Normal gastric motor function requires coordinated interaction between the central nervous system (CNS), the enteric nervous system (ENS), interstitial cells of Cajal (ICC), and smooth muscle cells (SMCs). Gastroparesis is a multifactorial neuromuscular disorder involving dysfunction at various levels of this network. For example, vagal dysfunction, disruption of the ENS (particularly loss of nitrergic neurons), smooth muscle abnormalities, and deficits in ICC networks have all been implicated in the pathogenesis of diabetic gastroparesis . Given this complexity, it is not surprising that our current therapies, whether pharmacologic, device-based (electrical stimulation), or surgical (gastrectomy or pyloromyotomy) fail to repair this functional interaction. An ideal approach would be to replace or replenish permanently impaired cells while maintaining the spatial and temporal function of the neuromuscular network. Given recent advances in stem cell biology, cell-based therapies offer promise.
In this chapter, we will discuss the progress that has been made in the discovery of stem cell populations that may be suitable for cell-based therapies for gastroparesis, and highlight challenges that remain to bring such therapies to patients. We will focus on CNS-derived neural stem cells (CNS-NSCs), enteric neural stem cells (ENSCs) derived from the gut, mesenchymal stem cells (MSCs) and progenitors of ICCs, and pluripotent stem cells ( Fig. 39.1 ). We will also discuss technical considerations to enhance stem cell engraftment.
Central nervous system-derived neural stem cells
Isolated from distinct regions of the brain, namely the hippocampus and subventricular zone , CNS-NSCs are multipotent “adult” stem cells that have remarkable plasticity when transplanted to other tissues—including the gut. They are capable of differentiating into nNOS-expressing neurons, a key neuronal subtype in the ENS, both in vitro and in vivo after injection into rodent stomachs . CNS-NSCs can be induced towards an enteric phenotype by co-culturing with dissected longitudinal muscle and myenteric plexus gut tissue . Upon differentiation, these neurons exhibited a negative resting membrane potential, the presence of voltage-gated sodium channels, and after-hyperpolarization current characteristic of enteric intrinsic primary afferent neurons (IPANs) . CNS-NSCs appear capable of homing to sites of injury in the gut. For instance, when injected intravenously into mice following intestinal surgery, they migrated to and differentiated into neurons at the surgical anastomosis . Importantly, in an animal model of gastroparesis (nNOS −/− mice) , CNS-NSCs transplanted into the stomach partially restored gastric function after 1 week . However, for unclear reasons, there was not sustained engraftment, with significant cell loss 2–4 weeks after transplantation. Unfortunately, the lack of a readily accessible source for CNS-NSCs, particularly given the limited potential of adult stem cells to be propagated in vitro , provides a significant challenge for their use in cell-based therapies.
Enteric neural stem cells
Resident gut NSCs, termed ENSCs, have been isolated from the embryonic and postnatal intestines of animals and humans and can give rise to new enteric neurons and glia when transplanted back into the intestine . Notably, ENSCs exhibited more robust migration, proliferation and differentiation into enteric neurons when transplanted into mouse colons compared to CNS-NSCs , suggesting they are a preferential NSC population for gut cell-based therapies. Similar to CNS-NSCs, ENSCs have a limited capacity to be expanded and propagated in vitro . However, Metzger et al. demonstrated that ENSCs could be derived from mucosal biopsies taken during endoscopy . This offers the potential for a replensishable source of stem cells, through a minimally invasive procedure, that is amenable to autologous transplantation. Nonetheless, for ENSCs to be a viable solution, there is a need to significantly scale up the numbers harvested from tissue. To this end, various methodologies have been developed to expand ENSCs in culture prior to their transplantation. While commonly cultured and transplanted as non-adherent neurospheres , Rollo et al. reported that using an initial monolayer culture prior to their enrichment with flow cytometry improved the yield of human ENSCs for transplantation . Supplementation of the culturing media with glycogen synthase kinase 3 inhibitor (CHIR-99021) , 5-HT 4 receptor agonist or glial cell line derived-neurotrophic factor (GDNF) was also found to expand ENSCs in culture.
ENSC-derived neurons can form functionally-relevant connections in host tissue. Neurally-mediated electrical activity and calcium transients have been observed following gut transplantation of human and mouse ENSCs . Investigators have also utilized optogenetics technology to demonstrate functional innervation of smooth muscle cells following their transplantation . Importantly, ENSC transplantation restored gut function in an animal model of gastrointestinal dysmotility. Injection of these cells into the colon of nNOS −/− mice restored nitrergic response and increased intestinal transit when compared to sham injected mice . Whether ENSCs can restore gastric function in animal models of gastroparesis remains to be explored.
