithelium. To investigate the mechanistic basis of this apparently epigenetic phenomenon, we are utilizing the Lgr5-GFP model established in the Fornace lab in addition to the Apc cancer-prone models. However, there are limited commercial tools for large-scale analysis of DNA methylation in the mouse genome. While whole-genome bisulfite sequence is possible, it is prohibitively resource-intensive. Our solution is the implementation of a large-scale targeted bisulfite sequencing platform capturing 69.9 Mb with 128,614 oligonucleotides. This provides a reasonable sampling of the mouse genome. After validating this system with high molecular weight DNA, we have initiated the analysis of DNA samples extracted from fixed tissues harvested from experimental and control mice. Results are excellent with clear delineation of CpG methylation status at the nucleotide level. We are now processing 24 samples and beginning the search for differentially methylated regions that may be related to radiation exposure.

Task book accomplishments on risk modeling using mouse tumorigenesis data:

Astronauts embarking on a trip to Mars and back or on other lengthy space exploration missions will encounter a multi-component mixture of ionizing radiations. These radiations damage cells in multiple ways, which can be roughly classified into two categories:

1. Targeted effects (TE), involving the consequences of direct traversals of cells by ionizing tracks leading to DNA double-strand breaks and other lesions.

2. Non-targeted effects (NTE, also called “bystander” effects) caused by the release of signals from cells directly hit by the tracks and the effects of these signals on other cells.

The presence of both TE and NTE can produce complex dose response shapes for space radiations, particularly in the low dose region that is most relevant for space missions.

Mechanistically-motivated radiation carcinogenesis models serve as important tools to describe these dose responses and make estimates and predictions of cancer risks from space exploration. Here we used our model that includes both TE and NTE to analyze updated mouse tumorigenesis data from the current NSCOR. The updated data set now includes more detailed information on space-relevant doses of multiple radiation types. A more detailed modified model formalism that uses a weighted Negative Binomial distribution to better describe the data variability was developed and implemented on this data set. This new information and improved modeling approach allowed a better estimation of dose-response shapes in the space-relevant dose region. Specifically, NTE parameters and relative biological effectiveness (RBE) and radiation effects ratio (RER) values at such doses of heavy ions and protons were estimated with more information than previously possible. Ultimately, these improvements provided updated and enhanced estimation of gastrointestinal cancer risks in human astronauts.

We continue to employ two models of GI cancers: the APC1638N/+ model, which was generated by a targeted modification of the Apc gene at a position corresponding to aa1638. These mice develop very few adenomas in the small as well as in the large intestine but post-radiation tumor numbers show marked increases. Radiation also induced substantial gastric tumorigenesis as well in this model. The second model is the CDX2P Apc flox/+ that has one APC allele inactivated almost exclusively in the colon leading to adenomas and subsequently a low rate of adenocarcinomas in the colon that is increased post-irradiation. Data being collected from these two relevant mouse models will be used to calculate risk estimates for NASA. Since low-LET risk estimates are available for human populations, we will then be able to model the relative risk for space radiation. In particular, this ratio can then be extrapolated to human exposure to low-LET radiation to then make an estimate of risk for colorectal cancer in humans exposed to space radiation compared to the standard risks based on atomic bomb survivors and other exposure populations.

In order to answer the critical question of risk for colonic tumorigenesis after acute and chronic space radiation exposures, in vitro human cell culture and animal models for colorectal cancer (CRC) initiation and progression are needed. We are investigating the effects of high-LET particle and high-energy protons in colonic epithelial cells with respect to the nature of damage and consequent tumor induction using biological cancer-related endpoints and systems biology (genomics, proteomics, and metabolomics) approaches. In addition, we propose to compare these results to the biological consequences of low dose rates and mixed fields of radiation which astronauts are likely to be continuously exposed. The studies of low- and high-LET radiation on colonocytes gain more relevance with increasing age when premalignant adenomas and other colonic lesions such as flat lesions and microscopic aberrant crypt foci become more prevalent. Pre-existing undetectable genetic or chromosomal aberrations in astronauts in addition to de novo mutations resulting from long-term irradiation may increase the chances of colonic stem cell transformation.

Plans for this proposal build on our past accomplishments with GI tumorigenesis models and are focused on studies at low doses with an expanded range of NASA's high priority beams. This proposal is also focused on assessing the effects of gender, dose-rate, age, and genetic heterogeneity in mouse models. Molecular events triggered by space radiation will be investigated in depth and the mechanistic basis for persistent aberrant cell-signaling after HZE radiation will be dissected. Genome-wide NGS combined with other omics approaches will contribute to a systems biology understanding of the effects of space radiation in the intestines and stomach. Both quantitative murine tumor data and results for underlying molecular events in mouse and human GI cells will contribute to modeling of human risks for GI cancer by space travel. The overall aims of this NSCOR are summarized below:

Aim 1 (Project 1). Quantitatively assess GI tumorigenesis in mouse models of GI cancer and collect samples for qualitative analysis

Aim 2 (Project 2). Dissection of the signaling events and consequences in GI cells of the persistent effects of space radiation

Aim 3 (Project 3): Characterization of IR(infrared)-induced neoplastic events in normal diploid human colonocytes

Aim 4 (Project 4). Development of systems biology (Project 4A) and mathematical modeling (Project 4B) approaches for GI cancer risk assessment.

