Prof Keith Meek
Research Topics and Related Projects:
Corneal Structure-Function Relationships
Transparency: Despite the fact that light entering the eye at the cornea’s periphery contributes significantly to vision, all previous studies on light transmission, including our own, have concentrated on making measurements at the centre of the cornea. We have corrected this anomaly and discovered that transmittance decreases peripherally across the cornea, that this can be modeled, and is caused by the peripheral increase in collagen fibril diameters. Functionally, this likely represents the need to compromise transparency in order to achieve the extra strength required to maintain the peripheral change in curvature of the cornea. It is intriguing, however, that the wavelength dependence remains constant across the cornea an obvious requirement to preserve colour balance. Extending our analyses outside the visible spectrum, we discovered that UV transmission through the cornea is lowest at the corneal periphery. UV can be damaging to cells, and the UV dose at the corneal endothelium is a critical clinical question because human corneal endothelial cells do not have replicative ability in vivo. The function of diseased or damaged cells is assumed by the expansion of adjacent healthy cells, however, if cell density drops below a certain threshold, the endothelium’s functional reserve is exceeded, the cornea swells and vision is impaired. The emerging concept, based on our work and in collaboration with Prof N Joyce, Harvard Medical School, is that the reduced UV transmittance at the corneal periphery is responsible for the relatively low levels of endothelial cell death at that location in the cornea. We have also studied the acquisition of corneal transparency during embryogenesis, and using chick and mouse models have shown that this is underpinned by structural changes in the collagen fibrils, possibly driven by the changing sulphation patterns of corneal glycosaminoglycans.
Ultrastructure: The arrangement of collagen fibrils within the corneal stroma has long been known to be responsible for its remarkable transparency. However, we still do not know what the arrangement of fibrils is, and how it is maintained in the normal cornea. We have been investigating this organization and how proteoglycans control it. Furthermore, at a higher structural level, we previously showed that the lamellae are not isotropically arranged but, rather, have preferred directions in different parts of the tissue (Figure 1). We have extended this work to determine lamellar organization at different depths in the cornea and limbus, and to try to understand the roles of the specific structural features we had previously described. This is particularly important because of the move towards lamellar corneal surgery (anterior and posterior) in place of full thickness penetrating grafts and the growing realization that the cornea is not a homogeneous structure with respect to depth.
Figure 1: X-ray scattering allows us to map the preferred orientations of the lamellae across the cornea and sclera.
Collagen is strongest along its length, thus the arrangement of fibrils and lamellae dictates corneal biomechanics. We previously showed that there are at least three distinct lamellar structures within the cornea/limbus: (i) preferred vertical and horizontal lamellae centrally; (ii) a tangential or circumferential arrangement near the limbus; (iii) anchoring lamellae that enter the stroma peripherally without crossing the pre-pupillary zone (Figure 2). This was an important finding because it can help guide surgeons when choosing where, in what direction, and to what depth an incision should be placed, and our results have swiftly found their place in influential corneal textbooks and lecture materials by leading educators.
Figure 2: X-ray scatter patterns from different regions in the cornea (insets) lead to a simplified model for the preferred lamellar directions in the cornea. However, it should be noted that changes in direction occur by lamella bifurcations rather than by bending.
We have exploited femtosecond laser technology to precisely cut the cornea into layers, parallel to the surface, to map in fine detail the lamellar arrangement in the individual layers and build a more detailed 3-D picture of the lamellar arrangement in the central cornea. In studies of the corneal-scleral junction a combination of X-ray scattering and multiphoton microscopy (utilizing both the second harmonic signal and two photon fluorescence) led us to discover that the limbal annulus is a distinct structure that is confined to deep stroma, and contains a separate population of elastic fibres. To understand how the distinct structures of the central human cornea and limbus integrate, we have exploited the high flux of the microfocus X-ray beamlines to show precisely how the orthogonal corneal fibrils change direction 1.0-1.5mm inside the limbus to integrate with the circumferential limbal fibrils. Our research also allowed us to characterize the age-related changes in corneal biomechanical properties in relation to corneal structural changes.
To try to understand further the reason for the predominantly inferior-superior and nasal-temporal lamellae in the central cornea, we carried out a comparative study. This included human, marmoset, horse, cow, pig, rabbit and mouse, and the results were very revealing, showing that the predominant orientation varies markedly between species, even between those that are closely related under taxonomic classification. Although fibril orientation showed no correlation with corneal size, shape, thickness or blink rate frequency, we identified that an excess of collagen directed towards one or both sets of opposing rectus muscles is a feature of animals that have higher visual acuity. This led us to propose a relationship between collagen fibril arrangement and the frequency of action and force generated by the various extraocular muscles during eye movement and image fixation.
