Comparison of acute proton, photon, and low-dose priming effects on genes associated with extracellular matrix and adhesion molecules in the lungs

Ionizing radiation (IR) includes photons, small packets of energy that carry electromagnetic radiation, as well as particle radiations such as protons. Among the forms of photon radiation, γ-rays have the smallest wavelength and the most energy of any other wave in the electromagnetic spectrum. In contrast, protons are subatomic particles with an electric charge of +1 and have greater biological impact compared to photons [1]. Virtually all forms of radiation are of concern to the National Aeronautics and Space Administration (NASA). Crew members on space missions are routinely exposed to low dose background radiation that is more than 150 times greater than on Earth [2] and are at risk for much higher doses during a solar particle event (SPE) that consists primarily of proton radiation [3]. Thus far, pulmonary abnormalities noted in astronauts have been attributed primarily to microgravity [4, 5]; risk for lung complications associated with space-relevant radiation are unknown. There have been essentially no investigations that directly compare acute proton and photon effects on normal lung tissue and possible modification of the outcome due to low-dose/low-dose-rate (LDR) γ-ray pre-exposure. Better understanding of radiation effects on normal lung tissue also has relevance in the clinical setting. Unfortunately, patients who receive radiotherapy for lung cancer develop side effects such as inflammation and sometimes even fibrosis [6, 7]. Protons, however, can be delivered to the tumor at a higher dose, while reducing the dose to normal tissue [8]. The physical advantage of a proton beam compared to conventional radiotherapy (X-rays) is that the beam can be modulated to deliver most of the dose to the intended target, that is, at the peak of the Bragg curve. With conventional radiation, the maximum dose is delivered within a few centimeters of the skin surface proximal to the target. Hence, proton radiation continues to be used with increasing frequency for the treatment of patients, including those with lung cancer [9].

The lungs are among the most radiosensitive organs in the body. Our previous investigations have shown that acute photon delivery resulted in profibrotic changes in the lungs of mice [10]. Lung repair is initiated immediately following injury and includes an acute inflammatory response, cytokine and growth factor release, activation of localized stem cells, and cell-cell and cell-matrix interactions mediated through cell adhesion molecules (CAM) [11]. Radiation-induced lung fibrosis (RILF), a major late effect of photon radiation damage [12], is characterized by loss of epithelia and excessive deposition of collagen and other extracellular matrix (ECM) components. CAM is believed to participate in fibrogenesis since relatively abundant CAM proteins and re-regulated mRNAs are detected in specimens of pulmonary fibrosis [13, 14]. CAM-mediated adhesive interactions that may be involved in the pathogenesis include cell-ECM and cell-cell interactions that are mediated through several CAM families, including the integrins, cadherins, selectins, and members of the immunoglobulin superfamily. Therefore, stable adhesion during these interactions is essential for adequate cell communication, epithelial integrity, and ECM homeostasis.

Several evidences have shown that transforming growth factor (TGF)-β plays a primary role in the fibrotic process. During fibrogenesis, epithelial cells lose their characteristic markers such as E-cadherin responsible for their adhesion, and the expression of α-smooth muscle actin (SMA), a myofibroblast marker capable of producing abundant collagen and other ECM molecules [15, 16], and Slug are enhanced. Slug (Snail 2) acts as a repressor of E-cadherin [17, 18]. However, whether the expression of these markers for fibrogenesis are affected by protons or combination of irradiation with LDR-γ rays has not been identified to date.

Interestingly, an increasing number of studies have shown that exposure to low-dose radiation can result in radioadaptation that can be beneficial in that it renders cells more resistant to a subsequent acute radiation event, as well as more resistant to cancer and certain other pathologies [19, 20]. The existence of this phenomenon, however, remains controversial. Thus, the overall focus of the present study was on lung parameters that may be altered due to excessive radiation exposure in the spaceflight environment, as well as nuclear/radiological events on Earth. Expression profiles of genes that are especially important in regeneration and remodeling of lung tissue after acute proton and photon irradiation were compared, both with and without pre-exposure to LDR γ-rays. In addition, protein profiles of a critical profibrotic cytokine and three markers for fibrogenesis were also compared between irradiated groups and the 0 Gray (Gy) control. The selection of day 21 post-irradiation rather than earlier time points was based in part on the premise that long-term changes in gene and protein expression patterns may have greater health-related significance than short-term changes that may simply be due to transient efforts to regain homeostasis. Day 56 was selected as the second time point to determine if the modulations observed in the lungs on day 21 were persistent. The present study is the first to investigate ECM and CAM response in the lungs under the conditions used.