After consultation with NASA, we have also carried out pilot studies to assess the potential protective effect of aspirin, which is associated with decreased frequency of human CRC. While aspirin has been reported to decrease spontaneous tumorigenesis in APC mutant mice, the experimental approach was developed to test effectiveness of aspirin in preventing/decreasing 28-Si-induced intestinal tumorigenesis in APC1638N/+ mice. We used 50 and 10 cGy 300 MeV/n 28-Si ions (LET: 70 keV/micron). We chose these doses because these are the lowest doses that demonstrated significantly higher tumorigenesis relative to sham irradiation. We have chosen two doses because 50 cGy radiation induced higher tumor relative to 10 cGy and at 10 cGy due to lower tumor number signal to noise ratio for the aspirin effect on tumorigenesis could be problematic. Aspirin was administered in diet and three doses used were equivalent to human daily doses of 75, 150, or 300 mg aspirin tablet. We used three color-coded diets containing 61.5 mg, 123 mg, 246 mg aspirin, respectively, per kg of diet. We have chosen 3 human doses of aspirin because CRC chemoprevention studies in humans have reported aspirin dose range from 75 to 300 mg/day and this dose range was also effective in removing existing adenoma. We used daily aspirin administration schedule in 6 to 8 weeks old APC1638N/+ male mice and aspirin administration was started 1 month before irradiation and was continued until the end of the study. Mice were sacrificed 150 d after radiation and intestinal and colon tumors counted. Data show that aspirin did alter the course of space radiation-induced intestinal tumorigenesis.

Task book accomplishments 2018 using CDX2P Apc flox/+ mice: Completed 3 beam fast switching experiments (30 cGy total exposure) and demonstrated that the order of beams matter (e.g., protons, then helium followed by silcon) was the most carcinogenic

1. 5 cGy of silcon did not increase carcinogenesis, nor did 20-30 cGy of proton or 5 cGy of helium. Combining 20cGy of proton and 5 cGy of helium did not increase cancer progression.

2. Using fast beam switching, 30 cGy total IR proton-helium-silicon increased cancer progression as did 27 cGy total IR (e.g., 2 cGy silicon vs 5 cGy silicon). However, 25.5 cGy of total IR (20 cGy proton, 5 cGy helium and 0.5 cGy silicon did not increase cancer progression in mouse models of cancer.

3. Using the fast beam switching and treating mice with 20 cGy protons, then 5 cGy helium and then waiting 24 hours before exposing the same mice to 5 cGy of silicon did not result in increased carcinogenesis. This suggests repair of the proton and helium damage over 24 hours was sufficient to reduce the cancer progression.

4. Pretreating mice with the countermeasure CDDO in chow for 2 days prior to the 3 beam fast switching (30 cGy total IR) was sufficient to reduce cancer progression. However, similarly feeding mice with aspirin had no effect and cancer progression.

5. Official GCR (galactic cosmic radiation) simulation: We were the first group to treat mice with an acute 50 cGy IR with 5 proton energies, 5 helium energies, and 5 heavy ions (NSRL 18B). This was followed in NSRL 18C with a 4 week protracted exposure to 50 cGy GCR simulation.

Results are in progress but initial results 100 day post IR suggest acute and protracted are equivalent in regards to chronic and persistent oxidative damage as measure by increases in lipid peroxidation. The overall goal of this project is to understand long-term effects of space radiation on normal gastrointestinal (GI) cells including stem cells. Coordinated epithelial cell migration is key to maintaining functional integrity and preventing pathological processes in gastrointestinal (GI) tissue and is essential for astronauts’ health and space mission success. The GI tract with its high cell turnover is sensitive to radiation, and the effects of radiation are evident when the replacement process of cells lost during normal turnover are deregulated. In small intestine, considered a model system to study GI epithelial cell turnover, differentiated epithelial cells from crypt-base stem cells migrate along the crypt-villus axis to replace cells that shed into intestinal lumen via apoptosis at the villus tip. The current study irradiated wild type (Wt) mice (C57BL/6J; male; 6 to 8 weeks old; n=10) with 50 cGy 56-Fe (energy: 1000 MeV/nucleon; LET: 148 keV/micron) using simulated space radiation source at the NASA Space Radiation Laboratory (NSRL), Brookhaven National Laboratory (BNL). Mice were sacrificed either 7 d, 60 d, or 12 m after radiation exposure and cell migration in small intestine was assessed quantitatively and qualitatively. Iron radiation data were compared to an equal dose of gamma-rays data. We demonstrated that IEC migration along the crypt-villus axis is decreased up to 2 months after a low dose of heavy ion radiation relative to gamma-rays. Heavy ion radiation-induced decreased migration was associated with perturbation of molecular pathways involved in coordinated intestinal epithelial cell migration. We used also small intestine as a model system to delineate space radiation effects on GI stem cells. The intestine is a self-renewing tissue with an active crypt-base proliferative stem cell compartment and is sensitive to radiation exposures. We studied space radiation effects on intestinal stem cells (ISC) in male Lgr5 (Lgr5-EGFP-IRES-creERT2) mice in C57BL/6J background using low doses (<1 Gy) 56-Fe (1000 MeV/n) ions. Lgr5 mice were exposed to 0.5 Gy of 56-Fe and GFP+ Lgr5 cells were sorted from isolate intestinal cell using flow cytometry. Intracellular ROS was assessed in sorted intestinal stem cells (ISCs) using fluorescent probe CellROX and flow cytometry. Data show increased reactive oxygen species (ROS) and mitochondrial superoxide in ISCs 2-month after heavy ion radiation. We also show increased gammaH2AX and 53BP1 foci in iron irradiated samples suggesting continued presence of sub-lethal label of DNA DSB (double strand breaks) in these live sorted ISCs. We stained ISCs for PCNA, a known cell proliferation marker and showed increased PCNA staining supporting ISC proliferation in presence of chronic stress and DNA damage with potential for transformation. Future plan involves understanding whether stem cell stress is propagated to progeny and developing countermeasure strategy to block space radiation-induced persistent stem cell stress.

