Plerixafor, a CXCR4 Antagonist, Mitigates Skin Radiation-Induced Injury in Mice

Abstract

Even with modern 3D conformal treatments skin radiation injury can be an inadvertent complication associated with clinical radiotherapy particularly at tissue folds. It is also of concern in the context of a radiological terrorism incident or accident, since skin irradiation lowers the lethal dose of whole body radiation. We hypothesize that radiation-induced skin injury originates from a loss of stem and progenitor cells, accompanied by excessive ROS production and proinflammatory cytokines. Plerixafor, a CXCR-4 antagonist, is one of the most efficient bone marrow stem cell mobilizers and these studies were designed to experimentally assess the potential of Plerixafor to reduce skin radiation injury. The right hind legs of groups of C57BL/6 mice were exposed to radiation alone or in combination with Plerixafor. Plerixafor was administered intraperitoneally at a dose of 5 mg/kg given in two doses separated by two days and started either on day 0, 4, 7, 15 or 24 after irradiation. The primary end point was skin injury, which was assessed three times a week for at least 2 months using a semi-quantitative scale. Secondary end points measured at selected time points included histology (primarily H&E) and cytokine levels (TGF-β and TNF-α). The acute and late skin injury in mice receiving Plerixafor was highly dependent on the timing of administration of the drug. The maximum benefit was observed when the drug was started 1 week after radiation exposure, and earlier or later administration of the drug decreased its efficacy. Secondary damage end points (cytokine levels and histologically assessed tissue thickness) provided confirmatory observations. In an attempt to gain insight into the effect of timing of administration of the agent on the mitigation effect, the ligand to CXCR4, stromal derived factor, SDF-1, was measured as a function of time after radiation exposure. Expression of SDF-1 monitored in skin as a function of time after a 30 Gy radiation exposure suggested a strong correlation between timing of administration of Plerixafor and expression of SDF-1 in irradiated skin: optimum drug administration timing coincided with maximal SDF-1 expression in the skin of irradiated mice. This report presents the first observation that CXCR4 antagonist improves both acute and late skin response to radiation exposure.

INTRODUCTION

Development of an effective mitigator of radiation injury after skin irradiation has the potential to benefit cancer patients receiving radiation therapy, as well as to victims of a radiological or nuclear attack. With victims there is the additional requirement that the mitigator is effective when administered after radiation exposure as prior knowledge of the radiological incident will likely not occur. Additionally, for victims of nuclear terrorism, the mitigator should be effective up to several days after the radiation exposure since that maybe the time required to deliver and distribute the agent.

Plerixafor is a bicyclam compound that inhibits binding of the ligand, stromal cell derived factor-1α, SDF-1α, to its chemokine receptor, CXCR4 (1). Plerixafor was originally developed as JM3100 by Johnson Matthey (JM) and later as AMD3100 by AnorMed (AMD) for potential use in the treatment of HIV because CXCR4 acts as a co-receptor by certain strains of HIV to attach to cells (2). During safety testing in normal volunteers, it was found that Plerixafor mobilized bone marrow-derived stem cells, white blood cells positive for CD34 (3). It has been shown that CXCR4 and its ligand SDF-1α are important in hematopoietic stem cell homing to the bone marrow and in hematopoietic stem cell quiescence (4). CD34⁺ cells mobilized by Plerixafor were capable of engraftment with long-term repopulating capacity up to 1 year in canine transplantation models (5). The drug (brand name Mozobil^®) was recently approved by the FDA for mobilizing hematopoietic stem cells to the peripheral blood for collection and subsequent autologous transplantation in patients with non-Hodgkin’s lymphoma and multiple myeloma in conjunction with G-CSF.

Since Plerixafor has been used to mobilize peripheral blood stem cells for autologous bone marrow transplantation, and radiation skin injury is believed to be a result of loss of a critical number of skin stem cells (6), we hypothesized that CXCR4 antagonist may reduce skin injury when administered after local irradiation.

