PEG-G-CSF and L-Citrulline Combination Therapy for Mitigating Skin Wound Combined Radiation Injury in a Mouse Model

Abstract

Radiation combined injury (RCI, radiation exposure coupled with other forms of injury, such as burn, wound, hemorrhage, blast, trauma and/or sepsis) comprises approximately 65% of injuries from a nuclear explosion, and greatly increases the risk of morbidity and mortality when compared to that of radiation injury alone. To date, no U.S. Food and Drug Administration (FDA)-approved countermeasures are available for RCI. Currently, three leukocyte growth factors (Neupogen^®, Neulasta^® and Leukine^®) have been approved by the FDA for mitigating the hematopoietic acute radiation syndrome. However these granulocyte-colony-stimulating factor (G-CSF) and granulocyte-macrophage colony-stimulating factor (GM-CSF) products have failed to increase 30-day survival of mice after RCI, suggesting a more complicated biological mechanism is in play for RCI than for radiation injury. In the current study, the mitigative efficacy of combination therapy using pegylated (PEG)-G-CSF (Neulasta) and L-citrulline was evaluated in an RCI mouse model. L-citrulline is a neutral alpha-amino acid shown to improve vascular endothelial function in cardiovascular diseases. Three doses of PEG-G-CSF at 1 mg/kg, subcutaneously administered on days 1, 8 and 15 postirradiation, were supplemented with oral L-citrulline (1 g/kg), once daily from day 1 to day 21 postirradiation. The combination treatment significantly improved the 30-day survival of mice after RCI from 15% (vehicle-treated) to 42%, and extended the median survival time by 4 days, as compared to vehicle controls. In addition, the combination therapy significantly increased body weight and bone marrow stem and progenitor cell clonogenicity in RCI mice, and accelerated recovery from RCI-induced intestinal injury, compared to animals treated with vehicle. Treatment with L-citrulline alone also accelerated skin wound healing after RCI. In conclusion, these data indicate that the PEG-G-CSF and L-citrulline combination therapy is a potentially effective countermeasure for mitigating RCI, likely by enhancing survival of the hematopoietic stem/progenitor cells and accelerating recovery from the RCI-induced intestinal injury and skin wounds.

INTRODUCTION

In a public health emergency radiation exposure scenario, such as a nuclear accident or terrorist threat, victims may suffer from radiation exposure accompanied by other types of injuries, such as wound trauma, hemorrhage, chemical or thermal burn, blast injury, or exposure to infectious agents or toxic chemicals. Termed radiation combined injury (RCI), these injuries could account for ~65% of injuries from a nuclear explosion and significantly increase the risk of morbidity and mortality compared to either type of injury alone, due to myelosuppression, immune system depletion, delayed traumatic wound healing, sepsis, multi-organ dysfunction syndromes and multi-organ failure (1). It is well recognized that the biological mechanisms underlying RCI are far more complicated than those for radiation injury alone, because of the exceedingly complex interaction and the synergistic effects exerted by radiation exposure and the accompanying trauma (2). Notwithstanding its complexity, for public health emergency preparedness, RCI has been identified by the National Institute of Allergy and Infectious Diseases (NIAID) as a key topic in need of further studies (3, 4).

Over the years, several RCI animal models have been developed to elucidate underlying mechanisms of observed synergistic effects of radiation exposure and combined injuries, characterizing the complicated pathophysiological responses to RCI and developing effective medical countermeasures for mitigating RCI. The RCI research group at the Armed Forces Radiobiology Research Institute (AFRRI) has developed a skin wound trauma combined with radiation injury mouse model, and has demonstrated that 30-day survival is decreased in RCI mice compared to those that received irradiation alone. This finding is likely due to increased susceptibility to infections, delayed wound healing, and an exacerbated acute radiation syndrome (5–8).

To be better prepared in the case of a radiation mass casualty event, there is an urgent need to develop medical countermeasures that mitigate the effects of, and provide treatment for, RCI victims. Therefore, products that potentially improve animal survival after RCI have been screened. Previously reported studies have shown that administration of synthetic trehalose dicorynomycolate (S-TDCM), a non-specific immuno- and hematopoietic modulator, in conjunction with systemic/topical gentamicin, administered immediately after RCI, increased 30-day animal survival (7). This outcome suggested that, for RCI, antibiotics are necessary to provide additional protection against infection, since RCI animals treated with S-TDCM alone did not survive (7). Furthermore, treatment with ciprofloxacin, a fluoroquinolone antibiotic, can help the recovery of blood cells from DNA damage, and increase hematopoiesis. This benefit is attributed to the ability of ciprofloxacin to modulate cytokine responses, especially by stimulating synthesis of colony-stimulating factors, thereby exerting immunomodulatory effects (9). To exploit these anti-bacterial and immunomodulatory effects, ciprofloxacin was administrated to mice after RCI or irradiation alone, once daily for 21 days starting at 2 h postirradiation, which resulted in a significant increase in the 30-day mouse survival after RCI, but not after radiation injury alone (10, 11). Owing to the increased susceptibility to infection seen in RCI in rodents, sepsis appears to be a key clinical feature of RCI and main cause of death after RCI. Therefore, ghrelin, effective in treating sepsis, was also tested for its mitigative effect after radiation injury and RCI. Similar to ciprofloxacin, ghrelin enhanced 30-day survival in mice after RCI, but not after irradiation alone (11). Conversely, some pharmacological countermeasures, such as pegylated granulocyte-colony stimulating-factor (PEG-G-CSF, Neulasta^®) (12), have been used to successfully mitigate radiation injury, but this product showed little or no efficacy for mitigating RCI. Therefore, it is reasonable to consider that the mechanisms of RCI may be, at least partially, different from radiation injury, and the therapeutic stratagem should be somewhat different between RCI and radiation injury alone.

