Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54

JPRR587_proof ■ 28 April 2015 ■ 1/23

Progress in Retinal and Eye Research xxx (2015) 1e23

Contents lists available at ScienceDirect

Progress in Retinal and Eye Research journal homepage: www.elsevier.com/locate/prer

Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells Q15 Q1

Q3

Nick Di Girolamo*, 1 School of Medical Sciences, University of New South Wales, Sydney, 2052 NSW, Australia

a r t i c l e i n f o

a b s t r a c t

Article history: Received 28 January 2015 Received in revised form 10 April 2015 Accepted 16 April 2015 Available online xxx

Lineage tracing allows the destiny of a stem cell (SC) and its progeny to be followed through time. In order to track their long-term fate, SC must be permanently marked to discern their distribution, division, displacement and differentiation. This information is essential for unravelling the mysteries that govern their replenishing activity while they remain anchored within their niche microenvironment. Modern-day lineage tracing uses inducible genetic recombination to illuminate cells within embryonic, newborn and adult tissues, and the advent of powerful high-resolution microscopy has enabled the behaviour of labelled cells to be monitored in real-time in a living organism. The simple structural organization of the mammalian cornea, including its accessibility and transparency, renders it the ideal tissue to study SC fate using lineage tracing assisted by non-invasive intravital microscopy. Despite more than a century of research devoted to understanding how this tissue is maintained and repaired, many limitations and controversies continue to plague the field, including uncertainties about the specificity of current SC markers, the number of SC within the cornea, their mode of division, their location, and importantly the signals that govern cell migration. This communication will highlight historical discoveries as well as recent developments in the corneal SC field; more specifically how the progeny of these cells are mobilised to replenish this dynamic tissue during steady-state, disease and transplantation. Also discussed is how insights gleaned from animal studies can be used to advance our knowledge of the fundamental mechanisms that govern modelling and remodelling of the human cornea in health and disease. © 2015 Published by Elsevier Ltd.

Keywords: Lineage tracing Limbus Cornea Stem cells Migration Transgenic mice

Contents 1.

Q2

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1. Structure and function of the cornea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2. Corneal development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3. Primary and secondary limbal niche structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4. Limbal epithelial stem cell markers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5. Corneal epithelial cell division and turnover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.6. Epithelial cell migration in the normal and wounded cornea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

00 00 00 00 00 00 00

Abbreviations: AP, Alkaline phosphatase; BM, Basement membrane; b-Gal, bgalactosidase; CFP, Cerulean fluorescent protein; ECM, Extracellular matrix; FP, Fluorescent proteins; GFP, Green fluorescent protein; HRP, Horseradish peroxidase; K, Keratins; LRC, Label-retaining cells; LESC, Limbal epithelial stem cells; LSCD, Limbal stem cell deficiency; RFP, Red fluorescent protein; SC, Stem cells; TAM, Tamoxifen; TDC, Terminally differentiated cells; TAC, Transient amplifying cells; VN, Vitronectin; YFP, Yellow fluorescent protein. * Tel.: þ61 2 93852538; fax: þ61 2 93851389. E-mail address: [email protected]. 1 Percentage of work contributed by each author in the production of the manuscript is as follows: Nick Di Girolamo: 100%. http://dx.doi.org/10.1016/j.preteyeres.2015.04.002 1350-9462/© 2015 Published by Elsevier Ltd.

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 2/23

2

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

2.

3.

4. 5.

1.7. Clinical observations of corneal epithelial cell migration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.8. Migration of conjunctival epithelial cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.9. Intrinsic and extrinsic factors implicated in corneal epithelial cell migration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Introduction to cell and lineage tracing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Methods to study cell and lineage tracing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Lineage tracing by genetic recombination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tracking corneal epithelial cell movement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Transplantation as a method of monitoring cell fate within the cornea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Lineage tracing corneolimbal epithelial cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. How limbal stem cells sustain the corneal epithelium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Future directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Funding sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Author contribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ethics approval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Uncited reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction Blindness from corneal disease affects 10-million individuals globally (Whitcher et al., 2001). The possibility of restoring vision with a trephined donor cornea was proposed by Erasmus Darwin (Charles Darwin's grandfather) nearly 250 years ago and successfully performed in humans for over a century (Zirm, 1906), and prior to that in animals (Bigger, 1837). For the more difficult to treat patients, including those with a disease called limbal stem cell deficiency (LSCD), a donor graft is not a viable option as their SC are depleted or their ‘niche’ residence destroyed by chemical, mechanical, or surgical trauma, or as a consequence of genetic and inflammatory diseases. The challenge for clinicians is to replenish the SC pool; previously achieved by transplanting large sectors of limbal tissue (Kenyon and Tseng, 1989). Today, ex vivo expansion and delivery of SC on biodegradable or synthetic scaffolds is standard practice. However, grafts still fail and success beyond 5e10 years is not well documented (Shortt et al., 2007a; Di Girolamo et al., 2009; Rama et al., 2010; Baylis et al., 2011; Sangwan et al., 2011; Bobba et al., 2015). Patients with LSCD suffer physical and psychological distress including falls, fractures, depression and premature mortality, and the economy is burdened from the loss of income and high health care costs (Geerling et al., 2002), particularly as the most common cause (chemical injury), typically occurs in patients of working age. Restoring corneal health and vision in patients with LSCD is a global initiative with researchers focused on improving current and developing new long-lasting therapies. Notably, a standardised therapeutic intervention has not been devised for this debilitating disease; this is not surprising as the disease varies in aetiology and severity. Therefore, in the future one might anticipate that a suite of procedures will become available to tailor treatment on a case-by-case basis. However, before we can adequately address this medical dilemma, a more comprehensive understanding of the basic biology of LESC is warranted. Of the external organs, the cornea is perhaps the perfect tissue to study the long-term fate of its epithelial cells; firstly because it's SC are spatially segregated from their differentiated progeny, and secondly because they are thought to contribute to life-long tissue maintenance. Other advantages include its translucency and accessibility, rendering cell dynamics through this clear ‘window’ readily observable with non-invasive high-resolution intravital microscopy.

00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

Over a century of research has been devoted to understanding the structural and functional relationship of the mammalian cornea and its resident cells, and despite the constant flow of literature, the field continues to be plagued with limitations and controversies especially in regards to how and from where the cornea receives its signals that direct structural organisation during embryogenesis, how it self-perpetuates during adult live, and how it's SC orchestrate the many activities that govern corneal health for exquisite vision. This article will address these voids in our knowledge base and showcase new scientific discoveries and technological advances within the field. However, the overarching goal is to highlight seminal studies that have informed modern-day fate mapping of cells within this highly specialised structure and to discuss how this data has been instrumental in detailing where SC reside, how they divide, how their daughters differentiate and migrate, how long they survive, and how they behave during steady-state, wound-healing and transplantation. While this information may seem trivial, it is anticipated clues will be divulged as to (i) whether the corneal epithelium contains one or multiple SC reservoirs, (ii) identification of novel SC markers and the signals that trigger cell egress from the niche, and (iii) identification of proteins that regulate ‘stemness’. 1.1. Structure and function of the cornea The adult mammalian cornea is composed of three distinct and functionally diverse cellular layers; the externalised anterior region consists of a multilayered non-keratinised squamous epithelium with proliferating basal cells, which in man rest on a thick (8e15 mm) basement membrane (BM)-like structure called Bowman's layer. Overlying these cells are suprabasal ‘wing’ epithelia, and above these cells are flat post-mitotic superficial epithelia. The posterior cornea contains a monolayer of specialised endothelial cells attached to Descemets' membrane which play a key role in fluid exchange across this tissue. Wedged between the epithelium and endothelium is an avascular connective tissue stroma that makes up 90% (~500 mm) of the cornea and comprised of an orderly array of collagen lamellae interspersed with resident keratocytes and sympathetic nerves with axons that pierce the BM and terminate between the epithelia. Circumscribing the cornea is a 1 mm wide transition zone otherwise known as the limbus; this region partitions the cornea from the conjunctiva and is the

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 3/23

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

presumed residence for SC, otherwise known as limbal epithelial SC (LESC). The limbus in man is architecturally striking compared to the unremarkable corresponding region in the mouse. Histologically, this zone coincides with the disappearance of Bowmans' layer and the appearance of a thick corrugated epithelium, recognised by papillae-like invaginations known as the Pallisades of Vogt (Goldberg, 1982; Townsend, 1991) (Fig. 1A and B). These structures are thought to be strategically placed, firstly as they located adjacent to a rich vascular network from which nutrients

3

and soluble signals diffuse to support LESC, and secondly their deep stromal location affords these cells added protection from environmental insults. Thirdly, their prevalence in the superior and inferior sectors compared to the nasal and temporal meridian is not well understood but it is speculated that the upper and lower eye-lids offer additional sanctuary. The mouse has no obvious Bowman's layer, and the limbus is recognised by an epithelium which tapers into a few layers as it merges with the adjacent conjunctiva (Fig. 2).

Fig. 1. Epithelial structures within the human limbus. Adult (A and B, GeJ) and foetal (CeF) human corneas were stained with either DAPI (A and B, GeJ) or H&E (CeF). Images in panels (A and CeF) were taken from transverse sections through corneal specimens. The Palisades of Vogt or papillae-like invaginations that characterise the adult corneolimbal zone are readily visualised by sectioning through the corneas (A). An alternative perspective of these structures is displayed in a tilled image generated from confocal scanning microscopy on a formalin-fixed, whole-mount cornea (B). In the developing human cornea, the limbal epithelium is dome-like in appearance (C, 12WG and E, 17WG, respectively) as indicated by the hatched black line, while the corresponding epithelium in the central cornea is unremarkable and consists of a bi-layered epithelium (D, 12WG and F, 17WG). 2Photon confocal imaging of a formalin-fixed flat-mount adult human cornea stained with DAPI (green). Images were taken at sequential depths spanning ~400 mm from the superficial epithelium to the depths of the stroma (GeJ). These structures have an opening to the surface epithelium (G, hatched white line), originate from clusters of small basal cells from deep within the stromal matrix (J, hatched white line), and seem to be cylindrical in shape. The image in panel (A) was taken at 100 and those displayed in panels (CeF) at 1000 final magnification under oil immersion. The excitation wavelength used to image the DAPI stained nuclei in panel (B) was 800 nm and the emitted signal was collected into a non-descanned detector equipped with SP 430/50 filter and imaged with 20/NA 0.7 glycerol immersion objective.

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 Q4 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 4/23

4

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

Fig. 2. K-14 Expression in the human and murine cornea. Immunofluorescence for K14 was performed on human (A) and mouse (B and C) corneas by incubating 5 mm tissue sections with either a mouse monoclonal (A) or rabbit polyclonal (B) antibody to K14, and immunoreactivity developed with Alexa488 (green) and Alexa596 (red)-conjugated secondary antibodies. The image in panel (A) was scanned on the Spectrum™ (Aperio Technologies) at 200 original magnification but displayed at 5 magnification in the main panel. Hatched rectangles encompassing the limbus and central cornea are magnified (A, insets). Arrows point to columns of K14þ limbal epithelia that seem to arise from basal epithelia. A representative cornea from a C57BL/6 mouse, showing strong K-14 immunoreactivity along a string of basal limbal epithelial cells (B) which diminishes to little or no expression in cells within the central cornea (C). Tissues in panels (B and C) were counterstained with DAPI and images taken at 1000 original magnification under oil immersion. The hatched white lines in panels (B and C) identify the epithelial basement membrane.

1.2. Corneal development The human cornea is ectodermal in origin and forms a primitive mono-layered epithelium which straddles the lens vesicle at 5 weeks gestation (Duke-Elder and Cook, 1963; Mann, 1964; Marshall and Grindle, 1978; Sevel and Isaacs, 1988; Davies et al., 2009). At this stage, the primordial surface epithelium is separated from the developing intraocular lens and cells of mesodermal origin invade from a periocular location, positioning themselves within the posterior cornea to form the endothelium (Mann, 1964). Several weeks later, the epithelium becomes bi-layered and a second flux of mesenchymal cells contributes to the formation of the stroma. In the mouse, a single insurgence of mesodermal cells is thought to occur which subsequently differentiate into either stromal keratocytes or corneal endothelium (Reneker et al., 2000). At this stage the limbus in man is visible as a protruding ‘dome’ (Fig. 2C and E; hatched line) and with increasing gestational age, SC which are initially widely distributed across the epithelial surface, begin to segregate into and around this structure (Davies et al., 2009). The signals that trigger migration and confinement into the

rudimentary niche are yet to be defined but may involve genetic program that govern eye-lid disjunction and exposure to amniotic fluid (Rodrigues et al., 1987), because at an undefined post-natal time point, the embryonic limbal dome regresses and evolves into the palisades that characterise the adult human limbus. Using electron microscopy, a mosaic pattern of light, intermediate and dark-coloured cells of variable size were identified on the human foetal corneal surface up to 21 weeks gestation, an observation harmonious with cell differentiation (Sellheyer and Spitznas, 1988). Eye-lid disjunction in humans occurs 25 weeks post-conception and coincides with thickening of the corneal epithelium and change in cell morphology (Davies et al., 2009), consistent with modifications that occur in the mouse and rat (Chung et al., 1992; Song et al., 2003; Zieske, 2004) where eye-lids separate 2 weeks after birth (Wider, 1963). Developmental changes associated with positioning of corneal epithelial cells are also likely to be influenced by matricellular proteins which contribute to tissue architecture. Tenascin-C is one such protein, widely distributed across the ocular surface during the early stages of cornea formation but consolidates to the limbus in

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 5/23

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

Q5

infants and adults (Maseruka et al., 2000). Although the reason for this spatialetemporal distribution is incompletely understood, it is concordant with the appearance and subsequent confinement of primordial epithelial cells within the limbal zone (Davies et al., 2009); suggesting matrikine extracellular matrix (ECM) proteins are involved in dictating quiescence and activation programs within the rudimentary SC lair. This theory is in line Schermer's (1986) seminal finding of the differential expression of keratin 3 (K3) in the rabbit corneolimbal region and Rodrigues et al. (1987) who showed that human basal limbal epithelia are more primitive than corresponding cells in the central cornea. This regional heterogeneity and specialisation was confirmed at the cellular level through keratin expression profiling by Wiley et al. (1991), Ryder and Weinreb (1990), and elaborated by Kurpakus et al. (1992). Several independent groups later investigated a myriad of BM/ECM proteins in the infant and adult human cornea to show their differential distribution across the ocular surface (Ljubimov et al., 1995; € tzer-Schrehardt et al., 2007). Not surKabosova et al., 2007; Schlo prisingly, and in line with the abovementioned reports, tenascin-C was among the proteins restricted to the limbal border (Kabosova €tzer-Schrehardt et al., 2007; Nakatsu et al., et al., 2007; Schlo 2013) and is a protein heavily implicated in corneal cell migration (Kaplony et al., 1991). Overall these data strongly suggest that BM constituents of the limbus influence LESC maturation and motility. 1.3. Primary and secondary limbal niche structures The concept that SC are dependent on their niche microenvironment was first proposed by Schofield (1978, 1983). Niches have a propensity to occur at tissue intersections (McNairn and Guasch, 2011), they are highly defined structural units, and provide SC with a sheltering and nurturing residence. They are strategically placed within a tissue to allow SC ready access to signals that either maintain quiescence or promote activation (Watt and Hogan, 2000; Moore and Lemischka, 2006; Lane et al., 2014). The limbal transition zone that partitions the cornea from the conjunctiva is an excellent example; the evidence supporting this site as the presumed repository for LESC is overwhelming, but is certainly not definitive. The key discoveries that reinforce this proposition include; (i) the absence of a key corneal differentiation marker in basal limbal epithelial cells (Schermer et al., 1986; Rodrigues et al., 1987), (ii) the discovery of slow-cycling basal limbal but not basal corneal epithelia (Cotsarelis et al., 1989), (iii) increased growth potential of limbal compared to corneal epithelia (Ebato et al., 1987), (iv) abnormal wound-healing after partial or total limbal ablation (Chen and Tseng, 1990, 1991; Huang and Tseng, 1991), (v) restoration of severely damaged ocular surface in recipients of a limbal tissue (Kenyon and Tseng, 1989) or limbal cell-based grafts (Pellegrini et al., 1997; Shortt et al., 2007a; Di Girolamo et al., 2009; Rama et al., 2010; Baylis et al., 2011; Sangwan et al., 2011; Bobba et al., 2015), and (vi) the limbal predilection of benign (Chui et al., 2011) and malignant ocular surface tumours (Di Girolamo et al., 2013). However, there are credible reports that contravene this firmly entrenched dogma. For example, Dua et al. (2009) identified a cohort of patients with LSCD that presented with persistent islands of healthy central corneal epithelium. Other investigators have shown that upon removing the limbal annulus, central corneal epithelia are able to migrate centrifugally (i.e. towards the limbus) to self-repair the defect (Chang et al., 2008). Likewise, Barbosa et al. (2009) noted that the rabbit limbal epithelium does not participate in healing central corneal wounds even after seven consecutive debridements (de Faria-e-Sousa et al., 2010), and ablating murine LESC then cauterising the central cornea, resulted in normal reepithelialisation (Vauclair et al., 2007) suggesting LESC are not essential for wound resolution. Following from Vauclair's

5

66 observations, Majo et al. (2008) discovered oligopotent SC in the 67 central cornea with the capacity to differentiate into corneal and 68 conjunctival lineages. They showed that the limbus was minimally 69 involved in steady-state maintenance of the cornea but played an 70 integral role in damage repair, results which do not align with those 71 published one year earlier by members of their research con72 sortium (Vauclair et al., 2007). Sun et al. (2010) cautioned Majo's 73 paradigm-shifting findings by proposing that the proliferative ca74 pacity of the central cornea may have arisen from its transient 75 amplifying cell (TAC) reservoir; a plausible explanation as TAC can 76 sustain the cornea under steady-state for greater than one month 77 when physically separated from the limbus (Kawakita et al., 2011). 78 The conventional and alternative hypotheses for adult corneal 79 epithelial maintenance are discussed at length in a recent review 80 (West et al., 2015). 81 In addition to the common clinically-observed, peripherally82 located Palisades of Vogt (Goldberg, 1982; Townsend, 1991), three 83 other less well defined anatomical structures with niche-like fea84 tures have been discovered and partially characterised in man. 85 Confocal and electron microscopy assisted Shortt et al. (2007b) to 86 identify ‘limbal crypts’ (distinct invaginations of epithelial cells 87 extending from the peripheral cornea into the limbal stroma) and 88 ‘focal stromal projections’ (finger-like projections that extend up89 wards into the limbal epithelium, surrounded by small tightly 90 packed basal cells). Dua and colleagues (Dua et al., 2005; 91 Shanmuganathan et al., 2007) found unique structures dubbed 92 ‘limbal epithelial crypts’ (L-shaped invaginations of epithelia that 93 appear perpendicular to the limbal epithelia, then turn to parallel 94 the epithelium). Dua's crypts are different to Shortt's; nonetheless 95 both harbour cells that express LESC antigens, they seem to be 96 prevalent in certain sectors of the cornea and are anatomically 97 modified in patients with ocular surface inflammation (Nubile 98 et al., 2013). We have potentially identified a fourth structure, 99 and tentatively named it the ‘limbal epithelial pit’; these pits are 100 cylindrical-shaped columns lined with tightly packed basal-like 101 epithelial cells which open to the surface and protrude deep into 102 the stroma (Fig. 1GeJ). Our structure bares a resounding resem103 blance to Shortt's focal stromal projections as en face optical sec104 tions through their projections disclosed that they too have an aperture to the surface (Shortt et al., 2007a,b). Our structures Q6 105 106 therefore require careful characterisation as it is plausible they are 107 identical to the previously described focal stromal projections, but 108 due to our tissue preparation, orientation and sectioning, we have 109 imaged them from an alternative vantage point. Notably, a SC110 harbouring anatomical structure within the mammalian central 111 cornea has not been unearthed. If indeed SC reside in this isolated 112 central location, several questions come to mind; firstly, how are 113 they protected from environmental and physical insults and sec114 ondly, where do they obtain their nutrients in the absence of a local 115 microvasculature and specialised stroma, both of which seem to be 116 critical SC support networks (Lane et al., 2014; Huang et al., 2015). 117 The concept that specialised tissues harbour duel SC depots is 118 not new and has been documented for the haematopoietic system 119 (Wilson et al., 2008), small intestine (Tian et al., 2011) and skin 120 where for example SC of the interfollicular epidermis are in a 121 quiescent state and function as ‘reserve’ cells which spring into 122 action upon injury (Levy et al., 2005, 2007; Li and Clevers, 2010;  et al., 2012; Alcolea and Jones, 2014), while those 123 Mascre 124 entrenched in the hair follicle bulge emigrate upwards to actively 125 contribute to the epidermal mass and downwards to support the 126 hair follicle during homoeostasis (Taylor et al., 2000). These ob127 servations raise an important question about whether the cornea 128 too contains ‘passive’ and ‘active’ adult SC occupants and whether 129 there is a hierarchical relationship amongst such populations 130 especially in relation to regenerative stress.

