SEARCH

SEARCH BY CITATION

Keywords:

  • cytokines;
  • eye rubbing;
  • keratoconus;
  • proteases;
  • tear film

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Background

Proteases, protease activity and inflammatory molecules in tears have been found to be relevant in the pathogenesis of keratoconus. We sought to determine the influence of eye rubbing on protease expression, protease activity and concentration of inflammatory molecules in tears.

Methods

Basal tears were collected from normal volunteers before and after 60 seconds of experimental eye rubbing. The total amount of matrix metalloproteinase (MMP)-13 and inflammatory molecules interleukin (IL)-6 and tumour necrosis factor (TNF)-α in the tear samples were measured using specific enzyme-linked immunosorbent assays (ELISA). Tear collagenase activity was investigated using a specific activity assay.

Results

The concentrations of MMP-13 (51.9 ± 34.3 versus 63 ± 36.8 pg/ml, p = 0.006), IL-6 (1.24 ± 0.98 versus 2.02 ± 1.52 pg/ml, p = 0.004) and TNF-α (1.16 ± 0.74 versus 1.44 ± 0.66 pg/ml, p = 0.003) were significantly increased in normal subjects after eye rubbing. The experimental eye rub did not alter significantly the collagenase activity (5.02 ± 3 versus 7.50 ± 3.90 fluorescent intensity units, p = 0.14) of tears.

Conclusion

Eye rubbing for 60 seconds increased the level of tear MMP-13, IL-6 and TNF-α in normal study subjects. This increase in protease, protease activity and inflammatory mediators in tears after eye rubbing may be exacerbated even further during persistent and forceful eye rubbing seen in people with keratoconus and this in turn may contribute to the progression of the disease.

Eye rubbing is a physiological response to uncomfortable eyes brought about by such things as fatigue, exposure to dust particles or allergens and also by stress, which can sometimes be relieved through the oculo-cardiac reflex.[1, 2] It is also common before or after sleep or during contact lens fittings.[3] Finger pads, knuckles or palms of the hand are mostly used for rubbing one or both eyes. The average duration of an episode of eye rubbing for most people is a few seconds.[4]

Keratoconus is a progressive, debilitating disease of the eye often linked to the habit of abnormal or forceful eye rubbing.[5, 6] A multivariate analysis for determining the risk factors for keratoconus indicated that eye rubbing was the only significant predictor for the disease.[7] Increased frequency of eye rubbing is also seen in other ocular conditions such as dry eye syndrome,[8] allergic,[9] bacterial or viral conjunctivitis,[10] misdirected eye lashes (trichiasis) and blepharitis.[11] Frequent eye rubbing is also common in dermatological conditions such as eczema and atopic dermatitis involving the eyelids.[12]

Ocular rubbing is considered abnormal, when there is a combination of amplified frequency, intensity and duration of rubbing episodes over an extended period of time. The technique used by many patients with keratoconus to rub their eyes is usually different from that used by people without keratoconus.[12] Vigorous ocular rubbing in keratoconus involves the use of finger tips or knuckles. The use of fingertips exerts pressure on the cornea. According to the observations made by Carlson,[13] the characteristic feature of eye rubbing in keratoconus is stroking the closed eyes with pressure using finger tips or middle knuckles in a circular motion restricted to the cornea (Figure 1A,B); however, these observations lack experimental evidence.

figure

Figure 1. Patients with keratoconus use their (A) finger tips or (B) knuckle to generate pressure localised to the cornea in a circular motion. Images reproduced with permission from Dr Alan Carlson.[13]

Download figure to PowerPoint

In contrast, a person with allergic or infective ocular disease tends to use the back of the hand, palm or finger pad to rub the eyes (Figure 2A–C).[13] The rubbing movement in these conditions is usually horizontal, involving the eyelids and caruncle, with minimal pressure spreading to the cornea (Figure 2A,C). This type of eye rubbing often lasts less than 15 seconds but the duration of rubbing of patients with keratoconus is much longer, usually extending from 10 to 180 seconds.[14] McMonnies and Boneham[4] have shown that people without allergy or keratoconus have a decreased frequency per day and duration (less than five seconds) of eye rubbing compared to atopic or keratoconic subjects. The method of eye rubbing might be used as an indicator for differentiating between normal or allergic subjects and those with keratoconus.

figure

Figure 2. Different methods adopted by allergic patients for eye rubbing using (A) the back of the hand, (B) the palm and (C) the finger pad involving the caruncle. Images reproduced with permission from Dr Alan Carlson.[13]

Download figure to PowerPoint

Several studies have investigated the role of proteolytic enzymes, such as matrix metalloproteinases (MMPs) in keratoconus. MMPs are involved in the degradation of extracellular matrix or activation of cellular apoptosis.[15] In the human cornea, MMPs are secreted by epithelial cells, stromal cells and neutrophils.[16] In keratoconus, the cornea is known to express elevated levels of MMP-1[17] and MMP-13.[18] The tear analysis in keratoconus has shown increased levels of MMP-1, MMP-3, MMP-7 and MMP-13.[19] Elevated gelatinolytic and collagenolytic activity has also been reported in the cornea[20-22] and tear film of patients with keratoconus.[19]