Considering the poly-traumatic detriment to multiple organs/tissues after RCI, a combination therapy that targets key pathways in a synergistic or an additive fashion is a better approach for mitigating RCI. PEG-G-CSF is one of a few drugs approved by the U.S. Food and Drug Administration (FDA) for enhancing survival in patients with hematopoietic acute radiation syndrome (H-ARS), and it is also clinically used in decreasing the incidence of infection after chemotherapy and/or radiotherapy. We hypothesized that survival after RCI will be improved using a combination therapy of PEG-G-CSF with another intervention directly against the common pathophysiological processes activated by radiation and the accompanying traumatic injuries, such as reversal of endothelial dysfunction (2). Endothelial dysfunction plays a critical role in the pathogenesis of radiation injury in multiple organs, and is also involved in the response to traumatic wound injury. Supplemental administration of L-citrulline, a neutral alpha-amino acid, has been shown to be effective in enhancing cardiovascular health through improving vascular endothelial function in cardiovascular diseases (13, 14). In addition, human umbilical venous endothelial cells treated with L-citrulline had delayed high-glucose-induced endothelial senescence (15). Furthermore, it is well known that plasma citrulline levels dramatically decrease after irradiation, due to enterocyte mass reduction in the gastrointestinal acute radiation syndrome (GI-ARS) (16), since plasma citrulline level is highly proportional to intestinal enterocyte mass (17). Therefore, in the current study we examined the mitigating effects of the combination therapy of PEG-G-CSF and L-citrulline after RCI. The results of this study indicate that this combination therapy is safe and effective for mitigating RCI, likely by enhancing proliferation/differentiation of the hematopoietic stem/progenitor cells and by accelerating recovery from RCI-induced intestinal injury and skin wounding.

MATERIALS AND METHODS

Mice

B6D2F1/J female mice (12–13 weeks old) were obtained from Jackson Laboratory (Bar Harbor, ME). Female mice were chosen for this study, because male mice are more aggressive when housed together and are therefore, more likely to sustain unnecessary injuries, as described elsewhere (11). Upon arrival, all mice were allowed to acclimate to their new surroundings for 72 h. Mice were randomized for each experimental group and housed 5 mice per cage in a facility accredited by the Association for Assessment and Accreditation of Laboratory Animal Care, International (AAALAC International). The animal room was maintained at 23°C ± 3°C with 50% ± 20% relative humidity on a 12:12 h light-dark schedule. Commercial rodent feed (Envigo Teklad Rodent Diet; Envigo Inc. Indianapolis, IN) and acidified water (pH 2.5–3.0), used to control opportunistic infections, were available ad libitum to all animals. All animal handling procedures were performed in compliance with guidelines from the National Research Council (2011). Animal studies were conducted according to a protocol approved by the Institutional Animal Care and Use Committee (IACUC) at Uniformed Services University of the Health Sciences (USUHS, Bethesda, MD).

Total-Body Irradiation

Mice were 14–15 weeks old, with an average weight of 24–25 g, at the time of irradiation. Mice received total-body irradiation (TBI) from a bilateral radiation field at AFRRI’s ⁶⁰Co facility. An alanine/electron spin resonance (ESR) dosimetry system (American Society for Testing and Materials, Standard E 1607; ASTI International, Philadelphia, PA) was used to measure dose rates (to water) in the cores of acrylic mouse phantoms. Prior to TBI, all mice were placed in ventilated plexiglass containers (four mice per box with separate compartments for each animal) and 40 animals were irradiated at once without anesthesia. Dose rate was determined by the distance between the irradiators and animals. A single, mid-line tissue dose of 9.5 Gy was delivered at ~0.4 Gy/min (6). Animals in control and wound-alone groups were sham-irradiated, treated in the same manner as the irradiated animals, but remained in the cobalt staging room (18).

Skin Wounding

The skin wounding procedure was performed as described elsewhere (18), and comprised two steps, depilation and skin wounding, both of which were performed under anesthesia by isoflurane inhalation. In preparation for skin wounding, the dorsal fur was shaved using an electric hair clipper two days before TBI. A 250–300-mm² circular wound was created by a 70% ethanol-sterilized steel punch in the anterior-dorsal skin fold and underlying panniculus carnosus muscle (between the shoulder blades) within 1–2 h after sham irradiation or TBI, for animals in wound-alone and RCI groups, respectively. Subsequent to the skin punch, these non-lethal wounds were left open to the environment and all animals were placed in autoclaved clean cages containing autoclaved bedding. All mice subjected to the skin injury were also given 0.5 ml of acetaminophen solution (150 mg/kg in saline, OFIRMEV^® injection, NDC 43825–102-01; Mallinckrodt™ Pharmaceuticals, Hazelwood, MO) intraperitoneally (I.P.), immediately after skin injury to alleviate pain. As a control, 0.5 ml saline was also administered I.P. to groups that received sham irradiation or TBI alone. Acetaminophen was used to alleviate the acute pain induced by skin wounding and minimize distress in wounded-alone and RCI mice, in accordance with the 3 Rs policy (refinement, reduction, replacement) of the USUHS IACUC. Since animals in the TBI alone group did not display any signs of acute pain, only saline was used as a control. We acknowledge that the use of analgesics is well documented for interacting with the immune system, and thus has the potential to interfere with the radiation countermeasure effects of the test reagents, such as PEG-G-CSF and L-citrulline used in this study. Therefore, when we compared the mitigating effects of different treatments, the data analysis for the groups receiving TBI alone was separated from that for the RCI groups. As supportive care, gentamicin sulfate 0.1% cream (G&W^® Laboratories Inc., South Plainfield, NJ), was topically administered to evenly cover the wounded area on all mice subjected to the skin injury, and was provided from day 1 to day 10 after wounding.

Preparation and Administration of Test Agents: PEG-G-CSF and L-Citrulline

PEG-G-CSF was prepared and administered as described elsewhere (12). Briefly, PEG-G-CSF (also known as Neulasta^® or PEG-filgrastim, 6 mg in 0.6-ml single-dose syringe; Amgen, Thousand Oaks, CA), was freshly diluted with its formulation solution, which is 0.6 ml of solution containing 0.35 mg acetate, 0.02 mg polysorbate 20, 0.02 mg sodium and 30 mg sorbitol in water, pH 4.0. PEG-G-CSF (1 mg/kg) was administered subcutaneously (s.c.) in a volume of 0.2 ml at 24 h, day 8 and day 15 post-TBI. The vehicle for PEG-G-CSF is its formulation solution.

