Article Text

The healing optic nerve in glaucoma: transforming growth factor β and optic nerve head remodelling
  2. P T KHAW
  1. Glaucoma and Wound Healing Research Units, Department of Pathology, Institute of Ophthalmology and Moorfields Eye Hospital, 11–43 Bath Street, London EC1V 9EL

Statistics from

Despite the many advances in the therapy and diagnosis of primary open angle glaucoma, and the recognition of intraocular pressure as the major modifiable risk factor,1 2 the pathogenesis of the disease still remains unclear. The most pathognomonic feature of glaucoma is optic disc cupping and the primary site of glaucomatous optic nerve damage appears to be at the optic nerve head.3 However, the exact mechanisms by which this damage occurs have not been elucidated. Changes in lamina cribrosa morphology and nerve fibre bundle pore morphometry4 5have been documented in this disease, in association with alterations in the surrounding extracellular matrix.6-11 The tortuous course of the individual nerve fibres may also play a role.12 In this issue of theBJO (p 209), Pena et al have found that the production of a growth factor, transforming growth factor β (TGF-β2) is considerably increased in the optic nerve heads of patients with open angle glaucoma but not in normal individuals. This finding is important because of its implications for the pathogenesis of glaucomatous damage. Like all interesting research findings, it raises further intriguing questions. Is TGF-β2 stimulated in response to nerve axon loss or changes in extracellular environment such as intraocular pressure induced compression, stretching, and shearing forces? Is TGF-β2 production neuroprotective or do its effects on extracellular matrix compromise nerve fibre function and integrity?

The main structural component of the optic nerve head is the lamina cribrosa through which pass all the optic nerve fibres exiting the eye. It is continuous with the sclera, and consists of stacks of fenestrated connective tissue plates (cribriform plates), each pore allowing passage of nerve fibre axons.13 It has long been accepted that the lamina cribrosa is the weakest part of the sclera. If the intraocular pressure (IOP) is raised for a prolonged period, as in cases of chronic glaucoma, the lamina cribrosa bows outwards producing a “cupped” optic disc.14 This bowing can also be induced in the short term at higher levels of IOP. These structural change are very similar even in patients with so called normal tension glaucoma, suggesting that physical forces at the nerve head do play a role in cupping, whatever the level of IOP. Histological analysis has demonstrated regional differences in lamina cribrosa structure with the superior and inferior areas being weakest and hence most susceptible to damage from raised IOP, with collapse and prominent posterior bowing in advanced glaucomatous disease.15 Differences in primary structure and cellular responses to “stress” including production of growth factors may explain in part the individual variation in the pattern of optic nerve damage seen with similar levels of IOP.

TGF-β is a multifunctional growth factor found throughout the body, and implicated in the processes of scarring.16-19 In the eye, TGF-β2 appears to be the predominant isoform.20-23Pena et al suggest the TGF-β2 in glaucomatous eyes is produced by astrocytes in the lamina cribrosa, which as a consequence take up a “reactive” phenotype, seen characteristically in various neurological disorders—Alzheimer’s disease,24 multiple sclerosis,25 and after neuronal injury. This phenotype is implicated in the development of glial scar formation,26 and in its remodelling. Although no gliosis is seen in glaucomatous optic neuropathy, they postulate that TGF-β may activate astrocytes to stimulate extracellular matrix remodelling of the lamina cribrosa. TGF-β is a prominent component of the healing response to damage in many parts of the body, and the optic nerve head may be no exception.

The cellular responses at the optic nerve head to changes in IOP are not yet known. However, the effects of different types of forces on a variety of cell types has been studied, most extensively in endothelium, where shear stress has been shown to produce vasodilatation, a process mediated by nitrous oxide and probably a change in cell-cell interactions via expression of integrins.27 Centrifugal tension has been shown to increase expression of growth factor receptors in fibroblasts, via stimulation of β1 integrin expression,28 and dermal fibroblasts alter production of matrix metalloproteases (MMP) in response to different tensile loads.29 It is therefore conceivable that changes in the IOP produce cellular responses in the lamina cribrosa, altering gene expression and the synthesis or degradation of extracellular components, and ultimately the support structure of the nerve fibres.

The idea that the optic nerve head may respond to dynamic physical force changes in its environment via alterations in TGF-β activity is very important. Historically, TGF-β was believed to stimulate scarring by inhibiting MMP production and stimulating tissue inhibitors of MMPs (TIMPs).30 Work in our laboratory suggests, however, that the cellular effects of TGF-β depend on interactions with the surrounding extracellular matrix31—TGF-β inducing different MMP and TIMP profiles in different extracellular environments. Cell-matrix interactions are mediated via integrin receptors,32 and TGF-β is known to directly affect integrin expression, which in turn determines MMP production.33 Hence, TGF-β is a potent modulator of extracellular remodelling.

Pena et al also comment on TGF-β having a neuroprotective effect in glaucoma. A number of possible mechanisms of neuroprotection have been reviewed by Flanders et al in the context of neurodegenerative diseases,34including antioxidant properties of TGF-β, its enhancement of mitochondrial potential, its maintenance of neuronal calcium homeostasis, and finally its inhibitory effects on apoptosis. Again, although these areas are currently being explored, the significance of these processes in glaucomatous optic neuropathy is not yet established.

As we approach the next millennium, this is an exciting period in glaucoma research. Technological advances in imaging have made it possible to visualise the lamina cribrosa pores in vivo.35 36 If we can now correlate in vivo features with changes at a cellular and molecular level, the processes occurring within the lamina cribrosa may provide us with an alternative tool not only to diagnose and monitor glaucoma but also to treat this fascinating but complex group of diseases.


MFC and PTK are supported by the Wellcome Trust (048474), Guide Dogs for the Blind Association and the Medical Research Council (G9330070).


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles