Characterization of Primary Cilia and Intraflagellar Transport 20 in the Epidermis

Steven H. Su

Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy under the Executive Committee of the Graduate School of Arts and Sciences

COLUMBIA UNIVERSITY

2020

Steven H. Su

Abstract

Characterization of Primary Cilia and Intraflagellar Transport 20 in the Epidermis

Steven H. Su

Mammalian skin is a dynamic organ that constantly undergoes self-renewal during homeostasis and regenerates in response to injury. Crucial for the skin’s self-renewal and regenerative capabilities is the epidermis and its stem cell populations. Here we have interrogated the role of primary cilia and Intraflagellar Transport 20 (Ift20) in epidermal development as well as during homeostasis and wound healing in postnatal, adult skin. Using a transgenic mouse model with fluorescent markers for primary cilia and basal bodies, we characterized epidermal primary cilia during embryonic development as well as in postnatal and adult skin and find that both the Interfollicular Epidermis (IFE) and hair follicles (HFs) are highly ciliated throughout development as well as in postnatal and adult skin. Leveraging this transgenic mouse, we also developed a technique for live imaging of epidermal primary cilia in ex vivo mouse embryos and discovered that epidermal primary cilia undergo ectocytosis, a ciliary mechanism previously only observed in vitro. We also generated a mouse model for targeted ablation of Ift20 in the hair follicle stem cells (HF-SCs) of adult mice. We find that loss of Ift20 in HF-SCs inhibits ciliogenesis, as expected, but strikingly it also inhibits hair regrowth. Closer examination of these mice reveals that Ift20 is crucial in maintaining HF-SC identity.

Specifically, ablation of Ift20 in HF-SCs results in loss of SOX9 expression in HF-SCs and results in ectopic expression of the IFE marker KLF5 in HF-SCs. Additionally, ectopic differentiation is observed in HF-SCs following loss of Ift20. Finally, using both in vitro and in vivo models, we also characterize the role of primary cilia and Ift20 in epidermal wound healing.

We find that loss of Ift20 slows collective keratinocyte migration in vitro and also slows HF-SC migration in vivo during wound repair. Interestingly our data suggests that Ift20 regulates keratinocyte migration in a primary cilia-independent manner. Instead, we find that Ift20 mediates focal adhesion (FA) turnover during keratinocyte migration. Specifically, Ift20 together with Rab5, regulates recycling of FA integrins and loss of Ift20 inhibits proper return of integrins to the keratinocyte surface. Overall, we demonstrate that the epidermis is highly ciliated throughout development and in postnatal skin. We show that Ift20 is crucial in maintaining HF-

SC identity and the telogen to anagen transition in HFs. We finally demonstrate that Ift20 regulates keratinocyte migration independent of its function in ciliogenesis and instead regulates recycling of FA integrins through a Rab5 dependent mechanism.

Table of Contents

List of Charts, Graphs, Illustrations ...... iv

Acknowledgments...... vi

Chapter 1: Introduction ...... 1

1.1 Mammalian skin: architecture and cell populations ...... 1

1.2 The hair cycle ...... 5

1.3 Homeostatic self-renewal of epidermal stem cell populations...... 9

1.4 The dynamics of epidermal wound healing ...... 11

1.5 Cell migration and focal adhesions ...... 15

1.6 Epidermal development during embryogenesis ...... 16

1.7 Primary cilia: function and pathophysiology ...... 20

1.8 Ciliogenesis and cilia disassembly ...... 24

1.9 Ift20 has both ciliary and non-ciliary functions ...... 28

Chapter 2: Characterization of primary cilia in embryonic and postnatal epidermis in a transgenic mouse model ...... 30

2.1 Introduction ...... 30

2.2 Characterization of Arl13b-mCherry; Centrin 2-GFP developing epidermis ...... 30

2.3 Characterization of Arl13b-mCherry; Centrin 2-GFP postnatal and adult epidermis ... 35

2.4 Cilia length, morphology, and live ex-vivo imaging ...... 39

2.5 Discussion ...... 44

Chapter 3: Ift20 and hair follicle homeostasis ...... 47

3.1 Introduction ...... 47

3.2 Postnatal loss of Ift20 and primary cilia results in gross hair regrowth defect in mice ...... 48

3.3 Wnt signaling is absent in IFT20 cKO hair follicles ...... 52

3.4 Loss of Ift20 in HF-SCs results in loss of SOX9 expression, stem cell identity and ectopic

expression of differentiation markers ...... 56

3.5 Discussion ...... 62

Chapter 4: Ift20 in Cell Migration and Wound Healing ...... 65

4.1 Introduction and background ...... 65

4.2 Ift20 is required for polarized migration of cultured primary keratinocytes ...... 65

4.3 Ablation of Ift20 does not affect Golgi polarization or overall structure in cultured

keratinocytes ...... 68

4.4 Loss of Ift20 leads to defects in mechano-chemical signal transduction that is independent

of the primary cilium ...... 71

4.5 Ift20 is required for focal adhesion reformation after microtubule-induced FA disassembly.

...... 75

4.6 Integrin surface expression is altered when Ift20 function is ablated ...... 79

4.7 Ift20 maintains its association with the Golgi apparatus during focal adhesion reformation

after MT-induced FA disassembly...... 82

4.8 Ift20-dependent FA reformation does not require Golgi function ...... 86

4.9 Ift20 is required to traffic β1 integrin through Rab5 endosomes ...... 86

4.10 Ift20 is required for the in vivo mobilization of epidermal stem cells in response to

epidermal injury ...... 91

4.11 Ift20 is required for stem cell migration from skin explants taken from wound biopsies 97

4.12 Discussion ...... 97

Conclusion ...... 103

References ...... 109

Appendix A: Methods and Materials ...... 116

Appendix B: Abbreviations ...... 128

iii

List of Charts, Graphs, Illustrations

Figure 1.1 Schematic of sagittal section of skin…………………………………………………..4

Figure 1.2 Illustration of the hair cycle……………………………………………………………8

Figure 1.3 Schematic of sagittal section of skin during wound healing…………………………14

Figure 1.4 Schematic of epidermal development………………………………………………..19

Figure 1.5 Schematic of intraflagellar transport…………………………………………………27

Figure 2.1 Characterization of E14.5 Arl13b-mCherry; Centrin 2-GFP epidermis……………..33

Figure 2.2 Characterization of E15.5 Arl13b-mCherry; Centrin 2-GFP epidermis……………..34

Figure 2.3 Characterization of P1 Arl13b-mCherry; Centrin 2-GFP epidermis…………………36

Figure 2.4 Wholemount of P250 Arl13b-mCherry; Centrin 2-GFP skin………………………..38

Figure 2.5 Live imaging of E14.5 Arl13b-mCherry; Centrin 2-GFP embryo…………………...43

Figure 3.1 Ablation of IFT20 in K15-CrePGR; IFT20 fl/fl; Rosa26-tdTomato fl/fl mice………...49

Figure 3.2 Loss of Ift20 in HF-SCs results in gross hair coat defect…………………………….51

Figure 3.3 IFT20 cKO hair follicles lack canonical Wnt signaling……………………………...54

Figure 3.4 IFT20 cKO hair follicles lose SOX9 expression……………………………………..57

Figure 3.5 IFT20 cKO hair follicles show ectopic expression of differentiation markers and

KLF5……………………………………………………………………………………………..60

Figure 4.1 Ift20 is required for the polarized migration of primary keratinocytes in response to in vitro wound repair………………………………………………………………………………..67

Figure 4.2 Ift20 loss does not alter Golgi polarity during cell migration or overall Golgi structure…………………………………………………………………………………………..70

Figure 4.3 Cultured primary keratinocytes are rarely ciliated during directional cell migration..73

Figure 4.4 Ift20 loss results in defective mechano-chemical signaling downstream of integrin engagement………………………………………………………………………………………74

Figure 4.5 Ift20 is required for focal adhesion reformation after Nocodazole washout and microtubule regrowth…………………………………………………………………………….77

Figure 4.6 Surface levels of recycled β1 integrin, previously present in FAs, are decreased upon

IFT20 knock out………………………………………………………………………………….80

Figure 4.7 Ift20 can be detected at primary cilia in keratinocytes……………………………….83

Figure 4.8 Ift20 dependent focal adhesion reformation is independent of its localization or function at the Golgi……………………………………………………………………………..84

Figure 4.9 Ift20 is required to transit β1 integrin through Rab5(+) endosomes during MT-induced

FA turnover………………………………………………………………………………………89

Figure 4.10 Ift20 is required for the polarized migration and invasion of hair follicle-derived stem cells and their clonal contribution to wound repair………………………………………...93

Figure 4.11 Telogen staged mouse skin immuno-labeled for Ift20……………………………...95

Figure 4.12 Quantification of the number of migrated Rosa-Tomato (+) HF-SCs in basal and suprabasal epidermal layers 7 days post wounding……………………………………………...96

Table 2.1 Length of primary cilia………………………………………………………………..40

Table 2.2 Bulged-tip primary cilia frequency……………………………………………………40

Acknowledgments

I would like to thank my PI Ellen Ezratty and former lab members for their support and guidance in conducting this research. I would also like to thank my thesis committee members

Drs. Gregg Gundersen, David Owens, and Ron Liem for their support in completing this PhD thesis. Finally, I would like to thank my friends and family members who provided encouragement throughout this endeavor.

Chapter 1: Introduction

Skin is an evolutionarily-conserved organ found on all mammals [1, 2]. Among the many critical functions of skin is its ability to serve as a tactile sensory organ as well as its ability to serve as a protective barrier for underlying tissues and organs [3-5]. Not only does mammalian skin prevent the passage of external microbes, toxins, and other environmental hazards from entering and damaging underlying tissues and organ systems it also serves as a physical barrier and shields underlying tissues from physical trauma [1, 2]. Skin also aids in thermoregulation of the body and serves as an external outpost of the immune system with its own distinct set of immune cells [1, 2].

1.1 Mammalian skin: architecture and cell populations

Mammalian skin is a multi-layered tissue a with a distinct cellular composition for each layer [1] and its architecture can be best appreciated through examination of sagittal sections

(Figure 1.1). Forming the exterior, outermost layer of the skin is the epidermis, an epithelium which predominantly consists of ectoderm-derived keratinocytes [1, 2, 6]. Underlying the epidermis is the mesoderm-derived dermis, consisting mostly of dermal fibroblasts [1, 6]. Closer examination of the skin reveals that interspersed within the epidermis are hair follicles (HFs) and associated sebaceous glands (Figure 1.1). Other appendages of the skin such as eccrine sweat glands can also be identified (Figure 1.1). Interestingly, the distribution of HFs and other structures (sweat glands, sebaceous glands, etc.) differ based on body site as does the thickness of the epidermis [1].

In addition to epidermal keratinocytes and dermal fibroblasts, the skin has an array of resident immune cells which enable skin-specific immunity mechanisms. Most notable are 1

Langerhans cells which are professional antigen-presenting cells that reside in the epidermis of skin [7]. In addition to Langerhans cells, the epidermis also has two specific subsets of T cells,

γδ T cells and CD8+ resident memory T cells (TRM) [7]. γδ T cells are a subset of T cells that are found in mouse epidermis but not human epidermis where as CD8+ TRM cells are non-circulating memory T cells that appear in both mouse and human epidermis after the resolution of skin inflammation [7]. The dermis also possesses a variety of innate immune cells including dermal dendritic cells, macrophages, mast cells and innate lymphoid cells [7]. During homeostasis a small number of neutrophils and monocytes survey the dermis for pathogens. In response to detection of pathogens or an inflammatory stimuli, these cells can mount a rapid immune response within the dermis which results in the rapid accumulation immune cells [7].

The epidermis can be broadly divided into 2 distinct partitions: the HF structure and the interfollicular epidermis (IFE) (Figure 1.1). Both of these elements undergo constant renewal under homeostatic conditions, with each element employing a distinct self-renewal program utilizing its unique stem cell population [1-4, 8] . The IFE is a stratified epithelium consisting of

4 layers situated on top of a basal lamina, rich in Type IV collagen and laminin, that demarcates the epidermal-dermal junction (Figure 1.1). The 4 layers of the IFE are the basal layer, spinous layer, granular layer, and corneal layer (Figure 1.1) [9]. Under steady state conditions, terminally differentiated keratinocytes found in the outer corneal layer are continually shed [1, 2, 9]. In order to replenish the cell population of the IFE, the Basal Stem Cells (BSCs) found in the basal layer continually regenerate through cell division and undergo differentiation to form the spinous and granular layers of the IFE [1, 9]. The keratinocytes found in the middle two layers are partially differentiated and will eventually undergo terminal differentiation to form the corneal layer [1, 9].

Under homeostatic conditions, the HF is also constantly regenerating. HFs continually cycle through stages of growth (anagen), regression (catagen), and quiescence (telogen) in what is termed the hair cycle [10, 11]. Long-term stem cells of the hair follicle (HF-SCs) reside in an area of the HF called the bulge, and in contrast with BSCs, spend much of their time in quiescence [1, 10, 11]. HF growth begins during early anagen when HF-SCs enter a short proliferative phase to generate shorter lived progeny cells in the hair germ which in turn fuel HF growth and generation of the new hair [10, 11]. The HF-SCs then return to quiescence by mid anagen and remain in this state until the start of the next hair cycle [11]. (See the following section for more details.) In mice, the cycling of HFs is well synchronized and the stage of the

HFs can be accurately estimated by the age of the animal [10]. Importantly, under homeostatic conditions, HF-SCs and BSCs remain segregated in their distinct niches: BSCs remain in the basal layer of the IFE and continually regenerate the IFE and HF-SCs remain in the bulge and regenerate HFs [2, 12-14].

Figure 1.1: Schematic of sagittal section of skin illustrating the layers of the skin, layers of the epidermis and other commonly seen structures like hair follicles and sweat glands.

1.2 The hair cycle

As previously described, HFs in adult skin continually cycle between telogen

(quiescence), anagen (growth), and catagen (regression) during homeostasis. During telogen,

HF-SCs are kept inactive by positive Bone Morphogenetic Protein (BMP) signals coming from surrounding dermal fibroblasts and adipocytes as well as JAK-STAT5 signaling within the HF-

SCs [10, 15]. Recent studies have demonstrated that during telogen, active JAK-STAT5 signaling is maintained by Oncostatin M (OSM) which is actively secreted by a subset of

TREM2+ dermal macrophages that appear in increased numbers during Telogen [10].

Interestingly, genetic deletion of the OSM receptor or STAT5 in HF-SCs can induce premature

HF-SC activation and premature transition to anagen [10].

Onset of anagen occurs when BMP inhibitors and activating FGFs from the dermal papilla (DP) overcome the positive BMP signals [4]. At the same time the number of TREM2+ macrophages surrounding the HF has decreased therefore reducing total OSM secretion and telogen-promoting effects [10]. These changes in BMP and OSM signals trigger cells in the hair germ (derived from HF-SCs) to become Wnt-high progenitor cells. These Wnt-high cells themselves proliferate and generate a population of Transit Amplifying Cells (TACs) which are responsible for the downgrowth of the HF and subsequent formation of differentiated layers of the HF [16]. At the same time the Wnt-high cells express Sonic Hedgehog (Shh) which triggers

HF-SCs in the bulge to proliferate and replenish hair germ cells which are depleted as Wnt-high cells in the hair germ differentiate into TACs [16]. At the same time, in a positive feedback mechanism, Shh from the Wnt-high progenitors also drives the DP to continue producing BMP inhibitors and FGFs which enables HF growth to continue (Figure 1.2). The TACs are limited in the number of cell divisions they undergo before they differentiate to form various components

5 of the hair. Eventually the supply of TACs is exhausted and the and the HF enters catagen, a destructive period in which the cells in the lower “cycling” portion of the HF undergo apoptosis and the HF regresses [17]. Molecules that promote the anagen to catagen transition include growth factors like FGF5 and EGF [17]. Factors that are known to maintain anagen include

SGK3 and Msx2 [17-19]. Notably, the pattern of BMP, Shh, and Wnt signaling during the telogen to anagen transition mimics that of embryonic HF morphogenesis.

Wnt signaling is of particular importance in HF-SC maintenance and the telogen to anagen transition. A variety of Wnt receptors including Frizzled 1, Frizzled 6, Frizzled 7 and

Frizzled 10 are expressed by HF-SCs and cells of the HF-SC-derived hair germ both during telogen and anagen [20]. Axin2, a downstream target in Wnt signaling, has been identified as a marker of quiescent HF-SCs [21]. Recent studies suggest autocrine and paracrine Wnt signals are required to maintain the HF-SC niche during telogen and also necessary to trigger the telogen to anagen transition [21]. Specifically, during telogen both Wnt ligands and Wnt inhibitors such as

Dkk, are secreted by HF-SCs and act in an autocrine and paracrine manner. This careful balance of positive Wnt and negative Wnt signals along with positive BMP signals (signals that inhibit

Wnt) from dermal fibroblasts allows for maintenance of the HF-SCs during telogen [21]. At the telogen to anagen transition, there is a significant increase in Wnt ligand secretion both from HF-

SCs as well as cells of the hair germ which is necessary to trigger the telogen to anagen transition

[21]. Consistent with this model, loss of Wnt signaling in HF-SCs has been shown to inhibit hair growth in adult mice and result in degeneration of HFs [21]. Preliminary work also suggests that loss of autocrine and paracrine Wnt signaling in quiescent HF-SCs results in HF-SC differentiation [21].

Recent studies suggest that adipocytes are crucial in HF cycling. As previously mentioned, during late catagen and early telogen, mature dermal adipocytes release BMP to inhibit activation of HF-SCs [15]. As the HF progresses through telogen, the amount of mature adipocyte wanes and there is expansion/proliferation of adipocyte precursor cells [15, 22]. These adipocyte precursor cells mature to an intermediate adipocyte lineage that expresses PDGF which acts on the DP to promote HF-SC activation [15, 22]. Loss of adipocyte precursor cells or inhibition of maturation of adipocyte precursor cells using PPARγ inhibitors can inhibit the activation of HF-SCs and hair growth [22].

Figure 1.2: Illustration of the hair cycle. During telogen, HF-SCs are inhibited by BMP signaling from surrounding fibroblasts. In early anagen, BMP inhibitors and FGF signals from the Dermal Papilla (DP) act on Wnt-high keratinocytes within the Hair Germ (HG). In response to these signals, the Wnt-high keratinocytes produce Shh which subsequently drives HF-SC proliferation and growth of the HF. Shh signaling also acts on the DP to increase BMP inhibitors and FGF signaling. The HF enters catagen when the proliferative capacity of the hair germ is exhausted.

1.3 Homeostatic self-renewal of epidermal stem cell populations

The maintenance of these both the HF-SC and BSC populations is dependent on a combination of extrinsic and intrinsic factors which defines the unique niches of each of these stem cell populations [2]. Extrinsic factors that govern each stem cell niche include the composition of neighboring cells, the extracellular matrix (ECM), growth factors, as well as physical parameters, all of which are location dependent [2]. Signal transduction pathways also participate in the regulation of these niches and the activation of these stem cell populations [2].

In particular, the niche of the HF-SCs (i.e. the bulge) and the HF cycle are regulated through the interplay of a variety of signaling pathways, including Wnt, Shh, BMP, and Transforming

Growth Factor Beta (TGFβ) signaling, that work in concert in both autocrine and paracrine manners [2, 17, 21, 23-25]. In contrast, maintenance of the BSC niche (the basal layer of the

IFE) and the differentiation of BSCs are thought to be mostly dependent on Wnt and Notch signaling, respectively [25-28].

Crucial intrinsic factors which regulate maintenance and activation of these 2 stem cell niches include stem cell-specific cell metabolism, gene expression, and chromatin arrangement

[2]. For example, lactate metabolism has been shown to regulate activation of HF-SCs [29].

During physiological hair growth, HF-SCs increase generation of lactate following activation

[29]. Inhibition of lactate generation via deletion of lactate dehydrogenase in HF-SCs inhibits their activation, while conversely, increasing lactate levels via deletion of mitochondrial pyruvate carrier in HF-SCs accelerates activation of HF-SCs and the subsequent hair cycle [29].

With regards to gene expression, HF-SC quiescence is maintained by the transcription factor

FOXC1 through activation of NFATC1 and positive BMP signaling [11, 30]. Loss of FOXC1 results in excessive HF-SC activation and shortened periods between hair growth cycles [11, 30].

In contrast, the transcription factor SOX9 maintains HF-SC identity and promotes HF-SC activation and hair growth [31-33]. Importantly, ablation of SOX9 in HF-SCs inhibits hair regrowth in adult mice and results in differentiation of HF-SCs such that they express differentiation markers typically only observed in differentiated suprabasal cells in the IFE [31,

32].

Interactions between neighboring cells are also crucial in activation and maintenance of stem cell populations during epidermal homeostasis. In the IFE, the transition of BSCs to suprabasal spinous cells during differentiation requires Notch signaling [27, 28]. Notch ligands reside in the basal layer BSCs while Notch receptors are expressed in suprabasal spinous cells [3,

27]. Loss of Notch signaling inhibits specification of spinous cell fate whereas excessive Notch signaling converts BSCs into spinous cells [27, 28]. Interestingly, canonical Notch signaling is also required for HF differentiation during development [27]. Physical interaction between the epidermis and dermis are also crucial in regeneration of epidermal stem cell niches following their ablation. In HFs, lineage tracing experiments show that both neighboring committed progenitor cells and non-HF derived epithelial cells can re-populate the HF-SC population and reform the bulge following ablation of the bulge niche such that the HF can continue regeneration of hair [34]. This regeneration process is contingent on the interaction between the remaining epithelium (i.e. hair germ) and the DP of the HF [34]. This interaction is likely necessary as the DP generates BMP inhibitors which allow HFs to enter anagen (hair growth)

[24].

ECM composition also regulates epidermal stem cell niches and homeostatic self- renewal. Loss of single ECM components can dramatically alter stem cell activity in the epidermis. For example, collagen 17a1 is a transmembrane protein located at the basal lamina

10 that lines the basal layers of the IFE and the HF structure. DNA-damage induced proteolysis of collagen 17a1 that occurs with age in the HF bulge leads to a reduction in HF-SCs through their terminal differentiation [35]. Physical aging also leads to loss of collagen 17a1 in the basal IFE which promotes transient hypertrophy of the IFE [36]. Interestingly, preserving or replenishing collagen 17a1 in the skin inhibits these aging phenotypes [35, 36].

Physical parameters such as tension and pressure also contribute to maintenance and activation of epidermal stem cell niches during homeostasis. In BSCs, a mechanosensory complex of Emerin, Nonmyosin-IIA, and Actin controls gene silencing, chromatin compaction, and subsequent lineage specification [37]. Briefly, application of forces on BSCs results in enrichment of Emerin at the outer nuclear membrane and subsequently leads to defective anchoring of heterochromatin to the nuclear lamina and a switch from H3K9me2,3 to

H3K27me3 at constitutive heterochromatin. Emerin enrichment is also followed by subsequent recruitment of Nonmyosin-IIA to the outer nuclear membrane which promotes local actin polymerization that reduces nuclear G-actin levels. This reduction in nuclear G-actin as results in reduction of transcription and subsequent accumulation of H3K27me3 at inactive/facultative heterochromatin and an overall downregulation of global gene transcription [37].