L-citrulline was purchased from MilliporeSigma (purity ≥ 98% by thin-layer chromatography; St. Louis, MO). L-citrulline was dissolved in Hanks’ Balanced Salt Solution (HBSS; Gibco^®, Gaithersburg, MD) at a concentration of 120 g/l. L-citrulline was orally administered via delivery to the cheek pouch within the oral cavity using an animal feeding needle to ensure accurate dosing at a lower stress level once daily for 21 days at the dosage of 1 g/kg in a volume of 0.2 ml, starting at 24 h post-TBI. The vehicle for L-citrulline was HBSS. The study design is shown in Table 1.

TABLE 1.

Experimental Design (20 Animals per Group, 16 Groups, 320 Mice in Total)

	Sham	Wound	Radiation injury	RCI
Vehicle (Veh)	Veh1a + Veh2b	Veh1a + Veh2b	Veh1a + Veh2b	Veh1a + Veh2b
PEG-G-CSF	PEG-G-CSF + Veh2	PEG-G-CSF + Veh2	PEG-G-CSF + Veh2	PEG-G-CSF + Veh2
L-citrulline	L-citrulline + Veh1	L-citrulline + Veh1	L-citrulline + Veh1	L-citrulline + Veh1
Combination	PEG-G-CSF + l-citrulline	PEG-G-CSF + l-citrulline	PEG-G-CSF + l-citrulline	PEG-G-CSF + l-citrulline

^aVeh1: The formulation buffer for PEG-G-CSF administered s.c.

^bVeh2: HBSS, used to dissolve l-citrulline, administered orally in cheek pouch.

Notes. PEG-G-CSF (1 mg/kg) and Veh1 (formulation buffer for PEG-G-CSF): Given subcutaneously in a volume of 0.2 ml on days 1, 8 and 15 after TBI. L-citrulline (1 g/kg) and Veh2 (HBSS): Orally administered once daily from day 1 to day 21 after TBI. They were delivered to the cheek pouch within the mouth cavity using the animal feeding needle in a volume of 0.2 ml.

There was concern that the excessive handling and stress of repeated oral dosing in irradiated mice can influence results. Indeed, if comparing results from the animals with less handling, the excessive handling and stress of repeated oral dosing in irradiated mice can influence results. However, in this study, animals in four injury groups (sham-irradiated, wounded alone, TBI alone and RCI) received the same extent of handling regardless of the treatments (vehicle, PEG-G-CSF, L-citrulline or combination), as shown above in Table 1. We considered an alternative approach such as adding the test agent (L-citrulline) into the rodent diet to reduce the stress of repeated oral dosing. However, during the 30-day period of the acute radiation syndrome, the GI-ARS could result in the loss of appetite, especially at the LD_70/30 radiation dose of 9.5 Gy, thereby leading to inaccurate dosing.

Determination of Thirty-Day Survival

After TBI and/or skin wounding, mice were closely monitored for 30 days by the researchers in addition to the regular health checks by vivarium staff. During the 30-day period, the USUHS IACUC Policy 020 (establishment of early end points in a mouse TBI model) was followed (19). Morbid animals were examined at least three times daily, and the moribund animals considered to have arrived at the end point were euthanized by CO₂ inhalation plus confirmatory cervical dislocation. The percentage of mice surviving the entire 30 days was recorded, and the Kaplan-Meier survival curve was plotted. For each group, n = 20, except for the RCI plus combination treatment group, which was n =19. In the RCI group that received combination treatment, one animal died on day 1 post-TBI and no extra RCI mouse was available to replace it. Survival data from the previous RCI studies indicate that onset of death for RCI mice is around one week post-TBI. Therefore, we surmised that the animal death on day 1 post-TBI was not related to RCI.

Body Weight and Wound Size Measurement

The basal body weight was measured immediately post-TBI (day 0), and on days 1, 3, 7, 14, 21 and 28 postirradiation.

Wound size was measured using digital caliper on days 1, 3, 7, 14, 21 and 28 postirradiation. The average area of each wound was calculated according to previously reported work (6): Wound area = π * A/2 * B/2 (A and B represent diameters at right angles to each other). The percentage of wound closure was calculated as: Percentage wound closure = 100% – (wound area on day X/wound area on day 1) * 100%; day X: days 3, 7, 14, 21 and 28 postirradiation. A wound closure of 100% indicates a fully-closed wound.

Blood Collection, Peripheral Blood Cell Count, Serum Preparation and Tissue Collection

On day 30 post-TBI, blood was collected from the surviving mice under deep isoflurane anesthesia, via cardiac puncture into a microtube containing EDTA for peripheral blood cell count, and into a microtube with serum separator additive for serum preparation as described elsewhere (20). Whole blood samples, at least 15 μl each in an EDTA-containing microtube, were analyzed using a clinical hematoanalyzer (Element HT5; Heska, Loveland, CO) following the manufacturer’s instructions. After at least 30 min coagulation at room temperature, sera were collected after centrifugation at 10,000g for 10 min, and immediately stored at −80°C for further analysis.

Cervical dislocation was performed after blood draw, and sternum, femur, spleen and jejunum were collected for further analysis. The number of animals used for blood and tissue collection was up to 6 per group.

Clonogenicity Assay

Clonogenicity of mouse hematopoietic stem and progenitor cells was assessed as described elsewhere (20). Briefly, bone marrow (BM) cells were extracted and isolated from pooled femurs collected from individual surviving mice on day 30 postirradiation. After erythrocytes were lysed with erythrocyte lysis buffer (QIAGEN, Hilden, Germany), total BM myeloid cell viability from each mouse was measured using trypan blue staining. Clonogenicity of mouse BM cells was quantified in standard semisolid cultures in triplicates using 1 ml of MethoCult™ GF+ system (including SCF, IL-3, IL-6 and erythropoietin) for mouse cells (STEMCELL™ Technologies Inc., Cambridge, MA) according to the manufacturer’s instructions. Bone marrow cells at a density of 1 × 10⁴ cells/dish were plated in 35-cm cell culture dishes (BD Biosciences, San Jose, CA). After 8–10 days of cell culture at 37°C in 5% CO₂, the colony-forming units (CFU) of erythroid (E), granulocyte-monocyte (GM), and granulocyte-erythrocyte-monocyte-megakaryocyte (GEMM) were scored for each plate. CFU-Total is the sum of CFU-GM, CFU-E and CFU-GEMM.