The goal of this project is to define the features of the tumor genomes arising in our murine models after gamma, proton, and HZE exposure. Using our well established next generation sequencing pipeline, we used a targeted to sequence the Apc and Trp53 regions to establish tumor purity and to identify the best samples for larger scale analysis. At the same time, we can identify any mutations occurring in these regions. Our data illustrate single nucleotide polymorphism analysis of the Apc region from a representative group of 12 tumors and 7 normal adjacent tissues. Allelic imbalance is observed in all tumors, but there is considerable variation in tumor purity. Over 50 tumors from the Fornace and Shay labs have been sequenced. In this fashion, we have identified the samples with the least dilution by normal DNA that are most suited to large scale sequencing. We are currently accessioning new samples in order to have sufficient material to test the hypothesis that HZE radiation has a recognizable mutation signature. From targeted sequencing, there is clear evidence that loss of the entire region is the primary mechanism of tumor initiation under all conditions. No single nucleotide mutations, small INDELS, or structural rearrangements in or flanking Apc have been identified. Excellent whole-genome copy number profiles can also be derived from this data. Our current analyses indicate at the dose levels used in this study, the tumor mutation burden from either chromosome rearrangements, aneuploidy, or small mutations (single nucleotide or small indels) is remarkably low. We have now initiated whole genome sequencing on selected samples to validate this conclusion at maximum resolution.

It has now been firmly established that HZE radiation leads to a persistent disturbance in the architecture of the gut epithelium. To investigate mechanistic basis of this apparently epigenetic phenomenon, we are utilizing the Lgr5-GFP model established in the Fornace lab in addition to the Apc cancer prone models. The Lgr5-GFP system presents the opportunity to carry out transcriptomic analysis of cells isolated from irradiated and control mice and flow sorted into groups which reflect stem cells (GFP high), proliferating cells (GFP low), and differentiated cells (GFP negative). Cells from control, gamma, and HZE irradiated mice were obtained in triplicate, sorted by GFP expression, and analyzed by RNA-Sequencing.

Analysis of the data suggests that distinct differences in gene expression are observed at 120 days post-irradiation within the stem cell (high GFP) samples. We are currently following up on this data with pathway gene set enrichment analyses and are preparing to validate and refine our results using single cell RNA-Seq techniques. To complement the transciptome studies, we plan to use DNA for targeted bisulfite sequencing of 850,000 CpG sites distributed across the mouse genome. Ultimately, the integration of transcriptional data with DNA methylation status could provide important insights into the mechanism and consequences of the persistent effects of HZE exposure. There may be potential to develop blood based assays to detect the signature of HZE DNA methylation.

Astronauts on long-duration space exploration missions such as a planned trip to Mars and back will be exposed to a complex mixture of radiations that is not encountered on Earth. This space radiation mixture includes heavy ions, which produce difficult to repair clustered damage in cells and can be much more carcinogenic than gamma rays. In such situations, non-targeted effects caused by release of signals from cells directly hit by heavy ion track cores are likely to be the dominant mechanism producing adverse health effects from heavy ion. Reliable estimates for the endpoint-specific dose-dependent weighting function, which take non-targeted effects into consideration, are needed to predict human cancer risks from heavy ion exposure during space missions, based on available data for gamma rays. We used a mechanistically-motivated mathematical model to analyze updated mouse tumorigenesis data from the current NSCOR and to generate convenient weighting function equations and estimates for several space-relevant heavy ion types, compared with gamma rays. The updated data set now includes two additional ion types and lower, more space-relevant, doses. The model-based weighting function estimates are dose-dependent and tend to be highest (>10) at low doses, where non-targeted effect terms dominate the radiation response.

We continue to employ two models of GI cancers: the APC1638N/+ model, which was generated by a targeted modification of the Apc gene at a position corresponding to aa1638. These mice develop very few adenomas in the small as well as in the large intestine but post-radiation tumor numbers show marked increases. Radiation also induced substantial gastric tumorigenesis as well in this model. The second model is the CDX2P Apc flox/+ that has one APC allele inactivated almost exclusively in the colon leading to adenomas and subsequently a low rate of adenocarcinomas in the colon that is increased post-irradiation. Data being collected from these two relevant mouse models will be used to calculate risk estimates for NASA. Since low-LET risk estimates are available for human populations, we will then be able to model the relative risk for space radiation. In particular, this ratio can then be extrapolated to human exposure to low-LET radiation to then make an estimate of risk for colorectal cancer in humans exposed to space radiation compared to the standard risks based on atomic bomb survivors and other exposure populations.

In order to answer the critical question of risk for colonic tumorigenesis after acute and chronic space radiation exposures, in vitro human cell culture and animal models for colorectal cancer (CRC) initiation and progression are needed. We are investigating the effects of high-LET particle and high-energy protons in colonic epithelial cells with respect to the nature of damage and consequent tumor induction using biological cancer-related endpoints and systems biology (genomics, proteomics, and metabolomics) approaches. In addition, we propose to compare these results to the biological consequences of low dose rates and mixed fields of radiation which astronauts are likely to be continuously exposed. The studies of low- and high-LET radiation on colonocytes gain more relevance with increasing age when premalignant adenomas and other colonic lesions such as flat lesions and microscopic aberrant crypt foci become more prevalent. Pre-existing undetectable genetic or chromosomal aberrations in astronauts in addition to de novo mutations resulting from long-term irradiation may increase the chances of colonic stem cell transformation.