1.4 The dynamics of epidermal wound healing

In contrast with self-renewal during homeostasis where BSCs and HF-SCs remain segregated in their distinct niches, during wound healing, these 2 stem cell populations mobilize outward from their niches and intermix with dermal and other skin cells during the tissue regeneration process. Wound healing in the skin can be divided into three distinct phases. The inflammatory phase occurs immediately following wounding during which a blood clot forms and immune cells infiltrate the wound site [2]. Following the inflammatory phase is a

11 proliferative phase during which cells of the epidermis and dermis begin proliferating and migrating into the wound bed to close the wound area [2]. The last phase is a resolution phase during which dermal cells deposit and restructure the ECM in the wound bed [2].

Recent studies have mapped the spatiotemporal dynamics of the epidermal component of wound healing using in vivo live cell imaging, lineage tracing and transcriptome analysis [38,

39]. These studies identified two concentric zones with distinct cellular activity and differentiation surrounding the wound. The concentric zone immediately surrounding the wound is characterized by rapid migration and differentiation of epidermal cells while the second concentric zone further from the wound is characterized as a zone of high epidermal proliferation and little migration along the basement membrane (Figure 1.3) [38, 39].

In the first zone (migration and differentiation), both basal and suprabasal cells of the epidermis migrate towards the wound. The keratinocytes in this zone are supplied from the second outer (proliferative) zone. The migrating and differentiating cells in this first zone show upregulation of genes associated with ECM remodeling and cell adhesion including integrin

α5β1, a fibronectin receptor which allows keratinocytes to migrate on the ECM deposited by fibroblasts [2]. Importantly, the rates of migration and differentiation in this first zone are tightly coupled so that tissue thickening at the leading edge is coordinated over time. The second outer

(proliferative) zone of keratinocytes not only supplies new keratinocytes through proliferation but also regulates the involvement of the surrounding unwounded epidermis during the re- epithelialization process. The cells in this second outer (proliferative) zone originate from a mixture of committed progenitor cells, BSCs, and HF-SCs from the surrounding unwounded epidermis that have migrated out of their homeostatic niches (Figure 1.3) [2].

Clonal lineage tracing suggests that most of the committed progenitor cells recruited for epidermal repair are highly proliferative and differentiate early in the wound healing process whereas the stem cell populations (BSCs and HF-SCs) become activated and increase the pool of stem cells in the outer (proliferative) zone later [2]. The clonal dynamics of the HF-SCs and

BSCs between proliferation and differentiation within each of the wound healing zones are similar. This suggests that proliferation and differentiation rates in each healing zone may be independent of where the stem cell originated [2, 39]. Overall, these studies suggest a model where the non-proliferating leading edge (zone of migration and differentiation) serve as a scaffold, preparing and enhancing the wound bed for efficient repopulation towards the wound center.

HFs and HF-SCs are thought to participate in and promote the acute phase of wound healing. Studies demonstrate that bulge-derived HF-SCs migrate out of the HF early during wound healing to participate in the immediate re-epithelization process (Figure 1.3) [12, 13]. The proportion of bulge-derived epidermal cells at the wound decreases and the proportion of non- bulge-derived epidermal cells at the wound increases during the duration of wound healing so that eventually the wound consists of mostly non-bulge derived epidermal cells [12, 13].

Interestingly mice lacking HFs are able to undergo a complete but delayed wound healing process, further evidence that HF-SCs may be required for the initial rapid re-epithelialization during wound healing [40].

Figure 1.3: Schematic of sagittal section of skin illustrating zones of wound healing and contribution from different stem cell populations. BSCs indicates basal stem cells and HF-SC indicates hair follicle stem cell (Adapted from Gonzales and Fuchs, 2017)

1.5 Cell migration and focal adhesions

Crucial in the process of wound healing is the migration of HF-SCs and BSCs out of their respective niches towards the wound bed. In addition to wound healing, cell migration is vital in numerous other physiological processes such as embryonic development and cancer metastasis.

A variety of extracellular and intracellular factors govern the ability of a cell to propel itself in a directed manner. Extracellular factors include the ECM surrounding migrating cells as well as various chemo attractants and mechanical signals derived from the extracellular environment

[41]. Intracellular components that control cell migration include dynamic actin and microtubule networks as well as protein trafficking and recycling pathways necessary to regulate the polarity and plasma membrane composition of a cell [41]. Linking the ECM to the intracellular cytoskeletal network are dynamic, integrin-based protein complexes referred to as focal adhesions (FAs). FA dynamics are crucial for directional cell migration in most cell types, as they are formed at sites of cell adhesion to the ECM and mediate mechano-transduction and integrin signaling [42].

The formation and maturation of FAs is a complex process. Briefly, following engagement with ECM ligands, a conformational shift in the intracellular tail of heterodimeric integrins occurs which promotes their linkage to the actin cytoskeleton through a multi-protein complex of linker proteins [43, 44]. These linker proteins include Talin, Vinculin, and α-Actinin

[44]. At the same time, engagement with ECM ligands promotes clustering of heterodimeric integrins and recruitment and activation (through phosphorylation) of scaffolding and signaling proteins to the FA such as Paxillin and Focal Adhesion Kinase (FAK), respectively [44].

Maturation of FAs is regulated by both tension force and local actin polymerization [43]. Rho

GTPase through downstream effectors such as Rho kinase (ROCK) and mDia stimulates both contractility and actin polymerization, therefore promoting focal adhesion maturation [43].

Polarized cell migration requires the dynamic assembly and disassembly of FAs. FA disassembly can be triggered by microtubule targeting of FAs [45] which induces clathrin- dependent endocytosis of FA-associated integrin [46, 47]. Endocytosed integrin derived from

FAs then transit through Rab5 (+) and Rab11 (+) endosomal compartments remaining in an active but un-liganded state. Rab5, Rab11, Src kinase, FAK, and PIPKIγ are required to recycle endocytosed integrin back to the plasma membrane, where it is then reassembled into FAs polarized toward the leading edge of migrating cells [48]. Interestingly, the endocytosed integrin is trafficked together with FAK, which enables it to remain in an active conformation, and Src kinases, which enables successful recycling back to the plasma membrane [48].

1.6 Epidermal development during embryogenesis

In addition to epidermal homeostasis and wound healing, stem cells of the epidermis are also crucial for epidermal development during embryogenesis. Interestingly, epidermal development utilizes similar signaling pathways and mechanisms as postnatal epidermal homeostasis. In mice, shortly following gastrulation, embryonic skin exists as a single layer of unspecified epithelial cells that soon begin to express Keratin 5 and Keratin 14 (K5 and K14), specific markers of BSCs in postnatal skin. As the epithelium develops, a scattering of cells within the epidermal plane begin producing a higher Wnt signal than their neighboring cells.

These Wnt-high cells cluster and assemble into placodes at embryonic day 13.5 (E13.5) through embryonic day 14.5 (E14.5) [49, 50]. If the mesenchyme underlying these placodes produces elevated levels of BMP inhibitors, the Wnt-high cells at the placode will eventually form a HF

[50]. If the mesenchyme underlying the placode produces a high BMP signal, the Wnt-high cells

16 eventually develop into a sweat gland [51]. The epidermal cells in the plane that do not receive a potent morphogen gradient will stratify to form what will eventually be the IFE [4].

In mice, HF morphogenesis begins around E14.5 after placodes have formed and continues until birth (Figure 1.4). The first cell divisions in the Wnt-activated hair placodes occur exclusively perpendicular to the epidermal plane and generate asymmetrically fated daughter cells such that following a perpendicular division, the basal daughter cell has high Wnt signaling

(Wnt-high) whereas the suprabasal daughter cell has low Wnt signaling (Wnt-low) [52]. Soon after, the basal Wnt-high daughter cells produces Shh which drives proliferation of the suprabasal Wnt-low daughter cells and hence expands this newly fated stem cell population

(Figure 1.4) [52]. These proliferative Wnt-low daughter cells develop into the outer root sheath

(ORS) as well as the bulge and HF-SCs [52]. The Shh produced by the Wnt-high basal daughter cells also signals the underlying mesenchyme to form the dermal papilla (DP) (Figure 1.4) [4].

In response to the continued Shh signaling from the Wnt-high daughter cells, the DP produces

BMP inhibitors, such as Noggin, which drive proliferation of the overlying Wnt-high basal cells.

These Wnt-high basal cells develop into the hair germ (HG) which gives rise to progenitors that form the inner root sheath (IRS) and subsequently the hair shaft [4]. Many of these steps and signaling pathways are recapitulated in the adult hair cycle when the HF cycles between hair growth, regression, and quiescence.

Similar to self-renewal of postnatal IFE during homeostasis, embryonic development of the IFE is also regulated by Notch signaling [53, 54]. As previously described, the IFE arises from regions of the single layer primordial epithelium which do not receive a potent morphogen signal [4]. At E15.5 these cells begin dividing perpendicularly with respect to the epidermal plane in what is the start of the stratification process (Figure 1.4) [53-56]. Each of these

17 perpendicular cell divisions generate differentially fated daughter cells through differential Notch partitioning. Specifically, following a perpendicular division, Notch signaling is elevated in the suprabasal daughter cell which begin forming a spinous layer. Inhibition of this asymmetric cell division inhibits proper Notch activation and stratification of the IFE during development [53,

54].

Interestingly, both HF morphogenesis and stratification of the IFE during embryogenesis are dependent on primary cilia of keratinocytes in the developing epidermis [53]. Loss of primary cilia in the developing epidermis results in loss of Notch signaling and defects in differentiation and stratification in the developing IFE as well as in most instances, hyperproliferation of basal cells of the IFE [53]. In addition to its effects on the developing IFE, loss of primary cilia in the developing epidermis also depletes the Shh signaling necessary for

HF morphogenesis and results in stunted HF development [53]. Remarkably, both Notch receptors and Notch-processing enzymes are found within the primary cilia of keratinocytes, and in other cell types, components of Shh signaling have been visualized within their primary cilia

[53, 57]. Despite the importance of primary cilia during skin development, more characterization of this cellular structure is necessary to understand its role in skin development as well as uncovering its role in postnatal skin homeostasis and wound healing.

Figure 1.4: Schematic of epidermal development during embryogenesis. Interfollicular epidermis and hair follicle development occurs simultaneously. (Adapted from Gonzales and Fuchs, 2017)

1.7 Primary cilia: function and pathophysiology

The primary cilium is a singular, non-motile, antenna-like structure that projects from the plasma membrane of most vertebrate cells including keratinocytes [57]. First discovered by A.

Ecker in the epithelium of the semicircular canals of sea lampreys in 1844, primary cilia were then rediscovered by Paul Langerhans in the epithelia of lancelets in 1876 and then by Karl

Wilhelm Zimmermann in human renal tubules in 1898 [58]. For almost one century primary cilia were thought of as rudimentary, vestigial structures serving no cellular function and as such were largely neglected in cell biology [58]. Only in the past 20 years has research into the function of the primary cilium intensified. This work has revealed that primary cilia are not rudimentary and instead serve as a signaling nexus through which a cell can translate extracellular chemical, biological and mechanical cues into intracellular functions.

In addition to the previously mentioned Notch and Shh signaling, primary cilia have also been implicated in governing Wnt and PDGF signaling in other cell types [53, 57, 59, 60].

Remarkably, receptors, downstream effectors, and or regulators of all of these signaling pathways have been found to co-localize with primary cilia in various cell types. For example,

Notch 3 receptor and PDGF receptor alpha (PDGFRα) have been found in the primary cilia of suprabasal keratinocytes and fibroblasts, respectively, [53, 59-61] whereas both Inversin/NPHP2, an inhibitor of canonical Wnt signaling, and Smoothened (Smo), a downstream effector of Shh, have been found in primary cilia of renal epithelia [62-64]. Together these signaling pathways are vital in a variety of cellular and tissue functions during development as well as during postnatal homeostasis [53, 57, 59, 60].

As previously described, loss of primary cilia inhibits Notch signaling and Notch- dependent differentiation of the IFE developing epidermis [53]. Regulation of Notch signaling by primary cilia has observed in both keratinocytes and retinal pigment epithelial (RPE) cells. In

RPE cells, Notch 1 receptor has been found to localize within the primary cilium and this localization has been found to be dependent on basal body proteins and endosomal trafficking

[65]. Loss of basal body proteins, including BBS1 and BBS4, inhibits trafficking of Notch 1 receptor into primary cilia and promotes accumulation of Notch 1 receptors in late endosomes where active Notch Intracellular Domain (NICD) can be produced [65]. This ultimately leads to an increase in Notch signaling [65]. In the developing epidermis, Notch 3 receptor as well as

Presenilin 2, the catalytic subunit of the γ-secretase that cleaves Notch receptor into its downstream effector, NICD, are found in primary cilia of suprabasal keratinocytes [53].

Remarkably, electron micrographs reveal that primary cilia of keratinocytes can invaginate into neighboring keratinocytes, suggesting possible physical engagement of Notch receptors with

Notch ligands in neighboring cells through the primary cilium [66]. Interestingly primary cilia have also been shown to regulate hematopoietic stem and progenitor cell specification through

Notch signaling in Zebrafish [67]. In particular, primary cilia in Zebrafish endothelial cells are thought to transduce Notch signals to hemogenic endothelium for proper hematopoietic stem and progenitor cell specification [67]. In this instance however, no particular Notch receptor or effectors have identified within the primary cilia of the endothelial cells [67].

Shh signaling is crucial during embryonic development and loss of Shh through ablation of primary cilia results in developmental defects such as polydactyl and skeletal dysplasia [68].

Briefly, in the absence of Shh ligand, the Shh receptor Patched (PTCH1) inhibits Smo. When

Smo is inhibited, GLI transcription factors are proteolytically processed into their repressor state

21 which inhibits downstream transcription of Shh target genes. Binding of Shh to PTCH1 relieves its inhibition of Smo allowing it to translocate into the primary cilium and promote conversion of

GLI into its transcriptionally active form [68]. Importantly, primary cilia are necessary for generation of both GLI in its repressor and activator forms [68].

PDGF signaling through the primary cilia of fibroblasts is thought to regulate chemotaxis in response to PDGF-AA ligand. In fibroblasts PDGFRα is compartmentalized within primary cilia. PDGF-AA ligand binding to ciliary PDGFRα triggers a phosphorylation cascade through

PI3K-AKT and MEK1/2-ERK1/2 pathways leading to activation of NHE1, a sodium hydrogen exchanger, subsequent alkalinization of the cytoplasm and finally actin nucleation at lamellipodium that generates directional migration [59-61]. Impressively, in in vitro chemotaxis experiments the primary cilia of migrating fibroblasts aligns in the direction of the PDGF-AA gradient and ablation of primary cilia eliminates the chemotactic migratory response of fibroblasts [60].

The regulation of canonical Wnt signaling by primary cilia remains somewhat controversial. Previous studies demonstrate that Inversin/NPHP2, an inhibitor of the key Wnt protein Disheveled (Dvl), localizes within primary cilia [63, 64]. Briefly, in the absence of Wnt ligands, the Axin/APC/GSK3-β “destruction complex” degrades β-catenin preventing it from entering the cell nucleus and enabling transcription of downstream Wnt genes. When Wnt ligands bind to Frizzled (Fzd) receptors, the “destruction complex” is recruited to the receptor by

Dvl inhibiting its degradation of β-catenin and allowing for β-catenin to enter the nucleus and engage transcription of downstream Wnt genes [69]. It is thought that ciliary Inversin/NPHP2 inhibits Dvl and as such retards active Wnt signaling [69]. Primary cilia-mediated Wnt signaling remains controversial as studies in mice have shown conflicting findings. In one study, loss of

22 primary cilia in developing mouse embryos through knockout of various Intraflagellar Transport

(IFT) proteins and Kif3a (all proteins necessary for ciliogenesis) did not produce any phenotypes typical of aberrant Wnt signaling [70]. However, in other mouse studies loss of primary cilia by deletion of the Mks1 protein results in overactivation of Wnt signaling [71, 72].

In addition to regulating signal transduction pathways, primary cilia are also thought to be mechanosensitive to external forces such as fluid flow and tissue compression/deformation in certain tissues such as mesenchymal stem cells (MSCs) and its derivatives (bone and cartilage) as well as during development [73-78]. In human MSCs fluid flow is sensed by primary cilia and transduced into increased expression of osteogenic genes, such as COX2 and BMP2 which ultimately leads to transient periods of increased proliferation [78]. In bone, the primary cilia of osteocytes are thought to sense fluid variations in canaliculi of bones and regulate bone remodeling [73, 76, 77]. Specifically, fluid flow increases expression of osteogenic genes such as

COX2 and OPN [79]. This process is mediated by Adenyl Cyclase 6 (AC6), which actually localizes to primary cilia of osteocytes, as well as cAMP [79]. Interestingly, the mechanosensitivity of primary cilia in osteocytes is regulated by its length as longer primary cilia generate a greater osteogenic response [76]. In chondrocytes, primary cilia are embedded within the ECM surrounding the cell and are sensitive to compressive forces and tissue deformation in vitro [74, 77]. This mechanosensitivity is mediated by ATP-induced calcium signaling and results in increased expression of proteoglycans in chondrocytes [74]. Primary cilia are also thought to mediate mechanotransduction during embryonic development. Specifically, to achieve left-right asymmetry, motile cilia of midline nodal cells are thought to beat and generate fluid flow that is sensed by the primary cilia of peripheral nodal cells [77, 80, 81].

Mutations in numerous primary cilia genes are known to cause diseases which have a spectrum of phenotypes across multiple organ systems. These diseases have been collectively termed “Ciliopathies” and include Polycystic Kidney Disease, Bardet-Beidl Syndrome, Meckel-

Gruber Syndrome, and Joubert Syndrome [82]. Because of the ubiquity of primary cilia across many organ systems and during development, many of these Ciliopathies share common phenotypes such as renal cysts, skeletal dysplasia, bronchiectasis, and defects in brain development [83]. Many of the phenotypes seen in ciliopathies are associated with defective cell migration during development [84, 85]. Interestingly, the majority of human ciliopathies show no notable phenotypes in the skin. The lack of dermatological phenotypes may be explained by skin defects being embryonic lethal during development such that the developing embryo does not survive gestation. Evidence for this lethality hypothesis is supported by work in mice showing that targeted knockdown of ciliary genes such as IFT74 and ARL3 in the embryonic epidermis results in immediate postnatal lethality for newborn mice likely due to skin barrier defects [53, 86]. The only ciliopathy reported in humans (to my knowledge) with a noticeable skin phenotype is Cranioectodermal Dysplasia which is caused by mutations in IFT43, IFT121,

IFT122, or IFT144 and results in hair sparsity and skin laxity [87]. Interestingly there may be differences in ciliopathy skin phenotypes between mice and humans. RPGRIP1L, a gene mutated in Joubert Syndrome and Meckel-Gruber Syndrome, disrupts HF morphogenesis in mice but human patients with mutations in RPGRIP1L do not show skin phenotypes [88-90].

1.8 Ciliogenesis and cilia disassembly

Ciliogenesis, or the formation of a primary cilium, requires Intraflagellar Transport proteins (IFTs) as well as Kinesin-2 and Dynein protein motors (Figure 1.5) [91, 92]. During ciliogenesis, a subset of IFT proteins, including Ift20, Ift74, and Ift88, cluster together to form

24 the IFT-B complex while a different subset of IFT proteins, including Ift43, Ift121, and Ift122, cluster together to form the IFT-A complex [91]. The IFT-B and IFT-A complexes form an IFT particle which traffics cargo into and out of the primary cilium along the cilium’s 9+0 doublet- microtubule axoneme using Kinesin-2 and Dynein motors, respectively [91, 92]. Interaction studies reveal that Kinesin-2 binds the IFT-B complex whereas dynein binds to the IFT-A complex [91]. Through IFT, the molecular cargoes necessary for the formation and maintenance of a primary cilium are trafficked into and out of the cilium [91]. Loss of any of these IFT components, including Ift20, Ift74, Ift88 and Kif3a (a subunit of Kinesin 2) can disrupt cilia function and/or inhibit ciliogenesis [53, 93]. Importantly, primary cilia typically assemble in non-dividing G0/G1 phase cells and disassemble with cell cycle entry [94].

Primary cilia disassemble through gradual resorption of ciliary material into the cell cytoplasm [95]. The disassembly process is mediated by Aurora A kinase (AurA) which induces cilia resorption partly through activation of Histone Deacetylase 6 (HDAC6)-driven deacetylation of microtubules in the ciliary axoneme [96]. Deacetylation of the microtubules destabilizes the axoneme promoting resorption of the ciliary structure [96]. Interestingly, recent studies suggest that disassembly of primary cilia and subsequent advancement of cell cycle may be regulated by ciliary ectocytosis [94, 97]. During ciliary ectocytosis, ciliary cargo accumulates in the tip of the cilium, forming a bulged-tip which is subsequently “decapitated” [94, 97].

Ectocytosis is actin dependent and governed by the ciliary phosphatase Inpp5e through phosphoinositide signaling within the ciliary membrane [94, 97]. In fibroblasts, cilia disassembly and subsequent transition of the cell from G0 to G1 is dependent on ectocytosis as inhibition of ectocytosis suppresses cilia disassembly and significantly delays a cell’s G0 to G1 transition

[94]. Interestingly, in kidney cells, ectocytosis is thought to serve as a physiological mechanism

25 by which cells can modulate signaling within the primary cilium as the contents of the ectosome

(i.e. the excised bulged tip) includes certain ciliary G-protein coupled receptors (GPCRs) sensitive to Hedgehog signaling [97]. These GPCRs are not recycled through IFT transport and instead require ectocytosis for ciliary exit [97]. In addition to GPCRs, the contents of ectosomes contain IFT proteins, Kinesin subunits, and ciliary proteins such as Arl13b and Inpp5e [94].

Figure 1.5: Schematic of Intraflagellar Transport in primary cilia. (Adapted from Valente et al., 2013)

1.9 Ift20 has both ciliary and non-ciliary functions

IFT proteins were initially thought to exclusively function at the primary cilium, however recent studies indicate that many of these proteins have extra-ciliary functions. Here we review the ciliary and extra-ciliary functions of Ift20.