Progenitors of interstitial cells of Cajal
Loss of ICC has been implicated in the pathogenesis of gastroparesis . Thus, regeneration of ICC networks by transplantation of ICCs, ICC progenitors, or mesenchymal stem stells (MSCs) has been proposed as a therapy. Kit + mesodermal precursors of ICCs have been characterized in embryos and in the immediate post-natal period . More recently, progenitors of ICCs have been found in adult guts of rodents that are Kit + and express the adhesion marker CD34 . While the potential of ICC precursors in cell-based therapies has not been explored, transplantation of mature ICCs to gut explants restored pacemaker activity in an animal model of ICC loss . Joddar et al. found that MSCs administered to stomach explants integrated into the intramuscular layer and differentied into cells expressing ICC marker c-kit . Interestingly, McCann et al. found that transplantation of ENSCs increased ICC numbers in the colon of NOS −/− mice . Whether restoration of ICC and neural networks in gastroparesis would require transplantion of ENSCs alone or in combination with MSCs or other ICC populations remains to be seen.
Pluripotent stem cells
Embryonic stem cells (ESCs) and induced pluripotent stem cells (iPSCs) are pluripotent stem cell populations capable of differentiating into all three germ layers. A major advantage of these stem cells is their abundance due to ease of propagation in culture. iPSCs, which are generated by reprogramming somatic cells into a pluripotent state , avoid the ethical controversy surrounding ESCs and provide an accessible source of immunologically-compatible stem cells for autologous transplantation. However, a drawback of pluripotency is an increased potential to form teratomas and difficulty differentiating to a desired cell fate . Fortunately, “nudging” pluripotent stem cells towards the desired lineage prior to transplantation appears to decrease tumorgenicity and improve engraftment . There has been significant progress in defining techniques and protocols to drive pluripotent stem cells towards a neural crest (NC) lineage . ESC-derived NC cells are capable of colonizing explanted gut and differentiating into glia and neurons, including enteric neural subtypes that express nNOS, 5-HT, and GABA . Notably, human ESC-derived NC cells transplanted into the colon of an animal model of Hirschprung’s disease improved survival . Of note, although the transplanted NC cells demonstrated extensive migration throughout the entire colon, they preferentially localized to the submucosal region . It remains unknown whether this would limit their utility for treating motility disorders like gastroparesis.
Optimization of cell transplants
Irrespective of which cell type is transplanted, several questions remain regarding how to achieve optimal stem cell engraftment. First, the best method for cell delivery in gastroparesis remains unclear. Most transplantation experiments involve direct injection of stem cells into the serosal side of the gut wall . In particular, Micci et al. directly injected the stem cells into the subserosa at the level of the pylorus . While such an approach might be appropriate for repopulating discrete regions of the ENS, it would likely require invasive surgery. Additionally, the extent to which injected cells can migrate over a large area of tissue, such as the human stomach, will need to be determined for clinical translation. Alternative delivery approaches including intraluminal , intraperitoneal , and intravenous administration, are less explored and warrant further investigation since they would be less invasive and more amenable to repeated stem cell administrations. In addition to the route of administration, another important factor in graft success is how the cells are administered. Stem cells have been coated in various materials including Matrigel , thermosensitive hydrogel , and fibrin matrix which appear to improve efficiency of transplant. Notably, a gelatin-alginate hydrogel enabled successful engraftment of MSCs when applied to the stomach lumen, suggesting that oral or endoscopic administration may be an option .
Investigators have also explored whether including other factors may benefit stem cell transplantation. Hotta et al. showed that supplementation of ENSCs with nanoparticles loaded with a 5-HT 4 receptor agonist promoted proliferation and neuronal differentiation following transplantation . Similarly, administration of CNS-NSCs in parallel with the 5-HT 4 agonist, mosapride citrate, enhanced neuron expansion and survival in a surgically anastomotic mouse ileum . Co-administration of a caspase-1 inhibitor with CNS-NSCs also improved graft survival . Finally, when ENSCs were pretreated with GDNF prior to their transplantation, they exhibited improved migration within the host tissue .
Lastly, it remains unexplored how factors in the host might affect stem cell engraftment. Immunosuppression might be required for transplantation of non-autologous stem cell populations (i.e. ESCs). Additionally, immune dysregulation has been implicated in the pathogenesis of diabetic and idiopathic gastroparesis which could potentially make the microenvironment less permissive to cell transplantation. The age of the host may be an additional consideration. We have found that aging causes neuroinflammation in the ENS and that elevated inflammatory cytokines, particularly IL-6, result in increased apoptosis of ENSCs in culture . Clearly, further examination of the impact of the host microenvironment on cell-based strategies using aged animals and disease models that more closely reflect the human condition (i.e. diabetic mouse models ) is warranted.
Given the number of clinical trials using stem cells in various organs, physicians will likely start seeing cell-based therapies as part of their arsenal in the near future. It is only a matter of time before we determine which stem cell populations are best for treating neuromuscular disorders of the stomach. However, several obstacles remain with regard to clinical application . What is the best method for delivering stem cells to the diseased tissue? Will they require endoscopic or surgical transplantation or can they reach the appropriate site following intravenous infusion or oral administration? Do they require embedding in a polymer or other matrix prior to transplantation? Will immunosuppressive therapies be required? Is prolonged engraftment possible or will repeat transplantations be required? Over time will these stem cells be predisposed to tumor formation? Despite these hurdles, novel cell-based therapies for treating debilitating disorders like gastroparesis will hopefully be realized in the near future.