Co- culture of iNeurons with primary human skin cells provides a reliable model to examine intercellular communication

Objective: The skin is a sensory organ, densely innervated with various types of sen - sory nerve endings, capable of discriminating touch, environmental sensations, pro - prioception, and physical affection. Neurons communication with skin cells confer to the tissue the ability to undergo adaptive modifications during response to en - vironmental changes or wound healing after injury. Thought for a long time to be dedicated to the central nervous system, the glutamatergic neuromodulation is in - creasingly described in peripheral tissues. Glutamate receptors and transporters have been identified in the skin. There is a strong interest in understanding the communica - tion between keratinocytes and neurons, as the close contacts with intra- epidermal nerve fibers is a favorable site for efficient communication. To date, various coculture models have been described. However, these models were based on non- human or immortalized cell line. Even the use of induced pluripotent stem cells (iPSCs) is posing limitations because of epigenetic variations during the reprogramming process. Methods: In this study, we performed small molecule- driven direct conversion of human skin primary fibroblasts

inates tactile sensation, distinguishing between noxious stimuli activating nociceptors and causing tissue damage, and innocuous stimuli such as social contact, pleasant touch, and proprioception. [12][13][14][15] The skin also has a neural circuit for sensitivity, to detect noxious stimuli, inflammatory cytokine releases, and pathogen-associated molecular patterns (PAMPs). 10,16,17 Glutamate is the most abundant excitatory neurotransmitter in the central nervous system. It is released by nerve cells and is responsible for sending signals between nerve cells. The skin also has a communication system that employs glutamate. 18 Glutamate receptors and transporters are expressed by skin cells, and studies describe the involvement of the glutamatergic system in major physiological skin processes such as skin renewal, hair growth in mice, skin pigmentation, and skin aging. [19][20][21][22][23] Finally, the existence of glutamatergic type communication between corneal epithelial cells and sensory neurons serving to initiate a healing process has been demonstrated. 24 Cell coculture provides a simple and powerful tool to examine intercellular communications. The literature review shows that numerous studies have been conducted to investigate the interaction between skin cells and sensory neurons. To date, studies used keratinocyte and neuronal cell lines. In particular, sensory neuronal preparations from non-human material for example, dorsal-rootganglion-derived cells, or neuronal cell lines such as PC12, ND7-23, and F11 cells were reported. [25][26][27] Even with IPSC technology, some limitations remain to overcome, as IPSCs are cells reprogrammed to an earlier epigenetic state, which generate aberrant methylation during reprogramming. 28,29 Recently, the direct cell-type conversion of human fibroblasts into induced neurons (iNeurons), opened the way to a new generation of neuron model. [30][31][32] As an alternative to iPSC reprogramming, iNeurons circumvent rejuvenation and preserve hallmarks of cellular aging, making them a particularly valuable tool for modeling the interactions with other skin cell types, that can be used for research in skin aging, neurodegenerative diseases, and aspects of psychological conditions that impact the skin. 33 In this study we generated iNeurons from human skin primary fibroblasts, and direct conversion was conducted as previously described, with modifications. 32

| Direct conversion of human dermal fibroblasts into neurons (iNeurons)
Fibroblasts isolated from foreskin were converted into functional neurons using a methodology adapted from Yang and collaborators. 32 Cells were seeded on poly-D-lysine coated chamber slide system (Thermo Fisher Scientific, Inc.), and maintained overnight in fibroblasts culture medium DMEM, with high glucose and 10% FBS (Gibco). Chambers were washed with PBS to remove all traces of FBS, and replaced for 3 days, by a neuronal induction medium, containing a cocktail of small molecules that directly convert terminally differentiated fibroblasts into neurons. Every 3 days during the neuronal induction, the neuronal induction culture medium is half replaced by alternating cocktails of small molecules, named "A" and "B". Composition of both cocktails are available in the reagents section. In our experiments, last renewal with small molecules cocktail B, and renewal with maturation medium was done by renewing only 30% of the medium (Neurotrophin-3 final concentration remained at 20 μg/mL). Changing only 30% of the media toward the end allowed us to have more living and healthy cells. We reduced the percentage of dimethyl sulfoxide (DMSO) to 0.7% in medium containing small molecules, which was more suitable for primary skin fibroblasts, and allowed to scale-up the process. These changes were found to be significant for the establishment of the subsequent cocultures with skin cells.