METHODS

Animal studies were performed according to an institutionally approved protocol governing the care and use of animals (IACUC) in an AAALAC accredited facility at Henry Ford Hospital. The studies were reviewed and approved by the IACUC committee at Henry Ford Hospital. Groups of C57BL/6 mice, 7–8 weeks old, were exposed to radiation alone or in combination with Plerixafor, also known as AMD-3100 or Mozobil^®. Mice acclimated for one week before radiation exposure, were housed in a temperature controlled, HEPA-filtered environment, offered food and acidified water ad libitum. Food was AIN-93G rodent chow (Land O’Lakes Purina Feed, Lansing, MI).

Pharmacological mitigation.

Plerixafor was purchased from Sigma-Aldrich (St. Louis, MO). Plerixafor was administered intraperitoneally at a dose of 5 mg/kg given in 2 doses separated by 2 days and started either on day 0, 4, 7, 15, or 24 after radiation exposure. The drug dose was chosen based on published rodent studies using drug doses ranging between 1–20 mg/kg, with the most common dose being 5 mg/kg. Two doses were chosen because clinically the agent is given on multiple days.

Irradiation.

Mice were anesthetized with an intraperitoneal injection of ketamine (80 mg/kg) and xylazine (8 mg/kg). Ten minutes after the anesthesia injection, animals were placed in a plexiglass jig and immobilized with masking tape. The jig allowed for localized radiation exposure of the right posterior legs of up to 10 mice at a time. Shielding was provided with a square primary collimator (12 × 12 cm) and a circular secondary Cerrobend collimator (three ½ value layers thick). The dose rate from a 6 MV X-ray linear accelerator (Varian Medical Systems, Inc., Palo Alto, CA) was approximately 5 Gy/min using 75 cm source to the surface distance. A 2 cm tissue equivalent bolus was used to bring the maximal dose to the skin surface. Dose was prescribed to the D_max. Doses were confirmed using micro-TLD dosimetry and ranged from 25–30 Gy.

End points.

Skin injury was assessed three times a week for at least 2 months using pooled data from two independent observers and a nonlinear semi-quantitative scale (7). Briefly, skin scores of 2.5 or less represented intact skin with increasing amounts of erythema from 1 (normal) to 2.5 (marked erythema) and increasing amounts of skin dryness from 1 (normal) to 2.5 (start of dry desquamation but intact skin). Whereas skin scores of 3 and higher represented skin breakdown of increasing severity. Skin breakdown accompanied by minimal dry crusting was assigned a score of 3, while skin breakdown accompanied by dry crusting and superficial minimal scabbing was assigned a score of 3.5. A score of 4 was for patchy moist desquamation and moderate scabbing, and a score of 4.5 and above was for more severe injury that would not heal. Histological examination of tissue sections after H&E staining was used to assess skin injury including tissue thickness. ELISA assays were used to determine cytokine levels of TGF-β (eBioscience^®, San Diego, CA; formerly Bender Medical System) and TNF-α (eBioscience) as described previously (8).

RESULTS

The acute and late skin injury in mice receiving Plerixafor was highly dependent on the timing of drug administration. The maximum benefit was observed when the drug was started one week after radiation exposure, because earlier or later administrations of the drug decreased its efficacy (Figs. 1 and 2).

FIG. 1.

Plerixafor given after radiation exposure mitigates skin injury. The hind legs of C57BL/6 mice were exposed to 25 Gy alone, n = 15, or 25 Gy with Plerixafor (5 mg/kg, i.p.) on day 7 and 9 after radiation exposure, n = 14 (Panel A) or 30 Gy alone, n = 16, or 30 Gy followed by Plerixafor (5 mg/kg, i.p.) on day 7 and 9 after the radiation exposure, n = 11 (Panel B). The insets at right illustrate representative maximum responses of mouse skin to radiation alone, and the typical corresponding response of mouse skin to radiation plus Plerixafor at approximately the same time. Dramatic mitigation of skin radiation injury is demonstrated. No error bars are shown for clarity, however SEM typically range between α0.4 and 0.9 on the semi-quantitative scale. By 5 weeks post irradiation, mitigation by Plerixafor is significant (P < 0.05, Kolomogorov-Smirnov non-parametric test).