Plans for this proposal build on our past accomplishments with GI tumorigenesis models and are focused on studies at low doses with an expanded range of NASA's high priority beams. This proposal is also focused on assessing the effects of gender, dose-rate, age, and genetic heterogeneity in mouse models. Molecular events triggered by space radiation will be investigated in depth and the mechanistic basis for persistent aberrant cell-signaling after HZE radiation will be dissected. Genome-wide NGS combined with other omics approaches will contribute to a systems biology understanding of the effects of space radiation in the intestines and stomach. Both quantitative murine tumor data and results for underlying molecular events in mouse and human GI cells will contribute to modeling of human risks for GI cancer by space travel. The overall aims of this NSCOR are summarized below:

Aim 1 (Project 1). Quantitatively assess GI tumorigenesis in mouse models of GI cancer and collect samples for qualitative analysis.

Aim 2 (Project 2). Dissection of the signaling events and consequences in GI cells of the persistent effects of space radiation

Aim 3 (Project 3): Characterization of IR-induced neoplastic events in normal diploid human colonocytes

Aim 4 (Project 4). Development of systems biology (Project 4A) and mathematical modeling (Project 4B) approaches for GI cancer risk assessment.

We have continued irradiation of the APC1638N/+ mouse model to obtain quantitative and qualitative tumorigenesis data at lower doses of space radiation beams spanning a range of LET and energies. We participated in two beam runs in the spring and fall of 2017 and we are breeding and genotyping mice for the spring 2018 run at NASA Space Radiation Laboratory (NSRL). These beam runs have allowed us to irradiate mice for fulfilling statistically required numbers decided in consultation with our Project 4B lead Dr. Brenner at Columbia University. We are also assessing the effect of chronic aspirin administration on tumor frequency; there are human data indicating that low-dose aspirin can have a protective effect against CRC. Our previous data in APC mutant mouse model demonstrated higher intestinal tumorigenesis after exposure to 12C, 28Si, and 56Fe radiation relative to gamma rays. We also observed important qualitative differences between heavy ions and gamma rays in both normal GI epithelial cells and tumors. However, our lowest dose for these experiments was 10 cGy. This year we exposed APC1638N/+ male and female mice to 5 cGy of gamma-rays, 16O (325 MeV/n), 4He (250 MeV/n), 28Si (300 MeV/n), and 56Fe (1000 MeV/n) with an aim to assess tumorigenesis at doses lower than the previous ones with a broader range of ions. We also exposed mice to 10 cGy of 16O (325 MeV/n) and 4He (250 MeV/n), which are new beams proposed in the renewed NSCOR, to get statistically required numbers. In an effort to assess effects of genetic heterogeneity on space radiation-induced intestinal tumorigenesis, we performed backcrossing of APC1638N/+ in C57BL/6J background into C3H/HeN and BALB/c background. The backcrossing was done for 10 generations and Sequencing of the region spanning the APC gene locus after backcrossing for 10 generations indicate that much of the chromosomal DNA flanking this locus is now either from C3H/HeN or BALB/c and not from C57BL/6J. We have scheduled beam time in 2018 fall to irradiate APC1638N/+ mice in C3H/HeN and BALB/c background to compare tumorigenesis in these vs. C57BL/6J background.

We expanded our GCR simulations (GCRsim) using both colon and lung cancer susceptible mouse model in comparison to wild type mice. In our initial experiments we used the following beams: protons (1H), helium (4He) and silicon (28Si). We used fast switching of these beams at the NSRL using a dose rate of 0.5 cGy/min and the large 60 x 60 beam. The total dose mice were exposed to was either 30 or 27 cGy. We used 20 cGy of proton (120Mev/n), 5 cGy of helium (250 MeV/n) and either 2 or 5 cGy of silicon (300 MeV/n). We previously showed that 30 cGy of proton and 5 cGy of silicon as single beam exposures did not increase polyp size or number of polyps in the CPC;APC mouse model. However, we observed 100 days post-IR that the number and polyp sizes were increased in mice exposed to the GCRsim in comparison to protons alone or unirradiated mice. Initial results indicate differential expression in normal colonic tissue in the CPC;APC mice 100 days after GCRsim. These included increases in senescence-associated inflammatory response genes and an increase in senescence markers (such as p16INK4a). We also varied the order of the different ions. While the trend was similar irrespective of beam order, GCRsim that begin with protons and end with silicon showed higher expression of specific inflammatory genes and larger and more polyps compared to unirradiated or proton only irradiated mice. These inflammatory genes have been previously shown to correlate with poorer overall survival in humans with colon cancer. We are currently testing a model of why the order of irradiation matters. In addition, we have completed a small pilot series of experiments to test aspirin or CDDO (an anti-oxidant/anti-inflammatory agent that is orally available and currently in clinical development) as effective countermeasures of IR-induced gene expression changes and tumor progression. We are also comparing the results in the colon susceptible mice to a mouse model of lung cancer. Initial results indicate that mice irradiated with GCRsim at 100 days post-IR have higher levels of lipid peroxidation in the serum as well as an increased number of premalignant lung lesions.