Ift20 is the smallest member of the IFT family of proteins and is evolutionarily conserved across vertebrates and invertebrates [93]. Homologs of Ift20 appear to be present in all ciliated organisms including Chlamydomonas, C. elegans, Drosophila melanogaster, Xenopus laevis, mice, and humans [93]. Ift20 is approximately 15 kDA and size and does not share significant homology with any other proteins [93]. Structurally Ift20 does not contain an conserved motifs except for a coiled-coil domain that makers up the C-terminal which accounts for approximately

75% of the protein [93]. As previously described, Ift20 is crucial in ciliogenesis where it is known to interact with other IFT proteins such as Ift52, Ift57, and Ift88 to form the IFT B complex which enables anterograde trafficking of ciliary cargoes into the primary cilium of a cell

[57, 93]. Loss of Ift20 can inhibit ciliogenesis both in vitro and in vivo [57, 93]. Interestingly, in addition to localizing to primary cilia, a significant proportion of Ift20 is also found in the Golgi complex of cells and is thought to participate in the trafficking of ciliary proteins from the Golgi complex to the basal body of the primary cilium where the cargo can be loaded into IFT complexes [93].

Ift20 has also been identified to regulate cancer cell invasiveness and collective cell migration independent of the primary cilium, through its role at the Golgi complex [98, 99].In non-ciliated cancer cells, loss of Ift20 has been shown to fragment the Golgi complex, disrupt trafficking and ultimately reduce cancer cell invasiveness and collective cell migration [98].

Alternatively, it has been proposed that the reduced collective cell migration observed in cancer

28 cells subsequent to knockdown of Ift20 is due to the inability of the Golgi complex to properly polarize due to its fragmented nature [99].

Additionally, Ift20 has also been identified to regulate recycling of cell surface receptors

[100, 101]. In T-cells, Ift20 regulates endosomal recycling of T-cell receptors crucial for immune synapse activation independent of primary cilia [100-102]. Specifically, Ift20 works in tandem with the Rab GTPase network to ensure proper recycling of T-cell receptors at the immune synapse [100-102]. Loss of Ift20 blocks recycling of T-cell receptors and results in entrapment of

T-cell receptors within Rab5-positive endosomes [101].

These examples demonstrate the multi-faceted roles for Ift20 within a cell: ciliogenesis, trafficking, and endosomal recycling.

Chapter 2: Characterization of primary cilia in embryonic and

postnatal epidermis in a transgenic mouse model

2.1 Introduction

Primary cilia regulate both epidermal development and postnatal epidermal homeostasis

[53, 103]. Previous studies show that loss of primary cilia in developing epidermis disrupts

Notch-dependent stratification of the interfollicular epidermis (IFE) and Shh-dependent hair follicle (HF) morphogenesis resulting in an epidermis with IFE that is non-stratified with hyperproliferative Basal Stem Cells (BSCs) and HFs that are short and stunted [53]. In postnatal epidermis, ablation of epidermal primary cilia results in hyperproliferation of BSCs in the IFE, ventral alopecia, as well as disruption of HF organization and structure in some instances [103].

These previous studies examining primary cilia in embryonic and postnatal skin have visualized primary cilia using antibodies against various ciliary proteins [53]. While antibody staining is useful in fixed tissue sections, it precludes live imaging and imposes limits on the number of cellular structures that can be visualized together. Here we utilize a double transgenic mouse expressing mCherry-tagged Arl13b and GFP-tagged Centrin 2 which mark primary cilia and centrosomes/basal bodies, respectively [104]. We characterize primary cilia in both developing and postnatal epidermis and demonstrate proof of concept for imaging primary cilia in live, ex- vivo developing epidermis.

2.2 Characterization of Arl13b-mCherry; Centrin 2-GFP developing epidermis

Both Arl13b and Centrin 2 have respectively been utilized as markers of primary cilia and centrosomes/basal bodies in previous studies [104-107]. Briefly, Arl13b is a member of the

ADP ribosylation factor-like sub-family of small GTPases which strongly and specifically localizes to the membrane of primary cilia [104-106]. Centrin 2 is a structural component of the 30 centrosome and as such, is subsequently found in basal bodies which derive from the mother centriole of the centrosome [107], To confirm the transgenic labeling and localization of the

Arl13b-mCherry; Centrin 2-GFP mice we first examined sections of 14.5 day old embryos

(E14.5) and co-labeled the sections with an Arl13b antibody (Figure 2.1 A). At E14.5, we can clearly identify nascent placode formations in the developing epidermis within which are keratinocytes with primary cilia that are co-labeled both by the Arl13b-mCherry transgene as well as the Arl13b antibody (Figure 2.1 A). Additionally, we can find keratinocytes surrounding developing placode (the future IFE) that also have primary cilia co-labeled by both the Arl13b transgene and antibody (Figure 2.1 A). Notably, the basal bodies as labeled by the Centrin2-GFP transgene are localized appropriately at the base of cilia (Figure 2.1 A). Examination of whole mount E14.5 Arl13b-mCherry; Centrin 2-GFP back skin reveals the highly ciliated nature of developing epidermis (Figure 2.1 B). Importantly, both transgene markers show strong fluorescent signal without the need for any additional antibody labeling.

We next examined E15.5 sections of the transgenic mouse where we observed significant placode development in the epidermis. At E15.5 the placodes have grown in size and have invaginated into the underlying dermis (Figure 2.2 A). Examination of E15.5 whole mount back skin shows that placodes are now easily identifiable and appear as circular “swirls” of keratinocytes (Figure 2.2 B). Also evident at E15.5 is the stratification of the regions surrounding the placodes which will eventually develop into the IFE (Figure 2.2 A). Notably, similar to at

E14.5, the keratinocytes within the placodes as well as in the surrounding regions (future IFE) are highly ciliated (Figure 2.2 A-B). We can clearly identify ciliated basal cells in non-placode

(future IFE) regions (Figure 2.2 A, right arrow) as well as ciliated cells in the newly stratifying

31 layer above the basal layer (Figure 2.2 A, left arrow). Primary cilia are also identifiable in the keratinocytes of the developing HF (Figure 2.2 A, arrowheads; Figure 2.2 B arrowheads)

Importantly, we observe no gross defects in the skin of E14.5 and E15.5 transgenic embryos. In both E14.5 and E15.5 skin we see epithelial cell-cell junctions as indicated by E- cadherin staining in the developing placodes and surrounding regions (i.e. future IFE) that is typical of skin at this embryonic timepoint (Figure 2.1 C and Figure 2.2 A). Overall, epidermal development in Arl13b-mCherry; Centrin 2-GFP mice is comparable to non-transgenic mice with regards to HF morphogenesis, IFE stratification, and tissue architecture.

Figure 2.1: Characterization of E14.5 Arl13b-mCherry; Centrin 2-GFP epidermis. A) Sagittal sections of E14.5 skin with antibody staining against Arl13b. Region of interest is magnified in inset. B) Whole mount E14.5 epidermis. C) Sagittal section of E14.5 future IFE region showing junctional staining with E-cadherin. Arrowheads indicate primary cilia in developing placode in A). Arrows indicate primary cilia in region surrounding placode or the IFE in A) and B). Scale bars are 10 μm in A), B) and C). Dashed lines indicate epidermal- dermal junction. Images in A), B), and C) are Z-projections of 5-10 confocal slices.

Figure 2.2: Characterization of E15.5 Arl13b-mCherry; Centrin 2-GFP epidermis. A) Sagittal sections of E15.5 skin. Magnified image of region of interest is shown on the right. Dashed lines indicate epidermal dermal junction. * indicates background fluorescence. B) Whole mount E15.5 skin. Developing placode is outlined with dashed line. Arrowheads indicate primary cilia in developing placode in A). Arrows indicate primary cilia in region surrounding placode or the IFE in A) and B). Scale bars are 10 μm in A) and B). Images in A) and B) are Z-projections of 5-10 confocal slices.

2.3 Characterization of Arl13b-mCherry; Centrin 2-GFP postnatal and adult

epidermis

We next investigated postnatal and adult skin of the Arl13b-mCherry; Centrin 2-GFP mice. Importantly, we did not observe any gross defects in the skin of newborn or adult mice.

Hair began appearing on newborn pups at postnatal day 7 to 8 (P7-P8) and a complete hair coat was observed by P10-P11, typical of most mice. To confirm that transgene expression was still present in newborn and adult skin, we examined postnatal day 1 (P1) and adult (P250) age skin

(Figure 2.3 and Figure 2.4). Sections of P1 back skin shows epidermis with fully-developed HFs and IFE that is well stratified (Figure 2.3 A). Primary cilia can be found throughout the HF structure as well as in both the basal and suprabasal layers of the surrounding IFE (Figure 2.3 A, arrow indicates ciliated basal cell). The prevalence of primary cilia is even more evident upon examining P1 whole mount back skin where we again find that both the HF and IFE are highly ciliated (Figure 2.3 B). Epidermal cell-cell junctions are well defined as indicated by E-cadherin staining (Figure 2.3 A-B).

Similarly, we find that both the IFE and HFs are highly ciliated in adult (P250) skin

(Figure 2.2 C). Compared with P1 neonatal skin, we find that the HFs have significantly elongated. Using whole mount preparations of P250 back skin, we were able to re-construct whole HFs through confocal microscopy (Figure 2.4). These reconstructions show that primary cilia are found in keratinocytes of the hair germ and bulge regions as well as other regions of the

HF (Figure 2.4). Grossly, the hair coat of adult mice showed no defects and was comparable to non-transgenic mice of the same age. Overall, the addition of the Arl13b-mCherry and Centrin 2-

GFP transgenes have no immediately visible deleterious effects on epidermal development or homeostasis in mice.

Figure 2.3: Characterization of P1 Arl13b-mCherry; Centrin 2-GFP epidermis. A) Sagittal sections of P1 skin. Magnified images of regions of interest are shown on the right. Dashed

36 lines indicate epidermal dermal junction. * indicates background fluorescence from corneal layer. B) Whole mount P1 skin. Hair follicle is outlined with dashed line. Arrowheads indicate primary cilia in hair follicle in A) and B). Arrows indicate primary cilia in the IFE in A) and B). Scale bars are 10 μm in A) and B). Images in A) and B) are Z-projections of 5-10 confocal slices.

Figure 2.4: Whole mount P250 Arl13b-mCherry; Centrin 2-GFP skin. A) Z projection of 15 confocal slices of hair follicle. Magnified images of regions of interest are shown on right. The upper and lower regions of interest are the outer root sheath and hair germ of the hair follicle, respectively. Arrowheads indicate primary cilia. The lower region of interest is Dotted line indicates epidermal dermal junction of the hair follicle. Scale bar is 10 μm.

2.4 Cilia length, morphology, and live ex-vivo imaging

Recent studies suggest that key ciliary functions including mechanosensation and ectocytosis are regulated by cilia length and morphology [74, 94, 97]. Cilia morphologies that have been previously characterized and identified in other cell types include straight, branched, and bulged-tip [94, 97, 108]. To better characterize primary cilia in developing and postnatal epidermis, we characterized the length and morphology of epidermal primary cilia at different developmental timepoints.

For this analysis of we used whole mount E14.5, E15.5, and P1 transgenic back skin and focused on primary cilia in the IFE (non-HF/non-placode) regions of back skin to enable easier length measurements and identification of ciliary morphologies. Overall, we do not observe any significant difference in the length of primary cilia between E14.5, E15.5 and P1 IFE (non- placode) regions of epidermis. The average length +/- standard deviation of primary cilia for

E14.5, E15.5, and P1 were, respectively, 2.2 +/- 1.0 μm, 2.1 +/- 1.1 μm, and 2.2 +/- 1.0 μm (see

Table 2.1). Statistical p-values computed using a Student’s t test were 0.40, 0.67, and 0.76 for

E14.5 versus E15.5, E15.5 versus P1, and E14.5 versus P1, respectively.

With regard to ciliary morphology, across all timepoints the majority of all primary cilia show straight morphology. We did not observe any primary cilia showing branched morphology.

There is however a high proportion of primary cilia that show bulged-tip phenotypes (Figure 2.1

A inset; Figure 2.1 B lower arrow; Figure 2.2 B arrow; Figure 2.3 B arrow). Quantification of bulged tipped primary cilia is shown in Table 2.2. Overall, we find a greater proportion of bulged-tip primary cilia at P1 compared to E14.5 or E15.5.

Table 2.1: Length of primary cilia. Average length of primary cilia in IFE (non-placode) regions of epidermis at various developmental time points. The total number cilia analyzed is indicated. For each time point, 2 or more back skins were analyzed.

Table 2.2: Bulged-tip primary cilia frequency. Proportion of primary cilia in IFE (non- placode) regions of epidermis with bulged-tip morphology at various developmental time points. The number of bulged-tip cilia and total number of cilia are indicated. For each time point, 2 or more back skins were analyzed.

Given the significance of bulged-tipped primary cilia and ectocytosis in regulating ciliary signaling and the cell cycle in vitro [94, 97], we wondered if ectocytosis occurs physiologically in vivo in the epidermis given the high proportion of primary cilia with bulged-tips (Table 2).

Here we choose to leverage our transgenic mice to determine if we could develop a method to image primary cilia in real time and determine if ectocytosis occurs in epidermis in vivo. We choose to focus on E14.5 embryos as their epidermis contains a significant proportion of primary cilia with bulged tips (Table 2.2) and the epidermis is not yet fully stratified making it amenable to imaging. Briefly, E14.5 transgenic embryos were extracted from the uterine sac of pregnant females. The whole embryos were placed in glass bottom imaging dishes such that the backskin of the embryos was flush against the glass bottom. The embryos were then mounted in place by partially embedding them in 1% low melting point agarose that was allowed to harden and submerging the remaining exposed embryo in keratinocyte culture medium. The embryos were then imaged using a spinning disk confocal microscope within a temperature and CO2 controlled environment.

Using this technique, we are able to observe the dynamics of primary cilia in live, developing embryos (Figure 2.5). Both the Arl13b-mCherry and Centrin 2-GFP signals are bright so that primary cilia are easily visualized and the morphology of a primary cilium is easily identifiable (Figure 2.5 A). Similar to fixed, whole mount preparations of E14.5 back skin, we are able to identify a significant proportion of epidermal primary cilia with bulged tips (Figure

2.5 A insets). Remarkably, we are able to visualize ectocytosis in vivo in real time wherein we can observe the bulged-tip of an epidermal primary cilium “decapitate” from the remaining cilium structure (Figure 2.5 B). The time frame of this ectocytosis event in vivo (approximately

80 minutes) is similar to previously reported kinetics in vitro [94, 97].

Despite the relatively high numbers of bulged-tip epidermal primary cilia in E14.5 embryos, observing ectocytosis was a relatively rare occurrence. We only observed ectocytosis occur in a total of two primary cilia across three E14.5 embryos in which 10 to 20 primary cilia were observed per embryo. This low frequency may be attributed to technical issues such as fluorophore photobleaching as well as focal drift. Additionally, unlike in previous in vitro studies, ectocytosis in developing epidermis cannot be triggered such that cilia undergo ectocytosis in a relatively synchronized or predictable manner.

Figure 2.5: Live imaging of E14.5 Arl13b-mCherry; Centrin 2-GFP transgenic embryo. A) Z projection of 10 confocal slices of E14.5 transgenic backskin. The upper and lower insets are magnified images of the right and left outlined regions of interest, respectively. Scale bar is 10 μm. B) Z projection of 10 confocal slices of an epidermal primary cilium undergoing ectocytosis. Arrow indicates separated ectosome. Scale bar is 3 μm.

2.5 Discussion

Overall, we find that the Arl13b-mCherry; Centrin 2-GFP transgenic mouse serves as an excellent tool for investigating primary cilia in the epidermis. Notably, epidermal primary cilia are easily visualized and identifiable without the use of additional antibody staining. General characterization of skin development and postnatal skin in these mice indicates that the transgenes do not appear to have any deleterious effects on skin development or skin architecture in embryonic or postnatal skin. Specifically, HF placodes appeared by E14.5 during embryogenesis and the development of HFs and IFE was similar to that of non-transgenic mice.

Skin architecture, cell-cell junctions, stratification of IFE, and hair growth appeared comparable to non-transgenic mice.

Using this mouse model, we find that both HFs and IFE are highly ciliated during development and postnatally, in adult mouse skin. We characterized the length and morphology of primary cilia in the IFE at various time points in development and in postnatal skin and found no significant differences in primary cilia length between any of the timepoints. Surprisingly, we observed a significant proportion of primary cilia with a bulged-tip morphology. To further investigate the physiological significance of the bulged-tip primary cilia we developed a live imaging technique for ex-vivo embryos and were able to observe ectocytosis of bulged-tip epidermal primary cilia in live embryos.

Previously ectocytosis had only been observed in vitro under strict serum induced conditions [94, 97]. This live imaging demonstrates that ectocytosis occurs physiologically in live embryos and is not an artifact of cell culture or serum induction. Our results also serve as proof of concept that live imaging of epidermal primary cilia ex vivo is feasible. Further refinement of this technique is necessary as the current methodology suffers from 2 main

44 technical issues, mainly focal drift and fluorophore photo-bleaching. These issues have limited us to imaging sessions of only 3-4 hours before the primary cilia signal loses fluorescent intensity or drifts out of focus. These issues should be resolvable with further refinement of the mounting mechanism and or change in fluorescence imaging settings.

Further studies into the contents of the ectosomes of epidermal primary cilia during development is also warranted. Previous studies have identified GPCRs, Hedgehog (Hh) signaling molecules such as Gli, ciliary proteins like Arl13b, Kinesin subunits, as well as core

IFT B and IFT A proteins within ectosomes of fibroblasts and kidney cells [94, 97].

Speculatively, the ectosomes of epidermal primary cilia likely contain core IFT proteins and ciliary proteins such as Arl13b and Kinesin subunits like the ectosomes of fibroblasts and kidney cells, but they may additionally contain keratinocyte-specific receptors and effectors that such as

Notch 3 or Presenilin 2, which both localize in epidermal primary cilia [53]. Another question that arises is what happens to the ectosome and its contents? Electron microscopy shows that primary cilia of keratinocytes can actually invaginate into neighboring cells suggesting that ectosomes could be deposited onto neighboring cells. Consistent with this observation, ectocytosis and ectosomes have been speculated by many as a possible method of non-cell- autonomous signaling (i.e. cell to cell communication) mediated by primary cilia [109].

Additional investigation into the role of ectocytosis and G0 to G1 cell cycle transition in vivo is also necessary. The Fluorescence Ubiquitin Cell Cycle Indicators (FUCCI) used to establish the relationship between ciliary ectocytosis and the G0 to G1 transition in vitro have been introduced into mice such that the cell cycle state of all cells in a mouse can be monitored.

Crossing the FUCCI mice with the Arl13b-mCherry; Centrin 2-GFP mice could serve as a powerful tool for answering the question of whether ciliary ectocytosis serves as a checkpoint for

45 the G0-G1 cell cycle transition in vivo. Specifically, it would answer the question of whether ciliated epidermal keratinocytes require ectocytosis for ciliary disassembly and subsequent transition from G0 to G1.

Chapter 3: Ift20 and hair follicle homeostasis

3.1 Introduction

Primary cilia are found in BSCs of the IFE and HF-SCs in both embryonic, and postnatal, adult mouse skin. As previously discussed, primary cilia are crucial in orchestrating stratification of non-placode (future IFE) regions of embryonic epidermis and for HF morphogenesis [53].

Loss of primary cilia in developing epidermis impedes differentiation and stratification of IFE

(non-placode) regions of epidermis and stunts HF morphogenesis due to loss of Notch and Shh signaling, respectively [53]. This results in epidermis with a hyperproliferative basal layer in the

IFE and shortened, malformed HFs [53].

Despite the importance of primary cilia during epidermal development, investigation into the function of primary cilia in postnatal, adult skin is limited. One previous study found that deletion of epidermal primary cilia in postnatal mouse skin results in hyperproliferative BSCs and ventral alopecia [103]. Additional phenotypes in postnatal epidermis such as tail skin with disorganized HF structures with excess sebaceous glands are observed if loss of primary cilia occurs during embryogenesis [103]. It should be noted however that the results were obtained using Keratin 14 (K14)-Cre mice to ablate primary cilia and resulted in loss of primary cilia in

BSCs of the IFE as well as the cells of the entire outer root sheath of the HF which includes outer

HF-SCs in the bulge [103]. The same study did find that ablation of primary cilia in Shh+ cells in the lateral disc of the hair matrix within the HF (using Shh-Cre mice) resulted in migration of

Shh+ HF cells to the IFE or possible conversion of the Shh+ cells into IFE [103]. Interestingly, loss of primary cilia in these Shh+ cells did not disrupt HF morphogenesis [103]. There has been minimal investigation into the role of primary cilia specifically in HF-SCs in postnatal, adult skin during homeostasis and what phenotypes, if any, occur following loss of this cellular

47 structure. This outstanding question is of particular interest as both the Wnt and Shh signaling pathways that govern HF homeostasis are thought to be regulated through primary cilia.

As such this chapter investigates the role of primary cilia in HF-SCs and HF homeostasis.

In particular we generate a mouse model where Ift20 is specifically ablated in HF-SCs to eliminate primary cilia in HF-SCs. We then develop a HF cycling assay to examine the effect of postnatal loss of HF-SC primary cilia in hair cycling.

3.2 Postnatal loss of Ift20 and primary cilia results in gross hair regrowth defect in mice

To investigate the role of primary cilia in HF-SCs in postnatal, adult skin, we generated

Keratin 15 (K15)-CrePGR; IFT20 fl/fl; Rosa26-tdTomato fl/fl mice. Ift20 is part of the IFT B complex and loss of Ift20 is known to inhibit ciliogenesis in keratinocytes in vitro. In these mice, following topical application of RU-486, K15-CrePGR specifically drives Cre-recombination in bulge HF-SCs. This enables specific ablation of IFT20 and activation of Tomato fluorescence in

HF-SCs (Figure 3.1 A-C). Importantly the Tomato fluorescence enables lineage tracing and clear identification of HF-SCs and any cells derived from HF-SCs.

Loss of Ift20 expression in K15-CrePGR; IFT20 fl/fl; Rosa26-tdTomato fl/fl mice was confirmed through both immunofluorescence staining (Figure 3.1 C) as well as qRT-PCR for

IFT20 mRNA in FACS sorted HF-SCs obtained from the back skin of treated mice (Figure 3.1

D-E). In both instances we observe a significant decrease in Ift20 expression in IFT20 cKO

(IFT20 fl/fl) mice compared to control (IFT20 +/fl) mice. Furthermore, as expected, loss of Ift20 inhibits ciliogenesis and we observe a significant reduction in ciliated HF-SCs (Figure 3.1 F).