FIG. 2.

Effect of Plerixafor timing on mitigation. Left panel: Effect of timing of Plerixafor administration relative to radiation exposure on skin response 2 months after irradiation. The data shown are a compilation of 5 experiments: 30 Gy alone, 30 Gy followed by 2 doses of 5 mg/kg, i.p. Plerixafor given on day 3 and 5, irradiation plus drug given on day 7, 9, 14 and 16, or day 23 and 25, with n = 5, 2, 16, 4, and 5, respectively. Error bars are SEM. The curve is a spline fit between 0 and 24 days. Best mitigation was seen when drug was given starting 7 days after irradiation. Right panel: Immunohistochemical staining of stromal derived factor 1 (SDF-1) is shown after 0 Gy alone (control), and 3, 7 or 14 days after 30 Gy irradiation. Examples of representative stained tissues are presented. Note that expression of SDF-1 is higher 7 days after irradiation than it is on day 3 or day 14.

Secondary end points of damage were assessed histologically, tissue thickness and cytokine levels provided confirmatory observations. Histological sections from C57BL/6 mouse skin stained with H&E demonstrated extensive radiation injury 2 months after irradiation compared with that of age-matched controls (Fig. 3). Skin from mice treated with Plerixafor administered 7 and 9 days after radiation exposure (5 mg/kg, i.p.) had a near normal appearance including the presence of hair follicles and sweat glands at 60–90 days after radiation exposure.

FIG. 3.

Plerixafor mitigates radiation injury – Histology/Cytokines. Histological and cytokine end points: Demonstration of mitigation of injury by Plerixafor 2 months after the radiation exposure. Irradiated and control skin was stained using H&E. Representative skin sections at 10× magnification of unirradiated control (panel A), 30 Gy radiation alone (panel B) and 30 Gy plus Plerixafor given on day 7 and 9 at 5 mg/kg i.p. per dose (panel C). Note that mice receiving drug had skin of relatively normal appearance including intact epidermis, and presence of hair follicles (shown by black arrow in panel A) and sweat glands (shown by blue arrow). Whereas skin from mice receiving radiation alone had thin or absent epidermis and a lack of hair follicles and sweat glands. A scab evident over a portion of the irradiated region is shown by a black arrow in panel B. Panel D illustrates a significant reduction in skin cytokines by ELISA when irradiated mice were administered Plerixafor (5 mg/kg, i.p.) on day 7 and day 9 after 30 Gy compared with irradiation alone. Cytokine concentrations were from multiple samples, n = 4 from each of control mice, mice receiving radiation alone and mice receiving combined drug and radiation.

Irradiated skin exhibits a significant increase in a number of cytokines including TGF-β and TNF-α, the expression of which were significantly reduced by the addition of Plerixafor (Fig. 3D). Cytokines are important regulators of tissue response to inflammation/injury and knowledge of their kinetics is key to understanding the mechanism of injury mitigation by Plerixafor.

In an attempt to understand the dependence of the magnitude of mitigation on the timing of drug administration, levels of the ligand to CXCR4, stromal derived factor, SDF-1, were measured at different times after radiation exposure. Expression of SDF-1 monitored in skin as a function of time after a 30 Gy radiation exposure suggested a strong correlation between timing of administration of Plerixafor and expression of SDF-1 in irradiated skin. Optimum drug administration timing coincided with maximal SDF-1 expression (Figs. 1, 2 and 3).