In 2005 we proposed a model to explain how the ectasia observed in keratoconus may arise due to tearing and slippage of interwoven lamellae. Later, we showed how structural changes in lamellae disposition in keratoconus could be broadly related to cornea’s shape change, and that changes in collagen orientation occur in both the anterior and posterior stroma. .We also showed how collagen fibril diameters and spacings and proteoglycan sulphation patterns change as a function of depth in keratoconus, and how the changes depend on the severity of the disease. The changes in proteoglycans are consistent with our slippage model for keratoconus, but to establish this we need to elucidate changes in the 3-D organization of these molecules.
Keratoconus has been a focus of our research into corneal pathology, but other conditions have been studied too. Our combined microfocus X-ray scattering/electron microscopy study of macular corneal dystrophy has added to our previous results by showing that the structural changes we reported occur in the deep stroma. This was attributed to the concept that the structural influence of sulphated keratan sulphate is high in this region of the cornea. Similarly, we have added to a sequence of studies into the pathogenesis of a group of proteoglycans storage diseases called the mucopolysaccharidoses. The recent study of Sly syndrome disclosed a number of similarities in the structural changes across the disease range even though the proteoglycans involved differ. We hope soon to be in a position to collate all the data to extract more information about how the presence of proteoglycans around collagen fibrils, or accumulated in the matrix, affects collagen structure and organization and thus corneal transparency. Stromal oedema can occur in a number of situations including bullous keratopathy, Fuchs dystrophy, keratoconus and during wound healing. Klf4 is a highly expressed transcription factor that plays an important role in maturation and maintenance of the ocular surface; its absence leads to stromal oedema. In conjunction with groups at the National Eye Institute, Bethesda and at the University of Pittsburgh we examined oedema-related changes in Klf4 conditional null corneal stroma. This indicated that corneal oedema in the null mouse corneas is characterized by significant changes in both the stromal collagen and by a reduction in proteoglycans, accompanied by up-regulation of matrix metalloproteinases.
Corneal wound healing is a hugely important topic, whether it is a wound following injury or surgery. By studying different types of corneal wound we have shown that swelling leads to increases in the collagen interfibrillar spacing and that this was more severe in trephine wounds (where tissue is removed) than from incisional LASIK-like wounds. This research has also indicated that the integrity of the epithelial basement membrane in murine debridement wounds is key in determining the resulting level of cytokine-driven corneal fibrosis and opacity, and that, structurally, this is underpinned by increased collagen fibril disorder and the deposition of abnormally large fibrils and proteoglycans. Furthermore, the gene-response profile clearly showed hitherto unreported increases in lumican and keratocan that preceded established fibrotic markers, indicating that the temporal aspects of proteoglycan control during wound healing is more complex than previously thought.
Tens of millions of people worldwide have undergone elective surgery to correct refractive error, in the form of LASIK. Despite the popularity of the procedure the LASIK wound does not properly heal, and as a result the anterior corneal flap is never properly adherent. We have made measurements of the force required to detach these flaps and have investigated a number of cytokines, stromal fibroblasts and cornea crosslinking, as potential biological “glues”. All of these increased flap adherence but only crosslinking preserved transparency. Critically, our studies aimed at future bespoke and autologous approaches to improving LASIK surgery have shown that oral mucosal cells (at least under organ culture conditions) can improve flap strength without compromising transparency. This has huge potential implications for the field of refractive surgery.
Riboflavin/UVA crosslinking is now being used extensively to treat keratoconus and other conditions. Despite its widespread clinical use, very little is known about how stability is conferred and, in particular, at what structural level crosslinking occurs. We have used X-ray scattering to see if, like other fixation methods, there are changes at the molecular level following this treatment, but found that molecular collagen structure is not affected by the treatment. This led us to hypothesize that the crosslinking occurs between the surfaces of the collagen fibrils and the surrounding matrix, and we are currently testing this hypothesis. We have also used absorption measurements to show that, contrary to some opinion, it is necessary to remove the epithelium to allow penetration of the riboflavin into the corneal stroma and that, although partial epithelial removal allows some penetration, this is uneven, and so is not a promising approach. We are currently investigating other novel techniques to carry out crosslinking without epithelial removal. It has been proposed by a number of clinicians that thinner keratoconus corneas, previously unsuitable for crosslinking treatment, may be operable in under hypo-osmolar conditions. However, we have shown using x-ray scattering that hypo-osmolar crosslinking results in an increase in the interfibrillar spacing and that this is not related to crosslinking but rather to tissue swelling. We are currently investigating if swelling, crosslinking, then dehydrating keratoconic corneas leads to any structural changes, in order to better understand the basis and safety of this new procedure.