Figure 3.1: Ablation of IFT20 in K15-CrePGR; IFT20 fl/fl; Rosa26-tdTomato fl/fl mice. A) Schematic of Tomato fluorescence-based lineage tracing. Following RU 486 treatment, tomato fluorescence is activated in bulge (Bu) HF-SCs and subsequent derivatives of HF-SCs such as the Hair Germ (HG), and Transit Amplifying Cells (TAC). B) Immunofluorescence images of Tomato (+) signal (red) labeling the HF-SCs of the bulge (Bu) and their transiently amplifying progenitors in the Hair Germ (HG). C) Immunofluorescence images of Ift20 in HF-SCs (red) of control (IFT20 +/fl) versus IFT20 cKO (IFT20 fl/fl) telogen staged mouse skin. D) FACS sorting profile of epidermal cells using CD34 and α6 Integrin surface labeling. E) qRT-PCR expression of IFT20 in FACS sorted HF-SCs. F) Quantification of the number of ciliated HF-SCs in control (IFT20 +/fl) versus IFT20 cKO (IFT20 fl/fl) skin. Data in histogram represents 20-30 hair follicles from n=3 mice. Dotted lines indicated epidermal-dermal junction in B) and C).

To investigate the effect of Ift20 and primary cilia loss in HF-SCs, we designed a HF cycling assay focused on the telogen to anagen transition. Briefly, the back skin of mice age P45-

P50 (i.e. mice in 2nd telogen) were shaved and a 1% topical solution of RU 486 was applied daily for 5-7 days. Following RU 486 treatment, hair-regrowth was monitored. Mice were harvested and analyzed at P100-P130 when wild type mice typically have transitioned out of telogen and are in mid-anagen (Figure 3.2 A) [110].

Using this assay, we observe a gross hair regrowth defect in IFT20 cKO (IFT20 fl/fl) mice compared to littermate controls (IFT20 fl/+) subjected to the same treatment. Specifically, while control mice are able to fully regrow their hair coat by the end of the assay, we observe minimal hair regrowth in IFT20 cKO mice (Figure 3.2 B). Importantly this phenotype is reproducible as it was observed in 6 out of 7 pairs of IFT20 cKO and control littermate mice treated. Histological sections of back skin obtained from the mice following the assay reveal that while the HFs in control mice have transitioned out of telogen and have entered anagen, the HFs in IFT20 cKO mice appear fixed in telogen (Figure 3.2 C). More specifically, the HFs of IFT20 cKO mice are short (i.e. telogen-like) in contrast with HFs of the control mice which are fully elongated and invaginate downwards into the underlying dermis and subcutaneous fat (Figure

3.2 C). Quantification of HF length reveals that HFs in IFT20 cKO mice are significantly shorter than those in control mice (Figure 3.2 D). Importantly, this HF phenotype appears robust and permanent as even when this assay was extended to up to P150 we saw very limited hair regrowth in IFT20 cKO mice.

Figure 3.2: Loss of IFT20 in HF-SCs results in gross hair regrowth defect. A) Induction assay protocol. B) Representative images of IFT20 cKO (IFT20) and control (IFT20 +/fl) mice at P50 and P120 during assay. C) Sagittal sections of Hematoxylin and Eosin stained P120 back skin. Scale bars are 200 μm. D) Quantification of hair follicle length following assay completion. Data represents 40-80 hair follicles from n=3 mice for each condition. Error bars are standard deviation. *** indicates p=2.2e-16 by Mann-Whitney U Test.

3.3 Wnt signaling is absent in IFT20 cKO hair follicles

The gross hair regrowth defect and short, telogen-phase HFs indicate that loss of Ift20 in

HF-SCs significantly delays and or inhibits the telogen to anagen transition. Crucial for this transition is the increase in autocrine and paracrine canonical Wnt signaling in HF-SCs, hair germ, and the dermal papilla (DP) of HFs [21, 111]. Because of this well-established role for

Wnt signaling during the telogen-anagen transition, we wondered if Wnt signaling was affected following loss of Ift20 in HF-SCs.

To investigate Wnt signaling, we examined sections of IFT20 cKO skin for Lef1 and nuclear β-catenin. Lef1 is a downstream transcription factor target of canonical Wnt signaling and nuclear localization is an indication of active Wnt signaling. Nuclear β-catenin is another indicator of active Wnt signaling. As expected, we observed robust nuclear Lef1 in the HF-SCs, hair germ, and DP of control (IFT20 +/fl) HFs (Figure 3.3 A). Similarly, we also observed nuclear β-catenin in the HF-SCs and hair germ of control (IFT20 +/fl) HFs (Figure 3.3 B). In contrast with control HFs, IFT20 cKO HFs did not show either of these nuclear markers: Lef1 staining was consistently absent in IFT20 cKO HFs while β-catenin staining was predominantly junctional (Figure 3.3 B). Quantitatively, 9/10 and 23/26 control HFs showed Lef1 and β-catenin staining, respectively, while only 0/19 and 12/41 IFT20 cKO HFs showed Lef1 and β-catenin staining, respectively. This difference in both Lef1 and β-catenin expression was significant as computed using a Fisher’s Exact Test (p=1.00x10-6 for Lef1 and p=2.90x10-6 for β-catenin). Lef1 data was obtained from 1 mouse for each condition while β-catenin data was obtained from 2 mice per condition. Overall, the lack of nuclear Lef1 and β-catenin are strong indicators that Wnt signaling is not active in IFT20 cKO HFs. Note that the Lef1 analysis was performed on P30-

P35 age mice during early second anagen after the mice been treated with RU486 during first

52 telogen (P20-P25) while β-catenin analysis was performed on third anagen age mice (P100-

P130) as previously described.

Figure 3.3: IFT20 cKO hair follicles lack canonical Wnt signaling. A) Immunofluorescence staining for LEF1 in control (IFT20 +/fl) and IFT20 cKO hair follicles during second anagen. Regions of interest are magnified in the insets. Scale bars are 50 μm. B) Immunofluorescence staining for β-Catenin in control (IFT20 +/fl) and IFT20 cKO hair follicles during third

54 anagen from sagittal sections. Regions of interest are magnified in the insets. Arrow heads indicate nuclear β-Catenin. Scale bars are 50 μm.

3.4 Loss of Ift20 in HF-SCs results in loss of SOX9 expression, stem cell identity and ectopic expression of differentiation markers

Both the dramatic hair follicle phenotype as well as the significant delay in transitioning from telogen to anagen observed in IFT20 cKO mice mimics the phenotype of mice in which the transcription factor SOX9 is ablated in HF-SCs [31, 32]. SOX9 is a “signature” gene of HF-SCs that maintains HF-SCs and is required for hair growth [31-33]. SOX9 is known to be expressed throughout the outer root sheath and in HF-SCs in the bulge in anagen phase HFs, while in telogen phase HFs, SOX9 expression is limited to HF-SCs in the bulge [31, 112]. Loss of SOX9 in HF-SCs inhibits hair regrowth and also results in ectopic differentiation of HF-SCs such that they express markers of differentiated IFE [32]. Given the similarity in phenotype between the

IFT20 cKO mice and SOX9 cKO mice, we examined SOX9 expression in the IFT20 cKO mice.

Surprisingly we find that SOX9 expression is reduced in IFT20 cKO HFs (Figure 3.4).

Specifically, we observe clear nuclear SOX9 staining in Tomato (+) cells in the outer root sheath regions of control (IFT20 +/fl) HFs and observe that SOX9 is absent in all Tomato (+) cells in

IFT20 cKO HFs (Figure 3.4). Quantitatively, we observed SOX9 staining in Tomato (+) cells in

19/27 control HFs and 0/26 IFT20 cKO HFs. This difference in SOX9 presence is significant as computed using Fisher’s Exact Test (p=2.40 x 10-8). This SOX9 data was obtained from 1 mouse per condition. Note that SOX9 analysis was performed on third anagen age mice (P100-P130) as previously described.

Figure 3.4: IFT20 cKO hair follicles lose SOX9 expression. A) Immunofluorescence staining for SOX9 in control (IFT20 +/fl) and IFT20 cKO hair follicles during third anagen. Regions of interest are magnified in the insets. Scale bars are 100 μm. * indicates background fluorescence.

In addition to loss of SOX9 expression, IFT20 cKO HF-SCs also express markers of differentiated epidermis, again mimicking the phenotype observed when SOX9 is ablated in HF-

SCs. In particular, we observe both Keratin 10 (K10) and Filaggrin expression in Tomato (+)

HF-SCs and HF-SC-derived cells within the hair germ (Figure 3.5 A-B). Quantification shows that expression of both of the markers (as measured by fluorescence intensity) is significant compared to controls (Figure 3.5 C-D). Importantly, both K10 and Filaggrin are markers of differentiated epidermis typically only found in suprabasal cells of the IFE [32, 33]. The presence of both suggests that the HF-SCs and HF-SC-derived hair germ cells may have switched stem cell lineages [33]. Note that this K10 and Filaggrin analysis was performed on

P30-P35 age mice during early second anagen after the mice been treated with RU486 during first telogen (P20-P25).

HF-SC lineage plasticity is known to occur physiologically in wound healing and pathologically in Squamous Cell Carcinomas (SCCs) [33]. This plasticity is transient in wound healing and sustained in SCCs, however in both instances it is characterized by SOX9 expression in HF-SCs being replaced by expression of the transcription factor KLF5 [33]. Just as SOX9 is thought of a “signature” gene for HF-SCs, KLF5 is thought of as a “signature” gene for BSCs of the IFE [33]. Mechanistically, KLF5 is known to suppress SOX9 and as such KLF5 expression allows for HF-SCs to switch their lineages to being like BSCs of the IFE [33]. Based on the loss of SOX9 and expression of differentiation markers in HF-SCs of IFT20 cKO HFs, we wondered if these HF-SCs had switched lineages and now expressed KLF5. Surprisingly, in IFT20 cKO mice of third anagen age (P100-P130), we find Tomato (+) cells in the hair germ and bulge regions of the HF that express KLF5 (Figure 3.5 E). This ectopic expression of KLF5 is absent in

58 control (IFT20 +/fl) HFs as expected (Figure 3.5 E). Quantitatively, we observed KLF5 in

Tomato (+) cells in 5/9 control HFs and 0/15 IFT20 cKO HFs. This difference in KLF5 expression is significant as computed by Fisher’s Exact Test (p=0.003). This KLF5 data was obtained from 1 mouse per condition. Taken together, the loss of SOX9 and ectopic expression of KLF5 and differentiation markers suggest that loss of Ift20 in HF-SCs results in a change in lineage from HF-SCs to BSCs.

Figure 3.5: IFT20 cKO hair follicles show ectopic expression of differentiation markers and KLF5. A) Immunofluorescence staining in sagittal sections for K10 in control (IFT20 +/fl) and IFT20 cKO hair follicles. Regions of interest are magnified in the insets. B)

Immunofluorescence staining in sagittal sections for Filaggrin in control (IFT20 +/fl) and IFT20 cKO hair follicles during second anagen. Regions of interest are magnified in the insets. C) Quantification of K10 fluorescence in HFs. D) Quantification of Filaggrin fluorescence in HFs. E) Immunofluorescence staining in sagittal sections for KLF5 in control (IFT20 +/fl) and IFT20 cKO hair follicles during third anagen. Regions of interest are magnified in the insets. Arrow heads indicate nuclear KLF5. Scale bars are 50 μm in A) B) and E). A.U indicates arbitrary units in C) and D). A Mann-Whitney U test was used to compute statistical values in C) and D). Dotted lines indicate dermal-epidermal junction.

3.5 Discussion

Our data indicates that Ift20 is crucial in maintaining homeostasis of HFs. Loss of Ift20 in

HF-SCs results in a hair re-growth defect where HFs appear to be trapped in telogen.

Specifically, IFT20 cKO HFs do not show markers of active canonical Wnt signaling, such as nuclear Lef1 or nuclear β-catenin, that typically mark HFs in anagen. This lack of active Wnt signaling is strong evidence of a delay and or inhibition of transition from telogen to anagen. It should be noted however that it is unclear if the delayed/inhibited telogen to anagen transition is due to the loss of Wnt signaling or if the absence of Wnt signals stems from the HFs remaining in telogen.

Closer examination of IFT20 cKO HF-SCs reveals they lose expression of SOX9, a signature gene of HF-SCs, and instead show ectopic expression of KLF5, a signature gene of

BSCs in the IFE. While SOX9 expression is more evident in anagen HFs as it is present throughout the outer root sheath and bulge, SOX9 expression is still seen in the bulge in telogen

HFs of wild type mice [31, 112]. This is in contrast with the IFT20 cKO mice in which SOX9 is absent in Tomato (+) cells in the bulge. Loss of SOX9 in HF-SCs has previously been reported to inhibit hair re-growth and promote ectopic differentiation of HF-SCs [31-33]. Our findings are in agreement with these previously reported results as IFT20 cKO HFs also express differentiation markers K10 and Filaggrin in HF-SCs and HF-SC-derived hair germ. Taken together, the loss of SOX9 and expression of differentiation markers and KLF5 in HF-SCs of

IFT20 cKO mice suggest that Ift20 is crucial in maintaining HF-SCs and that without Ift20, the

HF-SCs have switched to a BSC/IFE lineage. Interestingly, Ift57, which interacts with Ift20, has been identified as a downstream gene regulated by SOX9 in HF-SCs [32].There is however, no direct evidence of Ift57 regulating SOX9 expression.

Further investigation is necessary to determine if the observed lineage infidelity is directly attributable to the loss of primary cilia in HF-SCs or if it is due to the loss of a non- ciliary function of Ift20 in HF-SCs. Given our finding that loss of Ift20 inhibits trafficking of β1 integrin to the surface of keratinocytes and reduces overall surface levels of β1 integrin (see

Chapter 4), it is very tempting to attribute the differentiation and lineage infidelity phenotypes to deficient β1 integrin levels especially since previous in vitro studies have shown that activation of β1 integrin inhibits terminal differentiation in keratinocytes [113-115]. While no studies have specifically ablated β1 integrin in HF-SCs, in vivo loss of β1 integrin in BSCs of the IFE does not produce the expected differentiation phenotype [116]. Hence it is unclear if loss of β1 integrin would result in differentiation of HF-SCs.

Our data suggests that Ift20 functions upstream of SOX9 and HF-SC maintenance. As such, another enticing trafficking hypothesis is that Ift20 regulates trafficking of a Wnt receptor to the surface of keratinocytes. As previously described, HF-SCs and hair germ derived from

HF-SCs express a variety of Wnt receptors including Frizzled 1, Frizzled 2, Frizzled 7, and

Frizzled 10 both during telogen and anagen [20]. These receptors are crucial for the maintenance of HF-SCs through autocrine and paracrine Wnt signaling during telogen and also crucial for the

HF-SCs in responding to the increase in autocrine and paracrine Wnt signals that triggers the telogen to anagen transition [21]. Under trafficking hypothesis, loss of Ift20 would reduce surface expression of Wnt receptors and attenuate the maintenance of HF-SCs and retard the telogen to anagen transition [21]. Concordant with this hypothesis is that abolition of Wnt-ligand secretion in HF-SCs has inhibits hair growth in adult mice and results in degraded HFs [21]. It is however unclear if loss of SOX9 expression or HF-SC differentiation occur following loss of

Wnt secretion.

To definitively address whether the hair regrowth, HF-SC lineage infidelity and differentiation phenotypes are due to ablation of primary cilia or due to a non-ciliary function of

Ift20, ablation of primary cilia through a different ciliary protein is necessary. The IFT88 fl/fl mouse [117] is particularly attractive for this purpose. If the same phenotypes were observed when HF-SC primary cilia are ablated using the IFT88 fl/fl mouse, it would suggest that the primary cilium is responsible for all of the observed defects. However, if only a subset of the defects were observed, it could be concluded that those particular defects are due to loss of primary cilia and the other defects are due to loss of non-ciliary functions of Ift20.

Chapter 4: Ift20 in Cell Migration and Wound Healing

Note: The following chapter is from Su et al, Mol Bio Cell 2020 (in press).

4.1 Introduction and background

The skin serves as a protective barrier for underlying tissues and organ systems in all mammals. It defends against a constant barrage of physical, biological, and chemical insults and as such is frequently damaged. The skin’s remarkable regenerative capability allows it to continually repair itself when damage occurs such that most injuries and wounds to the skin are quickly resolved. Key to this regenerative capability is the ability to mobilize multiple stem cell populations during wound repair. As previously described, epidermal wound healing is characterized by the migration of both HF-SCs and BSCs out of their respective niches towards the wound. Together these two stem cell populations orchestrate the regeneration of epidermal tissue through concentric zones of proliferation and migration/differentiation around the wound

(Figure 1.3). Critical to the wound healing process is cell migration, and as such this chapter focuses on Ift20, primary cilia and keratinocyte migration in wound healing.

4.2 Ift20 is required for polarized migration of cultured primary keratinocytes

To determine whether Ift20 is generally required for epidermal motility, we established

Ift20 conditional knock-out (cKO) cultures of primary mouse keratinocytes (1MKs) and examined them for defects in cell migration (see methods). Briefly, IFT20 fl/fl; Rosa-tdTomato reporter basal stem cells were isolated from mouse epidermis and transduced with a lentiviral

(LV) Cre to establish Ift20 cKO keratinocyte cell lines (Figure 4.1 A). Transduction with LV-Cre induced bright Tomato reporter fluorescence, an indication of Cre-recombinase activity and likely floxing out of the IFT20 allele (Figure 4.1 B). Immunofluorescence (IF) of Ift20, which co-localized to the Golgi apparatus [93] was reduced in Tomato (+) keratinocytes in comparison

65 to neighboring, non-transduced cells (Figure 4.1 B, arrow heads). Ift20 protein was quantitatively reduced in keratinocyte cultures transduced with LV-Cre, in comparison to non-transduced cell lines (Figure 4.1 C). Ift20 cKO and control keratinocytes were subjected to an in vitro scratch wound and monitored for 18 hours to determine the rate of wound closure. Ift20 cKO primary mouse keratinocytes showed markedly delayed cell migration toward an in vitro wound (Figure

4.1 D-E). We conclude that Ift20 plays an essential function during the polarized migration of keratinocytes in response to in vitro wound repair.

Figure 4.1: IFT20 is required for the polarized migration of primary keratinocytes in response to in vitro wound repair. A) Schematic of keratinocyte isolation from IFT20 fl/fl Rosa 26- tdTomato skin and transduction of lentiviral-Cre (LV-Cre), which induced Tomato fluorescence. B) Immunofluorescence images of Ift20 and GM130 in IFT20 fl/fl Rosa-Tomato keratinocytes. Arrowheads point to Cre-transduced Tomato (+) cells that show reduction in Ift20 immuno-labeling. DAPI marks nucleus. Scale bar indicates 10 μm. C) Western blot of Ift20 protein from control (-Cre) or Cre-transduced (+Cre) keratinocytes. Tubulin is loading control. Histogram is quantification of bands shown in above. D) IFT20 fl/fl -Cre (control) or +Cre transduced keratinocytes were subject to scratch assay and migration was monitored for 18 hours. Dotted line indicates wound edge. Scale bar indicates 150 μm. E) Quantification of migration rate in +Cre or -Cre IFT20 fl/fl cells. Data in histogram represents n=5 independent experiments where the area of wound closure was measured after 18 hours of cell migration. * indicates p=0.03 by Student’s t-test. Error bars indicate standard deviation. A.U indicates arbitrary units.

4.3 Ablation of Ift20 does not affect Golgi polarization or overall structure in cultured keratinocytes

Previous studies of Ift20 function during cell migration have implicated it in regulating

Golgi polarization [98, 99]. We hypothesized that the cell migration defects in IFT20 null keratinocytes could similarly be due to defects in the Golgi apparatus. However, in cultured keratinocytes stimulated to migrate into an in vitro wound, polarization of the Golgi apparatus toward the leading edge is not robustly detected even under control conditions, and cKO of

IFT20 does not result in any statistically significant differences in Golgi polarization in comparison to controls (Figure 4.2 A-C). Ift20 has also been implicated in maintaining the structure of the Golgi, and cancer cells depleted of IFT20 via shRNA display a fractured and disorganized Golgi [98]. Quantification of Golgi fragment area per cell showed that cKO of

IFT20 did not significantly affect overall Golgi structure/fragmentation in comparison to control keratinocytes (Figure 4.2 D-E). Taken together, these results suggest that Ift20 is not required for

Golgi polarization or the maintenance of Golgi structure in 1MKs.

Figure 4.2: Ift20 loss does not alter Golgi polarity during cell migration or overall Golgi structure A) Immunofluorescence images showing GM130 (Golgi) and Actin in cells along the leading edge of a wound (dotted line) in both Cre (-) and Cre (+) IFT20 fl/fl keratinocytes. Scale bar is 10 μm. B) Diagram indicating the angular sectors in which Golgi positioning was quantified. Arrow indicates direction of migration. C) Quantification of proportion of Golgi (by area) in each sector in Cre (-) control and Cre (+) IFT20 fl/fl keratinocytes along the leading edge of a wound. Data is quantified form 17 or more leading edges per condition. Error bars indicate standard deviation. Statistical values were computed with a Mann- Whitney U test. D) Immunofluorescence of Ift20 and GM130. Scale bar is 10 μm. E) Quantification of average Golgi fragment size per cell. Data shows n=30 cells per condition.

Error bars are standard deviation. A.U indicates arbitrary units. Statistical tests were computed using a Mann-Whitney U test.

4.4 Loss of Ift20 leads to defects in mechano-chemical signal transduction that is independent of the primary cilium

To understand the Ift20-dependent mechanism that is required for optimal cell migration, we sought to determine whether there is an essential mechano- or chemo-sensory function of

Ift20/cilia required for keratinocyte motility. Previous studies have shown that primary cilia orient toward the wound edge of migrating fibroblasts [118]. It’s been hypothesized that cilia may sense extracellular cues that determine the speed or directionality of cell migration [60]. We analyzed ciliogenesis in 1MKs that have been stimulated to migrate toward an in vitro scratch wound, but <2% of these cells displayed primary cilium under growth conditions that are optimal for cell motility (Figure 4.3, also see [53]). These data suggest that Ift20 plays an extra-ciliary role during polarized migration of 1MKs.

Polarized cell migration requires activation of integrin mediated cell adhesion, which is translated to changes in the phosphorylation status of mechanically sensitive proteins [42]. Focal adhesion kinase, or FAK, is an adhesion-activated scaffolding protein that resides at sites of activated integrin signaling, the FA [119]. Auto-phosphorylation of FAK at Y397 is a sensitive readout of mechano-transduction via integrin-mediated adhesion signaling [120]. Interestingly, protein lysates from Ift20 cKO keratinocytes show reduced phosphorylation of Y397 FAK, indicating a possible defect in adhesion-mediated FAK activation (Figure 4.4 A-B). To determine whether there are any defects in cell adhesion, Ift20 cKO keratinocytes were stained with Vinculin, a resident FA protein. By IF, FAs appeared smaller and less numerous in Ift20 cKO cells in comparison to control keratinocytes (Figure 4.4 C). Quantification of total FA area per cell demonstrated a 2-3 fold decrease in the absence of IFT20 (Figure 4.4 D). FAs were also quantitatively smaller, as the average FA area/cell was significantly reduced in Ift20 cKO

71 keratinocytes in comparison to non-transduced control cells (Figure 4.4 E). The difference in average FA size cannot be attributed to a difference in cellular area, as cellular area remained unchanged between Cre (-) and Cre (+) Ift20 cKO keratinocytes (Figure 4.4 F). Taken together, these results suggest that the cell migration defects we observed could be a consequence of defective integrin mediated mechano-transduction at sites of cell-substrate contact, the FA.