Histological Examination of Mouse Bone Marrow and Jejunum

On day 30 post-TBI, the sternum and proximal jejunum (5 cm of the small intestine cut 7 cm and 12 cm distant from the stomach) harvested from the surviving animals were immediately fixed in 10% formalin for 24 h, washed with PBS and then stored in 70% ethanol. Tissue dehydration, paraffin embedding, sectioning into 5-μm-thick sections on charged glass slides, and then hematoxylin and eosin (H&E) staining were performed. Sternums were decalcified prior to paraffin sectioning. The bright-field images of H&E-stained slides were acquired using the Zeiss Axioscan.Z1 and analyzed using Zeiss Zen 2.5 (blue edition) (Carl Zeiss AG, Oberkochen, Germany), Microsoft^® Photoshop (Redmond, WA) and ImageJ (National Institutes of Health, Bethesda, MD) analysis program.

Citrulline Assay

Citrulline levels in serum were quantified in duplicate using a citrulline assay kit (Cell Biolabs Inc., San Diego, CA) according to manufacturer’s instructions. Briefly, samples were treated with SDS and proteinase K to release citrulline residues, and assay reagents were added to the samples, which reacted with citrulline to produce a chromophore. Absorbance was read at 540–560 nm.

Measurement of Spleen Weight and Splenocyte Counts

Spleens collected on day 30 post-TBI were weighed and placed in a plastic pouch containing 10 ml HBSS (Gibco; Life Technologies Inc., Carlsbad, CA). The pouch was inserted into a Stomacher^® 80 Biomaster Lab System (Seward Laboratory Systems, Port St. Lucie, FL) at high speed for 60 s. The contents were poured through a 70-μm cell strainer (Falcon™, Durham, NC), and centrifuged at 3,000 rpm (i.e.,1,960g) for 10 min. The cell pellet was resuspended in 10 ml ammonium-chloride-potassium (ACK) lysing buffer (Gibco; Life Technologies Inc) and incubated at 37°C for 10 min. Spleen cells were pelleted at 3,000 rpm (i.e.,1,960g) for 10 min and resuspended in 10 ml phosphate buffered saline (PBS). Splenocytes were counted using a Countess™ automated cell counter (Invitrogen™, Carlsbad, CA).

Statistical Analysis

Data presented here are the pooled results from two animal experiments. Data analysis was performed using GraphPad Prism version 7 (La Jolla, CA), and results are expressed as mean ± standard error of mean (SEM) unless otherwise stated. Kaplan-Meier survival curves were compared by log-rank (Mantel-Cox) test. If the group number was greater than two, difference among groups was analyzed by one-way or two-way analysis of variance (ANOVA) and Dunnett’s or Tukey’s multiple comparisons. If the group number equaled two, difference between groups was analyzed by two-way ANOVA, and Sidak’s multiple comparisons, or t test. A P value <0.05 was considered statistically significant.

RESULTS

Combination Treatment of PEG-G-CSF and L-Citrulline Mitigates Mouse Lethality after RCI

Mice receiving 9.5 Gy TBI and/or skin wound were treated with vehicle, PEG-G-CSF, L-citrulline, or combination of PEG-G-CSF and L-citrulline starting 24 h after TBI, and monitored for 30 days as described above. In the group that received TBI alone (Fig. 1A), the percentages of surviving mice on day 30 post-TBI and the median survival time (presented in parentheses) were: 25% (15 days) for vehicle; 70% (>30 days) for PEG-G-CSF; 35% (20 days) for L-citrulline; and 50% (26 days) for the combination therapy. PEG-G-CSF significantly increased 30-day survival after TBI alone compared to vehicle control (P=0.005). L-citrulline alone and the combination therapy did not provide significant survival benefit for mice that received TBI alone compared to the vehicle-treated group (P = 0.45 for L-citrulline; P=0.19 for combination). In the RCI group (Fig. 1B), the percentages of surviving mice on day 30 post-TBI and the median survival time (presented in parentheses) were: vehicle, 15% (14 days); PEG-G-CSF, 35% (15 days); L-citrulline, 40% (17.5 days); and combination therapy, 42% (18 days). PEG-G-CSF alone displayed no significant survival efficacy for RCI mice compared to vehicle (P = 0.329), although it was effective for TBI alone. In contrast, the combination therapy compared to vehicle significantly increased 30-day survival of mice after RCI (P = 0.007). L-citrulline displayed a trend towards enhancing survival in RCI mice compared to vehicle, although it did not reach significance (P = 0.057). Moreover, for RCI mice treated with the combination therapy, onset of death occurred after 13 days, while for RCI mice treated with each of other three treatments, the onset of death occurred 6–10 days postirradiation, which is consistent with the calculated median survival time for RCI mice (Table 2). Taken together, these data indicate that the combination therapy of PEG-G-CSF and L-citrulline is effective in mitigating lethality after RCI.

FIG. 1.

Effects of different treatments on mitigating lethality after TBI alone and RCI. B6D2F1/J female mice at 14 weeks old received TBI alone or RCI, and received vehicle, PEG-G-CSF, L-citrulline, or combination of the last two agents starting at 24 h after 9.5 Gy TBI. Kaplan-Meier survival curves for mice that received TBI alone (panel A) and RCI (panel B) with different treatments are shown (n = 19 for RCI with the combination therapy, n = 20 for other groups). **P < 0.01 for TBI alone + PEG-G-CSF vs. vehicle; **P < 0.01 for RCI + combination therapy vs. vehicle.

TABLE 2.

Comparison of Onset of Death (Day) and Median Survival Time (Day) of RCI Mice among Different Treatments

	Vehicle	PEG-G-CSF	L-citrulline	PEG-G-CSF + l-citrulline
Onset of death	10	9	6	13
Median survival time	14	15	17.5	18

To further elucidate the discrepancy between the effects of treatment for radiation injury alone and RCI, survival rates after TBI alone and RCI with four different treatments were compared (Table 3). With vehicle treatment, RCI mice had higher mortality than mice that received TBI alone, which is consistent with previously published findings related to this animal model (11). Interestingly, PEG-G-CSF treatment significantly increased survival of mice after TBI alone but not after RCI, while L-citrulline and combination therapy increased mouse survival after TBI alone and RCI at different levels. As alluded to earlier studies, products that are effective in mitigating radiation injury alone may have no efficacy for RCI (12); these data further support this.