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 6/23

6

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

1.4. Limbal epithelial stem cell markers The defining morphological and functional attributes of LESC include; (i) residence within a niche, primarily the Palisades of Vogt (Fig. 1A and B), (ii) their small size with heterochromatin-rich nuclei and associated melanin granules, (iii) retention of DNAlabels, indicative of their slow-cycling nature, (iv) ability to proliferate indefinitely, and (v) centripetal exodus of their progeny to maintain corneal epithelial mass. Biochemically, a series of markers are proposed to distinguish these cells from their differentiated descendants. In reality however, the discovery of a definitive LESC antigen has not been a straightforward assignment and there continues to be ambiguity regarding whether any one particular marker can discern a stem from progenitor or even an early TAC. From a biological point of view, this means we cannot pin-point their precise location and how they behave in situ. From a clinical perspective, SC grafts being prepared for therapeutic intervention may not be as effective as they should be. Meanwhile the list of biochemical markers continues to grow and although the current set of signature genes distinguishes LESC with some certainty in vivo, they may not be as reliable once these cells are ‘plucked’ from their native habitat and expanded in vitro (Vascotto and Griffith, 2006). Some of the more widely accepted LESC antigens include cytoskeletal intermediate filament proteins, for example €tzerK14 (Fig. 2) (Kurpakus et al., 1994; Harkin et al., 2004; Schlo Schrehardt and Kruse, 2005; Tanifuji-Terai et al., 2006; Figueira et al., 2007), K15 (Yoshida et al., 2006; Higa et al., 2009; Ordonez et al., 2013) and K19 (Pitz and Moll, 2002; Yoshida et al., 2006); and K3/K12 which are regarded indicators of corneal epithelial differentiation (Schermer et al., 1986; Rodrigues et al., 1987; Ryder and Weinreb, 1990; Wiley et al., 1991; Kurpakus et al., 1992). Other positive markers include mediators of WNT signalling (Figueira et al., 2007; Lu et al., 2012; Mei et al., 2014); integrins and adhesion molecules such as a6 (Hayashi et al., 2008), avb5 (Ordonez et al., 2013), and N-cadherin (Higa et al., 2009); transcription factors, DNp63a (Di Iorio et al., 2005) and C/EBPd (Barbaro et al., 2007); membrane transporter proteins, ABCG2 (De Paiva et al., 2005) and ABCB5 (Ksander et al., 2014), transmembrane proteins including LRIG1 (Nakamura et al., 2014) and Notch 1 (Thomas et al., 2007), and nerve growth receptors, TrkA (Qi et al., 2008a, 2008b) and p75 (Di Girolamo et al., 2008). Moreover, many of these factors interact at a molecular level, with K14 a transcriptional target for DNp63 (Romano et al., 2009) and the WNT-PAX6 axis dictating corneal epithelial fate (Ouyang et al., 2014). Of the abovementioned antigens K14 seems to be a robust marker but is by no means exclusive for LESC; this can be appreciated from the intense but expansive staining in basal cells of the limbal palisades, with expression receding in the peripheral zone and little or no staining detected in the paracentral and central cornea (Fig. 2A), results that were corroborated in the mouse (Fig. 2B and C). From the K14 immunolocalisation studies in human corneas, it appeared that some cells were in transit; i.e. in the process of detaching from the BM and travelling in convoy through the epithelium (Fig. 2A, arrows). Naturally, this notion is speculative, as histological evaluations can only provide ‘snapshots’ of events transpiring at a particular point in time. Nonetheless, these observations inspired us to use K14 as lineage tracing protein to follow the destiny of LESC within the murine cornea (see Section 3.2). 1.5. Corneal epithelial cell division and turnover The mammalian ocular surface is in a state of repetitive renewal. This means that cells lost from the surface epithelium must be promptly replaced by proliferating epithelial cells from below. For decades, the dynamics of this process and the precise region from

which this activity spawns has been an area of contention. The general consensus is that dividing basal corneal epithelia gradually move vertically with an out-ward trajectory as they fan out to replace sloughed cells. Cells reaching the surface are post-mitotic terminally-differentiated cells (TDC) or cells that are undergoing apoptosis (Ren and Wilson, 1996; Estil et al., 2000). An exquisitely coordinated program is required to achieve this finely tuned balancing act and any perturbation can have catastrophic consequences for corneal health and vision through a compromised epithelium. Through the assessment of semi-thin transverse plastic sections, Lamprecht's (1987, 1990) elegant morphological investigations concluded that cells positioned adjacent to the basal laminar divide both symmetrically (planar) and asymmetrically (perpendicular), meaning three possible outcomes; (i) both offspring cells can remain anchored to the BM or (ii) move into the superficial layers, or (iii) one sibling cell can remain anchored to the BM while the other detaches and is displaced vertically into the upper echelons of the epithelium. Beebe and Masters (1996) extended these observations by surveying the fate of BrdUlabelled epithelial cells in whole-mount rat corneas at successive time points. Using 3D-confocal microscopy they ascertained that more labelled cells existed in basal peripheral epithelium compared to corresponding region of the central cornea and that these cells formed clusters of 8e12 labelled cells eight days postBrdU injection. They also noted a similar pattern of synchronous cell division, with pairs of labelled daughter cells either remaining in a basal location or departing this site to emigrate and differentiate, thereby confirming Lamprecht's (1987, 1990) prior observations that a proportion of divisions are symmetrical. There is strong evidence in other organs (e.g. small intestine and interfollicular epidermis) that SC replacement follows a pattern neutral drift dynamics where clones expand and contract at random by relying on asymmetric division which predominates (80%) over symmetric (20%) mitosis (Clayton et al., 2007; Lopez-Garcia et al., 2010;  et al., 2010). If indeed this is the case, Snippert et al., 2010; Doupe then asymmetric division (vertical mitosis) results in two daughters with dissimilar phenotypic and functional attributes as one sibling is retained in the basal compartment to maintain the SC pool while the other is evicted from the niche and relocated into the superficial layers as a differentiated epithelial cell. Alternatively, following symmetric division (horizontal mitosis) both daughters can either remain basally and add to the SC reservoir or are ousted from this location as a pair of irreversibly committed cells as they journey to their final destination (Fig. 3). Defining the factors that determine SC fate in this process continues to be a challenge but it is likely that asymmetric mitosis is reserved for SC within the limbus while symmetric division predominates in more differen~ ozledo and tiated basal cells of the central cornea (Castro-Mun  mez-Flores, 2011). Go Proliferation of the corneal epithelium is influenced by circadian cycles (Scheving and Pauly, 1967; Burns and Scheving, 1975; Scheving et al., 1978; Haskjold et al., 1989; Lavker et al., 1991). However, the specific intrinsic factors that contribute to epithelial turnover have not been clearly defined. Glycogen is appreciably elevated in regenerating corneal epithelia, especially after wounding (Thoft and Friend, 1977), and stores of this biochemical fuel, as well as oxygen that dissipates from neighbouring limbal capillaries and tear-derived nutrients, are postulated to provide the energy to sustain this process (Lemp and Mathers, 1991). It is well accepted that the proliferative activity responsible for ceaseless epithelial turnover emanates from basal cells. Buschke et al. (1943) were among the earliest investigators to report that 1 in 250 cells within the basal and first suprabasal corneal epithelial layer were in mitosis, and after treatment with the mitotic inhibitor colchicine, they estimated that the duration of cell division in the rat cornea

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 7/23

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

7

Fig. 3. Model for symmetric and asymmetric division in the mammalian cornea. When a basal LESC (red) undergoes mitosis (A) it can pursue one of three fates. It can divide symmetrically and both daughter cells remain within their lair as SC (B), both daughter cells can detach from the BM and migrate vertically towards the superficial layers as differentiated epithelial cells (C) or divide asymmetrically producing a stem and committed daughter cell (D). The percentages given predict the rate of even and uneven division , 2010). Abbreviations; BM, basement membrane; SC, stem cells; TAC, transient amplifying cell; TDC, terminally differentiated cell. (Clayton, 2007; Snippert, 2010; Doupe

was approximately 1 h. Hanna and O'Brien (1960) corroborated these findings by injecting the DNA precursor 3H-thymidine into the anterior chamber of mouse, rat, guinea pig and dog to successfully traced the label over time in corneal sections using autoradiography. They observed; (i) greater mitotic activity in the peripheral compared to the central cornea with 1 in 75 cells labelled, (ii) groups or pairs of cells were labelled and displaced from the basal region into the superficial layers, and (iii) labelled cells were occasional present and in transit after 4 days. Bertalanffy and Lau (1962) confirmed that the replacement rate of epithelia in the rat cornea was one week and a similar weekly rate was recorded in organ cultured human adult corneas with 1 in 60 basal epithelial cells undergoing mitosis (Hanna et al., 1961). Cenedella and Fleschner (1990) used a similar approach however rats were pulse-injected with 3H-thymidine, and after recovering total DNA they noted increased label in the peripheral compared to the central corneal epithelium. From this data it was deduced that the rat corneal epithelium was completely replaced within 2-weeks, concurring with reports in rabbits (Haddad, 2000). Notably, Haddad (2000) identified rare DNA label-retaining cells (LRC) in the central cornea 3 months post-labelling and proposed that the proliferative capacity of centrally located cells was sufficient to guarantee corneal epithelial renewal under steady-state. These

findings contrast those published by Lavker et al. (1991) who noted that the central corneal epithelium was mitotically more active than the peripheral region. Yet others have shown no difference in mitotic activity between peripheral and central corneal epithelia (Szerenyi et al., 1994). These apparent inconsistencies in cell division and measures of epithelial turnover may reflect diverse labelling techniques and DNA-tagging agents used as well as interspecies and strain-specific differences related to corneal and limbal landscapes (Chan et al., 2004). However, in recipients of full thickness corneal grafts, donor epithelium can persevere for months before it is replaced by recipient cells (Khodadoust and Silverstein, 1969; Silverstein et al., 1970; Kaye, 1980; Lagali et al., 2009; Catanese et al., 2011), providing firm clinical evidence that epithelial turnover is slower than previously thought. Identifying the precise location of DNA label-retaining cells (LRC) has been another major limitation, particularly in the absence of a fiduciary anatomical limbal landmark in rodents. This was partially addressed by staining corneal cells in the green fluorescent protein (GFP) transgenic mouse with the fluorescence nuclear dye DAPI to demarcate a 100e200 mm annular perimeter which was found to contain the highest density of BrdUþ LRC even after a 10-week chase period (Zhao et al., 2009). Interestingly, 1 in 24 (4.2%) basal cells were LRC and were heterogeneously distributed

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 8/23

8

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

within this zone. This data was corroborated in BALB/c mice where approximately 3.6% of basal cells were LRC with the highest density found in the superior/inferior compared to nasal/temporal quadrants (Pajoohesh-Ganji et al., 2006). Cotsarelis et al. (1989) infused mice with 3H-thymidine via osmotic mini-pumps; their seminal study was the first to identify a subpopulation of basal limbal epithelial cells that retained DNA label and could be stimulated to proliferate in response to wounding. Since no such cells were found in the central cornea, it was rationalised that they were the precursor cells of the corneal epithelium. Closer inspection of these cells revealed that of the total basal cells, 200e400 were LRC that were exclusively located in the limbal fringe (Pajoohesh-Ganji et al., 2006; Zhao et al., 2009), inferring that a label-retaining stem/progenitor cell is responsible for renovating several circumferential degrees of corneal epithelium. Notably, however not all LRC are SC (Braun and Watt, 2004), therefore this interpretation should be considered with some reservation. 1.6. Epithelial cell migration in the normal and wounded cornea The controversy surrounding the rate at which the corneal epithelium turns over under steady-state is partially due to the difficulty in visualising this phenomenon in a living organism. Although the prevailing dogma is that peripherally located SC drive this process, this theory did not gain credence until DNA-labelling studies exposed the location of slow-cycling cells whose progeny, visualised by diluted label, were mobilised from the periphery under physiological conditions and in response to wounding (Cotsarelis et al., 1989). Corneal epithelial wounds provide a valuable model to survey cell fate and function because under these conditions cell movement is heightened, rendering the process easier to visualise and monitor. The possibility that cells moved about in wounded corneas was originally interrogated by Peters (1885) in frogs and later confirmed by Ranvier (1898) who described a ‘land-slide’ phenomenon of cells ‘filling-in’ the defect. Ida Mann's (1944) classic wound-healing studies in Dutch coloured rabbits were among the first to document cell movement as visualised through melanincontaining peripherally-located corneal epithelia. Her elegant experiments demonstrated that after a mild abrasion (i.e. without damaging the substancia propria), the limbal annulus in these animals became irregular and pigmented cells confined to this region and began to travel in a unidirectional manner, i.e. from the edge of the wound towards the inner part of the cornea, but not in the opposite direction i.e. conjunctival-bound. This ‘directed’ movement contrasted with the chaotic displacement of cells in rabbits that received a severe corneal injury. Buschke (1949) extended these findings by monitoring the closure of small epithelial defects provoked by a pin-prick injury prior to administering colchicine. During the first hour post-injury no change was noted in the wound bed, but within 2 h basal cells began to re-orientate with their long axis radially directed towards the defect and after 3 h wounds were sealed with cells having travelled at a rate of 0.25 mm/min. Although the origin of the cells that contributed to the initial phase of injury repair was not disclosed, DNA-label incorporation studies in superficial and penetrating wounds suggested each epithelial layer supplies cells to cover the fissure (Hanna, 1966). However, it is likely that the first cells entering the wound crater are suprabasal wing cells with minor contributions made by basal cells (Kuwabara et al., 1976). At this stage, proliferating cells are found some distance from the wound (Hanna, 1966; Kuwabara et al., 1976) and likely provide the impetus to ‘push’ the epithelial mass towards the defect, which in the mouse has been estimated to occur at 30e60 mm/h (Buck, 1979). Gipson and Keezer (1982) examined cell movement in wounded organ cultured rat corneas; their data

supported the hypothesis that leading edge cells ‘tow’ cells from behind. But to do so effectively they require traction which is achieved through de novo synthesis of proteins such as hemidesmosomes (Gipson and Kiorpes, 1982) and cell-surface glycoproteins (Gipson et al., 1984), both which cess to be produced upon wound closure. It is now well accepted that corneal epithelial healing occurs in 3 phases; an initial period of latency, followed by an intermediate period where cells migrate over the denuded zone, and a final proliferation stage to reconstruct the epithelial tiers. The second phase is thought to occurs via cells ‘sliding’ into the wound bed (Danjo and Gipson, 2002; Zhao et al., 2003). However an alternative mechanism has been proposed whereby cells ‘crawl’ or ‘leap-frog’ one another to reach the leading edge of the wound (Kuwabara et al., 1976). The energy expended to execute this process comes from glycogen stores (Kuwabara et al., 1976; Kinoshita et al., 1982) and the mode of movement is via lamellipodial extensions and through cell-to-cell junctional proteins and actin microfilament cables which are accentuated in cells at the wound forefront (Danjo and Gipson, 1998; Li et al., 2012). Computational modelling incorporating data from many of the abovementioned studies, as well as the involvement of LESC, represents a valuable tool for predicting cell kinetics within the cornea under wound healing conditions (Gaffney et al., 1999). 1.7. Clinical observations of corneal epithelial cell migration Clinical evidence in humans and laboratory animals suggests the limbus is the most likely residence for cells with regenerative activity. This notion was affirmed when guinea pig corneas were wounded adjacent to a pigment-bearing limbal epithelium and the melanin associated with these cells used as a visible marker to track their destiny from the limbus to the site of injury (Davanger and Evensen, 1971). In heavily pigmented individuals, the papillae which harbour these cells are clinically obvious as arching stria from which limbal epithelial cells egress to partake in centripetally directed movement, particularly after trauma (Mann, 1944; Cowan, 1963; Davanger and Evensen, 1971). In lightly pigmented individuals moving corneal epithelia are difficult to envisage; Auran et al. (1995) used scanning specular confocal microscopy to observe this movement and to measure the rate of centripetal migration of basal epithelial cells. Although these recordings were short, limited to one subject and time consuming, their estimate of 30 mm/day corresponded well with approximations in mice (Buck, 1985; Nagasaki and Zhao, 2003). If the cornea is damaged then irrigated with special dyes (e.g. fluorescein or Rose-Bengal) to discern the defect, wound closure in patients can be readily visualised. Dua and Forrester (1987) used fluorescein to monitor patients with corneal abrasions, noting wounds re-epithelialised in a similar manner irrespective of the type of injury. However, healing was not uniform in that neighbouring cell sheets converged in a radial manner to form interesting geometric shapes including ‘Y’ or ‘double Y’ patterns, which eventually developed into a whorl. Corneal epithelia from patients with cornea verticillata, a condition first described by Fleischer (1910), further documented by Denden (1966) and François (1966), and comprehensively characterised by Bron (1973), pursue similar migratory routes. This condition is sometimes referred to as ‘Fleischer's vortex’ or ‘Vortex keratopathy’, and can develop from toxicity to systemic or topically applied medications such as amphiphilic cationic agents including amiodarone, chloroquine, suramin and tamoxifen (Noureddin et al., 1999; Hollander and Aldave, 2004). These agents induce lipidoses and are suspected to reach the cornea via the limbal vasculature and become concentrated over the ocular surface by tear film proteins or by contact lens wear (D'Amico et al., 1981; Haug and

