Tumor suppressors inhibit reprogramming of African spiny mouse ( Acomys) fibroblasts to induced pluripotent stem cells

Background: The African spiny mouse ( Acomys) is an emerging mammalian model for scar-free regeneration, and further study of Acomys could advance the field of regenerative medicine. Isolation of pluripotent stem cells from Acomys would allow for development of transgenic or chimeric animals and in vitro study of regeneration; however, the reproductive biology of Acomys is not well characterized, complicating efforts to derive embryonic stem cells. Thus, we sought to generate Acomys induced pluripotent stem cells (iPSCs) by reprogramming somatic cells back to pluripotency. Methods: To generate Acomys iPSCs, we attempted to adapt established protocols developed in Mus. We utilized a PiggyBac transposon system to genetically modify Acomys fibroblasts to overexpress the Yamanaka reprogramming factors as well as mOrange fluorescent protein under the control of a doxycycline-inducible TetON operon system. Results: Reprogramming factor overexpression caused Acomys fibroblasts to undergo apoptosis or senescence. When SV40 Large T antigen (SV40 LT) was added to the reprogramming cocktail, Acomys cells were able to dedifferentiate into pre-iPSCs. Although use of 2iL culture conditions induced formation of colonies resembling Mus PSCs, these Acomys iPS-like cells lacked pluripotency marker expression and failed to form embryoid bodies. An EOS-GiP system was unsuccessful in selecting for bona fide Acomys iPSCs; however, inclusion of Nanog in the reprogramming cocktail along with 5-azacytidine in the culture medium allowed for generation of Acomys iPSC-like cells with increased expression of several naïve pluripotency markers. Conclusions: There are significant roadblocks to reprogramming Acomys cells, necessitating future studies to determine Acomys-specific reprogramming factor and/or culture condition requirements. The requirement for SV40 LT during Acomys dedifferentiation may suggest that tumor suppressor pathways play an important role in Acomys regeneration and that Acomys may possess unreported cancer resistance.


Introduction
Typically development occurs in a unidirectional, irreversible manner; however, this process can be reversed, and differentiated cells can be returned to an early embryo-like pluripotent state through transcription factor overexpression 1 . These induced pluripotent stem cells (iPSCs) recapitulate all characteristics of embryonic stem cells (ESCs) 2 and are believed to be essentially equivalent 3 . Although naïve pluripotent stem cells have been derived from a few species 4-6 , the signal requirements to support this state are not well defined for most species. Thus, the emergence of iPSCs has provided an alternative way to acquire PSCs, and these cells can then be used to identify critical signaling requirements. In species where embryos are not easily accessible, reprogramming presents a more convenient method of PSC generation 7 .
Since iPSCs retain characteristics of the species from which they are derived, they present useful in vitro models for biological phenomena 8 . For instance, thirteen-lined ground squirrel (Ictidomys) iPSCs exhibit cold adaptation 9 , while naked mole rat (Heterocephalus) iPSCs exhibit cancer resistance 10-12 . Another such trait of interest is regeneration, for which most adult animals, including Mus and humans, demonstrate a limited capacity. Though most models of regeneration are invertebrates or lower vertebrates 13 , the African spiny mouse (Acomys) presents a unique mammalian model of multi-organ regeneration 14 , and Acomys iPSCs may retain regenerative characteristics, enabling in vitro study of regeneration. Since the naïve pre-implantation epiblast is a stage of development exclusive to mammals, Acomys iPSCs would present the first and only PSCs from an organism both developmentally similar to humans and capable of such extensive regeneration.