TABLE 3.

Thirty-day Survival Rate Comparison of Radiation Injury and RCI Using Different Treatments

	RI % survival	RCI % survival	Survival rate: Drug-vehicle (P: drug vs. vehicle)		Survival rate: RI – RCI
			Radiation injury	RCI	(P: RI vs. RCI)
Vehicle	25%	15%	N.A.	N.A.	10% (P = 0.312)
PEG-G-CSF	70%	35%	45% (P = 0.005c)	20% (P = 0.329)	35% (P = 0.016b)
L-citrulline	35%	40%	10% (P = 0.45)	25% (P = 0.057)	5% (P = 0.874)
Combinationa	50%	42%	25% (P = 0.19)	27% (P = 0.007c)	8% (P = 0.977)

^aCombination therapy of PEG-G-CSF and l-citrulline.

^bP < 0.05

^cP < 0.01, log-rank (Mantel-Cox) test.

Note. RI = radiation injury.

Combination Treatment of PEG-G-CSF and L-Citrulline Attenuates Mouse Body Weight Loss after RCI

Mouse body weight was measured on days 0, 1, 3, 7, 14, 21 and 28 post-TBI. As shown in Fig. 2, mouse body weight dropped immediately after TBI alone and RCI, and reached the first nadir on day 3. The second nadir was around the critical period of animal death on day 21 for mice that received TBI only (Fig. 2A), and on day 14 for RCI mice (Fig. 2B). The body weight was increased by day 28 in all surviving mice after TBI only or RCI, regardless of the treatment, suggesting that the body weight recovery plays a key role in animal survival. Overall, the initial decrease in body weight was greater in RCI mice compared to TBI only mice, and the body weight loss was associated with animal morbidity.

FIG. 2.

Effects of different treatments on mitigating body weight loss after TBI alone and RCI. Body weight of TBI-only mice (panel A) and RCI mice (panel B), treated with vehicle, PEG-G-CSF, L-citrulline or combination of the last two agents are shown. Body weight was measured on days 0, 1, 3, 7, 14, 21 and 28 postirradiation. Data represent mean ± SEM. In panel A, **P < 0.01 for PEG-G-CSF (n = 14) vs. vehicle (n = 7) on day 21; in panel B, **P < 0.01 for combination of two (n = 8) vs. vehicle (n = 4), on day 21.

The difference in body weight among treatments at each time point in mice that received TBI only (Fig. 2A) and in RCI mice (Fig. 2B) was also compared; however, no differences in the basal body weight (day 0) were observed. As shown in Fig. 2A, TBI only mice treated with PEG-G-CSF displayed significantly higher body weights compared to those treated with vehicle on day 21 postirradiation (P = 0.0028), indicating PEG-G-CSF was able to maintain body weight after TBI. In addition, RCI mice treated with the combination therapy showed significantly higher body weights compared to those treated with vehicle on day 21 postirradiation (Fig. 2B, P = 0.0071), suggesting the combination therapy could accelerate body weight gain after RCI. On day 21 postirradiation, the differences of body weight between vehicle and PEG-G-CSF (P = 0.0646) or L-citrulline (P=0.0514)-treated RCI mice were noted, but were not statistically significant. The body weight gain from the PEG-G-CSF-treated TBI only group and the combination-treated RCI group could contribute to the survival benefit after TBI alone and RCI, respectively.

L-Citrulline Accelerates Wound Healing after RCI

The wound area and percentage wound closure after wounding alone or RCI were measured and calculated. The basal wound area (day 1 post-TBI) ranged from 230–270 mm² for all wound-alone and RCI groups. As expected, full wound closure took approximately 14 days for wound-alone mice, and ~28–30 days or more for RCI mice. Different treatments did not improve the wound healing process in wound-alone groups (Fig. 3A and C) and no animal death was observed within 30 days after wounding in these groups (data not shown).

FIG. 3.

Effects of different treatments on skin wound healing after wounding and RCI. Wound areas and the percentage of wound closure in wounded alone and RCI mice treated with vehicle, PEG-G-CSF, L-citrulline, or combination of the last two agents were measured and calculated, respectively. Skin wound areas were measured on days 1, 3, 7, 14, 21 and 28 after wounding alone (panel A) and RCI (panel B). Data represent the mean ± SEM (n = 4–15 per group). In panel B, *P < 0.05 and ***P < 0.001for L-citrulline vs. vehicle on day 14 and day 21, respectively. Percentages of wound closure in wounded alone mice (panel C) and RCI mice (panel D) were calculated. In panel D, *P < 0.05 and ****P < 0.0001for L-citrulline vs. vehicle on day 14 and day 21, respectively.

In the RCI group (Fig. 3B and D), no difference in wound area on day 1 was observed among the four treatments. However, compared to vehicle control, treatment with L-citrulline resulted in a significantly smaller wound area on day 14 (P = 0.0436) and day 21 (P = 0.0002). Consistently, higher-percentage wound closure on day 14 (P = 0.01) and day 21 (P < 0.0001) in the L-citrulline treated group was observed, indicating that L-citrulline significantly accelerated wound healing after RCI. On day 28, only 3/20 mice survived in the vehicle-treated RCI group and their wound healing levels were not significantly different compared with other groups.

Surviving RCI Mice Display a Mild Bone Marrow Hypocellularity Compared to Mice Receiving TBI Alone on Day 30 Postirradiation

It is evident that TBI-induced BM failure includes a decline in hematopoietic stem cells (hypocellularity), which consequently leads to the inability to produce proper blood cell numbers for maintaining the immune system (1). To evaluate the effects of different treatments on BM cellularity in surviving TBI-alone and RCI mice, histological examination was performed on H&E stained cross sections of mouse sternum harvested on day 30 after irradiation. Representative images in Fig. 4A show that, compared to sham-irradiated mice, surviving TBI-only mice in the vehicle-treated group had a severe BM hypocellularity, in which the marrow space was largely bereft of hematopoietic cells and mainly consisted of adipocytes. Treatment with PEG-G-CSF, L-citrulline or a combination of the two improved BM cellularity in TBI-only mice. Surprisingly, a mild depletion in the number of BM hematopoietic cells was observed in 30-day-surviving RCI mice, regardless of treatment, and a modest decrease in hematopoietic cells and lower number of adipocytes were observed in BM samples of these animals compared to mice that received TBI alone (Fig. 4A).

