This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
The direct conversion of somatic cells into another terminally differentiated cell type poses a major challenge to the widely held view that differentiated somatic cells cannot alter their final cell fate. Please review the literature to document the experimental approaches undertaken for the direct conversion of dopamine neurons and determine the distinctiveness of each study. Please comment if the dopamine neurons that have been obtained are suitable immediately for clinical applications. Outline what further improvements are needed, if any. (AP Chan Woon Khiong)
Direct conversion to dopaminergic neurons and its readiness for clinical translation
A series of research on direct conversion of somatic cells in vitro to generate functional dopaminergic neurons (DN) has been published just over the past year. Owing to the work pioneered by Vierbuchen et al. (2010), who reprogrammed mouse fibroblasts into neurons by transcription factors (TFs)-oriented approach, many groups have successfully improved the neuronal specification with similar technique, thereby highlighting the clinical potential of cell therapy for neurological disorders such as Parkinson's disease (PD). PD is characterized primarily by motor symptoms that result from loss of A9 dopaminergic neurons in the substantia nigra (SN).9
Conversion to DN can be achieved either directly (iDA)2,3,4 or via Vierbuchen's induced neuron (iN) en route to iDA.5,6 The common strategy begins with lentiviral-mediated transduction of a small pool of TFs, known to be essential for DN development, to the terminally differentiated non-neuronal cells, like fibroblasts. Scoring for the resulting phenotypic and genotypic changes was subsequently conducted in favour of minimal requirement of factors for successful dopaminergic conversion. Caiazzo et al. (2011) reported the use of three TFs (Ascl1, Nurr1, Lmx1a) to generate DN directly from not only mouse embryonic (MEFs) and adult fibroblasts but also from adult human fibroblasts, both healthy and Parkinsons' disease patients, interestingly with a comparable efficiency. Other reports by Kim et al. (2011) made use of three additional TFs (Pitx3, Foxa2 and EN1) while that by Liu et al. (2012) opted for Ascl1, Nurr1, Pitx3, Ngn2 and Sox2. Obtaining iN prior to iDA specification, Pfisterer et al. (2011) used Lmx1a and Foxa2, in conjunction with iN inducing factors (Brn2, Ascl1 and Myt1l); while Sheng et al. (2012) mixed iN's Ascl1 and Brn2 with three iDA factors (Lmx1b, Nurr1, Otx2). In general, Ascl1 (Mash1) has been identified as the fundamental and obligatory proneural gene to introduce, while other factors enhanced iDA phenotypic and functional maturity. In addition to TFs transduction, neutrophic factors such as Sonic hedgehog (Shh) and fibroblast growth factor 8 (FGF8) were also included to enhance the conversion efficiency.3
The resulting cells displayed typical dendritic neuronal morphology and were Tuj1-positive. Expression of several dopaminergic markers, including tyrosine hydroxylase (TH), dopamine transporter (DAT), vesicular monoamine transporter 2 (VMAT2), cytoskeletal protein MAP2 and particularly, En1 and Pitx3; and absence of adrenergic, serotonergic and cholinergic components further validated its iDA identity.2,3,4,5,6 While Kim's, Liu's and Pfisterer's group complemented the visual data with quantitative real-time PCR (qRT-PCR), Caiazzo et al. (2011) provided a more extensive coverage on gene expression profiles by performing global mRNA-array analysis. In agreement with upregulated dopaminergic markers and downregulated (lost) fibroblasts genes, the iDA was found to cluster with native adult mesencephalic dopaminergic (mDA) neurons.
To complement iDA phenotype, its functional maturity was further examined through a series of electrophysiological and biochemical tests such as patch-clamp recordings, pharmacological inhibition or stimulation as well as high performance liquid chromatography (HPLC).2,3,4,5,6 Briefly, the iDA displayed a stable resting membrane potential (-50 to -70 mV) and firing of mature action potential (AP) with amplitude similar to that of mDA neurons.2,3,4,5,6 Voltage-dependent sodium and potassium channels were also developed as demonstrated by abolishment of AP upon tetrodotoxin (TTX) blocker administration.2,3,4,5,6 Such electrically-responsive iDA was able to secrete its characteristic neurotransmitter, dopamine, upon potassium stimulation.2,3 Furthermore, Liu et al (2012) demonstrated 3H-dopamine uptake by iDA via D2 receptor (D2R), known to be responsible for negative feedback regulation of DN firing. Introduction of D2R antagonist, raclopride, disengaged the feedback loop resulting in enhanced AP production. Conversely, Caiazzo et al. (2011) showed reduced AP firing by iDA when treated with D2R agonist, quinpirole. Essentially, both Caiazzo et al. (2011) and Pfisterer et al. (2011) demonstrated the iDA potential to network with other neurons as shown by positively-stained synaptotagmin 1, synapsin and synaptophysin, respectively.