DISCUSSION

Medical management of the acute, sub-acute and late phases of skin injury mostly consists of symptomatic palliative care. Controlling cutaneous infection during the acute phase of moist desquamation is one of the critical determinants for a successful outcome. Various pharmacological approaches including inhibitors of COX1 and COX2 have been made to mitigate and treat the acute moist reaction of skin injury, but none achieved a significant outcome (9, 10). Experimentally, HDAC inhibitor, sodium butyrate and valproic acid, applied topically have shown promise in rodents (11). However, the mitigating effect was not shown when the drug was given systemically. In addition, it is not known whether the HDAC inhibitor would exert its mitigating effect on chronic injury. In an attempt to reduce the chronic dermal and subcutaneous fibrosis induced by radiation, we investigated the efficacy of ramipril, a FDA-approved ACE inhibitor on the radiation-induced skin and leg damage in mice. Ramipril was very effective in reducing the late effects of skin injury, while its mitigating effect on the acute and sub-acute injury was modest (7). However, the dose required to mitigate the late effects was pharmacologically high. Using another approach, we have shown significant mitigation of the acute skin injury with the use of adeno-associated virus encoding manganese SOD gene, when injected subcutaneously shortly after irradiation (12). However, difficulties in delivery, application and cost limit the use of this treatment strategy. Currently, there is no effective countermeasure against acute and sub-acute skin injury, although the combined pentoxifylline and α-tocopherol given to patients with severe subcutaneous fibrosis has shown some limited favorable clinical outcomes (13).

The concept that stem cells could be used to reduce normal tissue injury after irradiation has already been proposed (14). Bone marrow functions as a reservoir for progenitor cells including stromal progenitor cells (SPCs), endothelial cells (EPCs), and hematopoietic progenitor cells. In response to tissue injury, these cells are mobilized from the bone marrow and recruited to tissues where they contribute to tissue repair. Trauma-induced recruitment of EPCs promotes angiogenesis and/or vascular repair in damaged tissues, whereas recruitment of SPCs can result in the production of cells of many different lineages to replenish cells lost in the parenchyma (15). Numerous studies have provided the proof of principle of damage repair by the use of bone marrow cells. Bone marrow cells have been successfully used as a restorative therapy for stroke, myocardial infarctions and osteogenesis imperfecta (16–18). More recently, Chapel et al. have shown that the administration of human mesenchymal stem cells favored healing of the cutaneous radiation syndrome in a xenogenic transplant model (19). However, there are currently both practical and technical complications associated with harvesting, isolating and delivering these allogeneic cells to the desired sites. An alternative strategy for stem cell therapy is to stimulate mobilization of stem cells from bone marrow into the circulation, thereby circumventing these issues. Therefore, the more clinically attractive approach is to use endogenous bone marrow-derived stem cells. Coppes and his colleagues have recently shown that mobilization of bone marrow stem cells by granulocyte colony stimulating factor (G-CSF) induced repair processes, which improved function and morphology of radiation-induced damage to salivary glands in mice (20).

Others have found Plerixafor to be a strong inducer of “mobilization” of hematopoietic stem cells from the bone marrow to the blood stream as peripheral blood stem cells in mice, dogs and humans (5, 21). This FDA-approved agent produced significant mitigation of radiation skin injury in our mouse model when administered following irradiation. The results suggested a dose modifying factor (DMF) somewhat better than 1.2 since 30 Gy with drug yielded a slightly lower level of injury (score of 1.5–2.5 depending on the time; see red squares in Fig. 1A) than does 25 Gy without drug (score of 2 to 3 depending on the time; see green triangles in Fig 1B). However, more extensive studies are required that will include a range of radiation doses. Interestingly, the drug was the most effective in mitigation of acute skin reaction when given days after the single dose of radiation. More detailed kinetic studies are needed to optimize the timing of drug administration. Furthermore, it remains to be determined which subsets of progenitors cells, i.e., mesenchymal, endothelial or hematopoietic cells would be responsible for restoring and replenishing the functional units of injured skin tissue.