FIG. 4.

Effects of treatments on BM cellularity after TBI alone and RCI. Panel A: Representative H&E-stained images of cross sections of mouse sternum harvested on day 30 postirradiation. The area marked by the yellow square is ~110,000 μm²; three areas per animal were assessed. Panel B: Quantification of megakaryocyte and adipocyte in sham, wounded, TBI alone and RCI groups treated with vehicle. Quantification of megakaryocyte (panel C) and adipocyte (panel D) TBI-only and RCI groups with four interventions are shown. Data represent mean ± SEM (n = 3–7 per group). *^/#P < 0.05; **^/##P < 0.01; ***^/###P < 0.001; ****^/####P < 0.0001. Scale bar = 500 μm (black), 100 μm (yellow).

Numbers of megakaryocytes and adipocytes in BM samples were also counted. As shown in Fig. 4B, with vehicle treatment, megakaryocyte counts were significantly decreased in TBI-only mice compared to animals in the sham-irradiated (P = 0.0005), wound (P = 0.0004) or RCI (P = 0.0373) groups. Since an increase in adipocyte number in BM is secondary to a decrease in hematopoietic cells, mice that received TBI only also had markedly increased adipocyte counts, when compared to animals in the sham-irradiated (P < 0.0001), wound (P < 0.0001) or RCI (P = 0.0013) groups. No significant differences in either megakaryocyte or adipocyte counts were observed among sham-irradiated, wound and RCI groups. These data confirmed the observation that BM hypocellularity in RCI mice was mild and was significantly less than what was seen in TBI-only mice on day 30 after TBI. This phenomenon may be related to the synergistic effect induced by radiation and the combined traumatic injury. Further study is needed to explore the underlying mechanisms.

The effect of interventions on megakaryocyte (Fig. 4C) and adipocyte (Fig. 4D) counts in surviving mice was also assessed. In the RCI group, differences in megakaryocyte or adipocyte counts were not observed among different treatments (Fig. 4C–D). However, in the TBI-only group, PEG-G-CSF and combination therapy significantly reduced adipocyte numbers in the BM (P = 0.0019 for PEG-G-CSF versus vehicle; P = 0.0002 for combination versus vehicle), which indicates that both of the treatments are effective in preventing or accelerating the recovery from BM hypocellularity.

Thirty-day Surviving RCI Mice Exhibit a Better Recovery of Peripheral Blood Counts than Surviving TBI-Only Mice

It has been demonstrated that decreases in peripheral blood counts are correlated with a reduction in BM cellularity (21). Because of this, peripheral blood cell counts were measured on day 30 postirradiation. Data in Fig. 5 show that after TBI alone, white blood cell (WBC, Fig. 5A) and red blood cell (RBC, Fig. 5B) counts were significantly reduced in the vehicle, PEG-G-CSF, L-citrulline and/or combination-treated groups compared to those in the sham-irradiated group. In contrast, the WBC and RBC counts in 30-day surviving mice were not significantly different between RCI and sham-irradiated animals, regardless of treatments, although the WBC counts in the vehicle-treated RCI group were lower than in sham-irradiated mice. These results were consistent with a mild BM hypocellularity. RCI mice displayed higher counts for neutrophils (NEU, Fig. 5C) and monocytes (MONO, Fig. 5D) in those treated with vehicle, PEG-G-CSF, L-citrulline or the combination. However, treatment with PEG-G-CSF and the combination, but not L-citrulline alone, further significantly increased monocyte counts in compared to that in RCI vehicle-treated mice (Fig. 5D). Like the RCI mice, neutrophils and monocytes also recovered in vehicle- or drug-treated mice that had received TBI alone (Fig. 5C and D). Furthermore, in the groups that received either TBI alone and RCI, lymphocyte (LYM, Fig. 5E) and platelet (PLT, Fig. 5F) counts were significantly reduced, compared to the sham-irradiated group. No mitigation in these decreases was observed in mice treated with either individual drug or the combination.

FIG. 5.

Peripheral blood counts on day 30 postirradiation in sham-irradiated, wounded, TBI-only and RCI mice. Data represent mean ± SEM (n = 3–12 per group). Panel A: WBC (white blood cells). Panel B: RBC (red blood cells). Panel C: NEU (neutrophils). Panel D: MONO (monocytes). Panel E: LYM (lymphocytes). Panel F: PLT (platelets). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

L-Citrulline and Combination Treatment Augment the Bone Marrow Clonogenicity on Day 30 Postirradiation in Mice that Received TBI Alone or RCI

We further evaluated the function of mouse BM hematopoietic stem and progenitor cells isolated from femurs on day 30 postirradiation by colony-forming assay. Figure 6A shows that, in the vehicle-treated group, BM colonies, including CFU-GM, CFU-GEMM and CFU-Total, were significantly lower in TBI-only and RCI mice than those in sham-irradiated mice (P < 0.0001). Treatment with either L-citrulline or combination therapy exhibited an increase in BM clonogenicity in TBI-only and RCI mice. For TBI-only mice (Fig. 6B), the number of CFU-GM was higher in L-citrulline (P = 0.0169) and combination therapy (P = 0.0207) groups compared to vehicle control. For RCI mice (Fig. 6C), the combination therapy significantly increased the counts of CFU-GM (P < 0.0001), CFU-GEMM (P = 0.0476), and total colonies (P = 0.0029), compared to vehicle control, and colony numbers of CFU-GM (P = 0.032) and CFU-Total (P = 0.0491) were also significantly higher in the L-citrulline-treated group than in vehicle control. Overall, the data showed that L-citrulline and the combination therapy augmented the BM clonogenicity in TBI-only and RCI mice.

FIG. 6.

Effects of treatments on BM clonogenicity in sham-irradiated, wounded, TBI-only and RCI mice. In vitro colony-forming assay was performed on BM cells collected from 30-day-surviving mice. Panel A: Colony forming units (CFU) of GM (granulocyte-monocyte), E (erythroid), GEMM (granulocyte-erythrocyte-monocyte-megakaryocyte) and Total (sum of GM, E, GEMM) in sham-irradiated, wounded, TBI-only and RCI groups treated with vehicle. CFU of GM, E, GEMM and Total were also measured in TBI-only (panel B) and RCI (panel C) groups with four treatments. Data represent the mean ± SEM (n=6–9 per group). *^/#P < 0.05; **^/##P < 0.01; ***^/###P < 0.001; ****^/####P < 0.0001.

