**2.2 Telocyte network in gastrointestinal diseases**

*Innovations in Cell Research and Therapy*

(**Figure 4A**–**I**). This mixed TCs/ICCs meshwork and areas of simple apposition occurring between the TCs and the smooth muscle cells have suggested that the intramuscular TCs might support the spreading of the slow waves generated by the ICCs, which are electrically coupled to the smooth muscle cells, thus contributing to the regulation of gastrointestinal motility [32, 41]. In favor of this hypothesis, it has been recently reported that the "smooth muscle cells are electrically coupled to both ICCs and PDGFRα+ cells (i.e., the TCs) forming an integrated unit called the SIP syncytium" [42]. Another possible role attributed to the TCs located in the gut muscle coat is that they might eventually differentiate into ICCs. This hypothesis is mainly based on the existence of the ICCs/TCs mixed network, where the two interstitial cell types are often intercalated (**Figure 4I**) [32, 34, 37]. In support of this hypothesis, despite apoptotic ICCs have been described in the colon of human healthy subjects of different ages, no decrease in the number of ICCs was observed in relation to aging and no ICCs was ever seen undergoing mitosis [43], while mitotic TCs rich in rough endoplasmic reticulum can be detected in the interstitial spaces usually occupied by the ICCs (personal unpublished observation). Taken together, these data suggest that TCs might represent a pool of ICCs precursors being responsible for the physiological replacement of aged ICCs. Furthermore, it has been demonstrated that in culture stromal cells expressing CD34 (a typical marker of the TCs) proliferate and progressively lose their CD34-positivity to acquire the c-kit-positivity (a typical marker of the ICCs) [44]. Reasonably, in adulthood the TCs, wherever they are located, might be considered as a pool of mesenchymal stromal cells and, in the gut, to be important for ICCs renewal [37].

*TCs and ICCs in the human gastrointestinal tract. TCs and ICCs form intermingled networks in the muscularis propria of the human intestine. Representative images of human colon sections double immune-stained for: (A–C) CD34 (green) and PDGFRα (red), (D–F) CD34 (green) and c-kit (red), and (G–I) PDGFRα (green) and c-kit (red) are shown. Nuclei are counterstained with DAPI (blue). Merge images are shown in the right panels. All the TCs are CD34+/PDGFRα+ (A–C), while the ICCs are c-kit+ and negative for either CD34* 

