TGF-βs are expressed by immune cells and are thought to play an important immunoregulatory role. TGF-β1 is the most extensively described isoform in regard to immunity, mainly because TGF-β1 null mice develop a fatal systemic autoimmune disease [5]. While the role of TGF-β2 is currently considered to be insignificant in the regulation of the immune system because of its negligible expression in immune cells, TGF-β3 is emerging as a potentially important immunoregulator with functions that differ from those of TGF-β1 in many contexts [55].

Several cells of the immune system have been reported to produce TGF-β3 either at the mRNA or protein level, and the role of TGF-β3 has been reported to be both pro- and anti-inflammatory. Interleukin-17-producing helper T cells (Th17 cells) are generated from naïve CD4+ T cells via several cytokines, including TGF-β. Lee et al. found that while generation with TGF-β1 produced a type of Th17 cell that did not readily induce autoimmune disease, TGF-β3 generated a functionally distinct type of Th17 cell that was highly pathogenic [56]. In B cells, TGF-β3 is thought to enhance antibody production under certain circumstances and can possess bifunctional effects depending on concentration [57].

The anti-inflammatory role of TGF-β3 has been highlighted through its ability to inhibit differentiation of forkhead box P3-expressing CD4⁺ T cells [58], its potential to inhibit B cell proliferation and antibody production [59], and its role in lymphocyte-activation gene 3⁺ regulatory T cells (Treg) mediated immune suppression [60]. TGF-β3 was also observed to suppress the development of germinal center B cells and thereby inhibiting the T cell-dependent immune response, a function which was not associated with TGF-β1 [57].

Radiation fibrosis is a late complication of radiotherapy, with potentially debilitating effects. While it is widely accepted that the action of TGF-β1 is pro-fibrotic, and TGF-β2 is generally believed to aid in this action, conflicting reports exist regarding TGF-β3 [61,62,63,64]. Several studies have reported pro-fibrotic effects of TGF-β3: Sun et al. registered upregulation of TGF-β2 and TGF-β3, but not TGF-β1, in human idiopathic pulmonary fibrosis and non-alcoholic fatty liver disease. In addition, they demonstrated pro-fibrotic effects of TGF-β3 in mouse models of these fibrotic diseases [41]. Guo et al. proposed that TGF-β3 promoted liver fibrosis by downregulation of MMP13 [65]. Reggio et al. reported an increased expression of extracellular matrix and inflammatory genes after treating cultured adipocytes and endothelial cells with TGF-β3 [66].

Others, however, have argued the regulatory or anti-fibrotic roles of TGF-β3 [67,68,69,70]. Although they detected the increased expression of TGF-β3 in tissue samples from human myocardial infarction patients, Xue et al. found that TGF-β3 suppressed cardiac fibroblasts and decreased fibrotic markers in an in vitro model [71]. Wu et al. concluded that transplanted mesenchymal stem cells reduced skin fibrosis through the action of TGF-β3 [72]. Escansany et al. observed renal fibrosis in TGF-β3 deficient mice, in addition to impaired lipid metabolism [73].

Wound healing and scar formation are closely tied to the fibrotic response. Early studies indicated that scarless healing observed in mammalian embryos was due to elevated levels of TGF-β3 in fetal tissue [74,75]. Furthermore, Shah et al. demonstrated that cutaneous scarring in rats was reduced by the addition of exogenous TGF-β3 or by the inhibition of TGF-β1 or -β2 [76]. Indeed, recombinant human TGF-β3 under the trade name Avotermin was developed as an anti-scarring agent, showing potential to provide an accelerated and permanent improvement in scarring in several phase I/II trials [77,78,79], but failed the phase III trial [80]. Despite the clinical failure of Avotermin, several more recent studies have supported TGF-β3′s anti-scarring effects [72,81,82,83,84].

Shortly after its discovery, TGF-β1 was identified as a growth stimulator in fibroblasts, but a potent growth inhibitor in other cell types, including epithelial, lymphoid, and endothelial cells [85]. Early studies indicated identical functions for TGF-β2 and TGF-β3, but later research has elucidated isoform-specific variations in the growth-regulatory effects of TGF-βs. The addition of exogenous TGF-β3, but not TGF-β1, increased chondrocyte proliferation in neonatal murine cartilage in vitro [86]. Similarly, TGF-β3 increased chondrocyte proliferation in murine cranial suture-derived mesenchymal cells in vitro, while TGF-β1 had the opposite effect [87]. In contrast, mesenchymal stem cells were found to inhibit keloid fibroblast proliferation in vitro through a TGF-β3-dependent mechanism [72], and the addition of exogenous TGF-β3 reduced the proliferation of cultured taste bud epithelial cells [88]. Treatment with TGF-β3 also inhibited the proliferation of smooth muscle cells in vitro [89,90].

When considering the role of TGF-β3 in radiation cancer therapy, it is also of interest to consider its role in the disease itself. Research regarding the role of TGF-β3 in carcinogenesis and cancer progression is in many cases obscured by an assumption that the three TGF-β isoforms have identical functions and are interchangeable. This assumption is not necessarily true and can often be misleading. Furthermore, the majority of data referenced regarding the tumorigenic role of TGF-β3 originates from protein or mRNA analyses of tumor biopsies and are thus correlative [46]. Considerable caution should be applied before interpreting the data as causative.

In an animal model of cutaneous melanoma, TGF-β3, unlike TGF-β1 and TGF-β2, was induced in cells adjacent to tumor tissue by factors released by the malignant cells, suggesting a responsive rather than a causative role of TGF-β3 [91]. Another study observed decreased TGF-β3 expression in the skin overlying malignant melanoma and suggested that melanoma cells either suppressed TGF-β3 expression or that the lack of TGF-β3 expression itself promoted melanocyte proliferation [92].

In several studies identifying genes associated with prognosis in breast cancer patients, increased TGF-β3 mRNA expression was correlated with a longer interval to distal metastases [93,94,95]. However, increased TGF-β3 protein levels have been observed to correlate with the incidence of lymph node metastases [96], a decrease in survival time [97], and overall survival [98] in breast cancer patients. One study found that the increased expression of TGF-β3 mRNA could lead to ovarian carcinogenesis and tumor progression [99], while another concluded that the increased TGF-β3 expression induced by progestin treatment was associated with a lower risk of developing ovarian cancer in macaques [100]. An analysis of osteosarcoma biopsies revealed a correlation between TGF-β3 protein content and lung metastasis incidence [101]. Leiomyoma samples have been shown to have a higher expression of TGF-β3 mRNA and protein than normal myometrium, and the addition of TGF-β3 increased the proliferation of cultured leiomyoma cells [90,102,103].