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 9/23

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

Friedman, 1991). Chloroquines are particularly interesting because they are melanin-binding and can therefore affect basal limbal epithelia (Pulhorn and Thiel, 1976). Once lipid deposition occurs, cells become opaque and their journey to the midline of the cornea is revealed through slit-lamp examination as golden/brown streaks which disappear once medication is discontinued. Similar spiralshaped epithelial keratitis develops in patients treated with combination antibiotic and steroid preparations after penetrating keratoplasty (Mackman et al., 1983), as well as those on long-term topical steroid therapy (Dua et al., 1993) or in rigid gas-permeable contact lens wearers where the frictional forces of inserting and removing the lens is thought to accelerate corneal epithelial turnover (Dua and Gomes, 2000). Collectively, these data provide additional clinical evidence that the limbus spawns SC clones that progressively travel towards the central cornea. Fabry's dystrophy, independently described by Fabry (1898) and Anderson (1898) is also characterised by vortex keratopathy. This is a genetically inherited metabolic condition caused by a deficiency in the lysosomal enzyme a-galactosidase A, which results in the accumulation of lipids throughout the body including the corneal epithelium (Font and Fine, 1972). Corneal manifestations of Fabry's disease are indistinguishable from drug-induced vortex keratopathy. The striking similarities in epithelial markings and radial spoke arrangement led Bron (1973) to proposition that both conditions evolve through a similar process. The cell ribbons within these vortexes usually turn in a clockwise direction (Bron, 1973; Dua et al., 1993; Dua and Gomes, 2000) but occasionally patients present with vortexes that twist in different direction in either eye. It is currently not known whether the direction of gyration is completely random or influenced by inertial forces such as those exerted by the Coriolis Effect which accounts for differences in wind and fluid dynamics in the Southern and Northern hemisphere (Persson, 1998). 1.8. Migration of conjunctival epithelial cells The conjunctival epithelium is the cell type that shares a common embryological origin and somatic expression of the ocular morphogen PAX6 with the corneolimbal epithelium (Koroma et al., 1997). The close relationship between these two epithelia has been highlighted in recent studies where for example, ‘compound niches’ have been detected in the mouse limbal border which house K12þ goblet cells (Pajoohesh-Ganji et al., 2012) and clusters of corneal-like epithelia are seen to reside ectopically within the human conjunctiva (Kawasaki et al., 2006). Over 60 years ago Friedenwald (1951) proposed that corneal epithelial defects heal from the conjunctiva. Shapiro et al. (1981) subsequently provided supportive histological evidence and posited that conjunctival cells ‘transdifferentiate’ into a corneal phenotype after wound resolution. Others have suggested the limbus is a depot for bipotent SC that give rise to cells of both lineages, a claim supported by observations that conjunctival cells surge outwards from this region at a rate of 9e13 mm/day in the x and y-plane (Zajicek et al., 1995; Pe'er et al., 1996). Although these findings challenge the wellentrenched premise of LESC unipotency, there may be reasonable explanations. For example, long-term follow-up studies in animal models of partial (Chen and Tseng, 1990, 1991) and total (Huang and Tseng, 1991) limbal ablation have confirmed that epithelia of a conjunctival phenotype drift across the limbal margin to envelop the cornea. However, in less severe wounds the epithelia covering the defect may temporarily display biochemical features of conjunctival cells until limbal-derived TAC take control during the reconstruction phase (Kruse et al., 1990; Moyer et al., 1996); certainly these observations refute the transdifferentiation theory. It is now widely accepted that this expansive epithelia is

9

intrinsically divergent from its neighbour (Wei et al., 1996) and contains its own SC but in a less well-defined niche (Wei et al., 1995; Wirthschafter et al., 1999; Pellegrini et al., 1999). Nagasaki and Zhao (2005) used a GFP cell-tracking model to demonstrate that normal conjunctival epithelial cells are stationary in the lateral direction but able to move vertically through the cell layers in a similar fashion to how keratinocytes replenish the epidermis (Potten, 1975; Blanpain and Fuche, 2009). However, when the limbal barrier is compromised due to loss of LESC or aberrant functional disturbances in these cells or destruction to the limbal anatomy, centripetal insurgence of an inflamed fibrovascular conjunctiva occurs, which is a classical feature of pterygia (Fig. 4A and B) (Davanger and Evensen, 1971; Tseng, 1989; Chui et al., 2011), as well as partial and total LSCD (Fig. 4CeF, respectively) (Shortt et al., 2007a; Rama et al., 2010; Baylis et al., 2011; Sangwan et al., 2011). 1.9. Intrinsic and extrinsic factors implicated in corneal epithelial cell migration The mechanisms and molecules that direct centripetal flux of corneal epithelia have only been speculated, so there is a pressing need to delineate these factors in order to fully appreciate how the cornea sustains cell mass during steady-state, wound-healing and disease. Firstly, the actions of eyelids should be considered, as frictional forces are predicted to be greatest in the central cornea, implying centrally located cells are exfoliated at a greater rate than marginal cells (Mathers and Lemp, 1992). This renders superficial epithelia in the central human cornea to be smaller in size than those in the superior and inferior region. In considering this paradigm, if the blinking rate in humans is 20 blinks/min and the tear volume turnover is 0.31 ml/min (Mathers and Daley, 1996), then tear flow due to blinking is also expected to influence shear stress on the ocular surface. Clues as to how cells travel through the epithelium have come from studying actin microfilaments which are concentrated in the basal region of actively migrating cells (Gipson and Anderson, 1977), and ultra-structural studies that have indicated preferential radial arrangement of hemidesmosomes in basal epithelia (Buck, 1982), suggesting polarisation of these junctional proteins plays a key role in directionality of migration, at least under physiological conditions. After wounding, and depending on whether the BM is damaged, hemidesmosomes are not properly aligned and cell motility not uniformly directional until reepithelialisation is complete. Cell migration also involves change in cell shape and transient adhesion and de-adhesion along with the assembly and disassembly of integrins. Integrins are involved in stabilising connections between cells and ECM components. However, their role in corneal epithelial flux during homoeostasis is still unclear (Stepp et al., 1993, 1996), although after injury and wound recovery specific members are induced (Stepp and Zhu, 1997; Sta Iglesia et al., 2000; Blanco-Mezquita et al., 2013), some via mitogen activated protein kinase (Seomun and Joo, 2008) and focal adhesion kinase (Dreier et al., 2012) signalling cascades. Paradoxically, integrins are also involved in tethering LESC to their niche (Pajoohesh-Ganji et al., 2006), thereby preventing their exodus and subsequent differentiation. Other major drivers of centripetal migration in the cornea are likely to be chemotactic factors; a theory first put forth by Cowan (1963) and fortified through the recent identification of chemokine and chemokine receptor cross-talk (Xie et al., 2011; Ordonez et al., 2013) and neurotrophic factor gradients (Jones and Marfurt, 1996; Lambiase et al., 2000; You et al., 2000; Blanco-Mezquita et al., 2013). These neuronal effector molecules likely dissipate from sympathetic nerves which fan out circumferentially to

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 10/23

10

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

Fig. 4. Limbal border protection. Schematic diagrams (A, C, and E) and corresponding clinical images (B, D, and F) of eyes affected by a pterygium (A and B), or partial (C and D) and total (E and F) LSCD. In the schema, healthy SC (depicted in yellow) are distributed around the limbal circumference (red ring). When these cells are depleted or become functionally perturbed (shaded black), the limbal barrier is breached by an inflamed vascularised pannus of conjunctival tissue that migrates centripetally over the cornea to obscure vision.

innervate the corneal stroma with branches piercing the basal lamina and endings infiltrating basal and suprabasal epithelia (Masters and Thaer, 1994). In both man (Patel and McGhee, 2005) and mouse (McKenna and Lwigale, 2011) the sub-epithelial nerve plexus forms a radial swirl which is in a dynamic state; firstly during embryogenesis when corneal innovation appears at E16.5 and becomes pronounced thereafter, but is certainly not complete until the 3rd post-natal week in the mouse (McKenna and Lwigale, 2011), and secondly during adulthood where nerve branches migrate centripetally at a rate of 26 mm/week (Patel and McGhee, 2008). The centripetal extension of neuronal axons from the limbus to the apex of the cornea coincides with radially migrating epithelia, suggesting nerve fibres provide guidance cues for cell movement. During ageing terminal nerve endings are lost particularly from the periphery (Leiper et al., 2009). Whether this impacts LESC activity is not known, but it is tempting to speculate that neuronal-derived factors, to which these cells respond (Qi et al., 2008a, 2008b; Di Girolamo et al., 2008), are necessary for maintaining progenitor cells phenotype and function. This is not an unreasonable proposition as defects in innervation are known to interrupt corneal epithelial cell migration (Leiper et al., 2009). Substrate stiffness especially within the SC's immediate microenvironment is known to dictate cell lineage specification, anchorage, motility, and differentiation (Engler et al., 2006). Therefore, it comes to no surprise that compositional differences in the ECM across the cornea (Ljubimov et al., 1995; Kabosova et al., €tzer-Schrehardt et al., 2007) will affect tissue rigidity 2007; Schlo and compliancy (Hjortdal, 1996). This implies that mechanotransductional forces may also be involved in directing LESC migration and differentiation by establishing a ‘centripetal-stiffness’ gradient across the cornea (Eberwein and Reinhard, 2014). Corneal epithelial cells sensing these changes will respond through the expression of transcription factors that influence matrix adhesion and cytoskeletal contractile proteins required for cell

movement (Foster et al., 2014). Knowledge gained from these discoveries will have far reaching ramifications; for example in influencing the development of scaffolds which emulate the native niche and its topography for ex vivo expansion and therapeutic delivery of LESC (Levis et al., 2013). Others have suggested limbalderived epithelial clones travel on a permanent ‘rail network’ but this hypothesis has not been substantiated as the pattern of centripetal flux can deviate significantly from the original tracks after epithelial debridement (Hayashi et al., 2012). Curiously, when human stromal niche cells and limbal epithelial cells are co-cultured, a spiral migratory pattern develops, postulated to arise from stromal cell-derived signals which direct self-organisation of 3D spheroids that resemble limbal crypts (Mariappan et al., 2014). Corneal epithelial movement is also influenced by electric (Soong et al., 1990; Zhao et al., 2012; Li et al., 2012) and magnetic (Dua et al., 1996) fields as well as gravity (Miri et al., 2012); however the effects of these forces have only recently been investigated in vivo (Ghaffarieh et al., 2012). 2. Introduction to cell and lineage tracing Lineage tracing is a method for monitoring the destiny of a cell and its progeny in tissues. Its inception arose from the discovery that cells were derived from pre-existing cells through division (Virchow, 1858). This seminal finding likely inspired 19th Century developmental biologists including Whitman (1878, 1887) to follow the fate of dividing cells in the leech embryo, and Wilson (1892) who later coined the term ‘cell-lineage’, to monitor cell division in embryos of annelids, ascidians and molluscs. Sturtevant (1929) later combined cell-lineage and gene-function to create mosaics for studying Drosophila development. However, it was decades later when Sturtevant shared his observations with colleagues that an ontogenic fate map of the fly embryo was charted (Garcia-Bellido and Merrian, 1969). Because of their simple structural

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 11/23

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

organisation and developmental program, nematode embryos from Caenorhabditis elegans have been studied for over a century, originally in fixed specimens but more recently the fate of every living cell was mapped by observational time-laps Nomarski optics and information about cell interactions, movement and differentiation status was resolved (Sulston et al., 1983). The goal of effective fate mapping is to stably introduce a reporter into a cell so that when it moves, divides or differentiates the marker is faithfully passed on to its daughters and information about the cell's phenotype, movement, and function deduced after pin-pointing its location. If a tag is introduced into a SC and permanently transduced to its daughters in a lineage restrictedmanner, then a marked cell can be tracked as an expanding clone indefinitely. If the label is stably introduced into a differentiated daughter, for example a TAC, then it will eventually disappear from the trace. The advantage of lineage analysis using permanent genetic markers is that molecular, biochemical and functional determinants of labelled cells within a clone can be examined while they remain confined to the 3D native habitat that spawned their existence. This bypasses the disadvantages that complicate ‘artificial’ culture and transplantation systems where cells are introduced into an alien habitat, often after having undergone extensive ex vivo manipulations. 2.1. Methods to study cell and lineage tracing Early lineage tracing involved unperturbed experimental models that used ‘direct’ observational recordings of living organisms (Whitman, 1887; Wilson, 1892). In these studies, continuous monitoring could only be achieved in transparent organisms with a limited number of cells. In the early 20th century, Conklin (1905) followed the fate of cells into different lineages during embryonic development as eggs of some ascidian species are conveniently pigmented. In the early 1920s and 1930s ‘vital dyes’ were introduced as a means of labelling cells in amphibian (Vogt, 1929) and ascidian (Tung, 1932) embryos. However, these techniques were fraught with limitations as the dyes either disappeared from the trace through replicative dilution or contaminated the trace by diffusion into neighbouring cells. The advent of non-toxic lipophilic carbocyanine dyes, such as DiO and DiI that label cell membranes, was a significant technological advance as cell division in ascidian (Zalokar and Sardet, 1984) and sea urchin (Summers et al., 1996) embryos was more readily visualised. Summers et al. (1996) compared this methodology with microinjecting fluoresceinconjugated dextran (a high molecular weight sugar precluded from diffusing through gap junctions) and obtained comparable results. Other large molecules such as horseradish peroxidase (HRP), have been injected into cells of developing mouse embryos to successfully fate map morphogenetic cell movement during germ-layer formation (Lawson et al., 1991). It is noteworthy that trauma associated with microinjections may influence cell fate within blastomeres. SC are endowed with many characteristics that distinguish them from their differentiated descendants, one of which is their slowcycling nature under steady-state, a feature that has long been exploited to mark their location within tissues. Through repeated ‘pulse’ administration of the DNA-binding agent 3H-thymidine, and after a lengthy ‘chase’ period, label-retaining slowly-cycling cells were found amongst keratinocytes of the mouse skin (Bickenbach, 1981), hair follicle (Cotsarelis et al., 1990) and cornea (Cotsarelis et al., 1989). Although these cells possess SC-like qualities, the methodology used to identify them, prevent their isolation and expansion for further molecular investigations. When tissues are complex and inaccessible, direct injection of tracer elements may not be a viable option and the introduction of a genetic marker by

11

viral transfection may be more advantageous as a lineage beacon. Retroviral vectors carrying reporter genes can be effectively integrated into the genome of host mitotic cells and transduced to their progeny (Cepko et al., 1995). Sanes and colleagues (1986, 1989) pioneered the retrovirus-mediated gene transfer technology to trace cell-lineage in vitro and in vivo after disabling the viral replication machinery. They developed a recombinant retrovirus bearing the Escherichia coli-derived lacZ gene which when integrated into the cells' genome could be identified histochemically for the translated protein b-galactosidase (b-Gal) after applying the substrate X-galactose. Price et al. (1987) used a similar approach to lineage trace rat neural precursor cells after an intra-retinal injection of the construct into new born animals. More recently nonviral transfection of stem/progenitor cells has been documented with minimal effects on cell viability, proliferation and differentiation (Tinsley et al., 2006), problems readily encountered with retroviruses. One notable confounding concern in relation to faithfully mapping the destiny of marked cells irrespective of the technique employed is the rare but potential occurrence of spontaneous cell fusion which may cause a marker to be passed from the original cell to an adjacent cell of the same or different lineage (Sanges et al., 2011). Transplanting individual cells (Spemann and Mangold, 1924) or cells dissociated from tissues (Jensen et al., 2010) is an alternative means to study cell fate. If grafted cells harbour a permanent label they can be readily distinguished within the host, rendering this a major advantage. For example, cells from a GFP mouse can be transferred into a wild-type recipient; this has been performed using luminescent bone marrow-derived cells which emigrate to tissues as remote as the cornea (Sonoda et al., 2005) and retina (Müther et al., 2010). The obvious disadvantage of this model is the un-specified lineage signature of the trafficking GFPþ cells. In addition, the process of grafting can itself elicit a wound-healing and/or immunological response that interferes with fate determination especially when monitoring under steady-state. Another limitation of this procedure may be the requirement to re-wound the organism to assess cell behaviour, unless of course the site is readily accessible or cadaver tissue is harvested for analysis. The discovery of GFP from the jellyfish Aequorea victoria (Shimomura et al., 1962) and related fluorophores, and their incorporation into genomes of many species has significantly advanced and enhanced fate surveying. Meilhac et al. (2009) microinjected mRNAs encoding red fluorescent protein (RFP) into individual cells of a mouse blastocyst and using time-lapse microscopy noted that embryos developed normally and the formation of primitive endoderm involved cell-sorting and positional movements necessary for lineage segregation. Lineage tracing by genetic mosaics is another technique that has progressed our understanding of how cells partake in tissue and organ development. Landmark studies in Drosophila gynandromorphs [flies carrying a mixture of male (X0) and female (XX) tissue] have been instrumental in facilitating the construction of developmental fate maps on how progenitor cells in embryos contribute to different structures later in life (Garcia-Bellido and Merrian, 1969). Mouse mosaics are useful tools for investigating cell dynamics and tissue organisation, and can be generated by crossing a strain that contains a reporter with one that does not. This results in a jumbled assortment of marked and unmarked cells, useful for delineating cell proliferation and the origins of expanding clones (Giangreco et al., 2009). 2.2. Lineage tracing by genetic recombination Since its first use in mice (Lakso et al., 1992), genetic recombination has been refined and is now regarded the archetypal

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Q7 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 12/23

12

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

approach to permanently mark cells, allowing their dynamic behaviour to be investigated in a living organism. This technique works through site-specific recombinase enzymes typically derived from either the yeast Saccharomyces cerevisiae (Flippase; FLP-FRT) (Harrison and Perrimon, 1993) or the bacteriophage P1 (Cre-loxP) (Nagy, 2000). These enzymes must be induced to initiate recombination, ultimately resulting in the expression of a conditional reporter which is typically driven by a ubiquitous, lineage, or tissue-specific promoter. In a robust system where recombination is highly efficient, a permanent cell trace can be recorded. If SC are targeted, then their indefinite contribution to self-preservation and tissue-renewal can be gleaned throughout life. Typically the FLPFRT site-specific recombination system is used to pursue cells in Drosophila (Harrison and Perrimon, 1993) with single or multicoloured ‘Flybow’ (Hadjieconomou et al., 2011) or ‘dBrainbow’ (Hampel et al., 2011) transgenic cassettes available as powerful lineage tracking tools. The mammalian Brainbow system is perhaps the most elaborate; originally designed for neural circuit mapping in the mouse brain (Lichtman et al., 2008; Livet et al., 2007), modifications and variants of this technicolour approach have been developed (Brainbow 1.0, 1.1, 2.0 and 2.1) and used to investigate cell fate in different species e.g. Xenopus (Satoh et al., 2005) and Zebrafish (Pan et al., 2013), and in specific murine organs (Vasioukhin et al., 1999; Indra et al., 2005; Snippert et al., 2010;  et al., 2012; Schepers et al., 2012; Bardehle et al., 2013; Di Mascre Girolamo et al., 2015). Visualising the interactions that take place between cells and their native habitat can be particularly informative; especially to unearth the location of SC and to enhance our knowledge of how the niche influences their behaviour. The advent of multicolour genetic recombination strategies, such as the Confetti mouse, along with powerful optical imaging platforms has been instrumental in locating and distinguishing cells, and verifying their identity and activity in living organisms. This system operates by enabling the stochastic combinatorial expression of a transgene, in the case of Brainbow 2.1, two tandem palettes containing four spectrallydistinct fluorescent proteins (FP) separated by a pair of loxP sites. Cre-mediated DNA excision or inversion at these sites endows targeted cells the ability to express one or a combination of two FP indefinitely (Fig. 5). In cells that harbour a copy of the reporter on each allele (homozygous), Cre-mediated recombination results in the expression of one or two FP of the same or different colour, resulting in up to 10 possible combinations (Livet et al., 2007; Lichtman et al., 2008) (Fig. 5C). If one copy of the Brainbow cassette is expressed in a cell, then a single FP will be transcribed resulting in the production of 4 primary colours. If cells harbour the construct and for whatever reason, recombination does not proceed, they will not be illuminated (Fig. 5C). This is an advantageous system as individual cells and the boundaries between adjacent colonies that expand from a single cell can be discriminated on the basis of differing hues. Due to the direction of intervening LoxP sites, Cre is given a choice of whether to excise or invert DNA sequences; hence the FP expressed is entirely stochastic, akin to rolling a ‘molecular dice’. An additional level of complexity is engineered into this elegant system, by directing FPs to different sub-cellular locations; YFP and RFP to the cytoplasm, GFP to the cells' nucleus and Cerulean (CFP) to the plasma membrane (Fig. 5B and C). Using the neuron-specific Thy1 promoter to drive the Brainbow transgene, Livet and associates (2007) successfully circuit-traced neurons in the complex mouse brain, by monitoring their fate in cadaveric tissue or by surgically opening a tissue window, which itself can disturb the microenvironment. Moreover, due to the multitude of colours produced, spectral un-mixing may be necessary for optimal optical separation (Livet et al., 2007; Weissman et al., 2011; Ducros et al., 2009; Di Girolamo et al., 2015).

