Elsevier

Journal of Infection

Volume 49, Issue 4, November 2004, Pages 310-316
Journal of Infection

Acanthamoeba pathogenicity for corneal cells

https://doi.org/10.1016/j.jinf.2004.03.005Get rights and content

Abstract

Objectives. Comparison of the relative susceptibility of human keratocyte and corneal epithelial cells to Acanthamoeba castellanii.

Methods. Primary cultures of the mammalian cells were utilised at equivalent concentrations throughout the co-incubations. Preliminary experiments involved visual estimation of the effects of 10-fold dilutions of Acanthamoeba trophozoites after various time intervals. Formal quantative assessment was performed by image analysis of the respective effects of the Acanthamoeba at a concentration of 106/ml/well following 3, 6, 9 and 24 h of co-incubation.

Results. Epithelial cells were relatively resistant to the cytopathic effects of Acanthamoeba throughout the experiments course at all concentrations tested: the cut-off point, below which no effect was observed, was also greater. Formal assessment by image analysis confirmed these impressions while revealing that the relative resistance of epithelial cells was confined to the initial phases of co-incubation. For both mammalian cell types the first observable sign of cell damage was cell shrinkage with the formation of retraction fibres and gaps.

Conclusions. The relative resistance of epithelial cells to Acanthamoeba may be due to difficulty in initiating attack of a monolayer with tight junctions between cells. These results provide circumstantial evidence of the importance of the corneal epithelium in the prevention of amoebic invasion.

Summary. The relative susceptibility of cultured corneal epithelial cells and keratocytes were compared as regards both the concentration of amoebae required to produce obvious cell damage and the area of cells destroyed for a given concentration at various time intervals. The epithelial cells were more resistant than keratocytes with respect to both factors, at least early in the co-incubation period. This may be related to the different growth patterns of the cell types as epithelial cells form a monolayer connected by tight junctions and indirectly supports the idea that the epithelium forms an important barrier function against invasion in vivo.

Introduction

Acanthamoeba keratitis is painful and potentially sight threatening. The condition is also noted for its wide variety of clinical presentations.1., 2., 3., 4., 5. Indeed a recent confocal scanning study, confirmed by histology on epithelial scrapes, has suggested that amoebic keratitis may occur more commonly than currently recognised.6 This would imply that certain cases either resolve on non-specific therapy or are self-resolving. One possible explanation is that in some instances, perhaps due to less pathogenic organisms or in less favourable conditions, the organism fails to penetrate through the epithelium into the corneal stroma.

The association between Acanthamoeba keratitis and contact lens wear is well recognised. Microtrauma to the corneal epithelium is assumed to be an important factor in facilitating invasion of this small free-living amoeba. Indeed it has been demonstrated in vitro that injury increases exposure of mannose-glycoproteins which are known to be involved in binding of Acanthamoeba.7 It is also of note that no animal model has been developed which adequately mimics this process. Those that are available either involve direct inoculation of the organism into the stroma,8., 9., 10., 11., 12., 13. or prior abrasion of the epithelium.14., 15. Taken in conjunction, these factors suggest that the intact epithelium plays a very important role in preventing amoebic invasion.

It has been demonstrated that, for invasion to occur, the trophozoites must first bind to the corneal epithelial cells16 and then gaps appear in the cell junctions allowing the organism to penetrate, burrowing between cells.17 Initially the amoebae are presumably confined to this tissue but progressive stromal invasion occurs as the infection advances.18 Late disease is also associated with keratocyte depletion.18 It is therefore assumed that the amoebae are consuming these cells. Indeed it has been demonstrated that they are capable of utilising either keratocyte or corneal epithelial cells as a food source.19 The pathogenesis of this condition is, however, only incompletely understood.

In this paper we compare the relative susceptibility of keratocyte and corneal epithelial cells to the cytopathic effect produced by Acanthamoeba castellanii. Selected results are formally assessed by image analysis. The findings are discussed with regard to our understanding of the role of the epithelium with respect to the pathogenesis of the condition in vivo.

Section snippets

Cells

Human corneal epithelial cells and keratocytes were obtained by primary culture from corneal buttons removed at grafting from patients with keratoconus and from corneal buttons recovered from eyes enucleated for treatment of choroidal melanoma. Over a dozen different attempts were required to amass sufficient numbers of cells. This was mainly due to difficulties encountered with primary isolation of epithelial cells: they had more demanding growth requirements than keratocytes and were readily

Results

Results of the preliminary experiments measuring the effect of different concentrations of Acanthamoeba castellanii on epithelial cells and keratocytes are summarised in Table 1. A ‘cut-off point’ could be demonstrated for each of the mammalian cell types: for epithelial cells this was 105/ml/well while for keratocytes it was 104/ml/well, although the damage produced at these concentrations was slight. Concentrations of amoebae below these respective levels had no demonstrable effect. The

Discussion

Co-incubation of Acanthamoeba castellanii at a concentration of 106/ml or 105/ml with keratocytes (50 000/ml/well) led to progressive destruction of the majority of the ‘monolayer’ within 24 h. In both cases, however, destruction of the remaining cells occurred at a comparatively lesser rate. This may at least be partly due to the fact that the cells that survive longer tend to be relatively protected (i.e. physically more difficult to attack because of clumping). Interestingly, the reaction

Acknowledgements

This work was supported by grants from the Tennent Institute of Ophthalmology, the Ross Foundation, the Wallace Brown Memorial Trust and the Pocklington Trust. The author would like to thank J. Hay and G. Reid for instruction in amoebal and cell culture techniques respectively. The author is also grateful to W. Sehu for assistance with image analysis and illustrations and W.R. Lee for advice.

References (26)

  • M.A. Cote et al.

    Evaluation of the rabbit as a model of Acanthamoeba keratitis

    Rev Infect Dis

    (1991)
  • R.L. Font et al.

    An animal model of Acanthamoeba keratitis

    Invest Ophthalmol Vis Sci

    (1981)
  • R.L. Font et al.

    An animal model of Acanthamoeba keratitis: further studies with emphasis on the early phase destruction of the trophozoites

    Invest Ophthalmol Vis Sci

    (1982)
  • Cited by (8)

    • Serine-like proteolytic enzymes correlated with differential pathogenicity in patients with acute Acanthamoeba keratitis

      2011, Clinical Microbiology and Infection
      Citation Excerpt :

      Eventually, the pathophysiology of AK can result in corneal transplantation, blindness and eye enucleation [4,17]. The role of the corneal epithelium as an essential physical barrier in the prevention of amoebic infection has been described previously [18]. Most of the cytopathic factors involved in these infectious processes have been described as resulting from the activity of proteolytic enzymes secreted by trophozoites in the surface of host corneal tissue [4,5,7,19,20].

    • Laboratory diagnosis of ocular parasites

      2008, Revue Francophone des Laboratoires
    View all citing articles on Scopus
    View full text