Figure 4.3: Cultured primary keratinocytes are rarely ciliated during directional cell migration. Immunofluorescence images of Cre (-) control (left and center images) and Cre (+) (right image) IFT20 fl/fl keratinocytes. Arrow points to rare ciliated keratinocyte among Cre (-) control keratinocytes in left image. Primary cilia are labeled with Arl13b and nuclei are marked with DAPI. Center image shows magnified inset of Cre (-) control keratinocyte with primary cilia. Right image shows Cre (+) IFT20 fl/fl keratinocyte lacking primary cilia. Scale bars are 50 μm in the left image and 10 μm in the center and right images. Dotted line indicates wound edge.

Figure 4.4: Ift20 loss results in defective mechano-chemical signaling downstream of integrin engagement. A) Western blot of pFAK Y397 protein levels from -Cre control versus +Cre IFT20 fl/fl keratinocyte lysate. Tubulin shown as loading control. B) Quantification of pFAK Y397 signal from bands shown in (A). C) Immuno-labeling for Vinculin, Tyrosinated Tubulin, and nuclei (DAPI) in +Cre versus -Cre IFT20 fl/fl keratinocytes. Scale bar indicates 10 μm. D-E) Quantification of the total focal adhesion (FA) area/cell or average FA Area/Cell in IFT20 fl/fl keratinocytes. Histograms represent data from 2 independent experiments where FA size was measured in 50 or greater cells in each condition. Statistical values were computed using a Mann-Whitney U Test. F) Quantification of total cell area in both -Cre and +Cre keratinocytes. Statistical values were computed using a Mann-Whitney U Test.

4.5 Ift20 is required for focal adhesion reformation after microtubule-induced FA disassembly.

Since Ift20 is a microtubule-associated protein [121], we hypothesized that Ift20 may be required for focal adhesion disassembly induced by microtubule targeting. To test this hypothesis, a Nocodazole (NZ) washout assay to study microtubule (MT)-induced FA turnover was performed in Ift20 cKO versus control keratinocytes [122]. In this experiment, keratinocytes are treated with 10 μM Nocodazole for 3-4 hours to completely depolymerize the microtubule

(MT) cytoskeleton and induce Rho-dependent activation of contractility and subsequent FA formation. This is followed by a washout of NZ that induces MT polymerization, regrowth, and targeting to FAs to induce their disassembly. After 3-4 hours treatment with NZ, both control

Cre (-) and Ift20 cKO keratinocytes formed focal adhesions, ascertained by immuno-labeling for

Vinculin (Figure 4.5 A, 0 min timepoint). Consistent with reduced pFAK phosphorylation on

Y397, Ift20 cKO keratinocytes treated with NZ displayed FAs that were slightly less numerous, and smaller in size than those observed in control cells (Figure 4.5 B, C). However, a time course of NZ washout showed that FAs from Ift20 cKO cells still disassembled after 15-30 minutes of

MT regrowth during recovery from NZ (Figure 4.5 A-C). We note that this time course of FA disassembly is slightly faster than what we observed in NIH3T3 fibroblasts, and likely a consequence of reduced serum starvation conditions (see methods and [122]). The overall structure and organization of MTs appeared normal in Ift20 cKO interphase cells that were not treated with NZ (Figure 4.4 C). MT regrowth to the periphery of the cell membrane and focal adhesion targeting was still observed in the absence of Ift20 function (time course, Figure 4.5 D), and MTs still targeted toward focal adhesions in the absence of Ift20 (Figure 4.5 E). These observations suggest that MT-induced FA disassembly does not require Ift20.

Next, we evaluated the effect of IFT20 cKO at further timepoints after MT recovery post-

Nocodazole treatment, when FAs reform via a process that requires integrin recycling [48].

After 60-90 minutes of MT regrowth, FAs in control primary keratinocytes gradually reform toward the cell periphery (Figure 4.5 A, quantified in B and C), a phenomenon that has been extensively studied in fibroblasts and reported in other epithelial cell types [48]. However, FAs from Ift20 cKO cells fail to reform with the same kinetics, and both the number and size of FAs remain significantly reduced after 90 minutes of MT regrowth in the IFT20 cKO cells in comparison to control cells (Figure 4.5 A, quantified in B and C). These data suggest that Ift20 is required for FA reformation after MT-induced FA disassembly.

Figure 4.5: Ift20 is required for focal adhesion reformation after Nocodazole washout and microtubule regrowth A) Immunofluorescence images of Vinculin during time course of Nocodazole washout and microtubule regrowth (min indicates minutes) in – Cre control versus + Cre IFT20 fl/fl keratinocytes. DAPI marks nuclei. B-C) Quantification of total focal adhesion (FA) Area/Cell and Total FA Number/Cell during time course of microtubule regrowth in - Cre control versus + Cre IFT20 fl/fl keratinocytes. Data in histograms represents n=3 independent experiments where 63-127 cells were measured at each time point and each condition. *** indicates p<5.5e-15 via Mann-Whitney U test. D) Immunofluorescence images of Tyrosinated Tubulin during time course of Nocodazole washout and microtubule regrowth (min indicates minutes) in - Cre control versus + Cre IFT20 fl/fl keratinocytes. DAPI marks nuclei. White indicates + Cre Tomato signal. Note that microtubule regrowth occurs normally and toward the periphery in + Cre cells. E) Immunofluorescence images of - Cre control and + Cre IFT20 fl/fl keratinocytes 5 minutes after Nocodazole washout. Note that microtubules target focal adhesions during the focal adhesion disassembly process. Scale bars in both A) D) and E) denote 10 μm.

4.6 Integrin surface expression is altered when Ift20 function is ablated

Previous studies have shown that FA reformation after MT-induced FA disassembly is driven by integrin recycling to the plasma membrane. This process requires FAK, Src kinases,

PIPKIγ and the Rab5/Rab11 endosomal system [48]. We sought to determine whether integrin trafficking or recycling is perturbed in the absence of Ift20, specifically whether focal adhesion- associated β1 integrin is properly recycled back to the plasma membrane. First, we determined whether β1 integrin could be detected within the focal adhesions of cultured 1MKs. IF of β1 integrin shows convincing co-localization with Vinculin, a resident focal adhesion associated protein, in both untreated and NZ-incubated primary mouse keratinocytes (Figure 4.6 A).

Induction of MT-induced FA disassembly and FA reformation showed that the return of FA- associated β1 integrin is impaired when Ift20 is removed (Figure 4.6 B-C: 90 minute time point).

Next, we used the same antibody for FACS analysis of β1 integrin surface levels during FA disassembly and reformation during the NZ washout assay. Similar to cultured fibroblasts [48], surface levels of β1 integrin in primary keratinocytes decrease after 30 minutes of MT regrowth to induce FA disassembly (Figure 4.6 D). After 90-120 minutes of MT regrowth and FA reformation, β1 integrin surface levels increase significantly in control cells (Figure 4.6 D). This data is consistent with that shown in fibroblasts: surface integrin is removed from FAs via endocytosis during MT regrowth, and recycled back to the plasma membrane into newly forming peripheral FAs (Figure 4.6 D) [48]. In Ift20 cKO keratinocytes, integrin surface levels decrease at 30 minutes but do not return to the same extent as control cells after 90 minutes of MT regrowth (Figure 4.6 D). These results strongly suggest that Ift20 is required for the polarized recycling of FA-associated β1 integrin back to the cell surface.

Figure 4.6: Surface levels of recycled β1 integrin, previously present in FAs, are decreased upon IFT20 knock out. A) Immunofluorescence images of Vinculin (red) and β1 Integrin (green) in -Cre control versus +Cre IFT20 fl/fl keratinocytes treated with Nocodazole (NZ) for 3-4 hours or left untreated. Boxed regions are shown magnified in corner. Merged image with DAPI (nuclei) shows co-localization of β1 integrin and Vinculin signals. Scale bars indicate 10 μm. B) Representative images of β1 integrin immunofluorescence at various timepoints following Nocodazole washout (min indicates minutes). DAPI indicates nuclei. Scale bar indicates 10 μm. C) Quantification of β1 integrin immunofluorescence at focal adhesions at various time points following Nocodazole washout (min indicates minutes) in -Cre control and +Cre IFT20 fl/fl keratinocytes. Data in histogram represents n=60-81 cells per time point per condition from 2 independent experiments. Statistical values were computed using a Mann-Whitney U test. D) Normalized flow cytometry measurements of surface β1 integrin following Nocodazole washout. Data is from n=5 independent experiments with greater than 5000 cells measured per condition at each time point. Error bars indicate standard deviation. Statistical values were computed using a Student’s t-Test.

4.7 Ift20 maintains its association with the Golgi apparatus during focal adhesion reformation after MT-induced FA disassembly

In addition to its essential role in ciliogenesis, Ift20 functions at the Golgi apparatus where it contributes to the polarized trafficking of ciliary components toward the basal body

[93]. In non-ciliated T-cells, Ift20 is also required for Rab5 and Rab11-dependent recycling of T- cell Receptors to the immunological synapse independent of primary cilia [100-102].

Interestingly, integrin trafficking after MT-induced FA disassembly also requires Rab5/Rab11- dependent transit through the endosomal system [48]. A Golgi-retromer dependent pathway of integrin recycling has also recently been identified, although the reliance of this pathway on microtubules is less clear [123]. To decipher a role for Ift20 function via Rab5/Rab11-dependent versus Golgi/retromer trafficking mechanism(s), we began by determining the localization of

Ift20 in cultured keratinocytes during MT-induced FA disassembly and polarized FA reformation.

As previously reported, Ift20 localizes to the primary cilia (Figure 4.7) and the Golgi apparatus (Figure 4.1 B, Figure 4.2 D) in cultured primary keratinocytes [93]. During MT- induced focal adhesion disassembly and reformation, the Golgi is dispersed upon MT- depolymerization and reorganized upon NZ washout and MT-regrowth after recovery from NZ

(Figure 4.8 A). For the duration of the NZ washout and MT regrowth, Ift20 maintains co- localization with GM130 (a marker of the Golgi apparatus) (Figure 4.8 A). In Ift20 cKO keratinocytes, Golgi dispersion upon NZ treatment and reassembly during MT recovery both appear to be unaffected (Figure 4.8 B). We conclude that Ift20 maintains its association with the

Golgi apparatus during MT-induced FA disassembly and reformation.

Figure 4.7: Ift20 can be detected at primary cilia in keratinocytes. Acetylated tubulin (AcTub) labels primary cilia. Boxed region shown magnified in lower left corner. Scale bar indicates 10 μm.

Figure 4.8: Ift20 dependent focal adhesion reformation is independent of its localization or function at the Golgi. A-B) Immunofluorescence images of GM130, and Ift20 with DAPI marking nuclei of -Cre control (A) and +Cre (B) IFT20 fl/fl keratinocytes during microtubule induced focal adhesion disassembly and reformation at various timepoints after Nocodazole washout (shown in minutes). Scale bars indicate 10 μm. C) Immunofluorescence images of GM130 and Ift20 in control keratinocytes treated with Brefeldin A (BFA) or loading control (ETOH) for 3 hours. DAPI marks nuclei. Scale bar indicates 10 μm. D) Immunofluorescence images of Vinculin in control keratinocytes during various timepoints following Nocodazole washout in the presence of BFA or loading control (ETOH) after prior pretreatment with BFA or ETOH. DAPI indicates nuclei. Scale bars indicates 10 μm. E) Quantification of total focal adhesion (FA) area/cell in control keratinocytes during time course following Nocodazole washout in the presence of Brefeldin A (BFA) or loading control (ETOH) after prior pretreatment of cells with BFA or ETOH. Data obtained from n=40 cells per time point per condition from 2 independent experiments. Statistical values were computed using a Mann- Whitney U test.

4.8 Ift20-dependent FA reformation does not require Golgi function

Previous studies using NIH3T3 fibroblasts have shown that FA disassembly and reformation induced by MT regrowth after NZ washout does not require trafficking from the

Golgi apparatus [48]. Given Ift20’s known function(s) at the Golgi, as well as its localization at the Golgi during MT regrowth and FA disassembly/reformation, we questioned whether the

Ift20-dependent process of integrin recycling to promote FA reformation might proceed via a different, Golgi-dependent mechanism in our 1MKs such as the previously reported Retromer pathway [123]. To test the Golgi-dependence of Ift20 in regulating mechano-transduction at FAs, we treated keratinocytes with Brefeldin A (BFA) for 3 hours to disrupt Golgi structure and function. As expected, treatment with BFA fragmented the Golgi in comparison to controls

(Figure 4.8 C), and Ift20 maintained its association with the disrupted Golgi fragments as previously reported (Figure 4.8 C and [93]). To test the Golgi-dependence of Ift20 in regulating

FA disassembly and reformation, 1MKs were incubated for 3 hours with BFA and NZ prior to washing out the NZ and inducing FA disassembly and reformation in the presence of BFA.

Neither FA disassembly nor FA reformation after MT regrowth were affected by BFA treatment in 1MKs (Figure 4.8 D-E). We conclude that the Ift20 dependent defects in integrin trafficking and mechano-transduction at FAs (as measured using the FA turnover assay) are independent of its localization and function at the Golgi.

4.9 Ift20 is required to traffic β1 integrin through Rab5 endosomes

We next sought to determine whether Ift20 functioned in tandem with the Rab5 endosomal system to traffic β1 integrin during FA turnover. We focused our analysis at 0, 30, and 90 minutes of MT regrowth following NZ washout as the 30 and 90-minute timepoints represent timepoints of maximal FA disassembly and maximal FA reformation, respectively

(Figure 4.5 A-C). At 30 minutes of MT regrowth, we observe increased co-localization of Ift20 with Rab5 compared to 0 minutes in control keratinocytes (Figure 4.9 A, B). In particular, at the

30 minute timepoint, we can observe distinct puncta of Ift20 co-localizing with what appear to be small Rab5 (+) endosomal puncta (Figure 4.9 A, B). Following 90 minutes of MT regrowth, we observe sustained co-localization of Ift20 with Rab5, however the co-localization at 90 minutes occurs predominantly at larger Rab5 (+) compartments instead of the smaller Rab5 (+) endosomal puncta we observe at 30 minutes (Figure 4.9 A, B). Quantification of Ift20 and Rab5 co-localization during this time course confirms our observations, mainly that co-localization of

Ift20 with Rab5 increases at 30 minutes of MT regrowth and that this increase is sustained through 90 minutes of MT regrowth (Figure 4.9 C). We next determined the fraction of β1 integrin within Rab5 (+) endosomes by calculating the co-localization coefficient of Rab5 and β1 integrin staining at 0, 30, and 90 minutes of MT regrowth. As previously reported, β1 integrin co-localizes with peripheral Rab5 puncta after 30 minutes of MT-regrowth during recovery from

NZ treatment, and we observe a significant increase in the co-localization coefficient compared to 0 minutes (Figure 4.9 D-F) [48]. After 90 minutes of MT regrowth and FA reformation in control cells, fewer Rab5 (+) endosomes containing integrin are detected and the co-localization coefficient significantly decreases compared to the 30 minute time point (Figure 4.9 D-F). In

IFT20 fl//fl 1MKs transduced with Cre to ablate Ift20 function, we observe a significant increase of β1 Integrin co-localizing with peripheral Rab5 puncta at the 30 minute time point but importantly there is no significant decrease in co-localization between β1 Integrin and Rab5 puncta at the 90 minute time point (Figure 4.9 D-F). In fact, there is a statistically significant increase in the amount of internalized β1 integrin co-localizing with Rab5 after 90 minutes in

Ift20 cKO cells compared to control (Figure 4.9 F). Taken together, because we observe an

87 increased co-localization of Rab5 (+) endosomes with β1 Integrin after 30 minutes of MT regrowth that persists to 90 minutes only in Ift20 cKO cells, these data suggest that Ift20 is required for the transit of β1 integrin out of Rab5 (+) endosomes. Our in vitro data is consistent with the following model: in the absence of Ift20 function, integrin that has been endocytosed from disassembling FAs is not transferred out from Rab5 (+) endosomes and fails to be recycled back to the cell surface. This defective trafficking deleteriously affects FA turnover, mechano- transduction and overall cell migration efficiency.

Figure 4.9: Ift20 is required to transit β1 integrin through Rab5(+) endosomes during MT- induced FA turnover. A) Immunofluorescence images of Ift20 and Rab5A at indicated timepoints following Nocodazole washout. Scale bar indicates 10 μm and Min indicates minutes. Boxed regions are magnified in B). B) Magnified images of boxed regions shown in A). Scale bar indicates 5 μm and Min indicates minutes. Arrowheads indicate distinct puncta showing co-localization of Rab5 and Ift20. C) Quantification of co-localization of Rab5 and Ift20 as measured by a co-localization coefficient. Statistical values were computed using a Mann-Whitney U Test. Data represents more than 60 cells per condition obtained from 2 independent experiments. D) Immunofluorescence micrographs of -Cre control or +Cre IFT20 fl/fl keratinocytes at various timepoints during focal adhesion turnover (min indicates minutes) showing the localization of β1 integrin and Rab5 endosomal compartments. Boxed regions are shown magnified in E). Scale bar indicates 10 μm. E) Magnification of Rab5 endosomes and β1 integrin puncta. Arrow heads at 90 minutes point to co-localization of β1 integrin/Rab5 puncta in + Cre IFT20 fl/fl cKO cells. Scale bar indicates 5 μm. F) The co- localization coefficient between Rab5 and β1 integrin was calculated at various timepoints of FA disassembly/MT regrowth (min indicates minutes), data in plot represents 60-81 cells from 2 independent experiments. P values calculated using Mann-Whitney U test.

4.10 Ift20 is required for the in vivo mobilization of epidermal stem cells in response to epidermal injury

Ift20 function has previously been implicated in regulating the cell invasiveness of colon cancer cell lines in vitro [98, 99]. To study Ift20 function during cellular invasion and migration in a physiologically relevant model system, we turned to the tissue-resident stem cells of the epidermal hair follicle. Hair follicle stem cells (HF-SCs) transiently contribute to tissue regeneration during epidermal repair, when they are activated to leave their niche, invade the surrounding tissue and migrate toward the wounded epidermis [12, 124]. In order to efficiently invade and migrate, HF-SCs must sense mechano-chemical changes from the wound microenvironment and transduce these cues to the cell’s migration machinery [11]. Given our in vitro migration data, we hypothesized that Ift20 may play a conserved role in polarized migration and invasion of HF-SCs in vivo.

To study the function of Ift20 exclusively in HF-SCs we generated an inducible cKO mouse model where IFT20 could be specifically ablated in the HF-SC bulge compartment [125].

In particular, we mated K15-CrePGR mice to IFT20 fl/fl mice [126]. K15-Cre drives recombination specifically in HF-SCs, and these mice were maintained on a Rosa26-Tomato reporter background to monitor Cre-activation and for lineage tracing (Figure 4.10 A). Telogen- staged mice were treated topically with 1% RU-486 to induce K15-Cre-mediated recombination specifically in HF-SCs (Figure 4.10 B), which induced a strong Tomato signal in 80-90% of HFs

(Figure 4.10 C). Ift20 signal was uniformly reduced in Tomato (+) IFT20 cKO HF-SCs, but not cells of the interfollicular epidermis (IFE), indicating that K15-Cre-mediated floxing of IFT20 occurred specifically in the bulge stem cell compartment (Figure 4.10 D). High magnification imaging of telogen staged mouse skin reveals that Ift20 is peri-nuclear in its localization within

91 keratinocytes of the HF (Figure 4.11). Ciliogenesis was quantitatively inhibited, as expected, in the Tomato (+) HF-SCs (Figure 4.10 E). qRT-PCR from FACS isolated HF-SCs demonstrated a significant reduction in IFT20 expression (Figure 4.10 F-G). These data show that Ift20 expression and ciliogenesis can be conditionally ablated specifically in HF-SCs.

To test whether Ift20 is required for HF-SC migration in response to epidermal injury these mice were subject to a full-thickness epidermal punch biopsy and lineage tracing was used to measure the mobilization of cKO HF-SCs to the epidermis after 4 or 7 days of wound healing

(Figure 4.10 H-I and Figure 4.12). In IFT20 +/fl (essentially WT) mice, Tomato (+) cells were observed both in the wound-adjacent epidermis and in the wound bed, where they proliferated into epidermal clones (Figure 4.10 H-I). In contrast, in IFT20 cKO (IFT20 fl/fl) mice, Tomato

(+) cells were more rarely detected in the epidermis and significantly reduced in the wound bed

(Figure 4.10 H-I). Quantitative analysis demonstrated that loss of Ift20 impaired the mobilization of Tomato (+) HF-SCs to the injured epidermis after 4 days of wound healing (Figure 4.10 K).

Similar results are observed at 7 days of wound healing although with greater variation (Figure

4.12). These data suggest that Ift20 is required for the transient mobilization of HF-SCs to the epidermis and their clonal contribution to epidermal repair.

Figure 4.10: Ift20 is required for the polarized migration and invasion of hair follicle-derived stem cells and their clonal contribution to wound repair. A) Schematic of lineage tracing strategy showing the localization of K15-Cre Tomato (+) hair follicle stem cells (HF-SCs) in the bulge (Bu). Note that the Hair Germ (HG) and Transit Amplifying Cells (TAC) are derived from HF-SCs. B) Schematic of strategy used for RU486 treatment and Cre induction prior to wound healing and epithelial repair (see methods). C) Immunofluorescence images of Tomato (+) signal (red) labeling the HF-SCs of the bulge (Bu) and their transiently amplifying progenitors in the Hair Germ (HG). D) Immunofluorescence images of Ift20 in HF-SCs (red) of control (IFT20 +/fl) versus IFT20 cKO (IFT20 fl/fl) telogen staged mouse skin. E) Quantification of the number of ciliated HF-SCs in control (IFT20 +/fl) versus IFT20 cKO (IFT20 fl/fl) skin. Data in histogram represents 20-30 hair follicles from n=3 mice. F) Scatter plot from FACS purification of CD34 (high)/α6 integrin (high)/Tomato (+) HF-SCs G) qRT- PCR of IFT20 using mRNA isolated from FACS-purified HF-SCs from control (IFT20 +/fl) versus IFT20 cKO (IFT20 fl/fl) mouse skin. H) Sagittal section of skin isolated from control (IFT20 +/fl) Rosa Tomato (+) mouse subjected to a full-thickness punch biopsy wound. Immunofluorescence image shows HF-SCs (red) migrating toward, and contributing to, the epidermal wound bed. Arrows indicate direction of migration. I) Sagittal sections from control (IFT20 +/fl) or IFT20 cKO (IFT20 fl/fl) wounded epidermis demonstrating migrated HF-SCs (red). Keratin 10 (K10) labels suprabasal layer (green) and DAPI-labeled nuclei. Boxed region is magnified at right and shows examples of epidermal clones from control versus IFT20 cKO epidermis. K) Quantification of the number of migrated Rosa-Tomato (+) HFSCs and their contribution to basal or suprabasal layers of the wounded epidermis. Non- wounded epidermis (hatched bars) was used as a control. Histogram represents average data from 5 regions surrounding a wound from n=1 mouse per condition. Error bars are standard deviation. Statistical values were computed using a Student’s t-Test. L) Images of explanted skin from control (IFT20 +/fl) and IFT20 cKO (IFT20 fl/fl) mice 14 days after explant. Tomato fluorescent (red) cells are HF-SCs which have migrated. Scale bar is 167 μm. Dotted line demarcates border of explant. M) Quantification of HF-SC migration in explants. Data is from n=3 independent samples for each condition. * indicates p=0.013 by Student’s t-Test. Dotted line denotes dermal-epidermal border C), D), H) and I) and DAPI marks nuclei in C), D), H), I) and L).

