M-CSF-stimulated myeloid cells can convert into epithelial cells to participate in re-epithelialization and hair follicle regeneration during skin wound healing

Skin wound healing is a complex process which requires the interaction of many cell types and mediators in a highly sophisticated temporal sequence. Myeloid cells compose a significant proportion of the inflammatory cells recruited to a wound site and play important roles in clearance of damaged tissue and microorganisms. Myeloid cells have also been suggested to convert into fibro-blast-like cells and endothelial cells in participation of wound healing process. However, whether myeloid cells in wound skin can convert into epithelial cells and contribute to re-epithelialization and skin appendage regeneration is still unclear. In this study, we performed double immunofluorescent staining with antibodies for hematopoietic cells and keratinocytes as well as cell tracing technique to investigate hematopoietic cell conversion. The result show that during the healing process, some of the CD45-positive hematopoietic cells are overlapped with keratin 14, the markers of keratinocytes. Further, CD11b-positive myeloid cells seem the origin of converted epithelial cells. To confirm these results, we culture CD11b-posiitve myeloid cells from mouse splenocytes in a medium containing macrophage colony-stimulating factor (M-CSF) and dermal injection of these cells into the healthy skin when punch biopsy is created in mouse skin. Tracing injected labeled splenocyte-derived myeloid cells in skin, we confirm that myeloid cells able to convert into keratinocytes in repaired skin. Furthermore, our results from in vivo experiments provide new information on contribution of myeloid cells in hair follicle regeneration. In conclusion, this work highlights the myeloid cell contributions in wound repair and hair follicle regeneration in mice through conversion of M-CSF-stimulated CD11b+ myeloid cells into epithelial cells.


Introduction
Dermal wound healing is a dynamic process involving the coordination of epidermal keratinocytes, dermal fibroblasts, endothelial cells and infiltrated hematopoietic cells. After injury, it is thought that fibroblasts residing in the reticular dermis of wound edge contribute to the dermis replacement while basal keratinocytes and epithelial stem cells accomplish the reepithelialization through cell migration and proliferation [1]. It may be true for the healing of a small skin wound. However, a large full-thickness wound is unlikely to heal solely just through migration and proliferation of skin cells resided at the wound edge. Last few years, we have proposed and begun to investigate whether a subset of infiltrated hematopoietic cells have a capacity to be converted into skin cells or stem cells and thereby contributing to skin wound healing. In fact, in previous publication, our result showed that skin injection of M-CSF-cultured adherent hematopoietic cells can accelerate skin wound healing [2]. Other studies demonstrated that in addition to fibroblasts migrated from the wound edges, monocyte-derived fibrocytes from circulation [3] and residing macrophages converted into fibroblast-like cells also contribute to dermal tissue repair [4].
During the healing process in different tissues, it has been noticed that hematopoietic stem cells or hematopoietic cells have the capacity to be transdifferentiated into tissue specific cells such as neuronal cells, endothelium, skeletal muscle, hepatocytes and epithelial cells [5][6][7][8][9]. Although the capacity of hematopoietic stem cell or hematopoietic cell transdifferentiation has been questioned by some studies indicating cell fusion of hematopoietic cells with other cells rather than transdifferentiation in vivo, and mimic appearance of transdifferentiation from grafted hematopoietic cells after transplantation [10][11][12], new evidence from recently study by using cell tracing technique and single cell sequencing strongly support that hematopoietic cells specifically myeloid cells can convert to solid organ specific cells. For example, myeloid cells in wound bet have been demonstrated to convert to fibroblast-like cells [4] while csf1r-expressed erythro-myeloid progenitors can convert to endothelial cells in blood vessels during embryo development and in adult [13].
To further assess whether infiltrated hematopoietic cells in wound bed can convert to epithelial cells, contributing to epithelialization and skin appendage regeneration, here, we used antibodies of hematopoietic cells and keratinocytes for double immunofluorescent staining as well as cell tracing approach to address this question. The result showed that infiltrated myeloid cells could not only convert to keratinocytes for repairing damaged skin but also contribute to hair follicle regeneration in a mouse skin wound healing model.

Mice
Balb/C and C57BL/6 were purchased from The Jackson Laboratory (Bar Harbor, Marine). All experimental procedures were approved by the University of British Columbia Animal Care Committee. The methods were carried out in accordance with the approved guidelines.