Importantly, iPSCs could prove valuable in expanding the repertoire of tools to study Acomys. Since genetically modified iPSCs are capable of germline transmission, production of transgenic animals would be possible 15 , allowing for the interrogation of individual gene functions in Acomys regeneration. iPSCs would also allow for the generation of Acomys-Mus interspecies chimeras 16 , facilitating investigation of how cells from each species differentially contribute to wound healing. Furthermore, organoids generated through differentiation of iPSCs could allow for in vitro study of Acomys development and regeneration 17 . Organoid models for hair-bearing skin are especially attractive given Acomys's ability to regenerate skin 18 , but organoids from tissues across the body could enable study of organs whose regenerative capacity is yet to be assessed in Acomys.
In this work, we attempt to adapt Mus reprogramming protocols for use in Acomys; however, due to the unique physiology of this regenerative rodent along with a lack of available research resources developed for use in this non-traditional model, we encounter several biological and technical roadblocks impeding the generation of Acomys iPSCs.
Cells were frozen in N2B27 and DMSO (Applichem) in a 9:1 ratio at -80°C before transfer to liquid nitrogen for long-term storage.
Fibroblast reprogramming Effectene (Qiagen) was used for Acomys fibroblast transfections. One day prior to transfection, Acomys iPSCs were plated at 15,000 cells cm -2 in a 6-well plate in MEF medium. On the day of transfection, 180 μl Buffer EC, 1 μg total of all piggyBac plasmids of interest, 0.2 μg PBase plasmid, and 9.6 μl Enhancer were combined and incubated at room temperature for 4 minutes. Then, 30 μl Effectene reagent was added and incubated at room temperature for 10 minutes. Medium was replaced with 1 ml fresh MEF medium. Following incubation, mixture was combined with 1 ml MEF medium and added dropwise to cells. Medium was replaced the following day with KSR LIF supplemented with 1 μg/ml doxycycline to induce reprogramming.
Round Bottom Well: Cells were diluted to 16,500 cells/ml, and 30 μl was pipetted into each well of a non-adherent 96-well round bottom plate. Empty wells were filled with DPBS to minimize evaporation. After 3 days, cells were transferred to a non-adherent 10 cm dish for suspension culture in MEF medium.
Suspension Culture: 1,500,000 cells were transferred to an uncoated, non-adherent for suspension culture in MEF medium.
Hanging Drop: Cells were diluted to 16,500 cells/ml or 33,000 cells/ml, and 30 μl was pipetted onto the lid of an uncoated, non-adherent 10 cm dish then inverted for hanging drop culture. The 10 cm dish was filled with DPBS to minimize evaporation. After 3 or 5 days, cells were transferred to a non-adherent 10 cm dish for further suspension culture in MEF medium.

RT-qPCR
Fast SYBR Green Master Mix (Life Technologies) along with sample cDNA, and primers targeting both endogenous and exogenous expression of the genes listed in Table 2 were used to perform qPCR in technical triplicate reactions in an Applied Biosystems StepOne Real Time PCR system (Thermo Fisher). Default cycling parameters for SYBR Green regents were used (95°C hold for 20s, 40 cycles of 95°C for 3s, 60°C for 30s with data collection, then a melt curve was generated by a 15s hold at 95°C, 1 minute hold at 60°C, and gradual ramp up to 95°C with data collection).

Primer design
To analyze gene expression, we designed RT-qPCR primers in regions of the transcriptome shared between Acomys and Mus in order to verify proper primer binding and amplification using Mus ESCs as a positive control. We utilized a transcriptome assembled by sequencing early-stage At 1 day post-induction (dpi), several mOrange-fluorescent fibroblasts were present ( Figure 1B). While we initially observed proliferation of these cells, over longer time periods we found that this fluorescent population was lost either through transgene silencing or cell death. Following a group of fluorescent cells from 5 dpi to 8 dpi, it became clear that most cells expressing the reprogramming factors were dying, and specifically we observed widespread death around 6dpi ( Figure 1C). Few fluorescent cells remained after this period of cell death, and the surviving cells were non-proliferative, had not changed morphology, and were deemed senescent. Failure to induce dedifferentiation suggested there are roadblocks to reprogramming in Acomys.