In order to follow a cell and its derivatives, the reporter should be targeted to a ubiquitously expressed regulatory sequence; in the mouse the Rosa26 locus is ideal since it acts like a ‘gene-trap’ (Soriano, 1999) (Fig. 6). Furthermore, inserting a CAG promoter sequence within the construct enhances ubiquitous expression (Snippert et al., 2010), and adding a Neomycin (Neo) cassette upstream of the transgene provides a transcriptional ‘road-block’ (Fig. 6B). Temporal control of the transgene is achieved by fusing Cre recombinase to a mutated human oestrogen receptor (ER), a molecular complex that is maintained inactive in the cell's cytosol by heat-shock proteins. This modification renders its natural ligand 17b-oestradiol ineffective at signalling through this receptor but when the synthetic ligand tamoxifen (TAM) or its active metabolite 4-hydroxy-TAM is introduced, CreER is released from its chaperone proteins, facilitating it's translocation to the nucleus where Cre acts freely on loxP sites to initiate recombination (Fig. 6). To spatially control expression and to drive the transgene in a lineagerestricted, stem/progenitor cell-specific manner, an appropriate promoter targeting these cells must be inserted upstream of these critical elements. Vasioukhin et al. (1999) used one such construct under the control of the K14 promoter and because the long-term fate of epidermal keratinocyte was observed several months after TAM treatment, it was surmised that a SC population was being traced. Similar studies were conducted with the K14CreER-RosaYFP strain where persistent columns of K14-marked cells emerged from  et al., 2012), and cells of a similar the follicular epidermis (Mascre lineage gave rise to mammary gland development (van Keymeulen et al., 2011). The K14CreER-‘Rainbow’ mouse was also used to establish the epidermal SC lineage responsible for regenerating the murine digit tip (Rinkevich et al., 2011). Using a construct that was knocked into the Lgr5 locus and the Brainbow line, Snippert et al. (2010) carried out fate mapping of individual intestinal SC to show that initially each crypt was populated with multicoloured clones but with time a single clone predominated; in other words, the trace became monochromatic, suggesting homoeostasis in the murine intestine is supported by symmetric division of Lgr5þ SC. Selecting the Cre-driver is important as ‘off-target’ expression can occur (Murray et al., 2012) and promoters such as K14 and Lgr5 can push transgene expression in cells of multiple tissues (Vasioukhin  et al., 2012; van Keymeulen et al., 2011; et al., 1999; Mascre Rinkevich et al., 2011). Other considerations include whether tissue fixation and/or processing bleaches reporter proteins, and whether live animal imaging is a viable option for tracking cells (for additional commentary see Section 3.2). 3. Tracking corneal epithelial cell movement In an attempt to locate the regenerative zone of the corneal epithelium and track cell movement, Buck (1985) tattooed the peripheral murine corneal epithelium with India ink and thorium dioxide and observed the centripetal motion of dye-marked cells at 17 mm/day over 7 days. Later, Nagasaki and Zhao (2003) engineered a mouse line that ubiquitously expressed GFP and noted that 1 week after birth, mice displayed uniform GFP expression across the ocular surface but after 2 weeks, when eye-lids parted, a mosaic arrangement of GFPþ cells emerged that progressively developed into evenly distributed curvilinear spirals that migrated centripetally at a rate of 26 mm/day over the ensuing seven weeks. The patchwork and pin-wheel spirals that developed in many of the above-mentioned murine models have also been observed in the chimeric GFP rat (Iannaccone et al., 2012). Mystery continues to shroud this pattern of cell expansion and movement, especially since the tissue in which these spokes form is stationary and the shortest route to the apex of the cornea is a straight line along the meridian. However, if there is inherited uncertainty about the

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 13/23

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

13

Fig. 5. Stochastic recombination using the Brainbow 2.1 construct. The Brainbow 2.1 transgene is a powerful tool for tracing cell fate. This multicolour reporter harbours two randomly-inserted invertible palettes containing four fluorescent proteins (FP) each segregated by a pair of loxP sites (A). There are four possible recombination outcomes when the reporter is expressed on a single allele (B). Recombination occurs randomly and is initiated by an initial pulse of TAM which activates Cre through cleavage at specific loxP sites (34nucleotide sequences). DNA is excised between loxP sites of the same orientation and inverted between loxP sites of opposite orientation (B). Retracing can also be performed if mice are given a second pulse of TAM. In this scenario, the cassette can randomly ‘flip’ changing the colour of a GFPþ cell to one that begins to express YFP (B, 1), and so on (B, 2e4). In homozygous mice that carry one copy of the transgene on each allele, stochastic recombination can result in at least 14 colours (C) (Lichtman et al., 2008). If recombination occurs on both alleles, 10 colour combinations are produced (C, 1e10). If recombination proceeds on one allele, 4 primary colours are produced (C, 11e14), and if there is no recombination on either allele, K14þ cells will not be illuminated (C, 15). Notably, FP are directed to different subcellular compartments including the cell membrane (CFP), cytosol (YFP and RFP) and nucleus (GFP) (B and C).

direction of movement or cells encounter a force, then curvilinear streaking can develop (see Section 1.7). Epithelial movement has been monitored in chimeric LacZ Xinactivation mosaic mice which develop a series of blue and white radial stripes after corneas are extracted and stained ex vivo (Collinson et al., 2002). Notably, and in agreement with other reports (Nagasaki and Zhao, 2003), the striping configuration was not complete until about 5 weeks of age. If one postulates that a stripe comprises a clone of cells derived from a peripherally located progenitor, then the number of stripes should correspond to the number of SC. However, a significant limitation of Collinson's model is that one or more adjacent LacZ-expressing SC may have expanded and clones merged to form a thicker blue or white streak. The authors acknowledged this limitation and after correcting for this discrepancy, predicted that approximately 100 SC partake in the renewal process. If this is an accurate estimate, then a LESC is responsible for replenishing a 3e5 sector of corneal epithelium, which fits well with predictions from independent investigators (Pajoohesh-Ganji et al., 2006; Zhao et al., 2009). The authors also noted the formation of sharp boundaries between blue and white clones, implying there was little or no mixing among neighbouring colonies, as well as interesting geometric vortex patterns as clones converged at a midpoint, and an age-related decline in stripes, suggesting a time-dependent depletion or dysfunction of SC. Moreover, after inflicting a mild corneal abrasion, a regular striping pattern was restored with clones migrating at 750 mm/day to cover the defect (Mort et al., 2009), 40-fold faster than the velocity recorded in normal corneas (Buck, 1985; Nagasaki and Zhao, 2003;

Auran et al., 1995). Loss of clonal activity and distorted epithelial movement was detected in the LacZ-Pax6þ/ mouse (Collinson et al., 2004); an anticipated finding given Pax6 deficient mice have increased cell shedding rates and develop a thinner than normal corneal epithelium (Davis et al., 2003; Douvaras et al., 2013). Mutations in Pax6 cause aniridia and patients with this form of LSCD develop a spectrum of ocular defects, including corneal epithelial abnormalities which manifest gradually during life (Mackman et al., 1979; Skeens et al., 2011). Viral transfer of reporter genes can also be used to assess cell behaviour (see Section 2.1). This strategy was originally employed to transfect cultured rabbit keratolimbal explants with a retroviral vector carrying the LacZ gene to observe a steady trail of bluestained cells which emerged from biopsies (Bradshaw et al., 1999). Endo et al. (2007) extended these findings by injecting mice in utero with a lentiviral vector harbouring a GFP reporter targeting ectodermal tissues to observe GFPþ radial stripes emerging from the limbus after ex vivo detection with an anti-GFP antibody. Presumably immunostaining was necessary due to low reporter expression; nonetheless their results are comparable to those published by others using the same reporter in a different construct (Nagasaki and Zhao, 2003; Iannaccone et al., 2012). 3.1. Transplantation as a method of monitoring cell fate within the cornea Since Kenyon and Tseng's (1989) pioneering limbal tissue transplantation and Pellegrini et al.'s (1997) landmark study on the

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 14/23

14

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

Fig. 6. Cre recombination in the multicoloured Brainbow 2.1 reporter. Schematic of a Cre line engineered to be driven by a K14 promoter. In this construct bacterial Cre recombinase is fused to a mutated oestrogen receptor (ERT2). Following TAM administration, cytosolic CreER dissociates from its chaperone heat-shock protein (A) facilitating its entry into the cells' nucleus where it excises loxP flanked DNA sequences on the reporter construct to initiate recombination (B). The reporter construct is engineered on the ubiquitously expressed Rosa26 locus which acts as a gene-trap. Insertion of an upstream CAG sequence enhances reporter expression while the Neomycin (Neo) cassette regulates its expression by preventing transcription, until the roadblock is lifted after loxP sites are excised. This results in transgene expression in basal epithelial cells that actively express K14. When the Brainbow 2.1 construct is used as the reporter, one recombination scenario is the expression of cytoplasmic YFP and RFP in K14þ basal epithelia which can be visualised as an orange hue (C). Titration of the inductive agent TAM, results in low-frequency labelling of basal cells, heritable in daughter cells and traceable as luminescing (e.g. yellow or red, or other colour combinations) clones/streaks as they travel through the epithelium (D).

first autologous limbal epithelial cell graft, the field has progressed to the point where thousands of patients with LSCD have now received a cell-based graft of sorts for ocular surface rehabilitation (Shortt et al., 2007a; Di Girolamo et al., 2009; Rama et al., 2010; Baylis et al., 2011; Sangwan et al., 2011; Bobba et al., 2015). Although it is not possible to effectively fate map cells in autologous grafts, allogeneic cells can be identified by sex chromosome discrimination but only after resecting and histologically inspecting the graft; with implanted cells detected months-to-years posttransfer (Kinoshita et al., 1981; Stenevi et al., 2002; Reinhard et al., 2004; Djalilian et al., 2005; Daya et al., 2005). Still the major shortcoming of these studies is determining whether the transplanted cells actually partake in long-term maintenance of the corneal epithelium through engraftment, positioning, division, movement and differentiation. Recently the fate of ABCB5þ LESC was followed in mice with limbal SC failure, and despite the clinical success, without a permanent or genetically induced marker, the contribution of implanted cells could only be inferred (Ksander et al., 2014). To this end, Yin et al. (2013) investigated the distribution of surviving donor cells in reconstituted corneas by firstly isolating primary LESC from goats, transfecting them with a vector harbouring YFP, and cultivating them on denuded human amniotic membrane prior to grafting. Although no fluorescence was detected across the ocular surface, the transgene was detected by PCR across all epithelial layers, highest in the central and lowest in the peripheral cornea, suggesting retention of progenitor cells was not regionalised but wide-spread during the 3 month follow-up. Majo et al. (2008) assessed corneal epithelial reconstitution by transferring a segment of limbus from a LacZ-Rosa26 mouse to an athymic mouse. Curiously, cells from the graft did not emigrate into the recipient epithelium until the cornea was challenged with a chemical or physical insult. An eloquent inducible tissue-specific triple transgenic (K12rtTA/ rtTA /tetO-Cre/RosamTmG) line was engineered to resolve the destiny of murine vibrissae hair follicle bulge-derived SC that were grafted

into wild-type mice with experimentally induced LSCD (MeyerBlazejewska et al., 2011). In this model only K12þ cells were engineered to temporally express GFP upon doxycycline administration. In the absence of the inductive agent, all cells expressed the red fluorophore tdtomato. On the basis of this unique feature, K12GFPþ cells were identified in recipient mice confirming they were donor derived, had established residence in the basal epithelium and differentiated into cornea-like cells during the 5-week postgrafting period. Notably, corneas were monitored short-term and the outcome may have been transient; moreover host-derived nonfluorescing cells also contributed to re-epithelialisation implying either remaining or dormant recipient SC, alongside grafted cells contributed to revitalising the ocular surface (Meyer-Blazejewska et al., 2011). 3.2. Lineage tracing corneolimbal epithelial cells A major limitation of genetic tracing models is the unspecified identity of marked cells. This is because transgenes are either expressed during early embryogenesis, or are driven by ubiquitous or non-corneal specific promoters, rendering the lineage signature of marked cells undefined. To address this deficiency, a mouse line was generated whereby LacZ expression was driven by the K5 promoter which is known to be active in epithelial cells of the ocular surface. Notably, a mosaic pattern of stained and unstained cells developed across the cornea with the rare occurrence of b-Galþ streaks. This patchy blue pattern was anticipated since K5 is expressed in basal epithelial cells spanning the radial distance of the cornea (Lu et al., 2006). Other investigators have directed transgenes to cells of a corneal lineage in order to study the kinetics of epithelial maturation and migration (Tanifuji-Terai et al., 2006; Hayashi et al., 2010). In K12Cre/Cre/ZAP mice, the K12 promoter was used to drive ZAP, a cassette containing the LacZ gene flanked by loxP sites and an adjacent alkaline phosphatase (AP) reporter. In this model, corneal epithelial cells that lack K12

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 15/23

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

Q8

will express LacZ (blue) and were deemed immature. Alternatively, if LacZ is excised and AP (red) is induced, this signified corneal-specific differentiation. Results collated from this study showed accumulation of less differentiated LacZþ cells in the limbal fringe of mice older than 3 months, while mature APþ cells were widely distributed across the cornea. Moreover, after debriding the cornea, fewer LacZþ cells were observed in the periphery, while injury triggered the development of blue radial stripes from the periphery as well as sporadic patches of centrally located LacZþ cells whose function was not disclosed (TanifujiTerai et al., 2006). Other transgenic lineage tracing enquiries have led investigators to put forth the ‘Replacement Hypothesis’ where SC prevail during embryogenesis and position themselves throughout the epithelium as expanding clones that migrate vertically to form a mosaic pattern, but during the early postnatal period they are ‘replaced’ by invading limbal-specific SC clones (Hayashi et al., 2012). These results should be interpreted with caution as definitive SC markers were not used to define cell phenotype in either location. We recently capitalised on the relative abundant expression of K14 in basal limbal epithelia (Fig. 2). This finding, along with the number of reports that support K14 as a phenotypic signature gene for immature stem-like cell in the corneal perimeter € tzer-Schrehardt (Kurpakus et al., 1994; Harkin et al., 2004; Schlo and Kruse, 2005; Tanifuji-Terai et al., 2006; Figueira et al., 2007), provided the impetus for us to embrace a cutting-edge technology to generate a drug-inducible Cre-LoxP multicoloured double transgenic mouse (K14CreERT2-Confetti). As mentioned previously (see Section 2.2), this system is engineered to target the Brainbow 2.1 reporter to cells actively expressing K14 and their progeny (Di Girolamo et al., 2015). Upon ‘flicking on’ the molecular switch, marked polychromatic cells were visualised by intravital time-laps microscopy as they emerged from the periphery, initially as small clonal aggregates at 3e4 weeks post-TAM administration, and subsequently as multicoloured radial spokes that travelled at a rate of 11 mm/day along a linear path towards the central cornea (Fig. 7AeD). Individual spokes that spanned the radial length of the cornea contained greater than 1000 cells, suggesting the original K14þ founder cells underwent significant rounds of division before their descendants' trekked to the apex of the cornea. Moreover, high resolution confocal imaging of corneal flat-mounts (Fig. 7E and F) as well as conventional fluorescence microscopy (Fig. 7G) confirmed the limbal annulus as the region spawning these fluorescent streaks. But curiously, and in line with other reports (Tanifuji-Terai et al., 2006; Amitai-Lange et al., 2015), rare smaller central corneal epithelial clones were also evident (Fig. 7G, inset) however their role was not elaborated. It is thought that the murine corneal epithelium does not mature until several months after birth, meaning undifferentiated cells with stem-like activity may reside centrally (Collinson et al., 2002; Tanifuji-Terai et al., 2006). It is possible that these premature, centrally-located basal cells give rise to the rare, long-lasting smaller clones that we (Di Girolamo et al., 2015) and others (Amitai-Lange et al., 2015) have identified, the significance of which is yet to be elucidated but could explain some of Majo's (2008) paradigm-shifting observations. Certainly, investigations utilising K5LacZ/-transgenic mice have revealed the evolution of b-Galþ radial clones from two distinct locations (Douvaras et al., 2012). These include long clonal stripes which commonly forge a path from the limbus to the central cornea, and rare shorter stripes which are located a significant distance from the limbus. Bearing in mind the non-SC driver used in their study, it is tempting to speculate that stripes were produced by b-Galþ SC populations that reside in separate locations. However, the authors provided their own plausible explanation, suggesting that LESC have intermittent activity, i.e.