Understanding molecular underpinning of space radiation-induced adverse effects in normal gastrointestinal (GI) tissue is a key factor for pathophysiologic risk assessment. This year we focused on assessing long-term effects of heavy ion radiation on autophagy in intestinal epithelial cells two months after exposure. Autophagy is a complex catabolic process involved in removing and recycling damaged or unwanted cellular constituents in autophagolysosomal vesicles. Reduced autophagy is known to promote cellular stress not only due to increased accumulation of damaged proteins but also due to decreased removal of damaged mitochondria (mitophagy). Our data in wild type C57BL/6J mice demonstrate that exposure to 1.6 Gy of 56Fe radiation led to decreased autophagy and increased oxidative stress suggesting that heavy ion radiation is propelling cells into persistent dysregulated homeostasis. Deregulation of intestinal homeostasis is further strengthened by the fact that there is increased PI3K/Akt signaling invariably linked to increased reactive oxygen species (ROS) production as well as Akt-mediated activation of mTOR, which is known to downregulate autophagy. The generated data is expected to contribute towards developing GI risk estimates of heavy ion radiation, which is qualitatively high linear energy transfer (high-LET) in outer space relative to gamma-rays, which are low-LET. Currently, the persistent effects studies have been expanded to additional priority beams at lower radiation doses and dose rates more relevant to the space environment. Additionally, we have focused on intestinal stem cell (ISC) studies in male Lgr5 (Lgr5-EGFP-IRES-creERT2) mice in C57BL/6J background using low doses (<1 Gy) 28Si (300 MeV/n) or 56Fe (1000 MeV/n) ions. Since we observed persistent stress in intestinal epithelial cells, we are investigating to determine if persistent stress is originating from stem cell and propagating to differentiated epithelial cells. Lgr5 mice were exposed to 0.5 Gy of 56Fe and GFP+ Lgr5 cells were sorted from isolated intestinal cell using flow cytometry. Sorted cells were pelleted by cytospin and the pellet was paraffin embedded for sectioning and immunostaining. Our stem cell data show that 56Fe radiation exposure was associated with increased oxidative stress and DNA damage in intestinal stem cells. We also show that at least some of the ISC are senescent (senescent marker Glb1 positive) suggesting the possibility of senescence associated secretory phenotype associated inflammatory stress. Future plan involves understanding whether stem cell stress is propagated to progeny and developing countermeasure strategy to block space radiation-induced persistent stem cell stress.

We have also isolated colonocytes from normal and colon cancer susceptible mice. In addition to determining the best methods to culture these cells in 2D monolayers, we have establish 3D organotypic cultures from normal colonocytes and those from colonic polyps. The cells expand in size and number in organotypic culture (Matrigel) over a week period and can be trypsinized and re-cultured in 2D or 3D again. In initial experiments we established 3D cultures from WT and from the CPC;APC mouse model and irradiated them with either low or high LET IR and then placed them back in 2D culture. In comparison we irradiated intact mice (WT and CPC;APC mice) with the same doses and then removed colonocytes and established them in 2D and 3D cultures. These cells are being passaged and tested for cancer associated changes. In other experiments we initiated experiments to compare the CPC;APC mouse model to other cancer susceptible models of cancer such as a lung cancer mouse model. Initial results indicate that human bronchial epithelial cells irradiated in 2D culture form more colonies in soft agar compared to 3D-irradiated cells. These results indicate that estimates of cancer progression based on 2D culture may overestimate the cancer risks. Going forward, we plan to test a biological countermeasure, CDDO, to determine if the changes observed can be reduced or prevented by either providing CDDO prior to or after a GCR simulation.

The goal of the genomics project is to define the features of the tumor genomes arising in our murine models after gamma, proton, and HZE exposure. Using our well established next generation sequencing pipeline, we used a targeted approach to sequence the Apc and Trp53 regions to establish tumor purity and to identify the best samples for larger scale analysis. At the same time, we can identify any mutations occurring in these regions. Single nucleotide polymorphism analysis of the Apc region from a representative group of 12 tumors and 7 normal adjacent tissues was performed. Allelic imbalance is observed in all tumors, but there is considerable variation in tumor purity. A total of 34 tumors from the Fornace and Shay labs were sequenced. In this fashion, we have identified tumors with the least dilution by normal DNA that are most suited to large scale sequencing. We are currently accessioning new samples in order to have sufficient material to test the hypothesis that HZE radiation has a recognizable mutation signature. From targeted sequencing, it appears that loss of the entire region is the primary mechanism of tumor initiation under all conditions. No single nucleotide mutations, small INDELS or structural rearrangements in or flanking Apc have been identified. Excellent, whole genome copy number profiles can also be derived from this data.

Astronauts performing space exploration missions will be exposed to heavy ion radiation, which can be considerably more carcinogenic per unit dose than terrestrial radiations such as gamma-rays or x-rays. Our project within the space radiation and gastrointestinal cancer NSCOR focuses on developing biologically based mathematical models to predict heavy ion induced cancer risks at doses and exposure durations relevant for space missions. Specifically, our modeling efforts are aimed at addressing the following important question: do the risks from protracted heavy ion exposures in space differ from the risks of such irradiation delivered over a short time in terrestrial experiments? We proposed a model of heavy ion carcinogenesis and fitted it to radiation lung carcinogenesis data in humans and rats. The model predicts that, at doses expected during astronaut missions, protraction of heavy ion exposure does not have a major effect on lung cancer risks. In other words, the risks from small doses of heavy ion irradiation delivered over a short time or a long time are likely to be similar. Overall, we plan to generate in vivo data on tumor frequency and mechanisms of tumor development using NASA mandated additional priority space radiation beams at lower doses relevant to space environment. These data will be used to reduce uncertainty in CRC risk prediction and develop reliable risk prediction model for humans.