Since c-Myc plays a role in both apoptotic signaling and cellular senescence 24 , we reasoned this oncogene might be inhibiting reprogramming. Although reprogramming is possible without c-Myc in other species, the process is substantially delayed, efficiency is decreased, and germline transmissibility is compromised 25 . Thus, we sought a way to overcome the negative effects of c-Myc while still including it in the reprogramming cocktail. SV40 large tumor antigen (SV40 LT) has previously been used to combat c-Myc-induced cellular toxicity and increase reprogramming efficiency by inhibiting the p53 and Rb tumor suppressor pathways 10,12,26,27 . We investigated whether SV40 LT could similarly abrogate the toxic effects of c-Myc in Acomys.
We added a piggyBac, constitutively expressed SV40 LT construct to the reprogramming cocktail. Compared to cells expressing MKOS alone, we observed more robust proliferation with the addition of SV40 LT. By 3 dpi, colonies containing morphologically distinct cells began to emerge ( Figure 1D). These continued to expand over time, not showing the gradual loss of mOrange signal that we observed in cells expressing MKOS alone. Drawing comparisons to Mus fibroblast reprogramming in which highly proliferative, yet incompletely reprogrammed, intermediates arise soon after overexpression of MKOS, we assumed that these cells were likely to be pre-iPS-like cells. In Mus, these often exhibit an ESC-like morphology and show partial upregulation of select pluripotency markers while downregulating somatic markers 21 .
Acomys pre-iPSCs expressed total Oct4 and Klf4 exceeding Mus ESC levels, verifying reprogramming cassette expression ( Figure 1E). To assess pluripotency, we evaluated 4 wellcharacterized markers of Mus and human naïve pluripotency (Gbx2, Tcfp2l1, Tbx3, and Fgf4) using primers designed to amplify both Mus and Acomys transcripts (see Table 2 for primer details). Compared to fibroblasts, all 4 markers were upregulated to varying degrees in Acomys pre-iPSCs ( Figure 1F). Thus, overexpression of the Yamanaka factors combined with SV40 LT allows us to overcome the apoptosis and senescence caused by MKOS alone, permitting dedifferentiation of Acomys fibroblasts, and leads to slight upregulation

Data analysis
Representative microscope images are shown to illustrate qualitative changes in morphology and fluorescent protein expression. Brightness and contrast have been altered using Fiji 20 for the purpose of clarity. qPCR analyses of gene expression represent single experiments, and therefore have not been statistically analyzed.

SV40 LT facilitates the early stages of Acomys reprogramming
Initially, we attempted to reprogram Acomys fibroblasts to an iPSC identity through overexpression of the conventional Yamanaka factors, cMyc, Klf4, Oct4, and Sox2 (MKOS), combined with a media change to conditions supportive of naïve PSCs, as this is an effective protocol for the reprogramming of Mus fibroblasts 21 ( Figure 1A).
We transfected fibroblasts with a polycistronic cassette containing MKOS separated by self-cleaving 2A sequences along with an mOrange fluorescent protein connected via an IRES element (see Table 1 for plasmids), allowing for the expression of all 4 Yamanaka factors as well as mOrange under the control of a TetO promoter activated in the presence of doxycycline (dox) and rtTA 22 . The reprogramming cassette was flanked by piggyBac arms, allowing for random integration into the genome 23 . A constitutively expressed rtTA plasmid and non-integrating piggyBac transposase were also transfected. of some components of the pluripotency network, as would be expected from pre-iPSCs 21 .
2iL culture condition allows for conversion of Acomys pre-iPSCs to iPS-like colonies Pre-iPSCs represent an intermediate phase of reprogramming, but can be converted to fully pluripotency using small molecules 21 . Thus, we transferred our Acomys pre-iPSCs into replicate wells and applied a variety of culture conditions intended to encourage full reprogramming. One well was maintained in KSR LIF dox as a control (Figure 2A), but we removed dox in all other wells as acquisition of bona fide pluripotency is dependent upon transgene-independent self-renewal 28 .