Figure 4.11: Telogen staged mouse skin immuno-labeled for Ift20. Dotted line demarcates dermal-epidermal junction. Scale bar is 20 μm. Boxed region is magnified at right. Note the peri-nuclear staining of Ift20 (arrow heads).

Figure 4.12: Quantification of the number of migrated Rosa-Tomato (+) HF-SCs in basal and suprabasal epidermal layers 7 days post wounding. Non-wounded epidermis (hatched bars) was used as a control. Histogram represents average data from 6 regions surrounding a wound from n=1 mouse per condition. Error bars are standard deviation. Statistical values were computed using a Student’s t-Test.

4.11 Ift20 is required for stem cell migration from skin explants taken from wound biopsies

To test whether Ift20 is required for polarized cell migration of HF-SCs from skin explants, IFT20 cKO/Rosa-Tomato explants and IFT20 +/fl/Rosa-Tomato controls were grown from biopsied mouse skin isolated from in vivo wound healing studies. After 14 days of culture,

IFT20 +/fl Tomato (+) HF-SCs migrated out of the bulge cell niche to the surface of the epithelial explant whereas IFT20 cKO Tomato (+) HF-SCs are rarely observed migrating to the surface under these ex vivo conditions (Figure 4.10 L). These data demonstrate that Ift20 is required to mobilize SCs from their HF niche during injury repair, and suggests that Ift20- dependent trafficking of adhesion receptors may modulate the regenerative response and invasion of HF-SCs toward epidermal injury.

4.12 Discussion

Polarized cell migration is essential for embryonic patterning and development, drives tissue regeneration in response to wound repair, and contributes to numerous pathological conditions. Human ciliopathies, which are caused by a wide variety of defects in the primary cilium, display features associated with disrupted cell migration [127]. Defects in the formation or function of primary cilia can lead to migration-related disorders in mice, such as defects in skin wound healing and repair of the corneal epithelium [128]. However, whether defects in directional cell migration are a direct function of defective signaling through the primary cilium remains unclear.

In order to efficiently invade and migrate, epithelial cells must sense mechano-chemical changes from the wound microenvironment and transduce these cues to the cell’s migration machinery. The primary cilium is a microtubule-based cellular “antenna” that can detect

97 mechanical and chemical changes in the extracellular environment; and many of the signaling pathways that require primary cilia for their transduction (Shh, Notch, Wnt) are also essential for cell migration [129]. Previous studies have shown that primary cilia orient toward the wound edge during polarized cell migration of cultured fibroblasts [118], suggesting that the cilium could act as a mechano-sensor that determines the direction of cell migration [128]. This could be particularly important in vivo, where changes in extracellular matrix deposition or altered topology must be sensed as cells invade and migrate in response to regenerative signaling. How ciliary functions/signaling integrate with the cell migration machinery is not well understood, and whether IFTs also function in a non-ciliary context to regulate cell migration and invasion is an emerging area of investigation.

Data from this study indicate that Ift20 functions in an extra-ciliary role to drive cell migration by modulating FA dynamics and integrin trafficking. In contrast with previously reported results in fibroblasts [118], we find that the majority of wild type cultured primary keratinocytes lack a primary cilium during directed migration (Figure 4.3) suggesting that keratinocyte migration may not be hindered by loss of primary cilia to the extent of fibroblast migration. This difference in ciliation status may be a result of different cell types requiring different cues for cell migration. Fibroblasts have been shown to depend on primary cilia to sense chemical cues such as PDGF-AA during cell migration, as their primary cilia are embedded with a variety of receptors including PDGF receptor alpha (PDGFRα) [59, 61, 130].

Interestingly, keratinocytes are a major source of cutaneous PDGF, especially during wound healing, but they themselves likely lack PDGF receptors [131] suggesting a difference in migration cues between keratinocytes and fibroblasts. This difference in cues highlights possible differences in the extent by which primary cilia regulate migration between different cell types.

Interestingly, several studies have previously reported extra-ciliary functions for various

IFT proteins in driving aspects of collective cell migration, in particular a role for Ift20 in non- ciliated cancer cells [98, 99]. In these studies, it has been suggested that loss of Ift20 produces defects in the polarization of the Golgi apparatus or integrity of Golgi organization, that are at the root of Ift20-dependent defects in cell motility. More specifically, loss of Ift20 can disrupt polarization of the Golgi complex and subsequent organization of Golgi-associated microtubules during cell migration [99] or disrupt Golgi integrity and subsequent Golgi-dependent trafficking of a membrane receptor crucial for cancer cell invasion [98].

In keratinocytes, we find that Golgi structure and integrity as well as Golgi polarity during migration are not significantly altered following loss of Ift20 (Figure 4.2). Unlike cancer cells, in which the Golgi complex fragments subsequent to loss of Ift20 [98] the Golgi complex of keratinocytes appears peri-nuclear and non-compact even in wildtype keratinocytes, and there is no significant increase in Golgi fragmentation in IFT20-null keratinocytes (Figure 4.2 D, E).

Furthermore, the polarity of the Golgi complex in keratinocytes is not significantly altered in

IFT20-null keratinocytes compared to wild type during cell migration (Figure 4.2 A-C). Taken together, these findings suggest that Golgi fragmentation and loss of Golgi polarity does not fully elucidate the migration defect we observe in keratinocytes following Ift20 ablation. These differences in Golgi compactness and structure between keratinocytes and cancer cells may be simply intrinsic differences between the cell types. Other studies have reported similar findings with regard to the peri-nuclear scattering and relative non-compactness of the Golgi complex in human keratinocytes [132] [133].

Our data reveals a novel role for Ift20 in modulating mechano-transduction and FA turnover during keratinocyte migration. We find that IFT20 null keratinocytes demonstrate

99 smaller and fewer focal adhesions with reduced activation of FAK compared to wild type controls. Using NZ treatment followed by washout to induce MT regrowth we were able to interrogate the FA disassembly and reassembly processes during focal adhesion turnover. We find that FA reformation but not disassembly is impaired in IFT20 null keratinocytes. FA turnover (disassembly followed by reassembly of focal adhesions) has previously been characterized to depend on MTs and the recycling of active integrins together with FAK and Src kinases through Rab5 and Rab11 endosomal compartments [46, 48, 122]. Loss-of-function or dominant-negative mutations in Rab5 or Rab11 inhibit reassembly (reformation) but not disassembly of FAs [48]. Src kinase inhibitors have also been found to inhibit reassembly but not disassembly of focal adhesions [48].

In non-ciliated T-cells, Ift20 has been reported to regulate polarized trafficking of cell surface receptors at the immune synapse [100] [101]. This previous work demonstrated that Ift20 interacts with Rab5 and Rab11 to orchestrate the polarized recycling of surface T-cell receptors

(TCRs) at the immune synapse (IS) [134, 135]. Loss of Ift20 results in entrapment of TCRs in

Rab5 endosomes [101], similar to what we observe with β1 integrin during focal adhesion turnover. Likewise, the IS and FAs are also homologous in nature: both are formed via clustering of cell surface receptors at sites of cell adhesion, where accessory proteins and cytoskeletal components (such as actin) are recruited in response to mechano-transduction at the cell membrane. We postulate that Ift20 plays an analogous role in orchestrating the polarized recycling of an adhesion receptor (β1 integrin) in order to control assembly and mechano- transduction at FAs.

Consistent with this hypothesis, we find that loss of Ift20 reduces β1 integrin levels at focal adhesions, a finding in line with the smaller and fewer focal adhesions observed in these

100

IFT20-null cells (Figure 4.4). Furthermore, our data suggests that like fibroblasts, keratinocyte

FA turnover is dependent on recycling of integrins through a Rab5 early endosomal compartment and that Ift20 can co-localize at these compartments during this recycling process

(Figure 4.9). Quantitative microscopy analyses showed that β1 integrin becomes entrapped in

Rab5 (+) endosomes during focal adhesion turnover in IFT20-null keratinocytes (Figure 4.9).

Surface labeling experiments demonstrated that IFT20-null keratinocytes do not show the expected significant increase in surface β1 integrin as compared to control during the FA reassembly process (Figure 4.6 D). We conclude that Ift20 drives integrin trafficking through the early endosome en route to a recycling pathway required for subsequent integrin engagement into newly reforming focal adhesions. Further investigation is warranted to determine if integrins are recycled in an active conformation together with FAK and Src kinases during focal adhesion turnover [48] and what mechanistic role Ift20 might play in this process. To our knowledge this is the first study to specifically implicate Ift20 in integrin trafficking and signaling, although loss of primary cilia through knockout of Kif3a has been previously shown to depress FAK phosphorylation—a biochemical readout of integrin and adhesion mediated mechano- transduction—in Chondrocytes during bone development [136]. FAK has recently been shown to regulate activation of the Rab5 GTPase cycle at early endosomes [137]. It’s interesting to speculate that Ift20 may orchestrate integrin transit from early endosomes via a FAK/Rab5

GTPase dependent mechanism.

Consistent with our in vitro data, we observe both an in-vivo and ex-vivo migration defect when IFT20 is conditionally ablated in HF-SCs of the epidermis during wound healing. Wound healing is a complex process in which HF-SCs initiate repair of the wound by migrating towards the wound bed in an initial re-epithelialization of the wound; after which epidermal cells derived

101 from IFE complete the re-epithelialization process [12, 124]. It has been suggested that this initial re-epithelialization while not necessary for complete wound repair does accelerate wound healing [40]. As such further study into Ift20 and other cellular mechanisms that govern the HF-

SC migratory response is necessary. The extracellular signals that mobilize stem cells have not been well-established [4], but may involve Wnt signaling through GSK3 and ACF7 [138], and genomic regulation by SOX9 and its downstream target genes [139]. Interestingly, IFT57, which interacts with IFT20, was identified as a bona fide HF-SC signature gene regulated by SOX9

[32]. Further investigation into the ciliation status of these migrating HF-SCs in vivo and in explanted skin is warranted, but because we observe limited primary cilia in migrating cells in vivo (data not shown) we speculate that these migration defects in vivo may not be simply explained by loss of primary cilia. We propose, given our in vitro results, that the reduced migration and invasion observed in IFT20 cKO HF-SCs in vivo and in explanted skin results from defects in integrin recycling and mechano-transduction in the absence of an extra ciliary

Ift20 function.

Overall, we have demonstrated a novel role for Ift20 in regulating polarized cell migration and epidermal wound repair. Our data suggests that Ift20 regulates cell migration by governing FA turnover and integrin recycling in keratinocytes. These functions of Ift20 appear to be extra-ciliary and instead indicate a role for Ift20 in cellular mechano-transduction and cell motility through regulating surface integrin levels and focal adhesion dynamics. Ift20, and perhaps other ciliary or regulatory proteins, might be harnessed in order to promote stem cell migration in order to speed wound healing during tissue regeneration.

102

Conclusion

Mammalian skin is a dynamic organ. Often forgotten and overlooked, mammalian skin is actually a complex organ system consisting of multiple types of tissues and a diverse array of cell types. From mediating tactile sensory to releasing sweat for thermoregulation to generating melanin for skin color, the functions of mammalian skin are plentiful and crucial for human existence. Among its most important abilities is its ability to self-renew during homeostasis and regenerate following injury. These two remarkable properties distinguish skin from other organs and enables it to protect underlying organs and tissues within mammals. The ability to self-renew and regenerate highlights the importance of the multiple stem cell populations within skin.

In this work we have focused on the epidermal component of skin with a particular interest in HF-SCs. Specifically, we have examined the role of primary cilia and Ift20 in mouse epidermis during development, homeostasis, and in wound healing. Utilizing multiple mouse models, we have carefully characterized primary cilia of keratinocytes during development and postnatally, in adult mouse skin. We have also used generated mouse models to specifically ablate IFT20 and subsequently primary cilia in HF-SCs of adult skin and examined the consequent phenotypes. Finally, we investigated the role of Ift20 and primary cilia in keratinocyte migration and epidermal wound healing using both in vitro and in vivo models.

Using a transgenic mouse line with fluorescent markers for primary cilia and basal bodies we found that the epidermis (both the IFE and HFs) are highly ciliated during development as well as postnatally, in adult epidermis. We found that epidermal primary cilia length remains rather consistent through development and in postnatal, adult skin. Interestingly, we did observe a significant proportion of epidermal cilia with bulged-tip morphologies. Subsequent to this finding, we also demonstrated for the first time in live, developing epidermis that epidermal

103 primary cilia undergo ectocytosis, a phenomenon previously only observed in vitro, under strict serum induced conditions.

This finding is novel as it establishes ectocytosis as a possible physiological mechanism by which cells can possibly regulate ciliary disassembly, ciliary signaling, and/or cell cycle in tissues which are all functions of ectocytosis that have been established in vitro. Of course, further investigation into in vivo ectocytosis is necessary to determine if these functions translate from in vitro to in vivo models. Of particular interest, is the contents and final destination of the ectosomes. As previously described, ectosomes in vitro have been found to be enriched with

IFTB proteins, IFTA proteins, ciliary proteins like Arl13b and Kinesin subunits, as well as ciliary receptors specific to the cell type. It would not be unreasonable to speculate that the ectosomes of epidermal primary cilia contain all of these elements in addition to Notch 3 and

Presenilin 2 which are keratinocyte-specific receptors and effectors known to localize within epidermal primary cilia [53]. Further studies are necessary to study Notch 3 receptor and

Presenilin 2 exit from mammalian primary cilia. Are these Notch components similar to certain

GPCRs in that they cannot be recycled through retrograde IFT transport and are dependent on ectocytosis for ciliary exit? As previously mentioned, ciliary entry of Notch 1 receptor into primary cilia of RPE cells is mediated by basal body proteins (like BBS1 and BBS4) together with endosomal trafficking, however the mechanism by which Notch1 receptor (or Notch 3 in epidermal primary cilia) exits primary cilia is unknown [65]. Furthermore, given the physical proximity of epidermal primary cilia with neighboring keratinocytes such that they can actually invaginate into neighboring keratinocytes [66], it is very tempting to speculate that ectocytosis may serve as a mechanism of cell to cell communication. This function has been hypothesized by many in the primary cilia field since the in vitro characterization of ectocytosis. In this

104 hypothesis, the ectosomes would be endocytosed by neighboring cells such that the contents of the ectosome could possibly signal to the neighboring cell. Counter to this hypothesis is the idea that epidermal ectosomes may not be stable and may just degrade soon after separating from the primary cilia. This hypothesis leaves open the possibilities that the contents of the ectosome could signal in a paracrine manner.

Current limitations on our ex vivo live imaging technique for whole embryos limit our ability to track ectosomes. The technical issues are mainly fluorophore photobleaching and focal drift. These issues limit our live imaging to 4-5 hours maximum prior to loss of focal plane or significantly diminished fluorescence. Improving the agar-based mounting mechanism by which embryos are mounted in imaging dishes may alleviate both issues as it we currently image multiple Z planes at every timepoint in order to compensate for focal drift over time. Elimination or reduction in focal drift would allow for less Z planes to be acquired at every timepoint and therefore also reduce fluorophore photobleaching.

Using in vivo mouse models, we also demonstrated that Ift20 is crucial in HF regeneration. Loss of Ift20 in HF-SCs ablates primary cilia in HF-SCs, inhibits hair regrowth, delays the telogen to anagen transition, and results in lineage infidelity in HF-SCs such that they lose their HF-SC identity and differentiate. In particular we observe loss of SOX9 expression in

IFT20-null HF-SCs along with ectopic expression of the BSC marker KLF5 and differentiation markers K10 and Filaggrin. This finding is novel as it identifies Ift20 as a crucial regulator of HF homeostasis. Confirming our suspicion that the telogen to anagen transition was delayed in the

Ift20 cKO HFs, we find that they do not express markers of active Wnt signaling such as nuclear

Lef1 or nuclear β-catenin. As previously mentioned, it is unclear if the absence of Wnt signaling is the cause or effect of the delayed telogen to anagen transition.

105

It remains to be investigated if these observed HF phenotypes are due to loss of primary cilia in HF-SCs or due to loss of a non-ciliary function of Ift20 in the HF-SCs. As mentioned before deletion of HF-SC primary cilia through another means (such as the IFT88 fl/fl mouse) will delineate phenotypes caused by loss of primary cilia versus phenotypes caused by loss of a non-ciliary function of Ift20. Primary cilia are thought to regulate both Wnt and Shh signaling which are crucial in HF-SC growth and activation [69]. While primary cilia-regulated Wnt signaling remains somewhat controversial, previous studies suggest that primary cilia serve to attenuate canonical Wnt signaling through Inversin/NPHP2 and that loss of primary cilia results in increased Wnt signaling [69]. Because we observe decreased and not increased Wnt signaling in Ift20-null HFs, this is in conflict with the idea that loss of primary cilia may be responsible for this phenotype. Non-specific ablation of epidermal primary cilia during development is known to inhibit HF morphogenesis and Shh signaling in developing HFs [53]. Interestingly, targeted ablation of primary cilia in Shh+ matrix cells of the HF does not inhibit HF morphogenesis and instead, postnatally, cells-originating from the HF appear in the IFE [103]. It is unclear if these

HF cells migrated to the IFE or if it is indicative of HF conversion to IFE. Our findings suggest that loss of primary cilia in HF-SCs does result in HF-SCs however it is unclear if this is due to defective Shh signaling. Furthermore, deletion of primary cilia in Shh+ matrix cells of the HF did not result in any reported hair growth defects [103].

One possible non-ciliary mechanism that may explain our HF phenotypes in Ift20 cKO mice is that Ift20 regulates trafficking of a Wnt receptor to the surface of HF-SCs. Autocrine and paracrine Wnt signaling is thought to maintain HF-SC identity. Deletion of Wntless, a protein required for Wnt ligand secretion, in HF-SCs inhibits hair growth and results in degeneration of

HFs [21]. Additionally, differentiation of HF-SCs is reported [21]. While SOX9 and KLF5

106 expression was not analyzed, the hair growth defect, absence of Wnt signaling in HF-SCs, and differentiation phenotype observed in the Wntless cKO mice match that of Ift20 cKO mice. Ift20 has not been linked to canonical Wnt signaling, but interestingly it has been shown to function downstream of ROR2 a non-canonical Wnt receptor in cancer cell lines [98, 99].

We finally investigated the role of Ift20 and primary cilia in keratinocyte migration and epidermal wound healing. We find that loss of Ift20 slows keratinocyte migration both in vitro and in vivo in a manner that is independent of primary cilia. Instead, we find that Ift20 regulates

FA dynamics, specifically reformation, through Rab5-dependent recycling of integrins. Our findings implicate Ift20 in regulating mechano-transduction through FA turnover in a manner independent of it function at primary cilia. This discovery sheds light on non-ciliary functions of

IFT proteins and demonstrates the multi-faceted functions proteins typically associated with primary cilia may have. To our knowledge this is the first time that an IFT protein has been implicated in integrin signaling.

Interestingly, in non-ciliated T-cells, recycling of T-cell receptors at the immune synapse is not only dependent on Ift20 but also Ift52, Ift54, and Ift57. Loss of any of these IFT proteins inhibits proper accumulation of T-cell receptors at the immune synapse [101]. In non-ciliated T- cells, Ift20 has been shown to interact with Ift52, Ift54, and Ift57 suggesting that all of these IFT proteins may regulate recycling of the T-cell receptors together [101]. As we discussed previously, FAs are similar to the immune synapse in that they both consist of a clustering of receptors that engage with extracellular ligands. As such it is interesting to speculate whether recycling of keratinocyte integrins is also dependent on Ift52, Ift54, and Ift57. This question warrants further investigation into the possible non-ciliary function of these IFT proteins in tandem with Ift20 in recycling of surface integrins in keratinocytes.

107

In these concluding remarks I have proposed and speculated on numerous questions that stem from our findings regarding primary cilia and Ift20 in mammalian epidermis. In addition to these questions there remain many unanswered questions about primary cilia and Ift20 in mammalian skin which have not been discussed. However, it is evident from our findings and findings from others that primary cilia and Ift20 through both its ciliary and non-ciliary functions, play key roles in epidermal development, homeostasis, and wound healing. As such it is clear that future studies are necessary to fully elucidate the functions of primary cilia and Ift20 in the skin.