The lung is dynamically remodeled in response to injury, in which the change in ECM compositions can lead to either healthy or impaired regeneration. The tissue integrity and cell-cell or cell-matrix communications rely on cell adhesion molecules. Since protons and photons are two different forms of radiation, the biological effects generated by them are likely to be different. In this study, the changes in histology indicated that both types of radiations can induce profibrotic responses. Increased deposition of collagen in the lungs from irradiated animals (Figures 1 and 2) implies that activation of collagen producing cells is accelerated, leading to fibrosis-like change. Many studies have shown that low-dose radiation can induce radioadaptation that renders cells more resistant to a subsequent acute radiation event [20]. In this study, accumulation of more abundant collagen in both groups that received dual irradiation indicates that pre-exposure to LDR γ-rays did not render normal lung tissue more resistant to excessive production of this protein caused by acute radiation exposure.

In the present study, the differences between protons and photons in modulating expression of the genes related to ECM or CAM indicated that sensitivity of these genes to the two types of acute radiation regimens was different. More genes were affected by protons than by photons at both time points, suggesting that these relevant genes are more susceptible to protons. The markedly higher number of genes that were upregulated in the LDR + Photon group at both time points compared to the one that received only acute photons (Table 2) indicates that the protracted low-dose priming with γ-rays triggered mechanisms that rendered genes more responsive to acute photons. Only three genes (col1a1, mmp-14, and mmp-15) were upregulated in all irradiated groups on day 21, indicating that sensitivity of most genes was dependent on the type of radiation regimen. Col1a1 encodes one of the markers for mesenchymal cell lineage. In addition, overexpression of matrix metalloproteinase (MMP) family members such as MMP-14 and MMP-15 is a consequence of perturbation of the balance between synthesis and degradation of collagen and other ECM components [21]. MMP-14 is capable of proteolytic degradation of type I, II, and III collagens following the characteristic cleavage pathway [22]. MMP-14 also cleaves many membrane-anchored proteins such as E- and N-cadherin, integrins, CD44 (a hyaluronan receptor), and several cell surface proteoglycans and their receptors [22]. MMP-15 is a ubiquitously expressed enzyme with largely overlapping substrate specificity with MMP-14 [23]. The profibrotic change in histopathology post-irradiation indicates that all radiation regimens used in the present study resulted in perturbation of normal tissue remodeling and excessive production of collagen. The overproduction of collagen and/or ECM may initiate transcription of genes that downregulate MMP, that is, MMP inhibitor genes, timp-1 and/or timp-3. Of course, the final fate of lung status after irradiation depends on whether normal or aberrant repair takes place, a process that is largely determined by cell microenvironment.

In addition, cd44 and itgav (encoding a αv subunit of integrin) were significantly upregulated in the Proton and both combination groups on day 21. CD44 mediates cell-cell and cell-matrix interactions through its affinity for hyaluronic acid (HA), and also other ligands such as collagens and MMPs. Therefore, altered expression or dysfunction of CD44 may cause pathogenic phenotypes. The itgav gene encodes a receptor for many ligands, such as fibronectin, fibrinogen, laminin, and MMP-2. Moreover, the αv subunit can also activate TGF-β1 when it is associated with a β6 or β8 subunit on airway epithelial cells, leading to poor wound healing [24]. When cell surface proteins like E-cadherin and integrins that mediate epithelial connections to neighboring cells and the basement membrane, respectively, are replaced by N-cadherin and ectopic integrins, the cell may be primed for the mesenchymal phenotype by more transient adhesive properties [25]. In addition, we also noted that the two combined radiation groups had more upregulated integrin genes that participate in cell-matrix adhesion and/or are collagen and ECM structure constituents. This was especially evident on day 21 and suggests that the combined irradiation may also cause dysregulated repair. In addition, significant downregulation of itga4 by γ-rays, either at a high dose or a low dose, indicates that this gene was highly sensitive to γ-ray photons.