By using doxycycline-inducible lentiviral system, the exogenous TFs could be temporally regulated, such that the TFs were activated in the presence of doxycycline and vice versa. It was found that initial continuous supply of TFs, for at least three days, was required for direct conversion into neurons. Thereafter, removal of doxycycline did not significantly alter the neuronal phenotype and conductivity.2,3,5 The iDA stability could remarkably be maintained for at least 24 days.2 Endogenous expression of dopaminergic terminal selectors, stimulated by the transduced TFs, seems to be the reason behind such eventual independence.2
Clinical application of mature 'synthetic' iDA could only be substantiated with in vivo transplantation in suitable animal models. Caiazzo et al. (2011) transplanted the transduced fibroblasts of TH-GFP transgenic mice into neonatal mouse ventricle. Two- and six- weeks post-transplantation, GFP- and iDA markers-positive cells were observed being integrated in the host tissue. Further electrophysiological test showed that the iDA conductivity was still maintained. This implies that iDA maturation could occur in vivo. Similar grafting was also conducted by Kim et al. (2011) who, instead, transplanted Pitx3-eGFP-positive iDA neurons into the striatum of 6-hydroxydopamine (6-OHDA)-treated parkinsonian rodent. The iDA did not only survive but also get integrated with the host for at least eight weeks. Meanwhile, Liu et al. (2012) demonstrated even longer lifespan of human iDA (hiDA) that was xenogeneically transplanted in mice, up to 16 weeks. Most importantly, grafting in parkinsonian animal models resulted in elevated dopamine levels which correlated with significant stabilization of their rotational behaviours.
It is worth noting that another group, Addis et al. (2011) opted for astrocytes as the alternative cellular starting material, reasoning that the developmentally close relationship of astrocytes to neurons could improve the dopaminergic conversion efficiency. Additionally, astrocytes make up the bulk of central nervous system (CNS) thereby raising the potential of direct in vivo reprogramming without the need for transplantation.1 Even though the group has not quite addressed the proposed in vivo conversion, they have successfully reprogrammed postnatal mouse astrocytes cell line to become functional dopaminergic neurons in vitro by transducing doxycycline-regulated lentiviral vector carrying three polycistronic TFs (Ascl1, Nurr1 and Lmx1b).1 Expectedly, the conversion efficiency was higher than most of fibroblasts-derived iDA, perhaps due to equal stoichiometry provided by the polycistronic vector.1
Taken together, these studies highlight remarkable progress in deriving a particular cell type in a controlled manner. In comparison with stem cells-derived neurons, direct conversion seems to be the much preferred strategy for cell transplantation therapy. After all, iDA has been skilfully generated from mouse and human donors without involving potentially tumorigenic pluripotent intermediates. Evidently, the conversion process was so rapid that it could be achieved as early as three days.2 Moreover, bromodeoxyuridine (BrdU) signal was not detected in iDA cells, indicating its non-proliferative status.2,4 However, the downside of post-mitotic cells is limited amounts of converted iDA obtained, especially when the reprogramming efficiency is already low (generally less than 25%) to begin with.10,12
Judging from the low conversion rate, there are highly likely other factors that could influence the quantity of converted iDA. After all, the mechanisms behind TFs-mediated iDA conversion are still unknown. It is also unclear if direct dopaminergic conversion was simply a matter of finding the ideal TFs recipe. Even so, comparison across different studies would not be as straightforward for method standardization is lacking, such as the properties (age) of cells used, the culturing conditions, the day post-transduction certain measurement was conducted, the dopaminergic markers adopted, just to name a few. Apart from the instructive intrinsic determinants and extrinsic molecules, the conversion efficiency might be improved with a more precise temporal regulation such as adding the TFs in a sequential manner.7,8 Moreover, spatial adjustment (supportive growth substrate) to promote cell-cell interaction, thereby resembling the native environment, could have significant contribution to the iDA survival and functional maturity.8
Caiazzo et al. (2011) also showed the conversion efficiency of mouse fibroblasts to be higher than that of adult human pointing to species differences. Meanwhile, Pfisterer et al. (2011) found that the efficiency of postnatal human fibroblast was lower than that of embryonic cells, suggesting that age of the donor cells could have influence on the propensity to take in new identity. Therefore, further improvement is obviously needed to expand the converted iDA to a reasonable quantity for clinical use.11