Skin wound model
Two 8 mm punch biopsy excisional wounds were created on the dorsal skin of mouse after mouse was anesthetized by low-dose isoflurane inhalation. To prevent contraction of wounds, silicon splints were used to allow wounds to heal through granulation and re-epithelialization.

Immunofluorescent staining
Skin samples were fixed with 10% formalin. About 5 m thick sections were cut and used for immunofluorescent staining. Following deparaffinization and antigen retrieval, skin sections were incubated overnight with primary antibody after blocking non-specific binding with blocking solution containing 5% bovine serum albumin for one hour. For cell staining, cells were fixed with 10% formalin, blocked with blocking solution and incubated with primary antibody overnight without performing antigen retrieval. After washing three times with PBS or PBS-T (0.1% Triton X-100 in PBS) at room temperature, sections or cells were incubated with secondary, fluorescein-conjugated antibodies for another one hour. After washing three times, cells (or sections) were stained with 1.5% Vectashield mounting medium containing DAPI (Vector Laboratories). Images were captured using a Zeiss Axioplan 2 fluorescence microscope and AxioVision image analysis software. The primary and secondary antibodies used in this study were listed in Supplemental Table S1.

Culturing myeloid cells
Splenocytes were isolated from Balb/C or C57BL/6-Tg (UBC-GFP) mice. GFP mice were kindly provided by Mr. Darrel Trendall at Jack Bell research Centre, University of British Columbia. Splenocyte-derived myeloid cells were generated by culture in a medium containing 5 ng/ml M-CSF. Cells in passage two were used for cell injection to the mice.

Labeling cells in vitro by Dil prior to injection
Splenocyte-derived, M-CSF-cultured myeloid cells at passage 1-2 were used for this experiment.
Cells were detached by trypsin-EDTA, spun down and then re-suspended in 1 mL PBS containing 20 g/mL Dil and incubated for 15 minutes at room temperature. Unbound Dil was then saturated with 1 mL of FBS and removed by centrifugation and washing three-times with cell culture medium. Dil-labeled cells in PBS were ready for injection.

Dermal injection of myeloid cells to the healthy skin of wounded mice
When excisional wounds were created by punch biopsy, one million Dil-labeled or GFP expressing myeloid cells in 100 L PBS were dermally injected simultaneously at one spot on the dorsal healthy skin, one cm far from the edge of wounds.

Statistical analysis
All data were presented as the mean ± standard deviation. Statistical analyses were performed with GraphPad inStat software. P values were calculated using two-tailed unpaired student's t test. A p value of <0.05 was considered as statistically significant.

Distributions of hematopoietic cells in healed skin wounds
It is well known that hematopoietic cells such as neutrophils, lymphocytes and monocytes can infiltrate to the wound sites during the healing course after skin injury [14]. We first investigated whether hematopoietic cells are still present in the wounded skin after wounds are completely healed. To do this, we created dermal wounds by an eight mm size punch biopsy at dorsal skin of mice (Total 16 wounds were created from 8 Balb/C mice), used silicon splint to prevent wound contraction, and collected skin samples at day 14 (wounds were closed around day 10-12). As shown in Figure 1A, using an antibody for hematopoietic cell marker, CD45 to perform immunofluorescent staining in wounded skin of mice, a lot of CD45-positive cells were revealed at not only dermis but also epidermis of 3 of 16 complete epithelialized wounds. To compare the different of CD45-positive cell number in normal skin and wounded skin of mice. Five normal skin and five wounded skin were performed an immunofluorescent staining with CD45 and the ratio of CD45-positive cell number and the total cell number (DAPI stain, blue) was calculated.
A markedly higher number of CD45-positive cells in wounded skin was found as compared to that in uninjured normal skin (55.6 ± 12.3 % vs 4.5 ± 1.3 %; wounded skin vs Normal skin, n=5) ( Fig. 1B).

Conversion of hematopoietic cells to keratinocytes in repaired epidermis of skin
As shown in Figure 1A Figure 2A, some of the K14 positive keratinocytes were also positive for CD45, suggesting that some of CD45-positive hematopoietic cells really convert to epithelial cells to participate in epithelialization after injury.