Combination Treatment of PEG-G-CSF and L-Citrulline Inhibits Splenomegaly after TBI Alone and Recovers Splenocytes after TBI Alone and RCI

It is evident that TBI alone, but not RCI, induces spleen enlargement, namely splenomegaly (11). Total-body irradiation with shielded spleen results in significant enhancement of survival (22). As shown in Fig. 7A, TBI resulted in splenomegaly in 30-day surviving, vehicle-treated mice. Treatment with PEG-G-CSF fully blocked the splenomegaly, whereas treatment with L-citrulline partially inhibited it. Moreover, the combination therapy fully blocked it. As shown in Fig. 7B, wounding, TBI alone, and RCI did not alter the splenocyte counts. Treatment with either PEG-G-CSF alone or L-citrulline alone significantly reduced splenocyte counts in TBI-only and RCI mice. The combination therapy also resulted in decreases in splenocyte counts in both TBI-only and RCI mice compared to those in the sham-irradiated drug treatment group. However, the data in RCI mice, but not in TBI-only mice, showed recovery of the splenocyte count compared to sham-irradiated vehicle-treated mice (P = 0.125). The latter results warrant further exploration of the interaction between these two drugs.

FIG. 7.

Effects of different treatments on spleen weights and splenocyte counts. Spleens collected from 30-day surviving mice were weighed (panel A) and splenocytes of each spleen were counted (panel B). Data represent the mean ± SEM (n=3–12 per group). *P < 0.05; **P < 0.01 for TBI alone/RCI vs. respective sham group.

Combination Treatment of PEG-G-CSF and L-Citrulline Accelerates the Recovery of Intestinal Crypt Damage after RCI

To evaluate the extent of intestinal damage in 30-day-surviving mice and to determine the effect of interventions in mitigating intestinal injury, histological examination of mouse jejunums harvested from surviving mice on day 30 after irradiation was performed. Figure 8A shows representative images of H&E stained cross sections of mouse jejunum in sham-irradiated, wounded alone, TBI-only and RCI groups treated with vehicle, PEG-G-CSF, L-citrulline or combination of the two. Overall, the villi in irradiated mice, particularly after RCI, appeared to be shortened/blunted and enlarged; however, no significant differences in villus height (Fig. 8B) and width (Fig. 8C) were found among the different groups.

FIG. 8.

Effects of treatments on mitigating intestinal damage after sham irradiation, wounding, TBI alone, and RCI. Panel A: Representative H&E-stained images of cross sections of mouse jejunum harvested on day 30 postirradiation. Measurements of villus height (panel B), villus width (panel C), crypt depth (panel D) and crypt count (panel E) were performed using Zeiss Zen 2.5 software and ImageJ. Data represent the mean ± SEM (n = 3–7). Panel F: serum citrulline concentration on day 30 postirradiation in all four groups. n=3. *^/#P < 0.05; **^/##P < 0.01; ***^/###P < 0.001; ****^/####P < 0.0001. Scale bar: 100 μm (yellow).

The TBI-only and RCI mice showed significantly elongated crypts when compared to the sham-irradiated mice with the same interventions (Fig. 8D), which suggests a form of tissue remodeling in these surviving mice, such as amplification of the intestinal crypt cells, to accelerate recovery of the irradiated intestine. Combination therapy with PEG-G-CSF and L-citrulline further increased the crypt depth in RCI mice compared to vehicle-treated animals (P = 0.0277). The extent of intestinal absorptive dysfunction induced by radiation is correlated to a decrease in the intestinal crypt number (23). Figure 8E shows that crypt counts were significantly decreased in TBI-only and RCI mice treated with vehicle, PEG-G-CSF or L-citrulline compared to sham-irradiated mice. However, no differences in crypt counts were observed in these groups receiving the combination therapy. Our data suggest that the combination therapy may accelerate the recovery of intestinal crypt damage induced by RCI.

The citrulline level in serum has been widely recognized as a suitable biomarker for GI-ARS. It has been reported that serum citrulline immediately decreases after high-dose TBI, reaching a nadir around day 3.5 to day 7 after TBI (24). Based on this finding and the current data related to histological examination of mouse jejunum, the citrulline concentration was measured in serum collected on day 30 after irradiation. Results in Fig. 8F show that the serum citrulline levels were lower in the RCI group than in sham-irradiated, wound and TBI-only groups, and treatment with PEG-G-CSF or L-citrulline increased serum citrulline levels to a certain degree, but only the combination therapy significantly increased the serum citrulline concentration compared to the vehicle treatment in RCI mice (P = 0.0388). L-citrulline was administrated 1–21 days after TBI and the circulating level of citrulline was measured on day 30 after irradiation. According to the FDA, concerning the PK of L-citrulline in humans (25), the mean elimination half-life is 1 h for oral dosing of 2–15 g of L-citrulline in healthy adults (not dose-dependent). The mouse dose of 1 g/kg L-citrulline per day can be converted to the human dose of 3 g L-citrulline per day in a healthy adult, so its half-life in mouse is also estimated to be approximately 1 h. Considering the short half-life, it is not possible that the elevated circulating levels of citrulline could be attributed to the final dose of L-citrulline administrated 9 days prior. Therefore, this result suggests that the combination therapy improved the GI epithelium after RCI, likely by accelerating the recovery of RCI-induced crypt injury.

DISCUSSION

In the current study, the mitigative effects of the combination therapy (PEG-G-CSF and L-citrulline) on survival of mice after a lethal dose at 9.5 Gy followed by a full-thickness skin wound were demonstrated. These data suggest that attenuation of body weight loss, acceleration of wound healing, recovery from intestinal crypt damage, and augmentation of the BM clonogenicity created an overall benefit to improve survival in RCI mice treated with the combination therapy. Previously, the effort of developing countermeasures against RCI was mainly focused on the use of immune modulators, recombinant cytokines/growth factors, antimicrobial therapies and transplantation of stem cells (7, 10, 11, 26–34). However, whether RCI aggravates endothelium, as well as the role of endothelial dysfunction in mediating RCI-induced damage to multiple tissues/organs, are not well-characterized nor fully understood (2, 35). The current findings indicated that a combination therapy including L-citrulline, known as an agent for targeting reversal of endothelial dysfunction, is effective in mitigating RCI, which suggests that RCI might worsen endothelial dysfunction.