Despite the dopaminergic features displayed, iDA identity should still be accepted with scepticism. Most studies depended on cell markers such as Tuj1 or MAP2 expression to identify neurons while Yang et al. (2010) cautioned that these two proteins are not the definitive neuron markers. Similarly, Kim et al. (2011) argued for the superiority of Pitx3 marker due to its exclusive activity in the midbrain dopaminergic neurons, unlike TH which can also be expressed by norepinephrinergic and adrenergic cells. Distinctiveness of markers is crucial to obtain mesencephalic iDA, rather than generic iDA with diverse regional identity. After all, the general consensus is that only midbrain SN DN are able to replace the neurons lost in Parkinson's disease, probably owing to its unique target structure, the dorsolateral region of the host striatum.8 Robust markers are also needed to isolate a pure population of target cells, especially when not all Tuj1-positive neurons were dopaminergic and neurons of both excitatory and inhibitory GABAergic phenotypes were present in Pfisterer's culture. Essentially, Caiazzo et al. (2011) found that their iDA gene expression was not exactly identical with that of the native mDA neurons, suggesting that the fibroblastic epigenetic memory could still have been retained. This emphasizes the needs to study direct conversion mechanism from epigenetic perspective, which is obviously lacking at the moment.
In addition to iDA identity, completeness of the iDA conversion is also crucial. Similar to procedures consistency, the grading of neuronal maturity is yet to be established. Favourably, Yang et al. (2010) proposed standard criteria pertaining to the extent of neuronal reprogramming in general, distinguishing partially and fully reprogrammed iN cells. According to the suggested standardization, all of the currently reported iDA obtained by direct conversion were at most generic. Yang et al. (2010) placed synaptic functions as the most stringent criterion which was not adequately and electrophysiologically demonstrated in vivo by most groups. Although, elevated dopamine was observed, having transplanted the iDA within host striatum instead of SN, it is unclear if the damaged circuit has been properly reconstructed such that the grafted iDA could control the dopamine release in response to negative feedback.
One may argue that practical in vivo iDA transplantation in parkinsonian model should be sufficient to test iDA clinical readiness. However, the in vivo experiments conducted thus far are arguably not as comprehensive as the in vitro tests. Several factors have to be considered before jumping to conclusions, especially the quality of animal models employed in terms of etiological and pathological resemblance with human patients. It is apparent that the current diagnosis relies solely on the mice rotational behaviour, an oversimplification if it were to be compared with human patients' motor abnormalities that not only include resting tremors but also others such as postural instability and deformity, freezing and even non-motor malfunction.9 Furthermore, the extent of denervation caused by 6-OHDA lesion may only be singly represented, thereby overlooking the different Parkinson's disease staging.9 Rather than conducting the transplantation to mice of different ages, particularly old ages to assimilate the elderly Parkinson's patients, Kim et al. (2011) used three-month-old mice instead. While young niche was able to cooperate with exogenous iDA, the expectedly different neutrophic environment in older mice could have responded differently.14 Also, the quantity of iDA to be transplanted is another important issue because excessive dopamine-secreting neurons could result in dyskinesia.7,14
Having all these parameters not quite been tested meticulously, applicability of iDA in the clinical setting is still relatively early to be announced. In order for iDA to gain clinical acceptance, it will be critical to improve its conversion efficiency and functional maturation by refining the methodology employed, as well as to extend the in vivo transplantation experiment more in-depth. Safety concern about the use of highly efficient lentiviral system to deliver the exogenous TFs and its potential transforming integration event has relatively been a universal problem for cell therapy, be it stem cells-derived or transdifferentiation.10,11,12 Although there are still so much things to be done, iDA transplantation therapy for PD is still an optimistic goal. The clinical applicability of iDA, with the current drawbacks overcome, will certainly be much more beneficial compared to the existing short-term PD symptomatic relief.7,14
Word count: 2,026