Myeloid cells are the cellular origin of hematopoietic cell-converting to keratinocytes in wounded skin of mice
As myeloid cells have been suggested to be the cellular origin of converted fibrocytes [3], fibroblast-like cells [4] and endothelial cells [13], here, we examined whether these skin keratinocytes converted from infiltrated CD45-positive hematopoietic cells are originated from myeloid cells. By using double immunofluorescent staining with antibodies CD45 and myeloid cell marker CD11b to detect wounded skin of mice, we found the most of CD45-positive cells distributed in repaired epidermis were also positive for CD11b (Fig.3A). We then performed a double immunofluorescent staining with CD11b and K14 antibodies in wounded skin of mice.
The result showed that some of CD11b-posiitve cells located near basal membrane layer of epidermis were also stained K14 (Fig. 3B). Similar result was obtained in cultured skin cells from wounded skin of mice (Fig. 3C). These results suggest that CD11b-positive myeloid cells are the origin of keratinocytes converted from hematopoietic cells.
To provide evidence that M-CSF-cultured myeloid cells have the capacity to convert to keratinocytes in vivo, splenocytes from normal Balb/C mice were isolated, cultured in M-CSF medium and attached cells were labeled with 1,1'-Dioctadecyl-3,3,3',3'-Tetramethylindocarbocyanine Perchlorate (Dil). We also used a similar approach to culture myeloid cells from GFP mice but without Dil labelling. More than 94% cells cultured in this condition were CD11b positive while none of adherent cells were positive for Pro-col (fibroblasts) or K14 (keratinocytes) ( Figure S1). One million of either Dil-labeled or GFPexpressing syngeneic myeloid cells were then intradermally injected to the normal skin one cm far from the edge of wounds which were generated by 8 mm punch biopsy on the dorsal skin of either the Balb/C or C57BL/6 mice, respectively ( Figure 4A). The same strain of mice without cell injection were also wounded as controls. Skin samples from healed wounds were collected after 14 days, skin sections were deparaffinized (for skin of all mice) and further performed immunofluorescent staining with anti-GFP antibody (only for skin of C57/BL mice). As shown in Figure 4B, Dil-labeled myeloid cells (red color under the fluorescent microscope) migrated to the dermis and epidermis of epithelialized wounds but not in mice without cell injection. Similar result was obtained in C57BL/6 mice received GFP-expressing myeloid cells ( Figure 4C).
To further confirm that injected myeloid cells really converted to skin keratinocytes, we performed double immunofluorescent staining with GFP and K14 antibodies in healed wounds of C57/BL mice which have received GFP-expressing splenocyte-derived myeloid cells. As shown in Figure 5A, result revealed that some of K14 positive keratinocytes were also GFPpositive, suggesting these skin cells were from myeloid cell conversion. Furthermore, the similar result was obtained in cultured skin cells from wounded skin of mice which received GFPexpressed, M-CSF-cultured myeloid cells for 14 days ( Figure 5B).

Myeloid cells contribute to hair follicle regeneration in re-epithelialization mouse skin
Previous studies have demonstrated that skin injury could induce hair follicle regeneration in some types of adult mice when the size of skin wound is big enough (more than 1 cm) [16][17][18].
We have also noticed that many CD45-posiitve cells were found in the hair follicles at the edge of wounds and some follicle-like structures were formed in healed skin (Supplement Figure S2).
Since we have demonstrated that myeloid cells able to convert into keratinocytes, here, we were interested to further investigate whether myeloid cells participate in regeneration of hair follicles in adult C57BL/6 mice after skin injury. To address this question, two wounds with 8 mm diameter were created by punch biopsy on dorsal skin of each C57BL/6 mouse (total of 5 mice).
At the same time of wounding, those mice were dermally injected one million GFP-expressing splenocyte-derived myeloid cells cultured by M-CSF medium at one cm far from the edge of wounds. A silicone wound splint was used on the top of each wound to prevent contraction during the healing. After 4 weeks (a time line for hair follicle regeneration, indicated by previous studies [16][17][18]), wounded skin was collected for histological examination. We also collected wounded skin to isolate skin cells and hair follicles by collagenase digestion. As shown in Figure   6A Figure 7B). These data demonstrated that myeloid cells were able to regenerate hair follicles in wounded skin of mice.