We continue to employ two models of GI cancers: the APC1638N model, which was generated by a targeted modification of the Apc gene at a position corresponding to aa1638. These mice develop very few adenomas in the small as well as in the large intestine but post-radiation tumor numbers show marked increases. Radiation induced substantial gastric tumorigenesis as well in this model. The second model is the CDX2P Apc flox/+ that has one APC allele inactivated almost exclusively in the colon leading to adenomas and subsequent a low rate of adenocarcinomas in the colon that is increased post-irradiation. Data being collected from these two relevant mouse models will be used to calculate risk estimates for NASA. Since low-LET risk estimates are available for human populations, we will then be able to model the relative risk for space radiation. In particular, this ratio can then be extrapolated to human exposure to low-LET radiation to then make an estimate of risk for colorectal cancer in humans exposed to space radiation compared to the standard risks based on atomic bomb survivors and other exposure populations.

In order to answer the critical question of risk for colonic tumorigenesis after acute and chronic space radiation exposures, in vitro human cell culture and animal models for CRC initiation and progression are needed. We are investigating the effects of high-LET particle and high-energy protons in colonic epithelial cells with respect to the nature of damage and consequent tumor induction using biological cancer-related endpoints and systems biology (genomics, proteomics, and metabolomics) approaches. In addition, we propose to compare these results to the biological consequences of low dose rates and mixed fields of radiation which astronauts are likely to be continuously exposed. The studies of low- and high-LET radiation on colonocytes gain more relevance with increasing age when premalignant adenomas and other colonic lesions such as flat lesions and microscopic aberrant crypt foci become more prevalent. Pre-existing undetectable genetic or chromosomal aberrations in astronauts in addition to de novo mutations resulting from long-term irradiation may increase the chances of colonic stem cell transformation.

Plans for this proposal build on our past accomplishments with GI tumorigenesis models and are focused on studies at low doses with an expanded range of NASA’s high priority beams. This proposal is also focused on assessing the effects of gender, dose-rate, age, and genetic heterogeneity in mouse models. Molecular events triggered by space radiation will be investigated in depth and the mechanistic basis for persistent aberrant cell-signaling after HZE radiation will be dissected. Genome-wide NGS combined with other omics approaches will contribute to a systems biology understanding of the effects of space radiation in the intestines and stomach. Both quantitative murine tumor data and results for underlying molecular events in mouse and human GI cells will contribute to modeling of human risks for GI cancer by space travel. The overall aims of this NSCOR are summarized below.

Aim 1 (Project 1). Quantitatively assess GI tumorigenesis in mouse models of GI cancer and collect samples for qualitative analysis.

Aim 2 (Project 2). Dissection of the signaling events and consequences in GI cells of the persistent effects of space radiation.

Aim 3 (Project 3): Characterization of IR-induced neoplastic events in normal diploid human colonocytes.

Aim 4 (Project 4). Development of systems biology (Project 4A) and mathematical modeling (Project 4B) approaches for GI cancer risk assessment.

Task progress on GI tumorigenesis (GU). We have participated in three beam runs at NASA Space Radiation Laboratory (NSRL) after our current NSCOR award in 2015 with an aim to assess tumorigenesis at doses lower than the previous ones with a broader range of ions. Greater tumorigenesis was observed with 28Si ions in our previous studies relative to other ions and gamma rays. This year we exposed APC1638N/+ mice to 5 cGy of 28-Si (300 MeV/n) and we used male mice due to higher tumor frequency in male relative to female mice. This year we have also started experiments with 16-O (325 MeV/n) and 4-He (250 MeV/n) ions, which are NASA prioritized ions of lower Z value proposed in the NSCOR. In 2017, we are also approved for proton, 28-Si, 56-Fe, and 4-He beam time and we will use the time to irradiated APC1638N/+ mice to fulfill statistical requirements. Importantly, we completed animal numbers required for our low dose rate (0.33 cGy/min) studies, data has been compiled, and tumor data from low and high (50 cGy/min) dose rate exposures were compared. We observed that differences in dose rate did not affect tumor frequency or size. An important aspect of our proposal was to study effects of genetic heterogeneity on space radiation-induced intestinal tumorigenesis. Our backcrossing of APC1638N/+ of C57BL/6J background into C3H/HeN and BALB/c background for the genetic heterogeneity study is near completion. The backcrossing was done for 10 generations and the plan is to sequence APC gene locus for detection of the mutation. We expect APC1638N/+ mice in C3H/HeN and BALB/c background to be ready for space radiation exposure during fall 2017 beam run.

We simulated a solar particle event (SPE) at the NSRL to characterize the biological effects of low dose rate protons in vivo. Using the colorectal cancer susceptible (CPC;Apc) mouse model, we studied colonic tumorigenesis after whole-body exposure to a simulated SPE with varying energies (50-150 MeV/n) using a total dose of 2 Gy over a 2 hour period (at an average dose rate of 1.67 cGy/min). We also exposed mice to 2 Gy at 20 cGy/min of acute (50 MeV/n) proton (91% of the sSPE was at 50 MeV/n) or X-ray (250 kVp, 1 mA, 1.65 mm Al filter) at a dose rate of 20 cGy/min as a reference radiation. We observed that whole-body irradiation with simulated SPE was more effective in inducing invasive adenocarcinoma incidence compared to unirradiated controls and x-rays. We started a pilot GCR simulation study in 2016 using protons, and helium and silicon ions. Using the large beam and large boxes to contain 2-3 mice each we were able to irradiate about 90 mice in about 90 minutes with the new fast switching capabilities at the NSRL. We reduced the dose rate to 0.5 cGy per minute (instead of the usual 10-20 cGy per minute) with a total dose of 30 cGy. For NRSL16C we used 20 cGy of proton, followed by 5 cGy of helium and then 5 cGy of silicon. Since 5 cGy of silicon did not increase polyp formation, these experiments were conducted to determine if adding both helium and protons to 5 cGy of silicon would increase polyp formation. We also reversed the order of ions in the GCRsim since it may affect carcinogenesis initiation and/or progression. Results from these experiments will require about 9 additional months. Going forward we will add more mice to this GCRsim for NSRL 17A.