**74**

**Figure 4.**

*(D–F) or PDGFRα (G–I). Scale bar: 50 μm.*

Inflammatory bowel diseases (IBD), including Crohn's disease (CD) and ulcerative colitis (UC), are complex disorders in which chronic relapsing and inflammation progressively evolve into extensive fibrosis of the intestinal wall [41, 45]. Both CD and UC are characterized by abdominal pain and diarrhea, mainly as a result of the progressive fibrotic process that leads to a stiff intestine unable to properly carry out peristalsis and resorptive functions [45–47]. The evidence that intestinal dysmotility with a reduction in the number of ICCs is a feature of IBD and that the ICCs and TCs networks are intermingled in the gut neuromuscular compartment has prompted an investigation of the TCs distribution either in the terminal ileum of CD patients or in the colon of UC patients [32, 41, 48–51]. Interestingly, in both conditions, the gut wall fibrosis was strictly paralleled by a reduction in TCs [50, 51]. In fact, in the CD intestinal wall, which is histopathologically characterized by discontinuous signs of inflammation and fibrosis (referred to as "skip lesions"), TCs were normally distributed in all layers of the healthy-looking areas from the mucosa to the subserosa, while they were markedly reduced in the fibrotic areas displaying severe architectural derangement [50]. In particular, the network of TCs was discontinuous or even completely lost among smooth muscle bundles and around myenteric plexus ganglia [50]. As far as UC is concerned, TCs were investigated in tissue specimens from both patients in early and those in advanced phases of colonic wall fibrotic remodeling [51]. In the early phase, fibrosis is confined to the muscular mucosae and submucosa, while in the advanced one it extends affecting wide areas of muscle layers and the myenteric plexus. Of note, TCs were significantly reduced in the muscular mucosae and submucosa of both early and advanced fibrotic UC cases [51]. On the contrary, the intramuscular and myenteric plexus TCs networks were severely compromised in the advanced but not in the early fibrotic UC cases [51]. Through double immunofluorescence, it was possible to further reveal that in both forms of IBD the losses of TCs and ICCs occurred in parallel in the muscle layers and around the myenteric ganglia [50, 51]. Based on these findings, it has been proposed that the simultaneous reductions in the TCs and ICCs might significantly account for intestinal dysmotility in IBD. Several assumptions have also been made concerning the possible causes and pathophysiologic implications of this TCs impairment [41, 50, 51]. As reported in the failing human heart [41, 52], the progressive alteration in the ECM composition and entrapment of TCs in such fibrotic ECM may provoke profound cell sufferance and eventually lead to cell death. Both the ECM accumulation/rearrangement and the parallel reduction of TCs may profoundly impair their hetero-cellular networks with immune cells, fibroblasts, smooth muscle cells, ICCs, blood vessels, and nerve endings, thus hampering the TCs intercellular signaling functions [41]. Whether the loss of the TCs might even precede the onset of fibrosis rather than being merely a consequence of tissue fibrotic remodeling is difficult to demonstrate. In line with the proposed role of TCs as a guide for the correct tissue shaping during organ morphogenesis, it cannot be excluded that the loss of TCs might contribute to the altered 3D ECM organization in the fibrotic intestinal wall [46]. For instance, it has been proposed that the disappearance of TCs might favor an uncontrolled activation of ECM synthesizing fibroblasts and their transition to profibrotic α-smooth muscle actin (α-SMA)+ myofibroblasts [46]. Noteworthy, hetero-cellular contacts between TCs and fibroblasts/ myofibroblasts have been described in different organs, suggesting that the TCs could contribute to tissue homeostasis by controlling the synthetic activity of such partners through inhibitory signals [46]. In support of this hypothesis, in

UC colonic specimens the loss of TCs was paralleled by an increase in the number of α-SMA+ myofibroblasts [51]. This last observation suggests that during pathologic processes a subset of TC could undergo phenotypic changing, possibly contributing to the increase in the profibrotic myofibroblast population [41]. However, double immunolabeling for CD34 (as TC marker) and α-SMA did not reveal the presence of CD34+/α-SMA+ transitioning stromal cells in the colonic wall of UC patients, which makes unlikely the aforementioned hypothesis [51]. Also, electron microscopy investigations of different pathological tissues (e.g., failing human heart, fibrotic skin) have clearly shown that fibrosis is accompanied by TC degenerative processes rather than by activation/transformation into myofibroblasts [52, 53]. As the findings in the CD and UC intestinal tissues, a loss of TCs have also been reported in the neuromuscular compartment of the fibrotic gastric wall of patients with systemic sclerosis, where it likely contributes to gastric dysmotility clinically manifesting as delayed gastric emptying or gastroparesis [54].

Besides IBD, recent evidence suggests that the TCs might be a cell source of certain kinds of gastrointestinal stromal tumors (GIST) [55]. While ICCs hyperplasia has been identified as a crucial pathogenic feature of KIT-mutant GIST, it has been proposed that the TCs could represent the physiological counterpart of PDGFRa-mutant GIST and inflammatory fibroid polyps [55]. Indeed, a pathogenic relationship between TCs hyperplasia and both inflammatory fibroid polyps and PDGFRa-mutant GIST has been suggested. Moreover, the term "telocytoma" was proposed for defining inflammatory fibroid polyps, since it conveys both the pathogenic (neoplastic) and histotypic ("telocytary") nature of this tumor [55].