DISCUSSION
In this study, by using double immunofluorescent staining with antibodies for hematopoietic cells, myeloid cells and differentiated keratinocytes as well as cell tracing approach, we have demonstrated that myeloid cells able to convert to skin keratinocytes and participate in hair follicle regeneration. Our data further support previous observations that tissue injury could induce hematopoietic cells to transdifferentiate into solid organ-specific cells [4][5][6][7][8][9].
It has long been known that myeloid cells such as neutrophils, monocytes and macrophages can be recruited to the wound sites and play a critical role in wound healing [19][20][21]. Wounds in mice expressing the human diphtheria toxin (DT) receptor under the control of the CD11b promoter and treated with DT to deplete CD11b-positive monocytes/macrophages, result in stalled healing [22]. Similarly, when mice expressing DT receptor under the control of a myeloid cell-specific lysozyme M promoter are wounded and depleted myeloid cells in different phase of healing by DT treatment, the granulation formation and epithelialization are delayed or stop healing [23].
Further, experimental evidence show that application of autologous monocytes/macrophages can significantly improve wound healing through the stimulation of angiogenesis and reepithelialization in diabetic wounds in rats [24]. The mechanisms that myeloid cells improve wound healing are thought to be related to their scavenger ability, cytokines, chemokines, proteases and growth factors, released from these myeloid cells. However, here, our data strongly suggest a new mechanism of these myeloid cells in wound healing, in which myeloid cells may also improve wound healing via cell conversion.
The capacity of myeloid cells converting into endothelial cells, white adipocytes, osteoclasts and fibroblasts in vivo and in vitro has been documented [4,[25][26][27][28]. Here we found that myeloid cells could also convert to keratinocytes, under the condition of skin injury. It is understandable that cells such as endothelial cells, adipocytes and fibroblasts can be converted from myeloid cells as they are from the same germ layer, mesoderm. Surprisingly, here we found that myeloid cells could also convert to keratinocytes. Myeloid cells need to cross the germline barrier to convert into keratinocytes which are from the ectoderm. The possible explanation for this type of conversion is dedifferentiation of myeloid cells into multipotent stem cells before conversion to keratinocytes. Indeed, we and other have previously demonstrated that hematopoietic cells could be dedifferentiated into stage specific embryonic antigen (SSEA)-1 and -3 positive multipotent stem cells when hematopoietic cells are cultured in a medium containing M-CSF [29][30][31]. We also reported that hematopoietic cell-derived SSEA-1 and SSEA-3 positive multipotent stem cells are transiently present at skin wound beds of mice [2]. After injury, recruited myeloid cells in the wound beds can be induced to be dedifferentiation by M-CSF released from injurystimulating proliferating skin cells. We have also showed that topically application of M-CSF or its antibody in skin wounds can increase or delay healing process [2]. Therefore, it is likely that Hair follicles are thought to form only during development [32]. Loss of hair follicles are considered permanent. However, studies have shown that the hair follicles in some types of adult mice can be efficiently regenerated after skin injury [16,17]. The types of cells to regenerate new hair follicles are still unclear. One study suggested that cells from the outside of hair follicle stem cells may be the cellular origin of new regenerated hair follicles [15]. Macrophages have been indicated to promote hair follicle regeneration [18]. In a recent study, using CD11b-DT receptor transgenic mice, Rahman and colleagues reported that wound-induced hair follicle regeneration is dependent on CD11b + F4/80 + cells [33]. Here, we provide further evidence to support that CD11b-positive myeloid cells are the cellular origin of converting keratinocytes and involved in wound-induced hair follicle regeneration. We have noticed that CD45-posiitve cells are dominant in new forming hair follicles. Furthermore, by injection of GFP-labeled splenocytederived myeloid cells to skin injury mice, we clearly showed that GFP-positive cells are dominant in new forming hair follicles. Taken together, our data strongly support that myeloid cells are at least one type of cells contributing to hair follicle regeneration after skin injury.
In summary, this study provides a new mechanism regarding to skin wound healing. We demonstrate that myeloid cells are involved in skin tissue repair and appendage regeneration through their conversion to skin keratinocytes. Our data also provide further proof to support a previous controversial topic, injury-induced hematopoietic stem cell transdifferentiation. The information obtained from this study not only clarifies the fundamental cellular and molecular mechanism of tissue repair and regeneration, may also provide new therapeutic options for treatments for chronic healing wounds via myeloid cell conversion.       (blue) was used as a nuclear counterstain. Scale bars in all images was 50 µm.