It is worth noting that, in the group that received TBI only, the survival rate was lower (50%) with combination treatment than with Peg-G-CSF alone (70%). Whether there is a negative interaction between PEG-G-CSF and L-citrulline in survival of TBI only animals need further investigation. In addition, it has been reported that G-CSF improved wound healing in aged mice but not in young mice (36, 37). Previously published studies from this group used 14-week-old young mice and demonstrated that PEG-G-CSF treatment delayed wound healing (38). In the current study, we used the same young-aged mice and provided further evidence that treatment with PEG-G-CSF did not facilitate wound healing in this mouse model. Moreover, there is no available information on L-citrulline interactions.

Furthermore, RCI delays traumatic wound healing, excessively increases immune system depletion and the susceptibility to infection, and exacerbates intestinal injury compared with TBI alone (1, 2). Accumulating evidence suggests that the intestinal crypt microenvironment controls stem cells and determines daughter cell fate in response to radiation injury. For example, endothelial function regulates intestinal responses to radiation by controlling crypt stem cells (2, 39–42). Results from this study provided evidence that 30-day-surviving RCI mice treated with the combination therapy of PEG-G-CSF and L-citrulline had normal intestinal crypt counts, elongated crypts, increased serum citrulline concentrations and accelerated weight gain, compared to those observed in vehicle-treated RCI mice. These outcomes suggest that this combination therapy might assist in a better recovery from RCI-induced intestinal injury, via enhancing endothelial function and repairing the GI tight junction (34). More studies are required to further explore these possibilities.

In the current study, RCI exacerbated intestinal injury detected on day 30 postirradiation compared to what was observed after TBI alone, which is in agreement with previously reported studies (1, 2, 6, 34, 38). On the contrary, 30-day-surviving RCI mice exhibited a mild BM hypocellularity, normal spleen weights, and a better recovery of peripheral blood cell counts than surviving TBI-only mice, which is consistent with the previously published studies (7, 26, 29, 38) and suggests that the interaction between radiation and wound trauma may promote faster recovery from BM injury in RCI mice than that in TBI-only mice. The data presented here are consistent with what was demonstrated in an early study on thermal burn combined with whole-body X-ray irradiation, in which the investigators concluded that the increase in early mortality is due to a functional impairment in the GI system or in some other critical yet unidentified systems, but not the H-ARS (3, 43).

Despite the fact that BM cellularity was largely recovered in the surviving RCI mice, current data showed that peripheral lymphocyte counts remained extremely low, less than 0.7 × 10³ cells/μl, regardless of treatments (a normal range is approximately 3–4 × 10³ cells/μl). This is important, since lymphocytes play a large role in the immune response to infectious microorganisms and other foreign substances. In RCI mice, impaired immunity and diminished ability to repair causes delay in healing of an open wound (6, 44), thereby leading to increase in animal susceptibility to bacterial infection (45). In an early published study, Ledney, et al. mentioned that delayed wound healing and increased susceptibility to infection were the key features of mice exposed to radiation followed by wounding. He and others also pointed out that early onset of death in most of these animals resulted from infectious complications (18, 45). Indeed, animal studies, such as an open wound model in mice, along with clinical evidence collected from the Chernobyl accident, showed that morbidity/mortality after RCI is positively correlated with the wound size or percentage of skin surface affected (3, 46). Previously, it was documented that mesenchymal stem cells (29, 47, 48), ghrelin (30) and ciprofloxacin (10) accelerated wound healing in RCI mice, which potentially contributed to the improvement of 30-day survival in RCI mice. Herein, it was found that L-citrulline treatment accelerated wound healing in RCI animals. Treatment with either the combination of PEG-G-CSF and L-citrulline or L-citrulline alone increased the 30-day survival percentage, further reinforcing the concept that acceleration of wound healing could contribute to reducing RCI-induced mortality. However, further studies are needed to explore the mechanisms by which L-citrulline accelerated wound healing in RCI mice.

Although the synergistic effect of radiation and physical trauma, such as wound, burn and hemorrhage, is well-recognized (1), the biological mechanisms responsible for the synergism are not fully understood. Recently published studies have suggested that, compared to radiation injury alone, RCI further augments the activation of inducible nitric oxide synthase (iNOS) and the nitric oxide (NO) pathway, causes excessive production of cytokines/chemokines, induces a greater amount of DNA damage, and potentiates radiation-induced changes in gene expression in response to DNA damage and in cell adhesion, etc. (1, 6, 11). Exploration of the specific mechanisms for the mitigative effects of the combination of PEG-G-CSF and L-citrulline in RCI mice is ongoing. PEG-G-CSF stimulates white blood cell proliferation, and supplementation with L-citrulline has shown promising therapeutic effects on cardio-metabolic health because it can protect against endothelial damage, increase antioxidant capacity, and modulate inflammation as well as mitochondrial functions (12, 49). Therefore, it is hypothesized that this combination may exert its mitigative effects by increasing NO bioavailability, reducing reactive oxygen species (ROS) levels and pro-inflammatory cytokines, increasing anti-inflammatory cytokines, and enhancing mitochondrial oxidative capacity.

Over the past decade, significant progress has been made in characterizing pathophysiological responses to RCI. However, the management of RCI-induced injury, especially GI toxicity, remains underdeveloped. Messerschmidt et al. (50) suggested that the recommended management for RCI will be different, in many aspects, from the management of either physical trauma or radiation injury. Additionally, Hauer-Jensen (2) proposed that a promising intervention for RCI should target the common pathophysiological pathways shared by radiation and the accompanied injury. Elevated endothelial cell permeability and a procoagulant endothelial surface are shared responses to either wounding or radiation, which is also closely linked to multi-tissue/organ injury (3, 39). Results from this study suggest that interventions targeting reversal of endothelial dysfunction are promising as countermeasures for treating RCI.