15

switching between states of stimulation and quiescence, and if for some reason they are unable to continue with their tissue renewing task, centrally located ‘early’ TAC could resume their role. Under homoeostasis, these rare clones do not to expand like the conventional limbal-derived equivalents but their significance may come into play during damage repair. Currently, these observations have not been substantiated with functional evidence (West et al., 2015). These data provide the first direct evidence for the location of K14þ limbal stem/progenitor cells and the extent to which they contribution to corneal epithelial homoeostasis in a living mouse. Notably, this model also allowed us to mark conjunctival epithelia (Fig. 7E and H) as well as basal epidermal (Fig. 7I) and follicular (Fig. 7J) keratinocyte progenitors that actively express K14 to discern their location and vertical migratory pattern. Many of the cellular behaviour patterns described in our model were confirmed by Amitai-Lange et al. (2015), who in addition to documenting the slow corneal epithelial turnover and migration under steady-state, also noted the accelerated emergence of multicoloured limbal-derived clones in wounded corneas. In line with previous investigations (Collinson et al., 2002; PajooheshGanji et al., 2006; Zhao et al., 2009), adolescent K14CreERT2Confetti mice developed approximately 100 coloured streaks (Amitai-Lange et al., 2015), and although lineage specified, inaccurate counts can arise particularly if stochastic recombination results in the evolution of identically coloured neighbouring clones which merge to form a thicker monochromatic colony (Fig. 8). Counting individual streaks as being derived from an ancestral cell would therefore result in underestimating the number of ‘master’ adult SC within the tissue. The other obvious disadvantage of this model is that if SC other than those marked by this system exist, they will not be illuminated. If indeed they do exist, the question that begs an answer is whether they have similar or greater potential than those already identified. It is also worth highlighting the potential problems related to administering the inductive agent TAM; firstly the delivery route can delay recombination as was evident in peripheral corneal tissue (Di Girolamo et al., 2015), and secondly corneal toxicity can arise (Noureddin et al., 1999; Muftuoglu et al., 2006; Tarafdar et al., 2012), although this has not been reported in mice that receive small doses to induce transgenes. Moreover, small doses of TAM can result in low recombination efficiency, but as mentioned above this needs to be balanced with potential side-effects. Another drawback is the inability to decipher whether labelled cells contain a FP of the same colour on one or both alleles. It is however anticipated that future improvements in spectral separation techniques and detector sensitivity will address this ambiguity. Photostability of FP is also a potential problem; so too is the inability to use traditional immunostaining methods to detect FP as some (e.g. CFP, GFP, YFP) only differ by a few amino acids, endowing them antigenically identical. In the future, modifications to Brainbow transgenes could include epitope tags, rendering labelled cell detectable with conventional antibodies. The advantages of this system far outweigh the disadvantages. The ability to induce recombination and commence tracking at a single cell level, at a particular point in time and at a specific tissue location, greatly enhances experimental flexibility. Moreover, this model is conducive to experimental specificity as recombination is targeted to only corneal and conjunctival epithelia actively expressing K14 and no other cell type within the murine eye (Di Girolamo et al., 2015). The system uses DNA excision and/or inversion to randomly create a choice amongst 4 colour-coding genes within the Brainbow 2.1 cassette (Fig. 5). This creates a kaleidoscope of colours (Fig. 7E) allowing the origin of expanding clones to be readily visible and distinguishable from one another,

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 16/23

16

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

Fig. 7. Fluorescent clones emerge from the limbus. Intravital microscopy of a mouse cornea 12-weeks post-TAM administration, viewed under different filters (AeC) and images merged (D). Fluorescent clones are usually visible at approximately 4 weeks after TAM injection (not shown) which soon form streaks as they expand and migrate centripetally (AeD). Note, the intraocular lens autofluorescence. Confocal microscopy of a corneal flat-mount (E) showing multicoloured ribbons of cells exiting the limbus (L) and migrating towards the central cornea (CC). The hatched white line in (E) separates the limbus from the conjunctiva (Cj). This junction is the zone where streaks cease to exist and multicoloured patches of conjunctival cells are now evident. Higher magnification of a representative region allows the cellular distribution of several FP to be visualised (e.g. CFP to the cell membrane, and YFP and RFP to the cytosol). It is also notable that cells within a clone rarely mix with their neighbours, except for those on the very periphery. Transverse 2D tissue sections through various tissues (GeJ) reveal brilliantly coloured streaks arising from the basal limbal epithelium (G), with rare, small, centrally located colonies also evident (G, inset); the white arrow indicates the direction of the central cornea. The same system can be used to decorate K14þ conjunctival epithelial cells which develop into patches instead of streaks (H, and confirmed in panel E) as well as K14þ epidermal (I) and hair follicle (HF) keratinocytes (J). The hatched white line in panels (GeJ) signifies the BM. Images in panels (GeJ) were acquired under two filters with DAPI counterstaining.

and the ability to quantify cell behaviour by determining the velocity of expanding clones and the number of cells within a mobile colony. In essence an accurate ‘road-map’ of epithelial dynamics within the cornea is produced. Moreover, marked cells can be isolated from corneas and sorted on the basis of FP expression, thereby avoiding assays which utilise antibody tethering, and their properties interrogated. Finally, this approach can be used to gain valuable insights into epithelial behaviour after gene modifying perturbations. Many of these positive and negative attributes have been discussed in detail in recent reviews (Murray et al., 2012; Heffner et al., 2012); in the end it is up to the individual

researcher to decipher which driver and transgene to use for their specific application. 3.3. How limbal stem cells sustain the corneal epithelium The notion that the mammalian corneal epithelium has a predetermined number of SC which are defined shortly after birth, is certainly not new (Schermer et al., 1986; Rodrigues et al., 1987; Maseruka et al., 2000; Mort et al., 2009; Davies et al., 2009) but has been difficult to prove. The main reason is that an unequivocal marker for these cells remains elusive; therefore their location and

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 17/23

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

17

Fig. 8. Fate of LESC clones in the normal and diseased mammalian cornea. Multicolour reporters such as Brainbow may be useful tools for determining how SC and their progeny maintain the corneal epithelium under physiological (AeC) and pathological (D) conditions. In this model, LESC are evenly distributed into ‘positions’ along the BM and are illuminated with different colours from the stochastic genetic recombination that ensues in this model. If these cells undergo prolonged asymmetric division, SC and TAC will be produced and their long-term fate visualised as centripetally expanding clones that maintain a permanent streaking pattern (A). If LESC partake in both forms of mitosis, a daughter SC from a symmetric division could take up residency in an adjacent ‘slot’ and eventually a thicker clone will arise (B, blues and red clones). If LESC undergo symmetric mitosis with daughter cells differentiating and moving superficially, eventually they will be lost from the trace (C, yellow and orange clones) until an adjacent SC comes to occupy the vacant position. Loss of SC function, focal damage to the niche or sectoral ablation of LESC (D, white SC in the BM; upper panel) leads to a breach in limbal barrier function and subsequent invasion of conjunctival epithelium (D). Arrows in the upper schema indicate direction of cell migration.

abundance cannot be established with great precision, and without marking these cells and tracking their long-term fate, insights into how they orchestrate processes that maintain the cornea throughout life cannot be gleaned. It is tempting to speculate that for a hemispherical organ like the cornea, SC must be strategically distributed for tissue renewal to proceed evenly and in a circumferential manner. If LESC are programmed to take residence within their niche at a defined post-natal time point, they will come to occupy a ‘position’ along the BM. Whether this occurs randomly or different positions are available for LESC with ‘static’ or ‘active’ roles remains to be confirmed. Assuming all LESC play an equivalent role in maintaining corneal homoeostasis, then upon dividing they can pursue several fates. In this instance, an inducible multicolour genetic tagging approach, such as Brainbow, can be advantageous to visualise this type of cell behaviour. If a peripherally located LESC undergoes long-term asymmetric division, a permanent ‘finger-print’ of clonal expansion and migration will emerge, and this can be visualised through multicolour trace in presumed SC (Fig. 8A). If one applies the neutral drift model, symmetrically dividing basal LESC may compete for a position with neighbouring SC, and it is possible for one to displace the other resulting in the emergence of a thicker clone (Fig. 8B). If the tissue is monitored for a substantial period, a single monochromatic clone (monoclonal conversion) could eventually predominate. Alternatively, if a LESC divides symmetrically and both daughters detach from the niche, positions along the BM become vacant until another SC occupies the slot. This means that after several rounds of multiplication the original clone/streak will be lost from the trace (Fig. 8C). Likewise, loss of function or reduction in SC number, e.g. through ageing, trauma or disease, renders positions along the limbal BM obsolete, with remaining SC ‘overworked’ to maintain a wider sector of corneal epithelium. If this model prevails the outcome can be visualised as a sparse streaking pattern (Fig. 8C). Finally, and irrespective of the mode of SC division, focal or total limbal damage or loss of LESC renders the cornea susceptible to conjunctival invasion (Fig. 8D), a characteristic feature of pterygia (Chui et al., 2011) and LSCD (Chen and Tseng, 1990, 1991) (Fig. 4).

4. Future directions Understanding the basic biology of SC development, i.e. when they first become defined, when and how they segregate into their niche, how they behave throughout life, how they replicate, migrate, differentiate, how they replenish the cornea under normal and diseased states, and importantly what happens to these cells once they are transplanted, is information for which only circumstantial evidence currently exists. These questions cannot be answered in man; however lineage tracing using inducible multicolour genetic recombination holds great promise for progressing the field of LESC biology and corneal epithelial regeneration by allowing us to examine these paradigms in a live mammal. This experimental setup may assist us in resolving some of the uncertainties that have plagued the field. Genetic lineage tracing also lends itself to identifying physiological and pathological stressors that may provide clues as to the molecular programs that govern SC activity within the cornea. Ultimately this may result in the development of therapeutic opportunities directed at modifying signals that regulate SC function in situ. Moreover, data points collated from a plethora of studies that take into account ageing, wounding or disease could be distilled into quantitative mathematical models to provide accurate predictions of epithelial turnover and cell fate determinations within the cornea. Clinical trials in patients with LSCD are difficult due to patient heterogeneity, inconsistent inclusion/exclusion criteria, expensive culture and grafting techniques, and variable outcome and safety data. Transplantation studies in experimental models of LSCD using multicolour reporter such as Brainbow may provide clues as to whether graft failure is due to the type of SC or a decline in SC number. These data can inform clinicians as to why failure rates are still unacceptably high and provide the rationale to modify or develop alternative therapies for restoring vision and tissue function as a result of severe corneal disease. Certainly, there is data to suggest that cells which display certain functional and phenotypic attributes (e.g. holoclonal activity and high p63 expression) should be included in grafts, but in fact they are insufficient to guarantee 100% successful outcomes in patients

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 Q9 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 18/23

18

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

with LSCD (Rama et al., 2010). Furthermore, it is unknown how grafted cells interact with their new microenvironment and the positions they acquire prior to reinvigorating the corneal surface. If this information becomes available from lineage tracing models, we will be closer to understanding how such therapies work, how to improve current strategies and whether new interventions are needed. While the current article provides some commentary on LESC transplantation, it only touches upon tumourigenesis of the ocular surface. It has been postulated that benign and malignant tumours arise from altered LESC (Chui et al., 2011; Di Girolamo et al., 2013) but without a robust cell tracking system, spawning cancer cell clones on the ocular surface will be difficult to identify, until they have expanded to a detectable level. The K14CreERT2-Brainbow mouse model may be a useful tool for addressing this dilemma and with further manipulations including exposure to external stressors such as ultraviolet radiation and/or modifying tumour suppressor genes (Halliday et al., 2012; Hassan et al., 2014), some of the mechanisms involved in carcinogenesis of the ocular surface may be revealed. 5. Conclusions Lineage tracing is pushing the frontiers of medical science by making incredible in-roads into understanding the behaviour of SC in mammalian organs under physiological and pathological conditions. A ‘window of opportunity’ is now available to resolve many controversies that plague the SC field. With the advent of inducible multicolour genetic recombination in transgenic mice and high resolution imaging systems, we are beginning to appreciate the intricate interactions of SC with their microenvironment including how these cells meet the daily requirement of cell and tissue renewal during life. The external cornea is conducive to noninvasive continuous monitoring as the activity of genetically marked cells can be visualised in real-time by intravital microscopy. Moreover, transparent corneal whole-mounts can be prepared for confocal microscopy to gain a 3D perspective of the dynamics of a single founder cell from the time of labelling until the time of analysis. LESC must abide by the ‘rules of the game’ if they are to contribute tissue homoeostasis. Their fate with respect to cell division during this process is expected to be random, although they probably encounter guidance cues to specify directionality of movement. The challenge for researchers is to better understand the decisions these cells make and whether or not multiple SC populations exist within the corneal epithelium which function as either the ‘caretakers’ for steady-state tissue renewal or the ‘workhorses’ needed for injury repair. Over the past few decades corneal researchers have made amazing progress at the bedside, however if we are to improve on the current clinical success and lead the way in SC interventions, methods to visualise cell movement in the normal and wounded cornea, and how grafted SC manage to rebuild damaged tissue to restore vision are critically important. It is anticipated that lineage tracing using multicolour transgenic approaches will allow us to achieve some of these objectives. Funding sources University of NSW and Ophthalmic Research Institute of Australia. Author contribution The author performed the background research, conceptualised, wrote, and, designed and produced the figures for this article. He

collected data for analysis, interpretation, and assembly and provided some of the financial support. Ethics approval Institutional Human and Animal Research Ethics approvals were obtained to conduct all the studies described in this article. Uncited reference

Q14

Chow and Di Girolamo, 2014. Acknowledgements The author would like to thank James Zieske (Schepens Eye Research Institute, Harvard Medical School, USA) for his invaluable feedback on the manuscript, Renee Whan, Iveta Slapetova and Philip Nicovich (Biomedical Imaging Facility, University of New South Wales, Sydney, Australia) for their assistance with confocal imaging and analysis, Pierre Chambon (Institut  ne tique Biologie Mole culaire Cellulaire, Cedex, France) for Ge transgenic K14CreERT2 mice, Minas Coroneo (Prince of Wales Hospital, Randwick, Sydney, Australia) for the clinical image of pterygium, Stephanie Watson (Save Sight Institute, Sydney, Australia) for the clinical image of partial LSCD, Vani Raviraj and Guy Lyons (Department of Dermatology, University of Sydney, Sydney Australia) for their assistance with intravital microscopy and analysis. Q10 References Alcolea, M.P., Jones, P.H., 2014. Lineage analysis of epidermal stem cells. Cold Spring Harb. Perspect. Med. 4, a015206. Amitai-Lange, A., Altshuler, A., Bubley, J., Dbayat, N., Tiosano, B., ShalomFeuerstein, R., 2015. Lineage tracing of stem and progenitor cells of the murine corneal epithelium. Stem Cells 33, 230e239. Anderson, W., 1898. A case of “angeikeratoma”. Br. J. Dermatol. 10, 113e117. Auran, J.D., Koester, C.J., Kleiman, N.J., Rapaport, R., Bomann, J.S., Wirotsko, B.M., Florakis, G.J., Koniarek, J.P., 1995. Scanning slit confocal microscopic observation of cell morphology and movement within the normal human anterior cornea. Ophthalmology 102, 33e41. Barbaro, V., Testa, A., Di iorio, E., Mavilio, F., Pellegrini, G., De Luca, M., 2007. C/EBPd regulates cell cycle and self-renewal of human limbal stem cells. J. Cell Biol. 177, 1037e1049.  es, R.M., Haddad, A., 2009. Regeneration of the Barbosa, F.L., de Faria-e-Sousa, S.J., Go corneal epithelium after debridement of its central region: an autoradiographic study on rabbits. Curr. Eye Res. 34, 636e645. Bardehle, S., Krüger, M., Buggenthin, F., Schwausch, J., Ninkovic, J., Clevers, H., €tz, M., Snippert, H.J., Theis, F.J., Meyer-Luehmann, M., Bechmann, I., Dimou, L., Go 2013. Live imaging of astrocyte responses to acute injury reveals selective juxtavascular proliferation. Nat. Neurosci. 16, 580e586. Baylis, O., Figueirdo, F., Henein, C., Lako, M., Ahmad, S., 2011. 13 years of cultured epithelial cell therapy: a review of the outcomes. J. Cell. Biochem. 112, 993e1002. Beebe, D.C., Masters, B.R., 1996. Cell lineage and the differentiation of corneal epithelial cells. Investig. Ophthalmol. Vis. Sci. 37, 1815e1825. Bertalanffy, F.D., Lau, C., 1962. Mitotic rate and renewal time of the corneal epithelium in the rat. Arch. Ophthalmol. 68, 144e148. Bickenbach, J.R., 1981. Identification and behaviour of label-retaining cells in oral mucosa and skin. J. Dent. Res. 60, 1611e1620. Bigger, S.L., 1837. An inquiry into the possibility of transplanting the cornea, with the view of relieving blindness (hitherto deemed incurable) caused by several diseases of that structure. Dublin J. Med. Sci. 11, 408e417. Blanco-Mezquita, T., Martinez-Garcia, C., Proenca, R., Zieske, J.D., Bonini, S., Lambiase, A., Merago-Lloves, J., 2013. Nerve growth factor promotes corneal migration by enhancing expression of matrix metalloproteinase-9. Investig. Ophthalmol. Vis. Sci. 54, 3880e3890. Blanpain, C., Fuche, E., 2009. Epidermal homeostasis: a balancing act of stem cells in the skin. Nat. Rev. Mol. Cell Biol. 10, 207e217. Bobba, S., Watson, S., Di Girolamo, N., 2015. Clinical outcomes of xeno-free expansion and transplantation of autologous ocular surface epithelial stem cells via contact lens delivery. Stem Cells Res. Ther. (in press). Q11,12 Bradshaw, J.J., Obritsch, W.F., Cho, B.J., Gregerson, D.S., Holland, E.J., 1999. Ex vivo transduction of corneal epithelial progenitor cells using a retroviral vector. Investig. Ophthalmol. Vis. Sci. 40, 230e235.