108

References

1. Watt, F.M., Mammalian skin cell biology: at the interface between laboratory and clinic. Science, 2014. 346(6212): p. 937-40. 2. Rognoni, E. and F.M. Watt, Skin Cell Heterogeneity in Development, Wound Healing, and Cancer. Trends Cell Biol, 2018. 28(9): p. 709-722. 3. Fuchs, E., Finding one's niche in the skin. Cell Stem Cell, 2009. 4(6): p. 499-502. 4. Gonzales, K.A.U. and E. Fuchs, Skin and Its Regenerative Powers: An Alliance between Stem Cells and Their Niche. Dev Cell, 2017. 43(4): p. 387-401. 5. Fuchs, E., Epithelial Skin Biology: Three Decades of Developmental Biology, a Hundred Questions Answered and a Thousand New Ones to Address. Curr Top Dev Biol, 2016. 116: p. 357-74. 6. Samadikuchaksaraei, A., et al., Stem cells as building blocks, in Principles of tissue engineering. 2014, Elsevier. p. 41-55. 7. Kabashima, K., et al., The immunological anatomy of the skin. Nat Rev Immunol, 2019. 19(1): p. 19-30. 8. Blanpain, C. and E. Fuchs, Epidermal stem cells of the skin. Annu Rev Cell Dev Biol, 2006. 22: p. 339-73. 9. Sandilands, A., et al., Filaggrin in the frontline: role in skin barrier function and disease. J Cell Sci, 2009. 122(Pt 9): p. 1285-94. 10. Wang, E.C.E., et al., A Subset of TREM2(+) Dermal Macrophages Secretes Oncostatin M to Maintain Hair Follicle Stem Cell Quiescence and Inhibit Hair Growth. Cell Stem Cell, 2019. 24(4): p. 654-669 e6. 11. Lay, K., T. Kume, and E. Fuchs, FOXC1 maintains the hair follicle stem cell niche and governs stem cell quiescence to preserve long-term tissue-regenerating potential. Proc Natl Acad Sci U S A, 2016. 113(11): p. E1506-15. 12. Ito, M., et al., Stem cells in the hair follicle bulge contribute to wound repair but not to homeostasis of the epidermis. Nat Med, 2005. 11(12): p. 1351-4. 13. Cotsarelis, G., Epithelial stem cells: a folliculocentric view. J Invest Dermatol, 2006. 126(7): p. 1459-68. 14. Plikus, M.V., et al., Epithelial stem cells and implications for wound repair. Semin Cell Dev Biol, 2012. 23(9): p. 946-53. 15. Shook, B., et al., The Role of Adipocytes in Tissue Regeneration and Stem Cell Niches. Annu Rev Cell Dev Biol, 2016. 32: p. 609-631. 16. Hsu, Y.C., L. Li, and E. Fuchs, Transit-amplifying cells orchestrate stem cell activity and tissue regeneration. Cell, 2014. 157(4): p. 935-49. 17. Alonso, L. and E. Fuchs, The hair cycle. J Cell Sci, 2006. 119(Pt 3): p. 391-3. 18. Alonso, L., et al., Sgk3 links growth factor signaling to maintenance of progenitor cells in the hair follicle. J Cell Biol, 2005. 170(4): p. 559-70. 19. Ma, L., et al., 'Cyclic alopecia' in Msx2 mutants: defects in hair cycling and hair shaft differentiation. Development, 2003. 130(2): p. 379-89. 20. Reddy, S.T., et al., Expression of Frizzled genes in developing and postnatal hair follicles. J Invest Dermatol, 2004. 123(2): p. 275-82. 21. Lim, X., et al., Axin2 marks quiescent hair follicle bulge stem cells that are maintained by autocrine Wnt/beta-catenin signaling. Proc Natl Acad Sci U S A, 2016. 113(11): p. E1498-505.

109

22. Festa, E., et al., Adipocyte lineage cells contribute to the skin stem cell niche to drive hair cycling. Cell, 2011. 146(5): p. 761-71. 23. Rognoni, E., et al., Kindlin-1 controls Wnt and TGF-beta availability to regulate cutaneous stem cell proliferation. Nat Med, 2014. 20(4): p. 350-9. 24. Greco, V., et al., A two-step mechanism for stem cell activation during hair regeneration. Cell Stem Cell, 2009. 4(2): p. 155-69. 25. Choi, Y.S., et al., Distinct Functions for Wnt/beta-Catenin in Hair Follicle Stem Cell Proliferation and Survival and Interfollicular Epidermal Homeostasis. Cell Stem Cell, 2013. 13(6): p. 720-733. 26. Lim, X.H., et al., Interfollicular Epidermal Stem Cells Self-Renew via Autocrine Wnt Signaling. Science, 2013. 342(6163): p. 1226-1230. 27. Blanpain, C., et al., Canonical notch signaling functions as a commitment switch in the epidermal lineage. Genes Dev, 2006. 20(21): p. 3022-35. 28. Rangarajan, A., et al., Notch signaling is a direct determinant of keratinocyte growth arrest and entry into differentiation. EMBO J, 2001. 20(13): p. 3427-36. 29. Flores, A., et al., Lactate dehydrogenase activity drives hair follicle stem cell activation. Nat Cell Biol, 2017. 19(9): p. 1017-1026. 30. Wang, L., et al., Foxc1 reinforces quiescence in self-renewing hair follicle stem cells. Science, 2016. 351(6273): p. 613-7. 31. Vidal, V.P., et al., Sox9 is essential for outer root sheath differentiation and the formation of the hair stem cell compartment. Curr Biol, 2005. 15(15): p. 1340-51. 32. Kadaja, M., et al., SOX9: a stem cell transcriptional regulator of secreted niche signaling factors. Genes Dev, 2014. 28(4): p. 328-41. 33. Ge, Y., et al., Stem Cell Lineage Infidelity Drives Wound Repair and Cancer. Cell, 2017. 169(4): p. 636-650 e14. 34. Rompolas, P., K.R. Mesa, and V. Greco, Spatial organization within a niche as a determinant of stem-cell fate. Nature, 2013. 502(7472): p. 513-8. 35. Matsumura, H., et al., Hair follicle aging is driven by transepidermal elimination of stem cells via COL17A1 proteolysis. Science, 2016. 351(6273): p. aad4395. 36. Watanabe, M., et al., Type XVII collagen coordinates proliferation in the interfollicular epidermis. Elife, 2017. 6. 37. Le, H.Q., et al., Mechanical regulation of transcription controls Polycomb-mediated gene silencing during lineage commitment. Nat Cell Biol, 2016. 18(8): p. 864-75. 38. Park, S., et al., Tissue-scale coordination of cellular behaviour promotes epidermal wound repair in live mice. Nat Cell Biol, 2017. 19(2): p. 155-163. 39. Aragona, M., et al., Defining stem cell dynamics and migration during wound healing in mouse skin epidermis. Nat Commun, 2017. 8: p. 14684. 40. Langton, A.K., S.E. Herrick, and D.J. Headon, An extended epidermal response heals cutaneous wounds in the absence of a hair follicle stem cell contribution. J Invest Dermatol, 2008. 128(5): p. 1311-8. 41. Mayor, R. and S. Etienne-Manneville, The front and rear of collective cell migration. Nat Rev Mol Cell Biol, 2016. 17(2): p. 97-109. 42. Tapial Martinez, P., P. Lopez Navajas, and D. Lietha, FAK Structure and Regulation by Membrane Interactions and Force in Focal Adhesions. Biomolecules, 2020. 10(2). 43. Wu, C., Focal adhesion: a focal point in current cell biology and molecular medicine. Cell Adh Migr, 2007. 1(1): p. 13-8.

110

44. Parsons, J.T., A.R. Horwitz, and M.A. Schwartz, Cell adhesion: integrating cytoskeletal dynamics and cellular tension. Nat Rev Mol Cell Biol, 2010. 11(9): p. 633-43. 45. Krylyshkina, O., et al., Modulation of substrate adhesion dynamics via microtubule targeting requires kinesin-1. J Cell Biol, 2002. 156(2): p. 349-59. 46. Ezratty, E.J., et al., Clathrin mediates integrin endocytosis for focal adhesion disassembly in migrating cells. J Cell Biol, 2009. 187(5): p. 733-47. 47. Chao, W.T. and J. Kunz, Focal adhesion disassembly requires clathrin-dependent endocytosis of integrins. FEBS Lett, 2009. 583(8): p. 1337-43. 48. Nader, G.P., E.J. Ezratty, and G.G. Gundersen, FAK, talin and PIPKIgamma regulate endocytosed integrin activation to polarize focal adhesion assembly. Nat Cell Biol, 2016. 18(5): p. 491-503. 49. Ahtiainen, L., et al., Directional cell migration, but not proliferation, drives hair placode morphogenesis. Dev Cell, 2014. 28(5): p. 588-602. 50. Jamora, C., et al., Links between signal transduction, transcription and adhesion in epithelial bud development. Nature, 2003. 422(6929): p. 317-22. 51. Lu, C.P., et al., Spatiotemporal antagonism in mesenchymal-epithelial signaling in sweat versus hair fate decision. Science, 2016. 354(6319). 52. Ouspenskaia, T., et al., WNT-SHH Antagonism Specifies and Expands Stem Cells prior to Niche Formation. Cell, 2016. 164(1-2): p. 156-169. 53. Ezratty, E.J., et al., A role for the primary cilium in Notch signaling and epidermal differentiation during skin development. Cell, 2011. 145(7): p. 1129-41. 54. Williams, S.E., et al., Asymmetric cell divisions promote Notch-dependent epidermal differentiation. Nature, 2011. 470(7334): p. 353-8. 55. Lechler, T. and E. Fuchs, Asymmetric cell divisions promote stratification and differentiation of mammalian skin. Nature, 2005. 437(7056): p. 275-80. 56. Williams, S.E., et al., Par3-mInsc and Galphai3 cooperate to promote oriented epidermal cell divisions through LGN. Nat Cell Biol, 2014. 16(8): p. 758-69. 57. Singla, V. and J.F. Reiter, The primary cilium as the cell's antenna: signaling at a sensory organelle. Science, 2006. 313(5787): p. 629-33. 58. Bloodgood, R.A., From central to rudimentary to primary: the history of an underappreciated organelle whose time has come. The primary cilium. Methods Cell Biol, 2009. 94: p. 3-52. 59. Schneider, L., et al., PDGFRalphaalpha signaling is regulated through the primary cilium in fibroblasts. Curr Biol, 2005. 15(20): p. 1861-6. 60. Schneider, L., et al., Directional cell migration and chemotaxis in wound healing response to PDGF-AA are coordinated by the primary cilium in fibroblasts. Cell Physiol Biochem, 2010. 25(2-3): p. 279-92. 61. Clement, D.L., et al., PDGFRalpha signaling in the primary cilium regulates NHE1- dependent fibroblast migration via coordinated differential activity of MEK1/2-ERK1/2- p90RSK and AKT signaling pathways. J Cell Sci, 2013. 126(Pt 4): p. 953-65. 62. Corbit, K.C., et al., Vertebrate Smoothened functions at the primary cilium. Nature, 2005. 437(7061): p. 1018-21. 63. Otto, E.A., et al., Mutations in INVS encoding inversin cause nephronophthisis type 2, linking renal cystic disease to the function of primary cilia and left-right axis determination. Nat Genet, 2003. 34(4): p. 413-20.

111

64. Simons, M., et al., Inversin, the gene product mutated in nephronophthisis type II, functions as a molecular switch between Wnt signaling pathways. Nat Genet, 2005. 37(5): p. 537-43. 65. Leitch, C.C., et al., Basal body proteins regulate Notch signaling through endosomal trafficking. J Cell Sci, 2014. 127(Pt 11): p. 2407-19. 66. Ezratty, E.J., H.A. Pasolli, and E. Fuchs, A Presenilin-2-ARF4 trafficking axis modulates Notch signaling during epidermal differentiation. J Cell Biol, 2016. 214(1): p. 89-101. 67. Liu, Z., et al., Primary cilia regulate hematopoietic stem and progenitor cell specification through Notch signaling in zebrafish. Nat Commun, 2019. 10(1): p. 1839. 68. Bangs, F. and K.V. Anderson, Primary Cilia and Mammalian Hedgehog Signaling. Cold Spring Harb Perspect Biol, 2017. 9(5). 69. Wheway, G., L. Nazlamova, and J.T. Hancock, Signaling through the Primary Cilium. Front Cell Dev Biol, 2018. 6: p. 8. 70. Ocbina, P.J., M. Tuson, and K.V. Anderson, Primary cilia are not required for normal canonical Wnt signaling in the mouse embryo. PLoS One, 2009. 4(8): p. e6839. 71. Abdelhamed, Z.A., et al., Variable expressivity of ciliopathy neurological phenotypes that encompass Meckel-Gruber syndrome and Joubert syndrome is caused by complex de-regulated ciliogenesis, Shh and Wnt signalling defects. Hum Mol Genet, 2013. 22(7): p. 1358-72. 72. Wheway, G., et al., Aberrant Wnt signalling and cellular over-proliferation in a novel mouse model of Meckel-Gruber syndrome. Dev Biol, 2013. 377(1): p. 55-66. 73. Malone, A.M., et al., Primary cilia mediate mechanosensing in bone cells by a calcium- independent mechanism. Proc Natl Acad Sci U S A, 2007. 104(33): p. 13325-30. 74. Wann, A.K., et al., Primary cilia mediate mechanotransduction through control of ATP- induced Ca2+ signaling in compressed chondrocytes. FASEB J, 2012. 26(4): p. 1663-71. 75. Oliazadeh, N., et al., Identification of Elongated Primary Cilia with Impaired Mechanotransduction in Idiopathic Scoliosis Patients. Sci Rep, 2017. 7: p. 44260. 76. Spasic, M. and C.R. Jacobs, Lengthening primary cilia enhances cellular mechanosensitivity. Eur Cell Mater, 2017. 33: p. 158-168. 77. R, R.F., et al., The cilium as a force sensor-myth versus reality. J Cell Sci, 2019. 132(14). 78. Hoey, D.A., et al., Primary cilia-mediated mechanotransduction in human mesenchymal stem cells. Stem Cells, 2012. 30(11): p. 2561-70. 79. Kwon, R.Y., et al., Primary cilium-dependent mechanosensing is mediated by adenylyl cyclase 6 and cyclic AMP in bone cells. FASEB J, 2010. 24(8): p. 2859-68. 80. Shinohara, K. and H. Hamada, Cilia in Left-Right Symmetry Breaking. Cold Spring Harb Perspect Biol, 2017. 9(10). 81. Yoshiba, S., et al., Cilia at the node of mouse embryos sense fluid flow for left-right determination via Pkd2. Science, 2012. 338(6104): p. 226-31. 82. Anvarian, Z., et al., Cellular signalling by primary cilia in development, organ function and disease. Nat Rev Nephrol, 2019. 15(4): p. 199-219. 83. Ware, S.M., M.G. Aygun, and F. Hildebrandt, Spectrum of clinical diseases caused by disorders of primary cilia. Proc Am Thorac Soc, 2011. 8(5): p. 444-50. 84. Tobin, J.L., et al., Inhibition of neural crest migration underlies craniofacial dysmorphology and Hirschsprung's disease in Bardet-Biedl syndrome. Proc Natl Acad Sci U S A, 2008. 105(18): p. 6714-9.

112

85. Waters, A.M. and P.L. Beales, Ciliopathies: an expanding disease spectrum. Pediatr Nephrol, 2011. 26(7): p. 1039-56. 86. Bhattarai, S.R., et al., The ciliary GTPase Arl3 maintains tissue architecture by directing planar spindle orientation during epidermal morphogenesis. Development, 2019. 146(9). 87. Arts, H. and N. Knoers, Cranioectodermal Dysplasia, in GeneReviews((R)), M.P. Adam, et al., Editors. 1993: Seattle (WA). 88. Chen, J., et al., The ciliopathy gene Rpgrip1l is essential for hair follicle development. J Invest Dermatol, 2015. 135(3): p. 701-709. 89. Wolf, M.T., et al., Mutational analysis of the RPGRIP1L gene in patients with Joubert syndrome and nephronophthisis. Kidney Int, 2007. 72(12): p. 1520-6. 90. Brancati, F., et al., RPGRIP1L mutations are mainly associated with the cerebello-renal phenotype of Joubert syndrome-related disorders. Clin Genet, 2008. 74(2): p. 164-70. 91. Lechtreck, K.F., IFT-Cargo Interactions and Protein Transport in Cilia. Trends Biochem Sci, 2015. 40(12): p. 765-778. 92. Valente, E.M., et al., Primary cilia in neurodevelopmental disorders. Nat Rev Neurol, 2014. 10(1): p. 27-36. 93. Follit, J.A., et al., The intraflagellar transport protein IFT20 is associated with the Golgi complex and is required for cilia assembly. Mol Biol Cell, 2006. 17(9): p. 3781-92. 94. Phua, S.C., et al., Dynamic Remodeling of Membrane Composition Drives Cell Cycle through Primary Cilia Excision. Cell, 2017. 168(1-2): p. 264-279 e15. 95. Plotnikova, O.V., E.N. Pugacheva, and E.A. Golemis, Primary cilia and the cell cycle. Methods Cell Biol, 2009. 94: p. 137-60. 96. Pugacheva, E.N., et al., HEF1-dependent Aurora A activation induces disassembly of the primary cilium. Cell, 2007. 129(7): p. 1351-63. 97. Nager, A.R., et al., An Actin Network Dispatches Ciliary GPCRs into Extracellular Vesicles to Modulate Signaling. Cell, 2017. 168(1-2): p. 252-263 e14. 98. Nishita, M., et al., Ror2 signaling regulates Golgi structure and transport through IFT20 for tumor invasiveness. Sci Rep, 2017. 7(1): p. 1. 99. Aoki, T., et al., Intraflagellar transport 20 promotes collective cancer cell invasion by regulating polarized organization of Golgi-associated microtubules. Cancer Sci, 2019. 110(4): p. 1306-1316. 100. Finetti, F., et al., Intraflagellar transport is required for polarized recycling of the TCR/CD3 complex to the immune synapse. Nat Cell Biol, 2009. 11(11): p. 1332-9. 101. Finetti, F., et al., Specific recycling receptors are targeted to the immune synapse by the intraflagellar transport system. J Cell Sci, 2014. 127(Pt 9): p. 1924-37. 102. Galgano, D., et al., The T cell IFT20 interactome reveals new players in immune synapse assembly. J Cell Sci, 2017. 130(6): p. 1110-1121. 103. Croyle, M.J., et al., Role of epidermal primary cilia in the homeostasis of skin and hair follicles. Development, 2011. 138(9): p. 1675-85. 104. Bangs, F.K., et al., Lineage specificity of primary cilia in the mouse embryo. Nat Cell Biol, 2015. 17(2): p. 113-22. 105. Larkins, C.E., et al., Arl13b regulates ciliogenesis and the dynamic localization of Shh signaling proteins. Mol Biol Cell, 2011. 22(23): p. 4694-703. 106. Higginbotham, H., et al., Arl13b in primary cilia regulates the migration and placement of interneurons in the developing cerebral cortex. Dev Cell, 2012. 23(5): p. 925-38.

113

107. Higginbotham, H., et al., Transgenic mouse line with green-fluorescent protein-labeled Centrin 2 allows visualization of the centrosome in living cells. Transgenic Res, 2004. 13(2): p. 155-64. 108. Su, S., et al., Genetically encoded calcium indicator illuminates calcium dynamics in primary cilia. Nat Methods, 2013. 10(11): p. 1105-7. 109. Nachury, M.V. and D.U. Mick, Establishing and regulating the composition of cilia for signal transduction. Nat Rev Mol Cell Biol, 2019. 20(7): p. 389-405. 110. Muller-Rover, S., et al., A comprehensive guide for the accurate classification of murine hair follicles in distinct hair cycle stages. J Invest Dermatol, 2001. 117(1): p. 3-15. 111. Kishimoto, J., R.E. Burgeson, and B.A. Morgan, Wnt signaling maintains the hair- inducing activity of the dermal papilla. Genes Dev, 2000. 14(10): p. 1181-5. 112. Panteleyev, A.A., Functional anatomy of the hair follicle: The Secondary Hair Germ. Exp Dermatol, 2018. 27(7): p. 701-720. 113. Adams, J.C. and F.M. Watt, Fibronectin inhibits the terminal differentiation of human keratinocytes. Nature, 1989. 340(6231): p. 307-9. 114. Watt, F.M., et al., Regulation of keratinocyte terminal differentiation by integrin- extracellular matrix interactions. J Cell Sci, 1993. 106 ( Pt 1): p. 175-82. 115. Levy, L., et al., beta1 integrins regulate keratinocyte adhesion and differentiation by distinct mechanisms. Mol Biol Cell, 2000. 11(2): p. 453-66. 116. Raghavan, S., et al., Conditional ablation of beta1 integrin in skin. Severe defects in epidermal proliferation, basement membrane formation, and hair follicle invagination. J Cell Biol, 2000. 150(5): p. 1149-60. 117. Haycraft, C.J., et al., Intraflagellar transport is essential for endochondral bone formation. Development, 2007. 134(2): p. 307-16. 118. Christensen, S.T., et al., Analysis of primary cilia in directional cell migration in fibroblasts. Methods Enzymol, 2013. 525: p. 45-58. 119. Burridge, K., Focal adhesions: a personal perspective on a half century of progress. FEBS J, 2017. 284(20): p. 3355-3361. 120. Zhao, X. and J.L. Guan, Focal adhesion kinase and its signaling pathways in cell migration and angiogenesis. Adv Drug Deliv Rev, 2011. 63(8): p. 610-5. 121. Absalon, S., et al., Intraflagellar transport and functional analysis of genes required for flagellum formation in trypanosomes. Mol Biol Cell, 2008. 19(3): p. 929-44. 122. Ezratty, E.J., M.A. Partridge, and G.G. Gundersen, Microtubule-induced focal adhesion disassembly is mediated by dynamin and focal adhesion kinase. Nat Cell Biol, 2005. 7(6): p. 581-90. 123. Chen, K.E., M.D. Healy, and B.M. Collins, Towards a molecular understanding of endosomal trafficking by Retromer and Retriever. Traffic, 2019. 20(7): p. 465-478. 124. Ito, M. and G. Cotsarelis, Is the hair follicle necessary for normal wound healing? J Invest Dermatol, 2008. 128(5): p. 1059-61. 125. Bose, A., et al., Keratin k15 as a biomarker of epidermal stem cells. Int J Mol Sci, 2013. 14(10): p. 19385-98. 126. Jonassen, J.A., et al., Deletion of IFT20 in the mouse kidney causes misorientation of the mitotic spindle and cystic kidney disease. J Cell Biol, 2008. 183(3): p. 377-84. 127. Spasic, M. and C.R. Jacobs, Primary cilia: Cell and molecular mechanosensors directing whole tissue function. Semin Cell Dev Biol, 2017. 71: p. 42-52.