| Immuno-histological fluorescence
After removal of subcutaneous fat, the skin tissue was used to obtain

| Immuno-cyto-fluorescence
Cells were fixed with 4% paraformaldehyde (Sigma), permeabilized with 0.1% Triton X-100 (Thermo Fisher Scientific). About 1% bovine serum albumin was used as a blocking agent to prevent non-specific binding. Cells were incubated with primary antibodies for 1 hour at room temperature, then washed three times with PBS before being incubated with the secondary antibodies for 1 h at room temperature. Finally, Fluoromount-G™ with DAPI (Thermo Fisher Scientific) was used as mounting medium.

| Statistical analysis
All results were expressed as the mean ± SEM. Statistical analyses were carried out by Student's t-test. For all analysis, a p-value ≤0.05 was considered statistically significant (*), p-value ≤0.01 as very significant (**) and p-value ≤0.005 as highly significant (***).

| Ex vivo studies
To study nerve endings connected to the epidermis and particularly free nerve endings connected keratinocytes, we performed the im-  Figure 1D).

| In vitro studies
In our experiments, we observed very few cells refractory to conversion into neurons, whether young fibroblasts or senescent fibroblasts were used. Within a short period of time (15 days), primary fibroblasts were successfully converted into iNeurons ( Figure 2A). As expected, fibroblasts cultured in the medium without the chemical molecules necessary for the direct conversion were not converted and retained S100A4 expression, a marker specific to fibroblasts ( Figure 2A). In our study, we favored the use of foreskin fibroblasts whose donors were between 3 and 4 years old. Skin fibroblasts are easily accessible, and obtaining a pure primary culture is simple as shown by the expression of the marker S100A4 (S100 calciumbinding protein A4, also called FSP1) ( Figure 2B). The absence of expression of the neuronal nuclei antigen NeuN in primary fibroblasts was verified, to be certain that there was no contamination of the fibroblast culture by any NeuN-positive cells ( Figure 2C). Next, the expression of VGLUT1 in the cultured fibroblasts was investigated to determine if a glutamate transport activity in fibroblasts existed, as there was no such information available in the scientific literature.
Results showed no expression of the glutamate transporter VGLUT1 in cultured human dermal fibroblasts ( Figure 2D).
iNeurons obtained by direct conversion were characterized. Our observations showed that iNeurons expressed NeuN, the marker of mature neurons ( Figure 3A). We also observed the expression of PGP9.5, which is a neuron-specific cytoplasmic protein ( Figure 3A).
Besides, the subtype of the iNeurons was glutamatergic, as they expressed VGLUT1 ( Figure 3B). Other well-known pan-neuronal markers have been detected in the iNeurons, such as the structural neuronal marker TUJ1 (or neuron-specific beta-III tubulin), and the synaptic vesicle protein called synaptophysin (SYP) (Figure 4A,B).  Figure 6A). iNeuron-keratinocyte coculture was maintained for up to 6 days without any observation of cell death, whereas the iNeurons maintained in culture alone, began to decline at day 6. Within the coculture, iNeurons established many contacts with the keratinocytes, and we were able to observe that some neurites could be ensheathed by keratinocytes, giving the impression of neurites passing under or inside the keratinocyte cell body ( Figure 6A, mention N1). We also observed neurites passing at the junction of two keratinocytes ( Figure 6A, mention N2). Moreover, certain neurites appeared to follow the contour of the keratinocyte cytoplasm, and to connect to it ( Figure 6A, mention N3). Furthermore, we observed  Figure 6A, mention N4). Primary keratinocytes proliferate by forming islets. Interestingly, at higher magnification, we observed neurites entering an islet, passing between cell junctions to connect to a keratinocyte located in the middle of the islet ( Figure 6B).
iNeurons cocultured with melanocytes also showed the establishment of connections between the two cell types. Both cell types tended to make connections with the other type as we were able to observe melanocytes establishing contacts with neurites, and iNeurons making contact with melanocytes ( Figure 7A). Finally, the observation of the fibroblasts / iNeurons coculture showed connections of neurites with some fibroblasts. The number of dendrites was not as strong as observed in the keratinocytes / iNeurons coculture ( Figure 7B).