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 19/23

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23 Braun, K.M., Watt, F.M., 2004. Epidermal label-retaining cells: background and recent applications. J. Investig. Dermatol. Symp. Proc. 9, 196e201. Bron, A.J., 1973. Vortex patterns of the corneal epithelium. Trans. Ophthalmol. Soc. U. K. 93, 455e472. Buck, R.C., 1979. Cell migration in repair of mouse corneal epithelium. Investig. Ophthalmol. Vis. Sci. 18, 767e784. Buck, R.C., 1982. Hemidesmosomes of normal and regenerating mouse corneal epithelium. Virchows Arch. Cell Pathol. 41, 1e16. Buck, R.C., 1985. Measurement of centripetal migration of normal corneal epithelial cells in the mouse. Investig. Ophthalmol. Vis. Sci. 26, 1296e1299. Burns, E.R., Scheving, L.E., 1975. Cicadian influence on the wave form of the frequency of the labelled mitosis in mouse corneal epithelium. Cell Tissue Kinet. 8, 61e66. Buschke, W., 1949. Morphologic changes in cells of corneal epithelium in wound healing. Arch. Ophthalmol. 41, 306e316. Buschke, W.H., Friedenwald, J.S., Fleischmann, W., 1943. Studies on the mitotic activity of the corneal epithelium. Bull. Johns Hopkins Hosp. 73, 143e168. ~ ozledo, F., Go mez-Flores, E., 2011. Challenges to the study of asymmetric Castro-Mun cell division in corneal and limbal epithelia. Exp. Eye Res. 92, 4e9. Catanese, M., Popovici, C., Proust, H., Hoffart, L., Matonti, F., Cochereau, I., Conrath, J., Gabison, E.E., 2011. Fluorescent in situ hybridization (FISH) on corneal impression cytology specimens (CICS): study of epithelial cell survival after keratoplasty. Investig. Ophthalmol. Vis. Sci. 52, 1009e1013. Cenedella, R.J., Fleschner, C.R., 1990. Kinetics of cornea epithelium turnover in vivo. Investig. Ophthalmol. Vis. Sci. 31, 1957e1962. Cepko, C., Ryder, E.F., Austin, C.P., Walsh, C., Fekete, D.M., 1995. Lineage analysis using retroviral vectors. Methods Enzymol. 254, 387e419. Chan, C.K., Pham, L.N., Chinn, C., Spee, C., Ryan, S.J., Akhurst, R.J., Hinton, D.R., 2004. Mouse strain-dependent heterogeneity of resting limbal vasculature. Investig. Ophthamol. Vis. Sci. 45, 441e447. Chang, C.-Y., Green, C.R., McGhee, C.N.J., Sherwin, T., 2008. Acute wound healing in the human central corneal epithelium appears to be independent of limbal stem cell influence. Investig. Ophthalmol. Vis. Sci. 49, 5279e5286. Chen, J.J.Y., Tseng, S.C.G., 1990. Corneal epithelial wound healing in partial limbal deficiency. Investig. Ophthalmol. Vis. Sci. 31, 1301e1314. Chen, J.J.Y., Tseng, S.C.G., 1991. Abnormal corneal epithelial wound healing in partial-thickness removal of limbal epithelium. Investig. Ophthalmol. Vis. Sci. 32, 2219e2233. Chow, S., Di Girolamo, N., 2014. Vitronectin: a migration and wound-healing factor for human corneal epithelial cells. Investig. Ophthalmol. Vis. Sci. 55, 6590e6600. Chui, J., Coroneo, M.T., Crouch, R., Wakefield, D., Di Girolamo, N., 2011. Ophthalmic pterygia: a stem cell disorder with pre-neoplastic features. Am. J. Pathol. 178, 817e827. Chung, E.H., Bukusoglu, G., Zieski, J.D., 1992. Localization of corneal epithelial stem cells in the developing rat. Investig. Ophthalmol. Vis. Sci. 33, 2199e2206. , D.P., Klein, A.M., Winton, D.J., Simons, B.D., Jones, P.H., 2007. Clayton, E., Doupe A single type of progenitor cell maintains normal epidermis. Nature 446, 185e189. Collinson, J.M., Chanas, S.A., Hill, R.E., West, J.D., 2004. Corneal development, limbal stem cell function, and corneal epithelial cell migration in the Pax6þ/- mouse. Investig. Ophthalmol. Vis. Sci. 45, 1101e1108. Collinson, J.M., Morris, L., Reid, A.I., Ramaesh, T., Keighren, M.A., Flockhart, J.H., Hill, R.E., Tan, S.-S., Ramesh, K., Dhillon, B., West, J.D., 2002. Clonal analysis of patterns of growth, stem cell activity, and cell movement during the development and maintenance of the murine corneal epithelium. Dev. Dyn. 224, 432e440. Conklin, E.G., 1905. The organization and cell-lineage of the ascidian egg. J. Acad. Nat. Sci. (Phila.) 13, 1e119. Cotsarelis, G., Cheng, S.-Z., Dong, G., Sun, T.-T., Lavker, R.M., 1989. Existence of slowcycling limbal epithelial basal cells that can be preferentially stimulated to proliferate: implications on epithelial stem cells. Cell 57, 201e209. Cotsarelis, G., Sun, T.-T., Lavker, R.M., 1990. Label-retaining cells reside in the bulge area of pilosebaceous unit: implications for follicular stem cells, hair cycle, and skin carcinogenesis. Cell 61, 1329e1337. Cowan, T., 1963. Striate melanokeratosis in Negros. Trans. Am. Ophthalmol. Soc. 61, 61e74. D'Amico, D.J., Kenyon, K.R., Ruskin, J.N., 1981. Amiodarone keratopathy: druginduced lipid storage disease. Arch. Ophthalmol. 99, 257e261. Danjo, Y., Gipson, I.K., 1998. Actin ‘purse string’ filaments are anchored by E-cadherin-mediated adherins junctions at the leading edge of the epithelial wound, providing coordinated cell movement. J. Cell Sci. 111, 3323e3331. Danjo, Y., Gipson, I.K., 2002. Specific transduction of the leading edge cell of migrating epithelia demonstrates that they are replaced during healing. Exp. Eye Res. 74, 194e204. Davanger, M., Evensen, A., 1971. Role of the pericorneal papillary structure in renewal of corneal epithelium. Nature 229, 560e561. Davies, S.B., Chui, J., Madigan, M.C., Provis, J.M., Wakefield, D., Di Girolamo, N., 2009. Stem cell activity in the developing human cornea. Stem Cells 27, 2781e2792. Davis, J., Duncan, M.K., Robison Jr., W.G., Piatigorsky, J., 2003. Requirement for Pax6 in corneal morphogenesis: a role in adhesion. J. Cell Sci. 116, 2157e2167. Daya, S.M., Watson, A., Sharpe, J.R., Giledi, O., Rowe, A., Martin, R., James, E., 2005. Outcomes and DNA analysis of ex vivo expanded stem cell allograft for ocular surface reconstruction. Ophthalmology 112, 470e477.

19

de Faria-e-Sousa, S.J., Barbosa, F.L., Haddad, A., 2010. Autoradiographic study on the regenerative capacity of the epithelium lining the center of the cornea after multiple debridements of its peripheral region. Graefes Arch. Clin. Exp. Ophthalmol. 248, 1137e1144. Denden, A., 1966. Über die erblichkeit der dystrophia corneae verticillata. Albr. Von Graefes Arch. Klin. Exp. Ophthalmol. 169, 246e271. De Paiva, C.S., Chen, Z., Corrales, R.M., Pflugfelder, S.C., Li, D.Q., 2005. ABCG2 transporter identifies a population of clonogenic human limbal epithelial cells. Stem Cells 23, 63e73. Di Girolamo, N., Atik, A., McCluskey, P.J., Wakefield, D., 2013. Matrix metalloproteinases and their inhibitors in squamous cell carcinoma of the conjunctiva. Ocul. Surf. 11, 193e205. Di Girolamo, N., Bobba, S., Raviraj, V., Delic, N., Slapetova, I., Nicovich, P.R., Halliday, G.M., Wakefield, D., Whan, R., Lyons, J.G., 2015. Tracing the fate of limbal epithelial progenitor cells in the murine cornea. Stem Cells 33, 157e169. Di Girolamo, N., Bosch, M., Zamora, K., Coroneo, M.T., Wakefield, D., Watson, S., 2009. A contact lens-based technique for expansion and transplantation of autologous epithelial progenitors for ocular surface reconstruction. Transplantation 87, 1571e1578. Di Girolamo, N., Sarris, M., Chiu, J., Cheema, H., Coroneo, M.T., Wakefield, D., 2008. Localization of the low affinity nerve growth factor receptor p75 in human limbal epithelial cells. J. Cell. Mol. Med. 12, 2799e2811. Di Iorio, E., Barbaro, V., Ruzza, A., Ponzin, D., Pellegrini, G., De Luca, M., 2005. Isoforms of DNp63 and the migration of ocular limbal cells in human corneal regeneration. Proc. Natl. Acad. Sci. U. S. A. 102, 9523e9528. Djalilian, A.R., Mahesh, S.P., Koch, C.A., Nussenblatt, R.B., Shen, D., Zhuang, Z., Holland, E.J., Chan, C.-C., 2005. Survival of donor epithelial cells after limbal stem cells transplantation. Investig. Ophthalmol. Vis. Sci. 46, 803e807. , D.P., Klein, A.M., Simons, B.D., Jones, P.H., 2010. The ordered architecture of Doupe murine ear epidermis is maintained by progenitor cells with random fate. Dev. Cell 18, 317e323. Douvaras, P., Mort, R.L., Edwards, D., Ramaesh, K., Dhillon, B., Morley, S.D., West, J.D., 2013. Increased corneal epithelial turnover contributes to abnormal homeostasis in the Pax6(þ/-) mouse model of aniridia. PLoS ONE 8, e71117. Douvaras, P., Webb, S., Whitaker, D.A., Dora, N., Hill, R.E., Dorin, J.R., West, J.D., 2012. Rare corneal clones in mice suggest an age-related decrease of stem cell activity and support the limbal epithelial stem cell hypothesis. Stem Cell Res. 8, 109e119. Dreier, B., Raghunathan, V.K., Russell, P., Murphy, C.J., 2012. Focal adhesion kinase knockdown modulates the response of human corneal epithelial cells to topographic cues. Acta Biomater. 8, 4285e4294. Dua, H.S., Gomes, J.A.P., 2000. Clinical course of hurricane keratopathy. Br. J. Ophthalmol. 84, 285e288. Dua, H.S., Forrester, J.V., 1987. Clinical patterns of corneal epithelial wound healing. Am. J. Ophthalmol. 104, 481e489. Dua, H.S., Miri, A., Alomar, T., Yeung, A.M., Said, D.G., 2009. The role of limbal stem cells in corneal epithelial maintenance. Ophthalmology 116, 856e863. Dua, H.S., Shanmuganathan, V.A., Powell-Richards, A.O., Tighe, P.J., Joseph, A., 2005. Limbal epithelial crypts: a novel anatomical structure and a putative limbal stem cell niche. Br. J. Ophthalmol. 89, 529e532. Dua, H.S., Singh, A., Gomes, J.A.P., Laibson, P.R., Donoso, L.A., Tyagi, S., 1996. Vortex or whorl formation of cultured human corneal epithelial cells induced by magnetic fields. Eye 10, 447e450. Dua, H.S., Watson, N.J., Mathur, R.M., Forrester, J.V., 1993. Corneal epithelial cell migration in humans: ‘hurricane and blizzard keratopathy’. Eye 7, 53e58. Ducros, M., Moreaux, L., Bradley, J., Tiret, P., Griesbeck, O., Charpak, S., 2009. Spectral unmixing: analysis of performance in the olfactory bulb in vivo. PLoS ONE 4, e4418. Duke-Elder, S., Cook, C., 1963. Normal and abnormal development. Part 1: embryology. In: Duke-Elder, S. (Ed.), System of Ophthalmology. Henry Kimpton, London. Ebato, B., Friend, J., Thoft, R.A., 1987. Comparison of central and peripheral human corneal epithelium in tissue culture. Investig. Ophthalmol. Vis. Sci. 28, 1450e1456. Eberwein, P., Reinhard, T., 2014 Nov 20. The role of biomechanics in the limbal stem cell niche: new insights for our understanding of this structure. Stem Cells [Epub ahead of print]. Endo, M., Zoltick, P.W., Chung, D.C., Bennett, J., Radu, A., Muvarak, N., Flake, A.W., 2007. Gene transfer to ocular stem cells by early gestational intraamniotic injection of lentiviral vector. Mol. Ther. 15, 579e587. Engler, A.J., Sen, S., Sweeney, H.L., Discher, D.E., 2006. Matrix elasticity directs stem cell lineage specification. Cell 126, 677e689. Estil, S., Primo, E.J., Wilson, G., 2000. Apoptosis in shed human corneal cells. Investig. Ophthalmol. Vis. Sci. 41, 3360e3364. Fabry, J., 1898. Ein beitrag zur kenntnis der purpure haemorrhagica nodularis (purpure papulose haemorrhagica hebrae). Arch. Dermatol. Syph. 43, 187e200. Figueira, E.C., Di Girolamo, N., Coroneo, M.T., Wakefield, D., 2007. The phenotype of limbal epithelial stem cells. Investig. Ophthalmol. Vis. Sci. 48, 144e156. Fleischer, A., 1910. Über eine eigenartige bisher nicht bekannte hornhauttrübung (Ein hinweis auf die normale strukturder hornhaut). Albr. Von Graefes Arch. Ophthalmol. 77, 136e140. Font, R.L., Fine, B.S., 1972. Ocular pathology in Fabry's disease: histochemical and electron microscopic observations. Am. J. Ophthalmol. 73, 419e430. Foster, J.W., Jones, R.R., Bippes, R.R., Gouveia, R.J., Connon, C.J., 2014. Differential nuclear expression of Yap in basal epithelial cells across the cornea and substrates of differing stiffness. Exp. Eye Res. 127, 37e41.

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 20/23

20

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

François, J., 1966. Fleischer's vortex corneal dystrophyHeredo-familial corneal dystrophies. Trans. Ophthalmol. Soc. U. K. 86, 367e416. Friedenwald, J.S., 1951. Growth pressure and metaplasia of conjunctival and corneal epithelium. Doc. Ophthalmol. 5e6, 184e187. Gaffney, E.A., Maini, P.K., Sherratt, J.A., Tuft, S., 1999. The mathematical modelling of cell kinetics in corneal epithelial wound healing. J. Theor. Biol. 197, 15e40. Garcia-Bellido, A., Merrian, J.R., 1969. Cell lineage of the image disc in Drosophila gynandromorphs. J. Exp. Zool. 170, 61e75. Geerling, G., Liu, C.S.C., Collin, J.R.O., Dart, J.K.G., 2002. Costs and gains of complex procedures to rehabilitate end stage ocular surface disease. Br. J. Ophthalmol. 86, 1220e1221. Ghaffarieh, A., Ghaffarpasand, F., Dehghankhalili, M., Honarpisheh, N., Nirumandi, S., Tanideh, N., 2012. Effect of transcutaneous electrical stimulation on rabbit corneal epithelial cell migration. Cornea 31, 559e563. Giangreco, A., Arwert, E.N., Rosewell, I.R., Snyder, J., Watt, F.M., Stripp, B.R., 2009. Stem cells are dispensable for lung homeostasis but restore airways after injury. Proc. Natl. Acad. Sci. U. S. A. 106, 9286e9291. Gipson, I.K., Anderson, R.A., 1977. Actin filaments in normal and migrating corneal epithelial cells. Investig. Ophthalmol. Vis. Sci. 16, 161e166. Gipson, I.K., Keezer, L., 1982. Effects of cytochalasins and colchicine on the ultrastructure of migrating corneal epithelium. Investig. Ophthalmol. Vis. Sci. 22, 643e650. Gipson, I.K., Kiorpes, T., 1982. Epithelial sheet movement: protein and glycoprotein synthesis. Dev. Biol. 92, 259e262. Gipson, I.K., Kiorpes, T.C., Brennan, S.J., 1984. Epithelial sheet movement: effects of tunicamycin on migration and glycoprotein synthesis. Dev. Biol. 101, 212e220. Goldberg, M.F., 1982. Limbal palisades of Vogt. Trans. Am. Ophthalmol. Soc. 80, 155e171. Haddad, A., 2000. Renewal of the rabbit corneal epithelium as investigated by autoradiography after intravitreal injection of 3H-thymidine. Cornea 19, 378e383. Hadjieconomou, D., Rotkopf, S., Alexandre, C., Bell, D.M., Dickson, B.J., Salecker, I., 2011. Flybow: genetic multicolor cell labeling for neural circuit analysis in Drosophila melanogaster. Nat. Methods 8, 260e266. Halliday, G.M., Zhou, Y., Sou, P.W., Huang, X.X., Rana, S., Bugeja, M.J., Painter, N., Scolyer, R.A., Muchardt, C., Di Girolamo, N., Lyons, J.G., 2012. The absence of Brm exacerbates photocarcinogenesis. Exp. Dermatol. 21, 599e604. Hanna, C., 1966. Proliferation and migration of epithelial cells during corneal wound repair in the rabbit and rat. Am. J. Ophthalmol. 61, 55e63. Hanna, C., Bicknell, D.S., O'Brien, J.E., 1961. Cell turnover in the adult human eye. Arch. Ophthalmol. 65, 111e114. Hanna, C., O'Brien, J.E., 1960. Cell production and migration in the epithelial layer of the cornea. Arch. Ophthalmol. 64, 536e539. Harkin, D.G., Barnard, Z., Gillies, P., Ainscough, S.L., Apel, A.J.G., 2004. Analysis of p63 and cytokeratin expression in a cultivated limbal autograft used in the treatment of limbal stem cell deficiency. Br. J. Ophthalmol. 88, 1154e1158. Harrison, D.A., Perrimon, N., 1993. Simple and efficient generation of marked clones in Drosophila. Curr. Biol. 3, 424e433. Haskjold, E., Refsum, S.B., Bjerknes, R., 1989. Circadian variation in the mitotic rate of the rat corneal epithelium. Virchows Arch. B Cell Pathol. 58, 123e127. Hassan, N.M., Painter, N., Howlett, C.R., Farrell, A.W., Di Girolamo, N., Lyons, J.G., Halliday, G.M., 2014. Brm inhibits the proliferative response of keratinocytes and corneal epithelial cells to ultraviolet radiation-induced damage. PLoS ONE 9, e107931. Haug, S.J., Friedman, A.H., 1991. Identification of amiodarone in corneal deposits. Am. J. Ophthalmol. 111, 518e520. Hayashi, R., Yamato, M., Saito, T., Oshima, T., Okano, T., Tano, Y., Nishida, K., 2008. Enrichment of corneal epithelial stem/progenitor cells using cell surface markers, integrin a6 and CD71. Biochem. Biophys. Res. Commun. 367, 256e263. Hayashi, Y., Call, M.K., Liu, C.-Y., Hayashi, M., Babcock, G., Ohashi, Y., Kao, W.W.-Y., 2010. Monoallelic expression of Krt12 gene during corneal-type epithelium differentiation of limbal stem cells. Investig. Ophthalmol. Vis. Sci. 51, 4562e4568. Hayashi, Y., Watanabe, N., Ohashi, Y., 2012. The “Replacement Hypothesis”: corneal stem cell origin epithelia are replaced by limbal stem cells origin epithelia in mouse cornea during maturation. Cornea 31 (Suppl. 1), S68eS73. Heffner, C.S., Pratt, C.H., Babiuk, R.P., Sharma, Y., Rockwood, S.F., Donahue, L.R., Eppig, J.T., Murray, S.A., 2012. Supporting conditional mouse mutagenesis with a comprehensive cre characterization resource. Nat. Commun. 3, 1218. Higa, K., Shimmura, S., Miyashita, H., Kato, N., Ogawa, Y., Kawakita, T., Shimazaki, J., Tsubota, K., 2009. N-cadherin in the maintenance of human corneal limbal epithelial progenitor cells in vitro. Investig. Ophthalmol. Vis. Sci. 50, 4640e4645. Hjortdal, J.O., 1996. Regional elastic performance of the human cornea. J. Biomech. 27, 931e942. Hollander, D.A., Aldave, A.J., 2004. Drug-induced corneal complications. Curr. Opin. Ophthalmol. 15, 541e548. Huang, A.J.W., Tseng, S.C.G., 1991. Corneal epithelial wound healing in the absence of limbal epithelium. Investig. Ophthalmol. Vis. Sci. 32, 96e105. Huang, M., Wang, B., Wan, P., Liang, X., Wang, X., Liu, Y., Zhou, Q., Wang, Z., 2015. Roles of limbal microvascular net and limbal stroma in regulating maintenance of limbal epithelial stem cells. Cell Tissue Res. 359, 547e563. Iannaccone, S., Zhou, Y., Walterhouse, D., Taborn, G., Landini, G., Iannaccone, P., 2012. Three dimensional visualization and fractal analysis of mosaic patches in