Two collagen genes (col4a3 and col5a1) and selp (P-selectin, platelet) that were significantly upregulated on day 21 after protracted exposure to LDR γ-rays combined with acute radiation further indicates that synthesis of collagen may be enhanced. P-selectin is an inflammatory adhesion molecule found on endothelial cells and platelets. It enables the recruitment of leukocytes to the endothelium and activates platelets. Platelets contain a plethora of growth factors and cytokines, including high concentrations of TGF-β [26] that is well known to be a strong inducer of fibrogenesis. Thus, upregulation of selp can lead to the release of more cytokines and growth factors from activated platelets, potentially promoting a mesenchymal phenotype [27].

The enhanced appearance of major profibrotic proteins in response to the radiation regimens also indicates epithelial early injury. Here we noted excess production of collagen and ECM molecules after acute and combination irradiation. It is well established that TGF-β1 promotes differentiation of fibroblasts into activated myofibroblasts, enhances collagen synthesis, and reduces collagen degradation by downregulating proteases and upregulating protease inhibitors [28]. Increased expression of TGF-β1 in the samples from irradiated groups at both time points indicates that the mechanisms responsible for production of this cytokine are very sensitive to ionizing radiation, even low-dose-rate photons delivered over an extended time period.

Myofibroblasts are strongly associated with tissue repair and fibrosis. α-SMA, a major marker of myofibroblasts, is linked to cell-cell adherence sites and cell-matrix anchorages, the latter being essential for contractile as well as biosynthetic functions. The changes in histology and collagen staining in the irradiated groups may partly result from transition of more myofibroblasts from other cell types, for example, epithelial or adult stem cells. TGF-β1 is one of several profibrotic cytokines that activate myofibroblast progenitors [29]. As a result, excessive accumulation of collagen and ECM can occur. Reactive oxygen species (ROS) generated by radiation exposure are immediate activators of TGF-β1 [30, 31]. Evidence suggests that once activated, myofibroblasts may not require continued paracrine stimulation by fibrogenic cytokines, relying instead on positive autocrine feedback [29].

E-cadherin is broadly expressed only by epithelial cells. It is also a key component of adhesion junctions that play an important role in maintenance of epithelial integrity [32]. In this study, low ratio of E-cadherin to β-actin bands indicates that this epithelial marker was sensitive to radiation. Low E-cadherin is associated with breakdown of epithelial integrity and also poor repair capacity.

Jayachandran et al. [33] showed that TGF-β1 treatment increased the expression and nuclear accumulation of Slug, a transcription factor, in concert with induction of EMT in type II alveolar cells. However, in the present study, Slug expression was not markedly affected by ionizing radiation at the two time points of assessment, thereby implying that Slug protein, unlike TGF-β1, E-cadherin, and α-SMA, was not sensitive to irradiation. Further study will explore the expression of Slug gene post-irradiation.

This study provides new information on normal lung response to proton radiation. Both acute proton and photon radiation regimens induced overexpression of certain ECM/CAM genes and profibrotic mediators, leading to production of excess collagen and ECM. The present data also showed that the effect of protons on gene expression was more profound compared to photons. The differences in gene expression that were noted between the acutely irradiated groups are likely due to the different physical properties of protons and photons. Protons are particles with a positive electrical charge equal to that of electrons, whereas photons are a type of electromagnetic radiation consisting of elementary particles. Furthermore, pre-exposure to LDR γ-rays resulted in more modification in the gene profile when followed by acute photon irradiation compared to acute protons. These findings have implications related to human space exploration, as well as radiation exposure above normal on Earth. Future studies should include more time points and multiple doses after exposure to both forms of ionizing radiation. Furthermore, since fibrosis is an unlikely outcome after a 2 Gy dose as used here, future studies should include higher radiation doses localized to the lungs and longer intervals post-irradiation to determine if a difference exists between photons and protons in ultimate outcome and if LDR pre-exposure has modifying effects.