No colonies emerged from the KSR LIF condition, and untransfected fibroblasts in the well overgrew ( Figure 2B). In FAX, which supports primed ESCs that represent the post-implantation epiblast 29 , no colonies emerged, and fibroblasts again overgrew ( Figure 2C). We also tested two media conditions used to sustain naïve human PSCs: t2iL Gö XYaa 30 and PXGL 31 . Most cells died, and no colonies emerged in either condition ( Figure 2D,E). In addition to being used to culture ground state mouse ESCs 32 , 2iL conditions containing inhibitors of MEK/ERK and GSK3 signaling along with LIF can induce Mus pre-iPSCs to convert to full pluripotency 21,33 . We switched cells into KSR 2iL, KSR 2iL plus 0.5 μm 5-azacytidine, or KSR 2iL with a titrated amount of PD03, which has been shown to support naïve-like human ESCs 34 . The addition of 2iL caused an initial wave of cell death, but after 10 days, dome-shaped colonies emerged in all three KSR 2iL conditions and there was no noticeable difference between the different conditions ( Figure 2F-H). However, these conditions did not appear to be selective against the fibroblasts, potentially due to the presence of KSR in the media or expression of SV40 LT, and these fibroblasts overgrew. We mechanically picked colonies into new wells, but they collapsed soon after picking.
We also used serum-free N2B27 supplemented with 2iL to facilitate the transition to pluripotency. N2B27 2iL was much more selective than KSR 2iL, and almost no fibroblasts survived in these conditions, making it easy to identify the large, tightly-packed, dome-shaped colonies that emerged ( Figure 2I). Given the lack of proliferating fibroblasts in the culture, we attempted to enzymatically passage these N2B27 2iL colonies to a new well; however, the passaged cells did not survive.

Acomys iPS-like cells are transgene-dependent
To alleviate the problems with fibroblast overgrowth observed in the more permissive culture conditions, we mechanically picked a colony of Acomys pre-iPSCs and expanded it in KSR LIF dox as a 'pure' population devoid of fibroblasts. After multiple passages, tightly-packed colonies of small cells spontaneously emerged ( Figure 3A). We mechanically picked these colonies, believing them to represent a more advanced state in the reprogramming process. These cells might potentially represent a delayed, stochastic path to iPS generation that avoids becoming trapped in the pre-iPS stage 21 . However, when transferred to 2iL, these pre-iPSC colonies either differentiated to a primitive endoderm-like morphology or died ( Figure 3B,C).
Clearly, Acomys pre-iPSCs required sustained transgene induction to remain undifferentiated and survive. Thus, we transferred the aforementioned picked pre-iPSC colonies to 2iL dox and observed tightly packed, rounded colonies of Acomys iPS-like cells after 8 days. However, these iPS-like colonies exhibited cell death at their edges, and after 15 days, many colonies collapsed entirely ( Figure 3D). To support the remaining colonies we added Alk-5 inhibitor A83-01 and FGF receptor inhibitor PD173074, which are used to supplement 2iL to prevent differentiation of naked mole rat 10 and rat 35 iPSCs, respectively. This culture condition, termed 4iL dox, appeared to temporarily stabilize the iPS-like colonies ( Figure 3E). These Acomys iPS-like colonies survived passaging; however, they could not be maintained over multiple passages. Nevertheless, these experiments showed Acomys iPS-like cells could be derived from pre-iPSCs with sustained transgene induction.