| DISCUSS ION
Embryonically, the epidermis and the brain both originate from a single layer: the ectoderm. Hence the epidermis shares many similarities with the central nervous system. The two major organs are interconnected by bidirectional neural exchanges. In this axis that connects the two organs, the brain is the command center, and the skin is a sensory interface. The skin and the brain communicate via a circuit of neurons which produces molecules serving the two or- well documented that stress causes imbalances in skin homeostasis, and that chronic stress may exacerbate skin disorders with an inflammatory component, such as atopic dermatitis, psoriasis, and rosacea. 34 The innervation of the skin is far from being limited to its role in nociception, in which transient receptor potential channels (TRPs) play a central role. In our study, we highlighted the presence of fibers deficient in the TRPV1 nociceptor ( Figure 1A,B).
Some studies have shown a variety of implications for nerve endings in the skin, including proprioception and pleasurable touch sensation. [13][14][15] Other studies have also reported the implication of nerve endings in skin wound healing process. [35][36][37] To find therapeutic solutions in accompanying wound healing, there is a great interest in understanding how to promote re-epithelization, knowing the molecular pathways by which neurotransmitters modulate the immune system. We developed an in vitro coculture system composed of glutamatergic iNeurons, and the three major skin cell types (keratinocytes, fibroblasts, melanocytes) as it will provide a key platform of study for the skin processes modulated by glutamate. Indeed, glutamate signaling in the skin seems to be involved in major processes, for example, skin renewal, skin aging, skin pigmentation, and hair growth in mice. [19][20][21][22][23] Abnormalities in glutamate neurotransmission are part of the biological mechanisms of the stress response. 38 The impact of emotional stress on skin conditions has been described since decades.
Psychological pressure can aggravate inflammatory responses leading to the exacerbation of skin diseases, for example, psoriasis, atopic dermatitis, melisma, and connective tissue disorders. 39 With aging, a correlation seems to exist between diminished innervation and diminished repair capacity. Indeed, nerve endings density was reported to decrease with age, and age-related impairments in wound healing can happen. 40,41 After verifying the maturity of the iNeurons by monitoring the marker of neuronal maturation NeuN ( Figure 3A), the subtype by monitoring the vesicular glutamate transporters 1 (VGLUT1) ( Figure 3B) and the effective expression of pan-neuronal markers in iNeurons ( Figure 4A,B). We also observed that iNeurons expressed three receptors observed in skin neurites: the Piezo2 mechanoreceptor, the TRPV1 nociceptor, and the oxytocin receptor OXTR ( Figure 5A-C). 42,43 This suggests that the developed iNeurons model possesses mechanosensitive properties which is a fundamental physiological capacity of the skin and of skin cells. The presence of TRPV1 and OXTR may also be useful for the study of the nociceptive response, analgesia, and sensitive skin. 43 The modifications we F I G U R E 6 iNeurons coculture with primary keratinocytes. (A) Immunodetection of keratinocyte marker Pan-Cytokeratin (green), iNeurons marker PGP9.5 (red). Nuclei appear in blue. Scale bars = 16 μm; mentions N1 to N4 which appear in white on the photos identify neurites. (B) Immunodetection in of keratinocyte marker Pan-Cytokeratin (green), iNeurons marker PGP9.5 (red). Nuclei appear in blue. Scale bars = 4.9 μm; arrows in white on the photos show the path taken by the neurites in the islet of keratinocytes F I G U R E 7 iNeurons coculture with primary fibroblasts and primary melanocytes. (A) Immunodetection in of melanocyte marker PMEL17 (green), iNeurons marker PGP9.5 (red Nuclei appear in blue. Scale bars = 16 μm. (B) Immunodetection in of fibroblasts marker S100A4 (green), iNeurons marker PGP9.5 (red). Nuclei appear in blue. Scale bars = 16 μm introduced in the original protocol allowed us to control cell death for a longer time in culture, making it possible to obtain healthier iNeurons, and to allow the establishment of coculture with primary skin cells. In our coculture studies, the establishment of cellular connection between iNeurons and skin cells was observed. Keratinocytes stimulated neurite growth, and neurites ensheathment by keratinocytes was observed ( Figure 6A,B). This model is interesting to further investigate keratinocyte-sensory neuron communication, epithelial renewal, keratinocyte stem cells regulation, and other aspects of the epidermis physiology. The co-cultured system may be ideal for studying wound healing in vitro, compared to a keratinocyte monoculture. These co-culture models could be of great interest to study the influence of cutaneous innervation in processes that change with aging such as wound healing and scar formation. This model opens investigative perspectives to study further the implication of the nervous system in the pigmentation processes, as stress can trigger gray hair or skin depigmentation. 44,45 To conclude, non-human material has brought a lot to the advancement of science and the discoveries of new drugs. Non-human cellular material was often used from animal species because they are close to humans in terms of genetics, anatomy, physiology, and immunology. They produce results that will always remain less relevant than those obtained with human material. The difficulty with primary cells is that they undergo a process of senescence after a certain number of cell division, which depends on the cell type.

CO N FLI C T O F I NTE R E S T
All authors declare that they have no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data sharing is not applicable to this article as no new data were created or analyzed in this study.

E TH I C A L S TATEM ENT
For all tissue samples, written consents were obtained before surgical procedures, according to French regulation (agreement DC 2021-4617).