Funding at Cardiff:
- Meek, Quantock, Boote, Knupp. £1,732,354. “The ultrastructural basis of corneal dysfunction and the development and optimization of novel therapeutic strategies”. MRC Programme Grant. 2012-2017
- Meek, Purslow £111,451. “The ostracod carapace window as a biomimetic basis for the development of a novel eye shield” BBSRC JSBI initiative grant. 2012-2013
- Boote, Meek £149,950 “The role of the sclera in human glaucoma” Fight for Sight. 2012-2014
- Quantock, Meek, Young Tucker £843,195. “A Physical Characterisation of Assembly Mechanisms and Light Transmission in Cornea”. EPSRC. 2008-2011
- Meek. £75,000. “Corneal transparency, refractive status and their loss in pathological conditions” Royal Society/Wolfson Research Merit Award. 2007-2012
- Meek, Quantock £1, 351,742 “The collagen matrix in corneal pathology, and the effect of new therapies for loss of transparency and refractive status” MRC Programme Grant. 2007-2012
- Meek et al. £216,888. BBSRC DTA Award. 2006-2013
- Regini, Meek £69,000. “A structural investigation of age related protein degradation and its prevention “. Research into Ageing. 2006-2009
- Quantock, Meek, Yagi, Wess et al. £30 000. “Structural hierarchies in fibrous biopolymers”. BBSRC 2005-2008
- Meek, Quantock, Hodson, Caterson. £1, 083,276 “Corneal transparency, dioptric power and their alterations in pathological conditions” MRC Programme Grant. 2001-2006
- Quantock, Meek, Hodson £59,634 “Optical Transmission in Fibrous Biomaterials”. EPSRC 2001-2002
- Erichsen, Boulton, Meek et al. £18,031. MRC equipment grant. 2001
- Meek. £73,018 Royal Society/Wolfson Laboratory Refurbishment grant. 2000
- Meek, Hodson, Elliott £71,407 The Wellcome Trust Overseas Investigator Award 1999-2001
- Attenburrow, Meek £86,281 “The influence of pre-straining on the stiffness and fibre structure of leather”. EPSRC 1999-2002
- Meek, Attenburrow £117,349 “The influence of pre-straining on the stiffness and fibre structure of leather”. EPSRC 1999-2002
- Meek £117,312 “Synchrotron x-ray diffraction study of the circum-corneal annulus of collagen fibrils in the human limbus and its integration with the cornea in normal and pathological tissue” The Wellcome Trust. 1999-2002
Dr Sally Hayes
Dr Philip Lewis
Dr Barbara Palka
Mr Nicholas Hawksworth (Swansea) Corneal pathology
Mr Arun Brahma (Manchester) Corneal pathology and crosslinking
Mr Stephen Tuft (London) Corneal pathology
Professor Phil Stephens (Cardiff) Tissue engineering and wound healing
Dr John West (Edinburgh) PAX6 studies
Dr Lucia Kuffova (Aberdeen) Tissue engineering
Dr Paul Hocking (Edinburgh) Use of chick models in ocular research
Dr Chris Inglehearn (Leeds) Use of chick models in ocular research
Dr Manir Ali (Leeds) Use of chick models in ocular research
Professor John Marshall (London) Corneal biomechanics
Mr David O’Brart (London) Corneal crosslinking
Professor Victor Tybulewicz (London) Mouse models for Downs syndrome
Professor Peter Winlove (Exeter) Ostracod properties
Dr Ahmed Elsheikh (Liverpool) Corneal and ostracod biomechanics
Dr Rachel Williams (Liverpool) Ostracod properties
Dr Andrew Parker (London) Ostracod properties
Professor Anthony Bron (Oxford) Ostracod properties
Professor May Griffith (Linkopings) Tissue engineering
Dr Rafael Grytz (Birmingham, Alabama) Corneal modelling
Professor Farhad Hafezi (Geneva) Corneal crosslinking
Professor Jesper Hjortdal (Aarhus) Corneal pathology
Dr Petra Baziliska (Prague) Corneal pathology
Dr Peter Pinsky (Stanford,) Corneal modelling
Professor Shigeru Kinoshita (Kyoto) Corneal pathology
Professor James Funderburgh (Pittsburgh) Corneal wound healing and tissue engineering
Professor Vicky Nguyen (Baltimore) Corneal modelling