114

128. Veland, I.R., L. Lindbaek, and S.T. Christensen, Linking the Primary Cilium to Cell Migration in Tissue Repair and Brain Development. Bioscience, 2014. 64(12): p. 1115- 1125. 129. Goetz, S.C. and K.V. Anderson, The primary cilium: a signalling centre during vertebrate development. Nat Rev Genet, 2010. 11(5): p. 331-44. 130. Umberger, N.L. and T. Caspary, Ciliary transport regulates PDGF-AA/alphaalpha signaling via elevated mammalian target of rapamycin signaling and diminished PP2A activity. Mol Biol Cell, 2015. 26(2): p. 350-8. 131. Ansel, J.C., et al., Human keratinocytes are a major source of cutaneous platelet-derived growth factor. J Clin Invest, 1993. 92(2): p. 671-8. 132. Madison, K.C. and E.J. Howard, Ceramides are transported through the Golgi apparatus in human keratinocytes in vitro. J Invest Dermatol, 1996. 106(5): p. 1030-5. 133. Behne, M.J., et al., Human keratinocyte ATP2C1 localizes to the Golgi and controls Golgi Ca2+ stores. J Invest Dermatol, 2003. 121(4): p. 688-94. 134. Finetti, F., N. Capitani, and C.T. Baldari, Emerging Roles of the Intraflagellar Transport System in the Orchestration of Cellular Degradation Pathways. Front Cell Dev Biol, 2019. 7: p. 292. 135. Finetti, F., et al., The intraflagellar transport protein IFT20 controls lysosome biogenesis by regulating the post-Golgi transport of acid hydrolases. Cell Death Differ, 2020. 27(1): p. 310-328. 136. Song, B., et al., Development of the post-natal growth plate requires intraflagellar transport proteins. Dev Biol, 2007. 305(1): p. 202-16. 137. Arriagada, C., et al., Focal adhesion kinase-dependent activation of the early endocytic protein Rab5 is associated with cell migration. J Biol Chem, 2019. 294(34): p. 12836- 12845. 138. Wu, X., et al., Skin stem cells orchestrate directional migration by regulating microtubule-ACF7 connections through GSK3beta. Cell, 2011. 144(3): p. 341-52. 139. Nowak, J.A., et al., Hair follicle stem cells are specified and function in early skin morphogenesis. Cell Stem Cell, 2008. 3(1): p. 33-43. 140. Nowak, J.A. and E. Fuchs, Isolation and culture of epithelial stem cells. Methods Mol Biol, 2009. 482: p. 215-32. 141. Ku, A.T., Q. Miao, and H. Nguyen, Monitoring Wnt/beta-Catenin Signaling in Skin. Methods Mol Biol, 2016. 1481: p. 127-40. 142. Fischer, A.H., et al., Hematoxylin and eosin staining of tissue and cell sections. CSH Protoc, 2008. 2008: p. pdb prot4986. 143. Rizk, A., et al., Subcellular object quantification with Squassh3C and SquasshAnalyst. Biotechniques, 2015. 59(5): p. 309-12.

115

Appendix A: Methods and Materials

Generation of mouse lines

All mice were housed and bred according to Columbia University IACUC approved protocols. Arl13b-mCherry; Centrin 2-GFP mice (Jackson Laboratory, Stock No. 027967) were continually bred to be homozygous for both transgenic alleles. K15-CrePR1 mice (Jackson

Laboratory, Stock No. 005249) were bred with Ai9 mice (Jackson Laboratory, Stock No.

007909) to generate K15-Cre+; Rosa26-tdTomato fl/fl mice. These mice were then bred with

Ift20 fl/fl mice (Jackson Laboratory, Stock No. 012565) to generate K15-Cre+; Ift20 fl/fl;

Rosa26-tdTomato fl/fl mice. Genotyping was performed according to Jackson Laboratory recommendations using the following primer sets: Cre forward, 5’-

GATATCTCACGTACTGACGG-3’; Cre reverse, 5’-TGACCAGAGTCATCCTTAGC-3’; Ift20 forward, 5’-ACTCAGTATGCAGCCCAGGT-3’; Ift20 reverse, 5’-

GCTAGATGCTGGGCGTAAAG-3’; TomatoWT forward, 5’-

AAGGGAGCTGCAGTGGAGTA-3’; TomatoWT reverse, 5’-

CCGAAAATCTGTGGGAAGTC-3’; TomatoMut forward, 5’-

GGCATTAAAGCAGCGTATCC-3’;; TomatoMut reverse, 5’-CTGTTCCTGTACGGCATGG-

3’.

In-vivo ablation of Ift20 in hair follicle stem cells and wound healing experiments

All surgical protocols were approved by Columbia University IACUC. Mice were bred such that litters contained both K15Cre+; Ift20 fl/fl; Rosa26-tdTomato fl/fl (conditional knockout) and K15Cre+; Ift20 +/fl; Rosa26-tdTomato fl/fl (control) littermates. At postnatal day

21 or 22, the backskins of the mice were shaved and a topical solution of 1% RU486

(Mifepristone, Cayman Chemical 10006317) diluted in 70% ethanol was applied daily for 5-7

116 days. Between postnatal day 28 and 31, following the completion of RU486 treatment, the back skins of the mice were wounded with a full thickness 6 mm punch biopsy. Four days following the wound, the mice were euthanized and the backskin was dissected away, vigorously washed in

PBS, and either immediately frozen in O.C.T compound (Tissue-Tek 4583) or fixed in 4% PFA for 2-3 hours and then vigorously washed in PBS prior to being frozen in O.C.T compound.

In-vivo ablation of Ift20 in hair follicle stem cells and hair cycling assay

45 to 50, the back skins of the mice were shaved and a topical solution of 1% RU486

(Mifepristone, Cayman Chemical 10006317) diluted in 70% ethanol was applied daily for 5-7 days. The mice were then housed an observed for hair regrowth. Between postnatal day 95-150, the mice were euthanized and the backskin was dissected away, vigorously washed in PBS, and either immediately frozen in O.C.T compound (Tissue-Tek 4583) or fixed in 4% PFA for 2-3 hours and then vigorously washed in PBS prior to being frozen in O.C.T compound. For β- catenin and Hematoxylin and Eosin staining, the tissue was first fixed in 4% PFA for 2-3 hours and then paraffinized prior to staining. See https://www.protocolsonline.com/histology/sample- preparation/paraffin-processing-of-tissue/ for a detailed protocol.

Whole mount tissue preparation and antibody staining

For whole mount embryonic backskin, the entire embryo was first fixed in 4% PFA for 2-

3 hours. After fixation the backskin was carefully dissected away for subsequent antibody staining. For postnatal and adult backskins, the backskin was first carefully dissected away from the animal and then fixed in 4% PFA for 2-3 hours prior to antibody staining.

117

To stain whole mount backskins, the backskins were first vigorously washed with PBS to remove residual PFA. The backskins were then blocked in gelatin blocking solution (2.5% normal donkey serum, 2.5% normal goat serum, 1% BSA, 2% fish gelatin, and 0.3% Triton-X in

PBS) for 1-2 hours at room temperature. Backskins were then incubated overnight at 4°C with the primary antibodies diluted in gelatin blocking solution. After washing several times with PBS to remove residual unbound antibody, sections were incubated for 2 hours at room temperature in secondary antibodies diluted in gelatin blocking solution. After washing with PBS, the sections were mounted with coverslips using ProLong Gold Antifade Mountant with DAPI

(Invitrogen P36931). The following antibodies and dilutions were used: E-cadherin (1:200,

Invitrogen 13-1900) and Donkey anti-Rat Alexa Fluor 647 (1:500, ImmunoResearch 712-605-

153).

Isolation and culture of primary mouse keratinocytes

Primary mouse keratinocytes were isolated and cultured as previously described. Briefly, mouse pups between P0 and P4 were euthanized and their back skins removed and treated with

Dispase (50 mg/ml; Gibco 17105041) for 2 hours at 37°C or overnight at 4°C. The epidermal layer was peeled away from the dermal layer of the backskin. The epidermis was then floated in a 1:1 0.25% Trypsin EDTA (Gibco)/Versene (Gibco 1504066) mixture and incubated at room temperature for 15 minutes allowing for dissociation of the epidermal cells. Low calcium E media was then added to the mixture and vigorously pipetted. The cell mixture was then filtered through 70 um and 40 um strainers, centrifuged at 1100x g, and resuspended in low calcium E media prior to being plated on Mitomycin C treated 3T3 fibroblast monolayers. The

keratinocytes were placed in a 37°C, 7.5% CO2 incubator and continually passaged onto fresh

Mitomycin C treated 3T3 fibroblast monolayers as necessary. After 7 to 10 passages, the

118 keratinocytes gained ability to propagate without the fibroblasts. The keratinocytes were then continually cultured in low calcium E media or frozen in low calcium E media with 10% DMSO.

Mitomycin C treated 3T3 monolayers

3T3 fibroblasts were grown to confluency on 100 mm dishes in DMEM with 10% bovine calf serum at 37°C and 7.5% CO2. The dishes were treated with Mitomycin C (8 ug/ml, Fisher

BP25312) for 2 hours at 37°C after which the dishes were washed with PBS and given fresh media. Monolayers were maintained until ready for use up to 1 week following Mitomycin C treatment.

Generation of lentivirus and transduction of keratinocytes

Lentiviral NLS-iCre-GFP was a gift from Elaine Fuchs. Lentivirus was produced as previously described[86]. Briefly, the lentiviral plasmid was co-transfected with a pMD.2G envelope plasmid (Addgene #12259) and a psPAX2 packaging plasmid (Addgene #12260) using calcium phosphate/HBS into 293FT cells cultured in DMEM containing 10% FBS and

G418/Geneticin. One day following transfection the media was replaced with Ultraculture Media

(Lonza 12-725F) supplemented with 1% Penicillin/Streptomycin/L-Glutamine, 1 mM Sodium

Pyruvate, 0.075% Sodium Bicarbonate, and 5 mM Sodium Butyrate. 48 hours after transfection viral supernatant was collected from the cultures.

Transduction of keratinocyte cultures was performed as previously described [86].

Briefly, keratinocytes were grown to 70%-80% confluency in a 6-well plate. Immediately prior to viral transduction, 3 ul of Polybrene (10 mg/ml, Sigma Aldrich 107689) and 50-100 ul of viral supernatant was added per well. The plate was centrifuged at 1100xg at 37°C for 30 minutes after which the media was replaced with fresh low calcium E-media. Successful transduction was confirmed 2-3 days afterwards.

119

In vitro migration assays and quantification

Cells were grown in 2-well silicone inserts (Ibidi 80209) in 24 well plates until confluent.

The insert was then removed to form a 500 μm defined cell-free gap. The cells were then washed with PBS and given fresh media. Cell migration was imaged using an IX-83 Olympus inverted microscope at 0 hours and 18 hours after start of migration. Migration rate was calculated as the area covered over time.

Focal adhesion disassembly/reassembly assay

Focal adhesion disassembly/reassembly assays were performed as previously described

[46, 48, 122]. Briefly, keratinocytes were grown on fibronectin-coated coverslips until confluent and then serum starved for 18-24 hours in serum free low calcium E media. The keratinocytes were then treated with 10 μM Nocodazole (Sigma-Aldrich M1404) in serum free low calcium E media for 3 to 4 hours, washed with PBS, and placed in fresh serum free low calcium E media.

At the desired timepoints, the cells were then either fixed for 10 minutes with 4% PFA for subsequent immunofluorescence or scrapped off the cell coverslip with a cell scraper and incubated with antibodies for subsequent flow cytometry.

Immunofluorescence of fixed cells

Following fixation, coverslips were washed with PBS, and then permeabilized for 10 minutes with 0.3% Triton X in PBS. After washing with PBS, coverslips were blocked for 1 hour with 5% normal donkey serum (NDS), 5% normal goat serum (NGS) or a mix of 5% NGS/NDS in PBS at room temperature after which they were incubated at 4°C overnight with the primary antibody diluted in the blocking buffer. Coverslips were then washed with PBS several times and then incubated at room temperature with appropriate secondary antibodies diluted in the blocking buffer for 1.5 hours. After washing several times with PBS, coverslips were mounted on slides

120 with ProLong Gold Antifade Mountant with DAPI (Invitrogen P36931). The following antibodies and dilutions were used: Ift20 (1:200, Proteintech 13615-1-AP), Arl13b (1:200,

Proteintech 17711-1-AP), Acetylated Tubulin (1:1000, Sigma-Aldrich T7451), GM130 (1:200,

BD 610822), Vinculin (1:200, Sigma-Aldrich V4505), Phospho-FAK Y397 (1:200, Invitrogen

44-624G), Rab5A (1:400, Cell Signaling Technology 46449T), CD29/Beta1 Integrin (1:100,

Invitrogen 11-0291-82) Donkey anti-Rabbit Alexa Fluor 488 (1:500, Invitrogen A32790),

Donkey anti-Rabbit Alexa Fluor 647 (1:500, Invitrogen A32795), Donkey anti-Mouse Alexa

Fluor 488 (1:500, Invitrogen A32766), Donkey anti-Mouse Alexa Fluor 647 (1:500, Invitrogen

A32787), and Goat anti-Armenian Hamster Alexa Fluor 488 (1:500, Jackson ImmunoResearch

127-545-160).

Collection of protein lysate

Cells were grown to confluency on glass coverslips. Cells were washed with ice cold

PBS, scraped off the coverslip, and lysed in ice cold RIPA buffer with HALT protease and phosphatase inhibitors (Thermo Scientific 78440) for 30 minutes with regular agitation.

Afterwards the cell lysates were centrifuged and the protein-rich supernatant was collected. The total protein concentration was quantified using a BradfordUltra assay (Expedeon BFU05L) according to manufacturer’s instructions.

Western blotting

Western blots were performed using a NuPAGE electrophoresis system (Invitrogen) according to manufacturer’s instructions. Briefly, 20-30 ug of total protein was loaded per lane into precast NuPAGE Novex Bis-Tris gels (Invitrogen NP0335BOX) and run in either MES SDS or MOPS SDS (Invitrogen NP0002, Invitrogen NP0001) running buffers. The gels were then transferred onto 0.2 um pore size nitrocellulose membranes (Invitrogen LC2000) in NuPAGE

121 transfer buffer (Invitrogen NP0006). Membranes were then blocked in a solution of TBS with

5% BSA for 1 hour and then incubated overnight at 4C with primary antibody diluted in TBS with 0.2% Tween-20 (2X TBST) and 5% BSA. Membranes were then washed with 0.1% Tween

20/TBS (1x TBST) and incubated for 1 hour at room-temperature with secondary antibodies diluted in 2X TBST with 5% BSA. After washing with 1X TBST, the membrane was imaged using a LI-COR Odyssey Fc imaging system. The following antibodies and dilutions were used:

Phospho-FAK (1:1000, Invitrogen 44-624G), Ift20 (1:500, Proteintech 13615-1-AP), Alpha-

Tubulin (1:1000, Sigma-Aldrich T9025), Rab5A (1:1000, Cell Signaling Technology 46449T),

IRDye 800CW Donkey anti-Rabbit (1:10000, Li-Cor 926-32213) and IRDye 680RD Donkey anti-Mouse (1:10000, Li-Cor 926-68072).

Quantification of western blots

Intensities of western blot bands were quantified using the ImageJ Gel analysis function.

Intensities were normalized to control bands for comparison between conditions.

Cell surface labeling of cultured keratinocytes

Labeling of cell surface integrin was performed as previously described [46, 48, 122].

Briefly, cells were grown to confluency on glass coverslips. Cells were then scrapped into ice cold PBS and washed once with PBS. Approximately 1 x 105 cells per sample were incubated on ice with antibodies and diluted in PCN (PBS with 0.5% bovine calf serum and 0.1% NaN3) with Zombie Violet Dye (1:1000, BioLegend 423113) for 30 minutes. Samples were then washed once with PCN to remove unbound antibody and fixed in ice cold 4% PFA diluted in

PCN for 10 minutes. After washing samples with PCN to remove residual PFA, samples were analyzed on a Bio-Rad ZE5 Cell Analyzer and subsequent data was analyzed with FlowJo

122 software. Antibodies and dilutions used are as follows: CD29 (Integrin beta 1)-APC (1:10000,

17-0291-82).

Cell surface labeling analysis

The average fluorescent intensity of integrin in Zombie Violet-low cells (i.e. live cells) was quantified using FlowJo. The same fluorescent intensity was quantified in non-stained cells to measure average background fluorescence. To normalize intensities, the average background fluorescence was subtracted from each of the average fluorescent intensity measurements from the various stained samples. These values were then normalized to the desired time point (i.e. 0 min during Nocodazole washout).

Isolation of HFSCs through flow cytometry

HFSCs were isolated as previously described [140] with minor adjustments. Briefly, mice were sacrificed at appropriate time points and the backskins were shaved and removed. Using a blunted scalpel, the subcutaneous dermal fat of each backskin was scrapped away in a dish of cold HBSS. For telogen skin, the backskin was then directly incubated with 0.25% Trypsin

EDTA (Gibco) at 37°C for 1 hour with the epidermis facing upwards. For anagen skin, the backskin was first incubated with 0.25% (w/v) Type I Collagenase in HBSS for 40 min at 37°C after which the dermis was scrapped away with a blunted scalpel prior to floating the backskin epidermis upwards in 0.25% Trypsin EDTA (Gibco) and incubating for 30 min at 37°C. After adding cold D10 (DMEM with 10% FBS) to neutralize the Trypsin, the HFs were scrapped into the D10/Trypsin mixture using a blunted scalpel and any remaining dermis was removed. The

HF/D10/Trypsin mixture was then vigorously pipetted to break up clumps and then transferred into a bottle with a stir bar. After stirring the mixture for at 4°C for 30 min, the mixture was strained through a 70 μm filter and then a 40 μm filter. The HFSCs were then centrifuged at

123

1100xg and washed once with cold D10. At this point approximately 5% of the cells can be separated and used to prepare compensation staining samples if necessary. The remaining cells were then stained with CD34-eFluor660 (1:75, Invitrogen 50-0341-82) and CD49f (Integrin alpha 6)-FITC (1:100, Invitrogen 11-0495-82) in D10 on ice for 30-45 minutes wither intermittent agitation of the samples. The samples were then centrifuged at 1100xg and washed once with D10 media. The samples were then centrifuged again and then the resuspended in D10 media with DAPI (1:1000, Invitrogen D1306). Cells were then sorted on a BD FACS Aria II cell sorter. Live HFSCs were gated as being DAPI-low, CD34-high, CD49f-high, and tdTomato-high cells and sorted into TRIzol LS (Invitrogen, 10296028) and sorted into TRIzol LS (Invitrogen,

10296028). qPCR of HFSCs

RNA was extracted in TRIzol LS (Invitrogen, 10296028) per the manufacturer’s instructions. cDNA was generated from the RNA using the Verso cDNA Synthesis Kit (Thermo

Scientific, AB1453B) per the manufacturer’s instructions using all RNA. qPCR for each cDNA sample was performed in triplicate using Thermo Scientific Maxima SYBR Green/ROX qPCR

Master Mix (2x) (Thermo Scientific, K0221). The primers used for qPCR were as follows: Ift20 forward, 5'-TCCTGATTGCCACTGTCACC-3' and Ift20 reverse, 5'-

GTCCAACACTCGGAGCTTGT-3'.

Immunofluorescence of tissue sections

Tissues embedded in O.C.T compound were sectioned using a Cryostat (Leica) at 10 um thin sections onto glass slides. The sections were then stained according to previously described protocols (Reference). Briefly, unfixed sections were first fixed for 10 minutes in 4% PFA at room temperature. Sections were then blocked in gelatin blocking solution (2.5% normal donkey

124 serum, 2.5% normal goat serum, 1% BSA, 2% fish gelatin, and 0.3% Triton-X in PBS) for 1-2 hours at room temperature. Sections were then incubated overnight at 4°C with the primary antibodies diluted in gelatin blocking solution. After washing several times with PBS to remove residual unbound antibody, sections were incubated for 1-2 hours at room temperature in secondary antibodies diluted in gelatin blocking solution. After washing with PBS, the sections were mounted with coverslips using ProLong Gold Antifade Mountant with DAPI (Invitrogen

P36931). The following antibodies and dilutions were used: Ift20 (1:200, Proteintech 13615-1-

AP), Arl13b (1:200, Proteintech 17711-1-AP), Acetylated Tubulin (1:1000, Sigma-Aldrich

T7451), GM130 (1:200, BD 610822), E-cadherin (1:200, Invitrogen 13-1900), SOX9 (1:200,

Millipore Sigma), KLF5 (1:200, Proteintech), K10 (1:200, Invitrogen), Filaggrin (1:200,

Abcam), LEF1 (1:200, Proteintech), β-catenin (1:200, Millipore Sigma), Donkey anti-Rabbit

Alexa Fluor 488 (1:500, Invitrogen A32790), Donkey anti-Rabbit Alexa Fluor 647 (1:500,

Invitrogen A32795), Donkey anti-Mouse Alexa Fluor 488 (1:500, Invitrogen A32766), Donkey anti-Mouse Alexa Fluor 647 (1:500, Invitrogen A32787), and Donkey anti-Rat Alexa Fluor 647

(1:500, ImmunoResearch 712-605-153).

β-catenin staining as well as Hematoxylin and Eosin staining was performed on paraffinized tissue sections as previously described [141, 142].

Explant biopsy culture and quantification

The biopsy samples from the wounded mice were collected and plated on fibronectin- coated tissue culture dishes and covered in low calcium E-media. Culture media was replaced every 4-5 days. After 14 days of culturing, NucBlue Live Cell Stain (Invitrogen R37605) was added to the culture media to label nuclei and the sample was imaged using a fluorescence microscope. Migration of tdTomato-positive hair follicle stem cells was quantified by calculating

125 the number of tdTomato-positive cells that had migrated out of the tissue sample per circumferential length of the tissue sample (number of tdTomato-positive cells/circumference of biopsy).

Image Acquisition

Confocal images were acquired with an Olympus IX83 DSU unit with Hamamatsu Orca- r2 through 63x (N.A 1.4) oil or 4x (N.A. .10) Plan-Apochromat objectives and equipped with the following Chroma filter sets: 49008 ET TR C94094 (mRFP1), 49004 ET dsR C94093 (Cy3,

DyLight549), 41008 Cy5 (Cy5), 41001 FITC (AlexaFluor 488/GFP). For images of tissue sections, Z-stacks of 5-15 planes (0.5 μm) were captured, and either representative single Z- planes or maximum projections (3 images) are presented. For images of in vitro cells, Z stacks of

3-5 planes (0.5 μm) were captured and representative single Z planes are shown. Image acquisition was driven using Metamorph (Olympus). Image processing was performed using

ImageJ.

Focal adhesion and co-localization image quantification

Image quantification was performed using ImageJ. Briefly, the background of images was first subtracted using the rolling ball radius background subtraction function in ImageJ.

Individual cells were identified and saved as regions of interests (ROIs). Images were then converted to 8-bit images and thresholded according to the structures of interest (i.e. focal adhesions, endosomes, etc.). Using the analyze particles function in ImageJ, the structures of interest within each ROI were quantified by number and area. Average total area of focal adhesions per cell and average focal adhesion area were computed based on this data. Co- localization coefficients between Rab5A and Ift20 or Rab5A and 1 Integrin in cells were computed using the size-based co-localization coefficient defined in [143] (Area of Ift20 which

126 co-localizes with Rab5A / Total area of Ift20 or Area of Beta 1 Integrin which co-localizes with

Rab5A / Total area of Beta 1 Integrin).

Statistical Analysis

All statistical analysis was performed using Mann-Whitney U tests or Student’s t-tests using R statistical software. The test performed for each set of data is indicated in the figure legends.

127

Appendix B: Abbreviations

1°MKs: Primary Mouse Keratinocytes

BSCs: Basal Stem Cells

BMP: Bone Morphogenetic Protein

DP: Dermal Papilla

ECM: Extracellular Matrix

FA: Focal Adhesion

HF: Hair Follicle

HF-SCs: Hair Follicle Stem Cells

IFE: Interfollicular Epidermis

IFT: Intraflagellar Transport

NZ: Nocodazole

Shh: Sonic Hedgehog

128