rat chimeras: cell assortment in liver, adrenal cortex and cornea. PLoS ONE 7, e31609. , V., Bornert, J.-M., Messaddeq, N., Yaniv, M., Mark, M., Chambon, P., Indra, A.K., Dupe Metzger, D., 2005. Temporally controlled targeted somatic mutagenesis in embryonic surface ectoderm and fetal epidermal keratinocytes unveils two distinct developmental functions of BRG1 in limb morphogenesis and skin barrier formation. Development 132, 4533e4544. Jensen, K.B., Driskell, R.R., Watt, F.M., 2010. Assaying proliferation and differentiation capacity of stem cells using disaggregated adult mouse epidermis. Nat. Protoc. 5, 898e911. Jones, M.A., Marfurt, C.F., 1996. Sympathetic stimulation of corneal epithelial proliferation in wounded and nonwounded rat eyes. Investig. Ophthalmol. Vis. Sci. 37, 2535e2547. Kabosova, A., Azar, D.T., Bannikov, G.A., Campbell, K.P., Durbeej, M., Ghohestani, R.F., Jones, J.C., Kenney, M.C., Koch, M., Ninomiya, Y., Patton, B.L., Paulsson, M., Sado, Y., Sage, E.H., Sasaki, T., Sorokin, L.M., Steiner-Champliaud, M.F., Sun, T.T., Sundarraj, N., Timpl, R., Virtanen, I., Ljubimov, A.V., 2007. Compositional differences between infant and adult human corneal basement membranes. Investig. Ophthalmol. Vis. Sci. 48, 4989e4999. Kaplony, A., Zimmermann, D.R., Fischer, R.W., Imhof, B.A., Odermatt, B.F., Winthalter, K.H., Vaughan, L., 1991. Tenascin Mr 220,000 isoform expression correlates with corneal cell migration. Development 112, 605e614. Kawakita, T., Higa, K., Shimmura, S., Tomita, M., Tsubota, K., Shimazaki, J., 2011. Fate of corneal epithelial cells separated from the limbus in vivo. Investig. Ophthalmol. Vis. Sci. 52, 8132e8137. Kawasaki, S., Tanioka, H., Yamasaki, K., Yokoi, N., Komuro, A., Kinoshita, S., 2006. Clusters of corneal epithelial cells reside ectopically in human conjunctival epithelium. Investig. Ophthalmol. Vis. Sci. 47, 1359e1367. Kaye, D.B., 1980. Epithelial response in penetrating keratoplasty. Am. J. Ophthalmol. 89, 381e387. Kenyon, K.R., Tseng, S.C.G., 1989. Limbal autograft transplantation for ocular surface disorders. Ophthalmology 96, 709e723. Khodadoust, A.A., Silverstein, A.M., 1969. Survival and rejection of epithelium in experimental corneal transplants. Investig. Ophthalmol. 8, 169e179. Kinoshita, S., Friend, J., Thoft, R.A., 1981. Sex chromatin of donor corneal epithelium in rabbits. Investig. Ophthalmol. Vis. Sci. 21, 434e441. Kinoshita, S., Kiorpes, T.C., Friend, J., Thoft, R.A., 1982. Limbal epithelium in ocular surface wound healing. Investig. Ophthalmol. Vis. Sci. 23, 73e80. Koroma, B.M., Yang, J.M., Sundin, O.H., 1997. The Pax-6 homeobox gene is expressed throughout the corneal and conjunctival epithelia. Investig. Ophthalmol. Vis. Sci. 38, 108e120. Kruse, F.E., Chen, J.J., Tsai, R.J., Tseng, S.C.G., 1990. Conjunctival transdifferentiation is due to incomplete removal of limbal basal epithelium. Investig. Ophthalmol. Vis. Sci. 31, 1903e1913. Ksander, B.R., Kolovou, P.E., Wilson, B.J., Saab, K.R., Guo, Q., Ma, J., McGuire, S.P., Gregory, M.S., Vincent, W.J.B., Perez, V.L., Cruz-Guilloty, F., Kao, W.W.-Y., Call, M.K., Tucker, B.A., Zhan, Q., Murphy, G.F., Lathrop, K.L., Alt, C., Mortensen, L.T., Lin, C.P., Zieske, J.D., Frank, M.H., Frank, N.Y., 2014. ABCB5 is a limbal stem cell gene required for corneal development and repair. Nature 511, 353e357. Kurpakus, M.A., Maniaci, M.T., Esco, M., 1994. Expression of keratins K12, K4 and K14 during development of ocular surface epithelium. Curr. Eye Res. 13, 805e814. Kurpakus, M.A., Stock, E.L., Jones, J.C.R., 1992. The role of the basement membrane in differential expression of keratin proteins in epithelial cells. Dev. Biol. 150, 243e255. Kuwabara, T., Perkins, D.G., Cogan, D.G., 1976. Sliding of the epithelium in experimental corneal wounds. Investig. Ophthalmol. 15, 4e14. Lagali, N., Stenevi, U., Claesson, M., Fagerholm, P., Hanson, C., Weijdegard, B., 2009. Survival of donor-derived cells in human corneal transplants. Investig. Ophthalmol. Vis. Sci. 50, 2673e2678. Lakso, M., Sauer, B., Mosinger, B., Lee, E.J., Manning, R.W., Yu, S.-H., Mulder, K.L., Westphal, H., 1992. Targeted oncogene activation by site-specific recombination in transgenic mice. Proc. Natl. Acad. Sci. U. S. A. 89, 6232e6236. Lambiase, A., Manni, L., Bonini, S., Rama, P., Micera, A., Aloe, L., 2000. Nerve growth factor promotes corneal healing: structural, biochemical, and molecular analyses of rat and human corneas. Investig. Ophthalmol. Vis. Sci. 41, 1063e1069. Lamprecht, J., 1987. Mitosis in the corneal epithelium. A preliminary communication on the coexistence of differential and equivalent cell divisions. Cell Biol. Int. Rep. 11, 449e455. Lamprecht, J., 1990. Symmetric and asymmetric cell division in rat corneal epithelium. Cell Tissue Kinet. 23, 203e216. Lane, S.W., Williams, D.A., Watt, F.M., 2014. Modulating the stem cell niche of tissue regeneration. Nat. Biotech. 32, 795e803. Lavker, R.M., Dong, G., Chen, S.Z., Kudoh, K., Cotsarelis, G., Sun, T.-T., 1991. Relative proliferative rates of limbal and corneal epithelia. Investig. Ophthalmol. Vis. Sci. 32, 1864e1875. Lawson, K.A., Meneses, J.J., Pedersen, R.A., 1991. Clonal analysis of epiblast fate during germ layer formation in the mouse embryo. Development 113, 891e911. Leiper, L.J., Ou, J., Walczysko, P., Kucerova, R., Lavery, D.N., West, J.D., Collinson, J.M., 2009. Control of patterns of corneal innervation by Pax6. Investig. Ophthalmol. Vis. Sci. 50, 1122e1128. Lemp, M.A., Mathers, W.D., 1991. Renewal of the corneal epithelium. CLAO J. 17, 258e266.

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 21/23

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23 Levis, H.J., Massie, I., Dziasko, M.A., Kaasi, A., Daniels, J.T., 2013. Rapid tissue engineering of biomimetic human corneal limbal crypts with 3D niche architecture. Biomaterials 34, 8860e8868. Levy, V., Lindon, C., Harfe, B.D., Morgan, B.A., 2005. Distinct stem cell populations regenerate the follicle and interfollicular epidermis. Dev. Cell 9, 855e861. Levy, V., Lindon, C., Zheng, Y., Harfe, B.D., Morgan, B.A., 2007. Epidermal stem cells arise from the hair follicle after wounding. FASEB J. 21, 1358e1366. Li, L., Clevers, H., 2010. Coexistence of quiescent and active adult stem cells in mammals. Science 327, 542e545. Li, L., Hartley, R., Reiss, B., Sun, Y., Pu, J., Wu, D., Lin, F., Hoang, T., Yamada, S., Jiang, J., Zhao, M., 2012. E-cadherin plays an essential role in collective directional migration of large epithelial sheets. Cell. Mol. Life Sci. 69, 2779e2789. Lichtman, J.W., Livet, J., Sanes, J.R., 2008. A technicolour approach to the connectome. Nat. Rev. 9, 417e422. Livet, J., Weissman, T.A., Kang, H., Draft, R.W., Lu, J., Bennis, R.A., Sanes, J.R., Lichtman, J.W., 2007. Transgenic strategies for combinatorial expression of fluorescent proteins in the nervous system. Nature 450, 56e62. Ljubimov, A.V., Burgeson, R.E., Butkowski, R.J., Michael, A.F., Sun, T.-T., Kenney, M.C., 1995. Human corneal basement membrane heterogeneity: topographical differences in the expression of type IV collagen and laminin isoforms. Lab. Investig. 72, 461e473. Lopez-Garcia, C., Klein, A.M., Simons, B.D., Winton, D.J., 2010. Intestinal stem cells replacement follows a pattern of neutral drift. Science 330, 822e825. Lu, H., Zimek, A., Chen, J., Hesse, M., Bussow, H., Weber, K., Magin, T.M., 2006. Keratin 5 knockout mice reveal plasticity of keratin expression in the corneal epithelium. Eur. J. Cell Biol. 85, 803e811. Lu, R., Qu, Y., Ge, J., Zhang, L., Su, Z., Pflugfelder, S.C., Li, D.Q., 2012. Transcription factor TCF4 maintains the properties of human corneal epithelial stem cells. Stem Cells 30, 753e761. Mackman, G., Brightbill, F.S., Optiz, J.M., 1979. Corneal changes in aniridia. Am. J. Ophthalmol. 87, 497e502. Mackman, G.S., Polack, F.M., Sydrys, L., 1983. Hurricane keratitis in penetrating keratoplasty. Cornea 2, 31e34. Majo, F., Rochat, A., Nicolas, M., Jaoude, G.A., Barrandon, Y., 2008. Oligopotent stem cells are distributed throughout the mammalian ocular surface. Nature 456, 250e254. Mann, I., 1944. A study of epithelial regeneration in the living eye. Br. J. Ophthalmol. 28, 26e40. Mann, I., 1964. The Development of the Human Eye. British Medical Association, London. Mariappan, I., Kacham, S., Purushotham, J., Maddileti, S., Siamwala, J., Sangwan, V.S., 2014. Spatial distribution of niche and stem cells in ex vivo human limbal cultures. Stem Cells Trans. Med. 3, 1331e1341. Marshall, J., Grindle, C.F.J., 1978. Fine structure of the cornea and its development. Trans. Ophthalmol. Soc. U. K. 98, 320e328. , G., Dekoninck, S., Drogat, B., Youssef, K.K., Brohee , S., Sotiropoulou, P.A., Mascre Simons, B.D., Blanpain, C., 2012. Distinct contribution of stem and progenitor cell to epidermal maintenance. Nature 489, 257e264. Maseruka, H., Rigway, A., Tullo, A., Bonshek, R., 2000. Developmental changes in patterns of expression of tenascin-C variants in the human cornea. Investig. Ophthalmol. Vis. Sci. 41, 4101e4107. Masters, B.R., Thaer, A.A., 1994. In vivo human corneal confocal microscopy of identical fields of subepithelial nerve plexus, basal epithelial, and wing cells at different times. Microsc. Res. Tech. 29, 350e356. Mathers, W.D., Daley, T.E., 1996. Tear flow and evaporation in patients with and without dry eye. Ophthalmology 103, 667e669. Mathers, W.D., Lemp, M.A., 1992. Morphology and movement of corneal surface cells in humans. Curr. Eye Res. 11, 517e523. McKenna, C.C., Lwigale, P.Y., 2011. Innervation of the mouse cornea during development. Investig. Ophthalmol. Vis. Sci. 52, 30e35. McNairn, A.J., Guasch, G., 2011. Epithelial transition zones: merging microenvironments, niches, and cellular transformation. Eur. J. Dermatol. 21 (Suppl. 2), 21e28. Mei, H., Nakatsu, M.N., Baclagon, E.R., Deng, S.X., 2014. Frizzled 7 maintains the undifferentiated state of human limbal stem/progenitor cells. Stem Cells 32, 938e945. Meilhac, S.M., Adams, R.J., Morris, S.A., Danckaert, A., Le Garrec, J.-F., ZernickaGoetz, M., 2009. Active cell movements coupled to positional induction are involved in lineage segregation in the mouse blastocyst. Dev. Biol. 331, 210e221. € tzer-Schrehardt, U., Meyer-Blazejewska, E.A., Call, M.K., Yamanaka, O., Liu, H., Schlo Kruse, F.F.E., Kao, W.W., 2011. From hair to cornea: towards the therapeutic use of hair follicle-derived stem cells in the treatment of limbal stem cell deficiency. Stem Cells 29, 57e66. Miri, A., Hashmani, K., Al-aquba, M., Faraj, L.A., Fares, U., Otri, A.M., Said, D.G., Dua, H.S., 2012. The effect of gravitational force on limbal stem cell growth. Br. J. Ophthalmol. 96, 1034e1037. Moore, K.A., Lemischka, I.R., 2006. Stem cells and their niches. Science 311, 1880e1885. Mort, R.L., Ramaesh, T., Kleinjan, D.A., Morley, S.D., West, J.D., 2009. Mosaic analysis of stem cell function and wound healing in the mouse corneal epithelium. BMC Dev. Biol. 9, 4. Moyer, P.D., Kaufman, A.H., Zhang, Z., Kao, C.W., Spaulding, A.G., Kao, W.W., 1996. Conjunctival epithelial cells can resurface denuded cornea, but do not transdifferentiate to express cornea-specific keratin 12 following removal of limbal epithelium in mouse. Differentiation 60, 31e38.

21

Muftuoglu, O., Uçakhan, O.O., Kanpolat, A., 2006. Clinical and in vivo confocal microscopy findings in patients receiving tamoxifen citrate. Eye Contact Lens 32, 228e232. Murray, S.A., Eppig, J.T., Smedley, D., Simpson, E.M., Rosenthal, N., 2012. Beyond knockouts: cre resources for conditional mutagenesis. Mamm. Genome 23, 587e599. Müther, P.S., Semkova, I., Schmidt, K., Abari, E., Kuebbeler, M., Beyer, M., Abken, H., Meyer, K.L., Kociok, N., Joussen, A.M., 2010. Conditions of retinal glia and inflammatory cell activation after irradiation in a GFP-chimeric mouse model. Investig. Ophthalmol. Vis. Sci. 51, 4831e4839. Nagasaki, T., Zhao, J., 2003. Centripetal movement of corneal epithelial cells in the normal adult mouse. Investig. Ophthalmol. Vis. Sci. 44, 558e566. Nagasaki, T., Zhao, J., 2005. Uniform distribution of epithelial stem cells in the bulbar conjunctiva. Investig. Ophthalmol. Vis. Sci. 46, 126e132. Nagy, A., 2000. Cre Recombinase: the universal reagent for genome tailoring. Genesis 26, 99e109. Nakamura, T., Hamuro, J., Takaoshi, M., Simmons, S., Maruyama, K., Zaffalon, A., Bentley, A.J., Kawasaki, S., Nagata-Takaoka, M., Fullwood, N.J., Itami, S., Sano, S., Ishii, M., Barrandon, Y., Kinoshita, S., 2014. LRIG1 inhibits STAT3-dependent inflammation to maintain corneal homeostasis. J. Clin. Investig. 124, 385e397. Nakatsu, M.N., Vartanyan, L., Vu, D.M., Ng, M.Y., Li, X., Deng, S.X., 2013. Preferential biological processes in the limbus by differential gene profiling. PLoS ONE 8, e61833. Noureddin, B.H., Seoud, M., Bashshur, Z., Salem, Z., Shamseddin, A., Khalil, A., 1999. Ocular toxicity in low-dose tamoxifen: a prospective study. Eye 13, 729e733. Nubile, M., Curcio, C., Dua, H.S., Calienna, R., Lanzini, M., Iezzi, M., Mastropasqua, R., Agnifili, L., Mastropasqua, L., 2013. Pathological changes of the anatomical structure and markers of the limbal stem cell niche due to inflammation. Mol. Vis. 19, 516e525. Ordonez, P., Chow, S., Wakefield, D., Di Girolamo, N., 2013. Human limbal epithelial progenitor cells express avb5 and the interferon-inducible chemokine CXCL10/ IP-10. Stem Cell Res. 11, 888e901. Ouyang, H., Xue, Y., Lin, Y., Zhang, X., Xi, L., Patel, S., Cai, H., Luo, H., Zhang, M., Zhang, M., Yang, Y., Li, G., Li, H., Jiang, W., Yeh, E., Lin, J., Pei, M., Zhu, J., Cao, G., Zhang, L., Yu, B., Chen, S., Fu, X.-D., Liu, Y., Zhang, K., 2014. WNT7A and PAX6 define corneal epithelial homeostasis and pathogenesis. Nature 115, 358e361. Pajoohesh-Ganji, A., Ghosh, S.P., Simmens, S.J., Stepp, M.A., 2006. Integrins in slowcycling corneal epithelial cells at the limbus in the mouse. Stem Cells 24, 1075e1086. Pajoohesh-Ganji, A., Pal-Ghosh, S., Tadvalkar, G., Stepp, M.A., 2012. Corneal goblet cells and their niche: implications for corneal stem cell deficiency. Stem Cells 30, 2032e2043. Pan, Y.A., Freundlich, T., Weissman, T.A., Schoppik, D., Wang, X.C., Zimmermann, S., Ciruna, B., Sanes, J.R., Lichtman, J.W., Schier, A.F., 2013. Zebrabow: multispectral cell labeling for cell tracing and lineage analysis in zebrafish. Development 140, 2835e2846. Patel, D.V., McGhee, C.N.J., 2005. Mapping of the normal human corneal sub-basal nerve plexus by in vivo laser scanning confocal microscopy. Investig. Ophthalmol. Vis. Sci. 46, 4485e4488. Patel, D.V., McGhee, C.N.J., 2008. In vivo laser scanning confocal microscopy confirms that the human corneal sub-basal nerve plexus is a highly dynamic structure. Investig. Ophthalmol. Vis. Sci. 49, 3409e3412. Pe'er, J., Zajicek, G., Greifner, H., Kogan, M., 1996. Streaming conjunctiva. Anat. Rec. 245, 36e40. Pellegrini, G., Golisano, O., Paterna, P., Lambiase, A., Bonini, S., Rama, P., De Luca, M., 1999. Location and clonal analysis of stem cells and their differentiated progeny in the human ocular surface. J. Cell. Biol. 145, 769e782. Pellegrini, G., Traverso, C.E., Franzi, A.T., Zingirian, M., Cancedda, R., De Luca, M., 1997. Long-term restoration of damaged corneal surfaces with autologous cultivated corneal epithelium. Lancet 349, 990e993. Persson, A., 1998. How do we understand the Coriolis Force? Bull. Am. Meterol. Soc. 79, 1373e1385. Peters, A., 1885. Regeneration des epithels der cornea. Inaugural-Dissertation, pp. 1e32. Bonn. Pitz, S., Moll, R., 2002. Intermediate-filament expression in ocular tissue. Prog. Retin Eye Res. 21, 241e267. Potten, C.S., 1975. Epidermal cell production rates. J. Investig. Dermatol. 65, 488e500. Price, J., Turner, D., Cepko, C., 1987. Lineage analysis in the vertebrate nervous system by retrovirus-mediated gene transfer. Proc. Natl. Acad. Sci. U. S. A. 84, 156e160. Pulhorn, G., Thiel, H.J., 1976. Ultrastructural aspects of chloroquine-keratopathy. Albr. Von Graefes Arch. Klin. Exp. Ophthalmol. 201, 89e99. Qi, H., Li, D.-Q., Bian, F., Chuang, E.Y., Jones, D.B., Pflugfelder, S.C., 2008a. Expression of glial cell-derived neurotrophic factor and its receptor in the stem-cell-containing human limbal epithelium. Br. J. Ophthalmol. 92, 1269e1274. Qi, H., Li, D.-Q., Shine, H.D., Chen, Z., Yoon, K.-C., Jones, D.B., Pflugfelder, S.C., 2008b. Nerve growth factor and its receptor TrkA serve as potential markers for human corneal epithelial progenitor cells. Exp. Eye Res. 86, 34e40. Rama, P.P., Matuska, S., Paganoni, G., Spinelli, A., De Luca, M., Pellegrini, G., 2010. Limbal stem-cell therapy and long-term corneal regeneration. N. Engl. J. Med. 363, 147e155. e. Arch. d'Anatom. Ranvier, P.L., 1898. De la cicatrisation des plaies de la corne Microsc.