This year we focused on assessing long-term effects of heavy ion radiation on GI epithelial cell migration up to two months after exposure and compare the results to gamma radiation. The generated data is expected to contribute towards developing GI risk estimates of heavy ion radiation, which is qualitatively high linear energy transfer (high-LET) in outer space relative to gamma-rays, which is low-LET. Mice (C57BL/6J) were exposed to sham, or 0.5 Gy of gamma or 56Fe radiation and small intestine, which is organized into well-defined crypts and villi and is considered a model system to study GI cell migration, and collected 7 and 60 d after exposure. Rate of IEC migration was assessed using BrdU pulse labeling 24 h prior to sacrificing the mice. We analyzed key signaling molecules involved in cell polarity, microtubule dynamics, and cell adhesion dynamics using immunoblots, immunofluorescence, immunohistochemistry. Our data from this study suggests that heavy ion radiation could impede GI epithelial cell migration persistently. Overall, these studies in normal GI tissues indicate a profound and persistent perturbation of normal GI physiology and cellular homeostasis, and has implications for space travel-related GI pathophysiological changes.

At the genomics level we are using laboratory and bioinformatics tools of systems genomics to identify the features of gastrointestinal tumors arising in our murine models. Using next generation sequencing (NGS) based methods we are determining the global mutation burden, and mutation types including small mutations and structural variations in these tumors. To further inform mechanistic interpretation of HZE effects, we are planning to use NGS technologies to examine transcriptomic and epigenomic changes in colonocytes recovered after space radiation. Our overarching goal is to utilize integrative genomics analyses to identify the signatures of space radiation exposure, tumorigenesis, and the mechanistic consequences of these disturbances so that this information can be utilized to support risk assessment and mitigation studies. We are using formalin fixed paraffin embedded (FFPE) tumor samples for sequencing. As an initial step, FFPE sections were H&E stained for pathology evaluation and areas representing adenoma or carcinoma are marked for dissection. DNA is extracted from the dissected tissue using our standard protocols. We have developed a methodology for molecular quality control that will allow an estimate of sample purity and comprehensive sequencing for space radiation genomic signature identification.

Analyses of mouse model data of simulated space radiation, using mechanistically-motivated mathematical models of carcinogenesis, are needed to better understand and potentially mitigate these risks. We have used data on heavy ions (12-C, 28-Si, and 56-Fe) and gamma-ray induced intestinal tumor frequency in APC1638N/+ mice. Our mathematical model includes: (1) targeted effects (TE), caused by ionizations within/close to nuclear DNA; (2) non-targeted effects (NTE), caused by radiation damage to other cell structures (e.g., mitochondria) and/or activation of stress signaling pathways; (3) cytotoxic effects. We used the model to estimate the NTE contribution to the dose response for each ion. According to our model, NTE cause the steep initial rise of the heavy ion dose responses observed between 0 and 0.1 Gy. Because at low fluences of heavy ions many cell nuclei will not be traversed by the cores of ion tracks, the contribution of NTE to the dose response at low doses is expected to be high. Model-based estimates of NTE contributions were consistent with this expectation: they dominated the dose response below 0.1 Gy. Non targeted effects may therefore play an important role in low-dose carcinogenesis from heavy ions. Overall, we plan to generate in vivo data on tumor frequency and mechanisms of tumor development using NASA mandated additional priority space radiation beams at lower doses relevant to space environment. These data will be used to reduce uncertainty in CRC risk prediction and develop reliable risk prediction model for human.

We continue to employ two models of GI cancers: the APC1638N model, which was generated by a targeted modification of the Apc gene at a position corresponding to aa1638. These mice develop very few adenomas in the small as well as in the large intestine but post-radiation tumor numbers show marked increases. Radiation induced substantial gastric tumorigenesis as well in this model. The second model is the CDX2P Apc flox/+ that has one APC allele inactivated almost exclusively in the colon leading to adenomas and subsequent a low rate of adenocarcinomas in the colon that is increased post-irradiation. Data being collected from these two relevant mouse models will be used to calculate risk estimates for NASA. Since low-LET risk estimates are available for human populations, we will then be able to model the relative risk for space radiation. In particular, this ratio can then be extrapolated to human exposure to low-LET radiation to then make an estimate of risk for colorectal cancer in humans exposed to space radiation compared to the standard risks based on atomic bomb survivors and other exposure populations.

In order to answer the critical question of risk for colonic tumorigenesis after acute and chronic space radiation exposures, in vitro human cell culture and animal models for CRC initiation and progression are needed. We are investigating the effects of high-LET particle and high-energy protons in colonic epithelial cells with respect to the nature of damage and consequent tumor induction using biological cancer-related endpoints and systems biology (genomics, proteomics, and metabolomics) approaches. In addition, we propose to compare these results to the biological consequences of low dose rates and mixed fields of radiation which astronauts are likely to be continuously exposed. The studies of low- and high-LET radiation on colonocytes gain more relevance with increasing age when premalignant adenomas and other colonic lesions such as flat lesions and microscopic aberrant crypt foci become more prevalent. Pre-existing undetectable genetic or chromosomal aberrations in astronauts in addition to de novo mutations resulting from long-term irradiation may increase the chances of colonic stem cell transformation.