We returned to a population of Acomys pre-iPSCs a single passage after induction, and after culturing these pre-iPSCs in 2iL dox conditions, tightly-packed, dome-shaped colonies emerged again. Shortly after, these colonies were switched to 4iL dox conditions ( Figure 3F). After enzymatic passaging, we observed small, rounded iPS-like colonies ( Figure 3G); however, several flatter pre-iPSCs remained and continued to proliferate, eventually overgrowing. When we instead mechanically picked primary iPS-like colonies into new wells, we obtained a pure population of iPS-like colonies devoid of flat pre-iPSCs ( Figure 3H). These iPS-like colonies had well-defined edges and were composed of cells with a high nucleus-to-cytoplasm ratio, characteristic of PSC colonies.
Surprisingly, however, these Acomys iPS-like cells did not exhibit upregulated naïve pluripotency marker expression ( Figure 3I). To determine whether these cells were functionally pluripotent despite not expressing expected pluripotency markers, we performed embryoid body (EB) differentiation. Though we attempted three different EB differentiation protocols using varying cell numbers, we were unable to obtain any differentiating EBs. In all attempts, the Acomys iPS-like cells aggregated but failed to proliferate and differentiate. Cell debris was observed in the media, and the aggregates appeared necrotic ( Figure 3J). The lack of pluripotency marker expression along with the failure to form EBs indicated these Acomys iPS-like cells, though morphologically similar to Mus PSCs, were not pluripotent.

EOS-GiP system does not report pluripotent identity in Acomys
Since reliance upon morphological criteria to ascertain pluripotency of Acomys cells proved unsuccessful, we sought a fluorescent reporter to give a visual indication of pluripotency.
We utilized a piggyBac EOS-GiP plasmid containing an EOS expression cassette driving expression of GFP and puromycin (puro) resistance. The EOS cassette is composed of a mouse early transposon (ETn) promoter, which is specific to PSCs, combined with Oct4-and Sox2-binding motifs found in PSC-specific enhancers 36 . Thus, only PSCs should express GFP and survive puro treatment, allowing us to visually monitor and select for fully reprogrammed iPSCs.
We first knocked EOS-GiP into previously generated Acomys iPS-like cells, conjecturing that there might exist a small population of iPSCs hidden among a majority of non-pluripotent cells. Following transfection, puro selection was applied.
After 5 days, GFP-positive cells emerged, and after 10 days, GFP-positive colonies were picked and passaged. After a single passage, we had a pure population of GFP-positive Acomys iPS-like cells ( Figure 4A); however, these cells still did not express any of the pluripotency markers that we checked for ( Figure 4B). We hypothesized that the selective pressure from addition of puro immediately following introduction of the EOS-GiP construct led to selection for a population of cells containing aberrantly activated EOS-GiP, resulting in spurious GFP expression.
After performing transfections with EOS-GiP added to the reprogramming cocktail, we observed spurious GFP expression in transfected fibroblasts in KSR LIF dox ( Figure 4C). Past studies using a similar Oct4-GFP reporter system found reporter activation is not necessarily indicative of pluripotency in serum-containing media; however, transition of reprogramming intermediates to 2iL allowed them to progress to full pluripotency 21 , and we would not expect to see spurious GFP expression in 2iL.We transitioned these EOS-GiP cells first to 2iL dox and then 4iL dox upon emergence of colonies. Indeed, select dome-shaped colonies were GFP-positive ( Figure 4D); however, some flat pre-iPSCs that clearly did not have an ES-like morphology also expressed GFP, suggesting that the EOS-GiP was not accurately reporting pluripotency ( Figure 4E). Nevertheless, we picked 24 dome-shaped colonies into separate wells and expanded them. Only one of the picked colonies remained GFP-positive so we passaged this colony and applied puro selection. The cells that survived selection had a distinctive morphology, growing in loose clumps of floating cells rather than in tightly packed, adherent colonies ( Figure 4F). Unsurprisingly, these EOS-GiP cells did not strongly express any of the pluripotency markers we assessed ( Figure 4G).