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

JPRR587_proof ■ 28 April 2015 ■ 22/23

22

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

Reinhard, T., Spelsberg, H., Henke, L., Kontopoulos, T., Enczmann, J., Wernet, P., Berschick, P., Sundmacher, R., Bohringer, D., 2004. Long-term results of allogeneic penetrating limbo-keratoplasty in total limbal stem cells deficiency. Ophthalmology 111, 775e782. Ren, H., Wilson, G., 1996. Apoptosis in the corneal epithelium. Investig. Ophthalmol. Vis. Sci. 37, 1017e1025. Reneker, L.W., Silversides, D.W., Xu, L., Overbeek, P.A., 2000. Formation of corneal endothelium is essential for anterior segment development- a transgenic mouse model of anterior segment dysgenesis. Development 127, 533e542. Rinkevich, Y., Lindau, P., Ueno, H., Longaker, M.T., Weissman, I.L., 2011. Germ-layer and lineage-restricted stem/progenitors regenerate the mouse digit tip. Nature 476, 409e413. Rodrigues, M., Ben-Zvi, A., Krachmer, J., Schermer, A., Sun, T.-T., 1987. Suprabasal expression of a 64-kilodalton keratin (no. 3) in developing human corneal epithelium. Differentiation 34, 60e67. Romano, R.A., Ortt, K., Birkaya, B., Smalley, K., Sinha, S., 2009. An active role of the DN isoform of p63 in regulating basal keratin genes K5 and K14 and directing epidermal cell fate. PLoS ONE 4, e5623. Ryder, M.I., Weinreb, R.N., 1990. Cytokeratin patterns in corneal, limbal, and conjunctival epithelium. Investig. Ophthalmol. Vis. Sci. 31, 2230e2234. Sanes, J.R., 1989. Analyzing cell lineage with a recombinant retrovirus. Trends Neurosci. 12, 21e28. Sanes, J.R., Rubenstein, J.L.R., Nicolas, J.-F., 1986. Use of a recombinant retrovirus to study post-implantation cell lineage in mouse embryo. EMBO J. 5, 3133e3142. Sanges, D., Lluis, F., Cosma, M.P., 2011. Cell-fusion-mediated reprogramming: potency or transdifferentiation? Implications for regenerative medicine. Adv. Exp. Med. Biol. 713, 137e159. Sangwan, V.S., Basu, S., Vemuganti, G.K., Sejpal, K., Subramaniam, S.V., Bandyopadhyay, S., Krishnaiah, S., Gaddipati, S., Tiwari, S., Balasubramanian, D., 2011. Clinical outcomes of xeno-free autologous cultivated limbal epithelial transplantation: a 10-year study. Br. J. Ophthalmol. 95, 1525e1530. Satoh, A., Sakamaki, K., Ide, H., Tamura, K., 2005. Characteristics of initiation and early events for muscle development in the Xenopus limb bud. Dev. Dyn. 234, 846e857. Schepers, A.G., Snippert, H.J., Strange, D.E., van den Born, M., van Es, J.H., van de Wetering, M., Clevers, H., 2012. Lineage tracing reveals Lgr5þ stem cell activity in mouse intestinal adenomas. Science 337, 730e735. Schermer, A., Galvin, S., Sun, T.-T., 1986. Differentiation-related expression of a major 64K corneal keratin in vivo and in culture suggests limbal location of corneal epithelial stem cells. J. Cell Biol. 103, 49e62. Scheving, L.E., Burns, E.R., Pauly, J.E., Tsai, T.H., 1978. Circadian variation in cell division of the mouse alimentary tract, bone marrow and corneal epithelium. Anat. Rec. 191, 479e486. Scheving, L.E., Pauly, J.E., 1967. Circadian phase relationships of thymidine-3H uptake, labelled nuclei, grain count, and cell division rate in rat corneal epithelium. J. Cell Biol. 32, 677e683. €tzer-Schrehardt, U., Kruse, F.E., 2005. Identification and characterization of Schlo limbal stem cells. Exp. Eye Res. 81, 247e264. €tzer-Schrehardt, U., Dietrich, T., Saito, K., Sorokin, L., Sasaki, T., Paulsson, M., Schlo Kruse, F.E., 2007. Characterization of extracellular matrix components in the limbal epithelial stem cell compartment. Exp. Eye Res. 85, 845e860. Schofield, R., 1978. The relationship between the spleen colony-forming cell and the haemopoietic stem cell. Blood Cells 4, 7e25. Schofield, R., 1983. The stem cell system. Biomed. Pharmacother. 37, 375e380. Sellheyer, K., Spitznas, M., 1988. Surface differentiation of the human corneal epithelium during prenatal development. Graefe's Arch. Clin. Exp. Ophthalmol. 226, 482e488. Seomun, Y., Joo, C.-K., 2008. Lumican induces human corneal epithelial migration and integrin expression via ERK 1/2 signaling. Biochem. Biophys. Res. Commun. 372, 221e225. Sevel, D., Isaacs, R., 1988. A re-evaluation of corneal development. Trans. Am. Ophthalmol. Soc. 86, 178e207. Shanmuganathan, V.A., Foster, T., Kulkarni, B.B., Hopkinson, A., Gray, T., Powe, D.G., Lowe, J., Dua, H.S., 2007. Morphological characteristics of the limbal epithelial crypt. Br. J. Ophthalmol. 91, 514e519. Shapiro, M.S., Friend, J., Thoft, R.A., 1981. Corneal re-epithelialization from the conjunctiva. Investig. Ophthalmol. Vis. Sci. 21, 135e142. Shimomura, O., Johnson, F.H., Saiga, Y., 1962. Extraction, purification and properties of aequorin, a bioluminescent protein from the luminous hydromedusan, Aequorea. J. Cell. Comp. Physiol. 59, 223e239. Shortt, A.J., Secker, G.A., Notara, M.D., Limb, G.A., Knaw, P.T., Tuft, S.J., Daniels, J.T., 2007a. Transplantation of ex vivo cultured limbal epithelial stem cells: a review of techniques and clinical results. Surv. Ophthalmol. 52, 483e502. Shortt, A.J., Secker, G.A., Munro, P.M., Khaw, P.T., Tuft, S.J., Daniels, J.T., 2007b. Characterization of the limbal epithelial stem cell niche: novel imaging techniques permit in vivo observation and targeted biopsy of limbal epithelial stem cells. Stem Cells 25, 1402e1409. Silverstein, A.M., Rossman, A.M., de Leon, A.S., 1970. Survival of donor epithelium in experimental corneal xenografts. Am. J. Ophthalmol. 69, 448e453. Skeens, H.M., Brooks, B.P., Holland, E.J., 2011. Congenital aniridia variant: minimally abnormal irides with severe limbal stem cell deficiency. Ophthalmology 118, 1260e1264. Song, J., Lee, Y.G., Houston, J., Petroll, W.M., Chakravarti, S., Cavanagh, H.D., Jester, J.V., 2003. Neonatal corneal stromal development in the normal and lumican-deficient mouse. Investig. Ophthalmol. Vis. Sci. 44, 548e557.

Sonoda, K.-H., Nakao, S., Nakamura, T., Oshima, T., Qiao, H., Hisatomi, T., Kinoshita, S., Ishibashi, T., 2005. Cellular events in the normal and inflamed cornea. Cornea 24 (Suppl. l), S50eS54. Soong, H.K., Parkinson, W.C., Bafna, S., Sulik, G.L., Huang, S.C.M., 1990. Movements of cultured corneal epithelial cells and stromal fibroblasts in electric fields. Investig. Ophthalmol. Vis. Sci. 31, 2278e2282. Soriano, P., 1999. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat. Genet. 21, 70e71. Snippert, H.J., van der Flier, L.G., Sato, T., van Es, J.H., van den Born, M., KroonVeeboer, C., Barker, N., Klein, A.M., van Rheenen, J., Simons, B.D., Clevers, H., 2010. Intestinal crypt homeostasis results from neutral competition between symmetrically dividing Lgr5 stem cells. Cell 143, 134e144. Spemann, H., Mangold, H., 1924. Uber induction von embryonalanlagen durch implantation artfemder organisatoren. Arch. Entwicklungsmech. Org. 100, 599e638. Sta Iglesia, D.D., Gala, P.H., Qiu, T., Stepp, M.A., 2000. Integrin expression during epithelial migration and restratification in the tenascin-C-deficient mouse cornea. J. Histochem. Cytochem. 48, 363e376. Stenevi, U., Hanson, C., Claesson, M., Corneliusson, E., Ek, S., 2002. Survival of transplanted human corneal stem cells: case report. Acta Ophthalmol. Scand. 80, 105e108. Stepp, M.A., Spurr-Michaud, S., Gipson, I.K., 1993. Integrins in the wounded and unwounded stratified squamous epithelium of the cornea. Investig. Ophthalmol. Vis. Sci. 34, 1829e1844. Stepp, M.A., Zhu, L., 1997. Upregulation of a9 integrin and tenascin during epithelial regeneration and debridement in the cornea. J. Histochem. Cytochem. 45, 189e201. Stepp, M.A., Zhu, L., Cranfill, R., 1996. Changes in b4 integrin expression and localization in vivo in response to corneal epithelial injury. Investig. Ophthalmol. Vis. Sci. 37, 1593e1601. Sturtevant, A.H., 1929. The claret mutant type of Drosophila simulans: a study of chromosome elimination and of cell lineage. Z. Wiss. Zool. 135, 323e356. Sulston, J.E., Schierenberg, E., White, J.G., Thomson, J.N., 1983. The embryonic cell lineage of the nematode Caenorhabditis elegans. Dev. Biol. 100, 64e119. Summers, R.G., Piston, D.W., Harris, K.M., Morrill, J.B., 1996. The orientation of first cleavage in the Sea Urchin embryo, Lytechinus variegatus, does not specify the axes of bilateral symmetry. Dev. Biol. 175, 177e183. Sun, T.-T., Tseng, S.C., Lavker, R.M., 2010. Location of corneal epithelial stem cells. Nature 463, E10eE11. Szerenyi, K., Wang, X.W., Gabrielian, K., LaBree, L., McDonnell, P.J., 1994. Immunohistochemistry with 5-bromo-2-deoxyuridine for visualization of mitotic cells in the corneal epithelium. Cornea 13, 487e492. Tanifuji-Terai, N., Terai, K., Hayashi, Y., Chikama, T.-I., Kao, W.W.Y., 2006. Expression of keratin 12 and maturation of corneal epithelium during development and postnatal growth. Investig. Ophthalmol. Vis. Sci. 47, 545e551. Tarafdar, S., Lim, L.T., Collins, C.E., Ramaesh, K., 2012. Tamoxifen keratopathy as seen with in-vivo confocal microscopy. Semin. Ophthalmol. 27, 27e28. Taylor, G., Lehrer, M.S., Jensen, P.J., Sun, T.-T., Lavker, R.M., 2000. Involvement of follicular stem cells in forming not only the follicle but also the epidermis. Cell 102, 451e461. Thoft, R.A., Friend, J., 1977. Biochemical transformation of regenerating ocular surface epithelium. Investig. Ophthalmol. Vis. Sci. 16, 14e20. Thomas, P.B., Liu, Y.-H., Zhuang, F.F., Selvam, S., Song, S.W., Smith, R.E., Trounsdale, M.D., Yia, S.C., 2007. Identification of Notch-1 expression in the limbal basal epithelium. Mol. Vis. 13, 337e344. Tian, H., Biehs, B., Warming, S., Leong, K.G., Rangell, L., Klein, O.D., de Sauvage, F.J., 2011. A reserve stem cell population in small intestine renders Lgr5-positive cells dispensable. Nature 478, 255e259. Tinsley, R.B., Faijerson, J., Eriksson, P.S., 2006. Efficient non-viral transfection of adult neural stem/progenitor cells, without affecting viability, proliferation or differentiation. J. Gene Med. 8, 72e81. Townsend, W.M., 1991. The limbal palisades of Vogt. Trans. Am. Ophthalmol. Soc. 89, 721e755. Tseng, S.C.G., 1989. Concept and application of limbal stem cells. Eye 3, 141e157. Tung, T.C., 1932. Experiences de coloration vitale sur l'oeuf d'Ascidiella aspersa. Arch. Biol. 43, 451e469. van Keymeulen, A., Rocha, A.S., Ousset, M., Beck, B., Bouvencourt, G., Rock, J., Sharma, N., Dekoninck, S., Blanpain, C., 2011. Distinct stem cells contribute to mammary gland development and maintenance. Nature 479, 189e193. Vascotto, S.G., Griffith, M., 2006. Localization of candidate stem and progenitor cell markers within the human cornea, limbus, and bulbar conjunctiva in vivo and in cell culture. Anat. Rec. A 288A, 921e931. Vasioukhin, V., Degenstein, L., Wise, B., Elaine, F., 1999. The magical touch: genome targeting in epidermal stem cells induced by tamoxifen application to mouse skin. Proc. Natl. Acad. Sci. U. S. A. 96, 8551e8556. Vauclair, S., Majo, F., Durham, A.-D., Ghyselinck, N.B., Barrandon, Y., Radtke, F., 2007. Corneal epithelial cell fate is maintained during repair by Notch 1 signaling via the regulation of vitamin A metabolism. Dev. Cell 13, 242e253. Virchow, R., 1858. Die cellularpathologie in ihrer begrundung auf physiologische und pathologische gewebelehre. A Hirschwald, Berlin. €rtlicher vitalfa €rbung. II Vogt, W., 1929. Gestaltungsanalyse am amphibienkeim mit o Teil. Gastrulation und mesodermbildung bei urodelen und anuren. Arch. Entw. Mech. 120, 384e706. Watt, F.M., Hogan, B.L.M., 2000. Out of Eden: stem cells and their niches. Science 287, 1427e1430.

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

JPRR587_proof ■ 28 April 2015 ■ 23/23

N. Di Girolamo / Progress in Retinal and Eye Research xxx (2015) 1e23

Q13

Wei, Z.G., Cotsarelis, G., Sun, T.T., Lavker, R.M., 1995. Label-retaining cells are preferentially located in fornical epithelium: implications on conjunctival epithelial homeostasis. Investig. Ophthalmol. Vis. Sci. 36, 236e246. Wei, Z.-G., Sun, T.-T., Lavker, R.M., 1996. Rabbit conjunctival and corneal epithelial cells belong to two separate lineages. Investig. Ophthalmol. Vis. Sci. 37, 523e533. Weissman, T.A., Sanes, J.R., Lichtman, J.W., Livet, J., 2011. Generating and imaging multicolor Brainbow mice. Cold Spring Harb. Protoc. 7, 763e769. West, J.D., Dor a, N.J., Collinson, J.M., 2015. Evaluating alternative stem cell hypotheses for adult corneal epithelial maintenance. World J. Stem Cells 7, 281e299. Whitcher, J.P., Srinivasan, M., Upadhyay, M.P., 2001. Corneal blindness: a global perspective. Bull. World Health Org. 79, 214e221. Whitman, C.O., 1878. The embryology of Clepsine. Quant. J. Microsc. Sci. 18, 215e315. Whitman, C.O., 1887. A contribution to the history of germ layers in Clepsine. J. Morphol. 1, 105e182. Wider, S., 1963. Effect of hormones on eyelid disjunction in the rat. Ophthalmologica 146, 268e274. Wiley, L., SundarRaj, N., Sun, T.-T., Thoft, R.A., 1991. Regional heterogeneity in human corneal and limbal epithelia: an immunohistochemical evaluation. Investig. Ophthalmol. Vis. Sci. 32, 5944e6602. Wilson, A., Laurenti, E., Oser, G., van der Wath, R.C., Blanco-Bose, W., Jaworski, M., , P., Macdonald, H.R., Offner, S., Dunant, C.F., Eshkind, L., Bockamp, E., Lio Trumpp, A., 2008. Hematopoietic stem cells reversibly switch from dormancy to self-renewal during homeostasis and repair. Cell 135, 1118e1129. Wilson, E.B., 1892. The cell-lineage of Nereis. A contribution to the cytogeny of the annelid body. J. Morphol. 6, 361e481. Wirthschafter, J.D., Ketcham, J.M., Weinstock, R.J., Tebesh, T., McLoon, L.K., 1999. Mucocutaneous junction as the major source of replacement palpebral conjunctival epithelial cells. Investig. Ophthalmol. Vis. Sci. 40, 3138e3146.

23

Xie, H.-T., Chen, S.-Y., Li, G.-G., Tseng, S.C.G., 2011. Limbal epithelial stem/progenitor cells attract stromal niche cells by SDF-1/CXCR4 signaling to prevent differentiation. Stem Cells 29, 1874e1885. Yin, J.-Q., Lui, W.-Q., Liu, C., Zhang, Y.-H., Hua, J.-L., Liu, W.-S., Dou, Z.-Y., Lei, A.-M., 2013. Reconstruction of damaged corneal epithelium using Venus-labeled limbal epithelial stem cells and tracking of surviving donor cells. Exp. Eye Res. 115, 246e254. Yoshida, S., Shimmura, S., Kawakita, T., Miyashita, H., Den, S., Shimazaki, J., Tsubota, K., 2006. Cytokeratin 15 can be used to identify the limbal phenotype in normal and diseased ocular surfaces. Investig. Ophthalmol. Vis. Sci. 47, 4780e4786. You, L., Kruse, F.E., Volcker, H.E., 2000. Neurotrophic factors in the human cornea. Investig. Ophthalmol. Vis. Sci. 41, 692e702. Zajicek, G., Perry, A., Pe'er, J., 1995. Streaming of labeled cells in the conjunctival epithelium. Cell Prolif. 28, 235e243. Zalokar, M., Sardet, C., 1984. Tracing of cell lineage in embryonic development of Phallusia mammillata (Ascidia) by vital staining of mitochondria. Dev. Biol. 102, 195e205. Zhao, J., Mo, V., Nagasaki, T., 2009. Distribution of label-retaining cells in the limbal epithelium of a mouse eye. J. Histochem. Cytochem. 57, 177e185. Zhao, M., Chalmers, L., Cao, L., Vieira, A.C., Mannis, M., Reid, B., 2012. Electrical signaling in control of ocular cell behaviors. Prog. Ret. Eye Res. 31, 65e88. Zhao, M., Song, B., Pu, J., Forrester, J.V., McCaig, C.D., 2003. Direct visualization of a stratified epithelium reveals that wounds heal by unified sliding of cell sheets. FASEB J. 17, 397e406. Zieske, J.D., 2004. Corneal development associated with eyelid opening. Int. J. Dev. Biol. 48, 903e911. Zirm, E., 1906. Eine erfolgreiche totale keratoplastik. Graefes Arch. Ophthalmol. 64, 580e593.

Please cite this article in press as: Di Girolamo, N., Moving epithelia: Tracking the fate of mammalian limbal epithelial stem cells, Progress in Retinal and Eye Research (2015), http://dx.doi.org/10.1016/j.preteyeres.2015.04.002

25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48