H Y P O T H E S I S L E T T E R3. Although loss of Nf1 is required for neurofibroma development, it is not sufficient and tumor microenvironment plays a key role in this process. Poorly regulated wound healing has been described in NF1 haploinsufficient tissues, S8 again highlighting a possible relationship between wounding and tumor initiation.
| HYPOTHESISIn the skin, terminal SCs ensheathe peripheral nerve endings.
S9These SCs are able to dedifferentiate into a progenitor-like state in response to trauma or other insults. 6-8 Furthermore, recent data from our group suggest terminal SCs may behave as bona fide dermal stem cells. 9 Based on all of these premises, we hypothesize that dedifferentiation of SCs could originate human dNFs.
| EVIDENCE SUPPORTING THIS HYPOTHESISWe analysed the differentiation stage of dNFs of 8 patients diagnosed with NF1 (Data S1), looking at the expression levels of genes characteristic of the different dermal stem cell stages by qRT-PCR of isolated tumor tissue (Fig. 1a,b). 9 We found that SC markers SOX10 and S100B were upregulated in 100% and 75% of tumors, respectively (Fig. 1c). Interestingly, NGFR, NCAM1 and SOX2, all of them dermal stem cell markers were present in tumors although in variable numbers (Fig. 1d). In contrast, MCAM and SOX9 which mark more differentiated stages were mostly nonexpressed in tumors (Fig. 1e). CDH19, a marker of Schwann cell precursors and adult dermal stem cells, was upregulated in 100%(8/8) of the cases (Fig. 1f). Levels of neurofibromin (NF1) were unaltered (75%) or slightly upregulated (25%; data not shown).To confirm CDH19 expression at the protein level, we analysed dNF and unaffected skin sections of NF1 patients as well as healthy donor skin by immunofluorescence. We confirmed upregulation of CDH19 in dNFs as compared to controls (Fig. S1). The difference was especially clear in the lower dermis, where most lesional NGFR + , S100β + , and SOX10 + cells are present ( Figs S2-S4). We also observed that S100β + /NGFR + early stage microneurofibromas were present in "unaffected" NF1 patient skin and that, interestingly, CDH19 was also HypotHesis Letter 902 | upregulated in those areas (Fig. S5). Finally, we pursued characterization of CDH19 expression by immunofluorescence in dermal spheres isolated from control foreskin and dNFs. As expected, CDH19 + cells were detected in both normal and dNF-derived spheres (Fig. S6). In these animals, we predict that dedifferentiation of SCs would produce dNFs even in the absence of resident dermal stem cells.
| HOW TO TEST THIS HYPOTHESIS
| RELEVANCE AND PERSPECTIVESOur preliminary results are consistent with dermal regeneration via SC dedifferentiation playing a role in dNF initiation. Further work is needed in order to confirm these preliminary results in a higher number of patients and to elucidate the molecular mechanisms behind the SC dedifferentiation process, to further illuminate our understanding of human dNF development.
ACKNOWLEDGEMENTS
CONFLICT OF INTERESTSThe authors have declared no conflicti...