EOS-GiP does not provide a reliable readout of the pluripotent state in Acomys cells, though it is unclear why since this system has been used to track acquisition of pluripotency   To overexpress Nanog, we integrated a piggyBac plasmid containing Nanog downstream of a dox-inducible TetO promoter into our reprogramming cocktail. Substitution of SV40 LT with Nanog in the reprogramming cocktail was insufficient to prevent widespread apoptosis and senescence ( Figure 5A). We then attempted to use SV40 LT, MKOS-imO, and Nanog in combination to reprogram Acomys fibroblasts. Nanog overexpression works synergistically with the DNA methyltransferase inhibitor 5-azacytidine (5-aza) to promote the final stages of reprogramming in pre-iPSCs 33 , so at 11 dpi, we added 1 μM 5-aza to the media.
Nanog overexpression had a noticeable effect on the morphology of early reprogramming intermediates. Acomys pre-iPSCs without Nanog formed looser colonies composed of larger cells, with heterogeneous levels of mOrange. In contrast, Acomys pre-iPSCs expressing transgenic Nanog (iNanog) formed colonies with defined edges composed of very small, tightly packed cells ( Figure 5B) with more consistent mOrange expression. At 15 dpi, we switched the Acomys cells to 2iL dox with 1 μM 5-aza. iNanog Acomys pre-iPSCs formed mostly tightly packed, dome-shaped colonies composed of small cells, whereas pre-iPSCs without Nanog formed many looser colonies composed of larger cells ( Figure 5C).
In past experiments, we passaged cells in bulk or picked reprogramming colonies then pooled them together. In Mus and human contexts, properly reprogrammed cells outcompete non-reprogrammed cells which eventually senesce, so it is not necessary to pick and characterize individual colonies 40 .
However, since our non-reprogrammed cells could be immortalized by the SV40 LT and therefore remain in culture indefinitely, we picked 24 iNanog Acomys iPS-like colonies and cultured them as separate clonal lines. After picking, we observed that these lines exhibited a range of morphologies ( Figure 5D).
Only 11 of the iNanog clones survived mechanical passaging, of which 9 were successfully expanded for RT-qPCR analysis. Acomys iNanog iPS-like cell Tbx3 levels were slightly upregulated compared to Acomys fibroblast levels and similar to Acomys pre-iPS levels ( Figure 5E). Remarkably, levels of Gbx2, Tfcp2l1, and Fgf4 were all highly upregulated in Acomys iNanog iPS-like cells compared to both pre-iPSCs and fibroblasts, and these dramatic increases were consistent across all 9 clones assayed ( Figure 5F-H). Mus ESCs, used as a positive control for these RT-qPCR reactions, appeared to show far higher relative expression of Tbx3, Gbx2 and Tfcp2l1; however, it should be noted that direct cross-species comparisons are difficult to interpret as we do not know the absolute level of expression of these factors, or the housekeeping gene being normalized to, and they may differ significantly between Acomys and Mus naïve cells. Nevertheless, inclusion of Nanog in the reprogramming cocktail clearly induces upregulation of several naïve pluripotency markers that were not strongly expressed following reprogramming with the Yamanaka factors alone or in combination with SV40 LT in Acomys. Future work will elucidate whether these iNanog Acomys iPS-like cells are functionally pluripotent through differentiation and chimera assays.

Discussion
Our data show traditional reprogramming protocols developed in Mus cannot be directly applied to Acomys. Nevertheless, this preliminary work provides several avenues for future investigation.
The requirement for SV40 LT during reprogramming suggests a hyperactive tumor suppressor response in Acomys. Immortalization increases reprogramming efficiency in Mus and human 41 and also greatly enhances reprogramming in Heterocephalus 10,12 . Tan et al. found Heterocephalus cells require SV40 LT to undergo reprogramming, mirroring our findings in Acomys 12 . Lee et al. independently found adult Heterocephalus fibroblasts could be reprogrammed without SV40 LT; however, colonies emerged at day 43, which was longer than we cultured our Acomys cells 10 . It is possible Acomys reprogramming requires more time, though this is unlikely given the extensive apoptosis and senescence we observed relatively early compared to these timescales.