Plans for this proposal build on our past accomplishments with GI tumorigenesis models and are focused on studies at low doses with an expanded range of NASA’s high priority beams. This proposal is also focused on assessing the effects of gender, dose-rate, age, and genetic heterogeneity in mouse models. Molecular events triggered by space radiation will be investigated in depth and the mechanistic basis for persistent aberrant cell-signaling after HZE radiation will be dissected. Genome-wide NGS combined with other omics approaches will contribute to a systems biology understanding of the effects of space radiation in the intestines and stomach. Both quantitative murine tumor data and results for underlying molecular events in mouse and human GI cells will contribute to modeling of human risks for GI cancer by space travel. The overall aims of this NSCOR are summarized below.

Aim 1 (Project 1). Quantitatively assess GI tumorigenesis in mouse models of GI cancer and collect samples for qualitative analysis.

Aim 2 (Project 2). Dissection of the signaling events and consequences in GI cells of the persistent effects of space radiation.

Aim 3 (Project 3): Characterization of IR-induced neoplastic events in normal diploid human colonocytes.

Aim 4 (Project 4). Development of systems biology (Project 4A) and mathematical modeling (Project 4B) approaches for GI cancer risk assessment.

Task progress on GI tumorigenesis (GU). Since the award of the this NSCOR in April 2015, we have participated in one beam run (Fall 2015) at the NASA Space Radiation Laboratory (NSRL) and currently are scaling up APC1638N/+ mice breeding in preparation for our spring 2016 beam run. We have already demonstrated higher tumorigenesis with space radiation relative to gamma rays as well as important qualitative differences in the IR effects in both normal GI epithelial cells and GI tumors. The goals going forward are to assess tumorigenesis at more relevant lower doses and varied dose rates with a broader range of ions. Since maximal effects have been seen with 28-Si ions in our previous studies, we have continued experiments with 28-Si (300 MeV/n) albeit at new lower doses (5 and 1 cGy). We have also started studies using 16-O (325 MeV/n) and 4-He (250 MeV/n) ions, which are NASA prioritized ions of lower Z value proposed in this NSCOR. Additional mice will be irradiated in spring 2016 to fulfill statistical requirements. Apart from acute exposures (dose rate 10 to 50 cGy/min), we have also initiated studies using low dose rate (0.33 cGy/min) exposures and will continue in spring 2016. For comparison, we have exposed mice to gamma radiation at doses equal to HZE ions. Mice irradiated in 2016 fall will be ready for data collection 150 days after radiation exposure. Alongside, we have started backcrossing APC1638N/+ of C57BL/6J background into C3H/HeN and BALB/c background for the genetic heterogeneity study proposed in the project and is expected to take about a year to fully establish the model in these backgrounds.

Proton radiotherapy induces more precise DNA damage at the tumor site with reduced side effects to adjacent normal tissues. However, protons emerging from solar flares are considered to be one of the major acute risk factors for astronauts. Even though some shielding is possible, the long-term biological effects of proton irradiation in cancer initiation and progression compared with conventional photon irradiation remain poorly characterized. In this series of studies, using a human familial adenomatous polyposis syndrome susceptible mouse model, we show that whole-body irradiation with simulated solar particle event (SPE) protons are more effective in inducing senescence-associated inflammatory responses (SIRs), which are involved in colon cancer initiation and progression. After simulated SPE irradiation, a subset of SIR genes (Troy, Sox17, Opg, Faim2, Lpo, Tlr2, and Ptges) and a gene known to be involved in invasiveness (Plat), along with the senescence-associated gene (P19Arf) are markedly increased. Following these changes, loss of Casein kinase Ia (CKIa) and induction of chronic DNA damage and TP53 mutations are increased compared with the same doses of acute proton or x-ray irradiation. Simulated SPE irradiation (2 Gy, 92% 50 MeV/n, 8% 60-150 MeV/n) also increases the number of colonic polyps, and invasive adenocarcinomas compared to 2 Gy (50 MeV/n) single dose proton or 2 Gy (250 kVp) x-ray. Thus, exposure to SPE irradiation elicits significant changes in colorectal cancer initiation and progression

Considering that the responses were more pronounced after HZE irradiation relative to gamma radiation and 56-Fe and 28-Si were tested at moderately high doses in previous studies, the current project will expand persistent effect studies to additional priority beams, and lower radiation doses and dose rates more relevant to the space environment. To this end, we have irradiated male C57BL/6J and Lgr5-EGFP-IRES-creERT2 (Lgr5 mice) during 2015 fall run at 5 and 10 cGy and samples collected 2 months after exposure for molecular analysis, which is ongoing. Later, samples will also be collected 12 months after radiation to determine persistence of effects. The goal is to characterize persistent effects of space radiation at low doses and dose rates on GI epithelial and stem cells. Irradiation of mice is performed concurrently with the Project 1 mice. Work is also ongoing on human colonic epithelial cells (HCECs) with plans to irradiate these during the 2016 beam runs at NSRL. The work on the genomic project has been started with tumor samples from a moderate dose (1.6 Gy) of 56-Fe radiation archived from previous NSCOR. The goal is to standardize the methodology for signal detection using a relatively high dose of HZE radiation before analyzing lower doses (<1 Gy) of HZE radiation. Additionally, tumorigenesis data from the previous NSCOR is also used to calculate relative biological effectiveness (RBE) and develop modeling approach for colorectal cancer risk estimates in astronauts.