Heterocephalus has a stable epigenome that resists de-differentiation, characterized by histones marked more by H3K27me3 repressive marks than H3K4me3 activating marks, and expression of SV40 LT opened previously closed reprogramming factor promoters 12 . Since the epigenetic landscape is reset to facilitate reprogramming 42 , it is possible Acomys possesses a similarly stable epigenome. Past studies showed Acomys skin exhibits resistance to UV radiation-induced DNA damage and age-related senescence, drawing further parallels with Heterocephalus, a model of cancer resistance and longevity 43 . This suggests that more extensive epigenetic remodeling may be required to fully revert Acomys cells to a pluripotent identity. Acomys could possibly be resistant to tumorigenesis, similar to Heterocephalus and the regenerative axolotl salamander 44 , warranting further study into cancer in Acomys.
It is possible the non-pluripotent Acomys iPS-like cells we generated without Nanog represented transformed cells akin to cancer stem cells as many of the same mechanisms control reprogramming and oncogenesis 45  In order to improve the reprogramming of Acomys cells, several considerations should be made. A better understanding of Acomys reproductive biology would give a point of reference to guide reprogramming efforts. Although a transcriptome exists for gene expression during the earliest stages of embryonic development 19 , a thorough understanding of the gene expression network in the pre-implantation naïve epiblast will be important for a more comprehensive characterization of putative Acomys iPSCs. Differences between Mus and Acomys reproductive biology make it difficult to assume development occurs similarly. Significant differences exist in the hormones needed to stimulate superovulation as well as the timing of ovulation in Acomys 52 . Strikingly, Acomys is the only known rodent capable of menstruation 53 , and embryonic genome activation in Acomys is more human-like than that of Mus 19 . Continued study of the Acomys reproduction and development will inform efforts to improve iPSC generation.
The choice of starting cell can also have a significant impact on reprogramming. We used Acomys neonatal fibroblasts due to their ease of acquisition; however, it has been shown that human postnatal fibroblasts exhibit lower reprogramming efficiency compared to embryonic fibroblasts 26 . Unfortunately, acquiring embryonic starting materials is difficult since Acomys embryonic development is not well characterized. Furthermore, somatic stem cells reprogram more efficiently than differentiated cells, potentially because they do not express as many lineage specific genes, which inhibit reprogramming 54 . Thus, it may be of interest to derive tissue-specific stem cells to be used as a starting material in the future.
Given the deceptive appearance of non-pluripotent Acomys iPS-like cells, morphology cannot be used as a reliable indicator of pluripotency. Thus, we sought to use an EOS-GiP reporter to monitor achievement of pluripotency in vitro; however, we found widespread spurious activation in Acomys cells. This exogenous reporter integrates randomly in the genome; however, an endogenous reporter would present a more accurate reflection of gene regulation since it is placed within the appropriate chromatin context 55 . Unfortunately, without an annotated genome, it would be extremely difficult to develop an Acomys endogenous pluripotency reporter line, further highlighting difficulties in working with this non-traditional model organism.
There remains a possibility that additional factors may be required to induce pluripotency in Acomys. For instance, Lin28 increases the kinetics of reprogramming in a cell proliferationdependent manner, similar to the effects of p53 knockdown 56 . Furthermore, iPSCs have been successfully generated using a cocktail combining the Yamanaka factors with Nanog and Lin28 in several species 57 . Thus, future experiments adding Lin28 or other factors to the reprogramming cocktail might enhance reprogramming further.
Generation of transgene-dependent Acomys iPSCs would subsequently allow for the screening of chemical compounds to determine the species-specific culture conditions necessary to maintain Acomys PSCs, independent of exogenous transgene expression. A similar approach was previously used to identify the culture conditions supportive of the human naïve state 58 .
Signaling requirements for pluripotency maintenance vary from species to species 59 , but Rattus, Heterocephalus, and Tokudaia iPSCs can be cultured transgene-free in 2iL conditions with only slight modifications 10,37,60 , suggesting the same may hold true for Acomys.
The work presented here identifies Acomys-specific obstacles to reprogramming and provides the preliminary work necessary to successfully reprogram Acomys cells. The requirement for SV40 LT during initial dedifferentiation of Acomys fibroblasts suggests tumor suppressor mechanisms might tightly control cell identity change during Acomys regeneration. We also showed that overexpression of Nanog induces upregulation of several pluripotency markers in Acomys iPS-like cells. In summary, there are several avenues of exploration that could potentially lead to improved generation of Acomys iPSCs.
If successful, bona fide Acomys iPSCs would allow for the development of transgenic animals, chimeras, and organoid models, all of which would contribute greatly to our understanding of Acomys regeneration. Overall, continued study of this emerging, non-traditional model organism could have broad implications in the fields of wound healing, oncology, and cellular plasticity. The African spiny mouse (Acomys) is an emerging mammalian model for scar-free regeneration, and further study of Acomys could advance the field of regenerative medicine. The work presented by Dr. Silva and their colleagues identifies Acomys-specific obstacles to reprogramming and provides the preliminary work necessary to reprogram Acomys cells successfully. The authors found that SV40 LT is required during initial dedifferentiation of Acomys fibroblasts upon classic Yamanaka MKOS-mediated reprogramming, which brought up an interesting point suggesting tumor suppressor mechanisms might tightly control cell identity change during Acomys regeneration. The authors also showed that overexpression of Nanog induces upregulation of several pluripotency markers in Acomys iPS-like cells. While the final outcome and the status of bona fide Acomys iPSC are far from satisfactory at the moment, the current study does provide several avenues of exploration that could potentially lead to the improved generation of Acomys iPSCs. The multiple trials and painstaking troubleshooting efforts the authors put into this study are highly appreciated by this reviewer. Continued study of this emerging, non-traditional model organism could have broad implications in wound healing, oncology, and cellular plasticity. If successful, bona fide Acomys iPSCs would allow for the development of transgenic animals, chimeras, and organoid models, all of which would contribute significantly to our understanding of Acomys regeneration.

Data availability
One potential weakness of this study is the lack of a "last kick in front of the door to score the final goal" when the authors state, "Future work will elucidate whether these iNanog Acomys iPS-like cells are functionally pluripotent through differentiation and chimera assays." While chimera assays are time-consuming, I am eager to know whether these iNanog Acomys iPS-like cells are functionally pluripotent through EB differentiation, which should be relatively quick and straightforward to perform. However, this is certainly something on the authors' agenda and could be part of their next report. determined.
As the authors stated, Acomys iPSCs would be a really interesting research tool to understand regeneration capacity of Acomys. Particularly the generation of chimeric mice would be exciting. As shown with rat ESCs, mice with specific Acomys organ could also be generated by using tissue specific master transcription factors knockout mouse embryos as a recipient embryos for chimera.
This work has demonstrated that the generation of Acomys iPSCs is not straightforward, but probably possible with further optimization. Up-regulation of some of pluripotency genes is promising, while deeper characterization, and perhaps more optimization, would be needed. If the cell lines presented here are pluripotent, they could be a useful tool to identify optimal culture condition, even if they are exogenous factor dependent, as the authors described.
As a minor point, the definition of "pre-iPSCs" were not clear to me. What is the difference between pre-iPSCs and iPSC-like cells without pluripotentcy? I believe there is no clear definition in the research field, but some definitions within this manuscript could have been useful for the readers.

If applicable, is the statistical analysis and its interpretation appropriate? Yes
Are all the source data underlying the results available to ensure full reproducibility? Yes Are the conclusions drawn adequately supported by the results? Yes