Keratoconus is defined as a non-inflammatory disease of the cornea; however, inflammatory molecules have been shown to be over-expressed in the corneas[23-26] and tear film of patients with keratoconus.[25] For example, increased concentrations of interleukin (IL)-4, -5, -6, -8 and tumour necrosis factor (TNF)-α, -β have been reported in the tears of patients with keratoconus.[19]

The impact of experimental eye rubbing on the biomechanical properties of the cornea,[27] thickness of the layers of the cornea,[28, 29] the corneal curvature[30] and tear levels of IL-8, epidermal growth factor (EGF) and hepatocyte growth factor (HGF)[31] have been investigated by a number of groups. There have been no reports in the literature examining the influence of eye rubbing on MMP-13, collagenase activity and inflammatory markers, such as IL-6 and TNF-α in tears. The present study examined the effects of experimental eye rubbing on the levels of MMP-13, IL- 6, TNF- α and collagenase activity in tears.

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Ethics approval was granted by the Human Research Ethics Advisory Panel at the University of New South Wales, Sydney, Australia. Written informed consent was obtained from each subject after briefly outlining the nature and possible consequences of the study. All the procedures were conducted in accordance with the 2000 Declaration of Helsinki.

Subjects

A group of 17 subjects was recruited to study the levels of MMP-13, IL-6 and TNF-α. The total protein levels and collagenase activity in the tears were examined in 14 subjects and 11 subjects respectively, from the same group. Subjects with keratoconus were not included in the study for ethical reasons. Contact lens wearers were also excluded. Other exclusion criteria included subjects with a history of dry eye disorders, ocular or systemic allergic conditions, ocular surgery and use of topical or systemic medications. The demographics of the subjects recruited are illustrated in Table 1.

Table 1. Demographics of the study
GroupsTotal proteinMediator levelsCollagenase activity
  1. Age is expressed as mean and standard deviation.  17 subjects were recruited in total but subsets of these were examined for total protein and collagenase activity in tears.

Total14/1717/1711/17
Men676
Women8105
Mean age32.36 ± 7.0832.58 ± 7.9533.18 ± 7.70

Assessment of the corneal curvature

The corneal curvature was mapped on all the subjects using a Medmont corneal topographer E300 (Medmont Pty Ltd, Melbourne, Victoria, Australia) on all the subjects. The topographic maps were subjected to the modified Rabinowitz-McDonnell test to exclude the presence of keratoconus.[32-35]

Eye rubbing technique

The required experimental rubbing technique was demonstrated to the subjects by one of the investigators. Subjects were then encouraged to follow the keratoconus eye rubbing technique for 60 seconds without interruption on both eyes, one eye at a time. The technique and duration of the eye rubbing was supervised and monitored by the investigator. Using three finger tips (index, middle and ring), all the subjects performed the keratoconus eye rubbing. The right hand was used to rub the right eye and vice versa. The amount of force applied for rubbing was limited so as not to cause any discomfort to the eyes. Subjects were advised to exert equal pressure on both eyes. Further care was taken by the subjects to adapt to the circular eye rubbing movement, applying constant pressure toward the centre of the cornea. Instructions were also given to the subjects to maintain a steady central fixed gaze position with the contra-lateral eye opened during eye rubbing to control fixation and assist in achieving central corneal rubbing.[29]

Tear collection

Basal tears were collected, using glass capillary tubes before and immediately after 60 seconds of rubbing the right eye followed by the left eye. The method of collection of basal tears was based on a previous report.[19] A minimum of 10 μl of basal tears was collected before and after eye rubbing. Before rubbing (BR) and after rubbing (AR) tears from both eyes were pooled separately for each subject to increase the sample volume for the analysis.

Total tear protein concentration

Bicinchoninic acid (BCA) protein assay[36] was used to determine the total protein concentration of tear samples before and after rubbing. Bovine serum albumin (BSA) standard curve was obtained by preparing serial dilutions of BSA. The BCA solution (Pierce BCA kit, Thermo Scientific, Melbourne, Victoria, Australia) was added and the absorbance was measured at 562 nm using a spectrophotometer (Tecan Spectrofluoro Plus; Tecan Group Ltd, Männedorf, Switzerland). The levels of protein in the tears before and after rubbing were estimated against the standard BSA curve.

Quantification of MMP-13, IL-6 and TNF-α

The levels of MMP-13 in BR and AR tear samples of individual subjects were determined by using the commercially available enzyme-linked immunosorbent assay (ELISA) kit (Ray Biotech, Inc., Norcross, GA, USA). Similarly, the tear samples were assayed for IL-6 and TNF-α with commercially available ELISA kits (Quantikine® HS, R&D Systems, Inc, Minneapolis, MN, USA). The minimum detection limits of the assays were 6.0 pg/ml for MMP-13, 0.016 pg/ml for IL-6 and 0.038 pg/ml for TNF-α. Tear samples were diluted 1:50 with the assay diluents and the analyses were performed according to the manufacturers' protocols. The mediators, MMP-13, IL-6 and TNF-α and collagenase activity were chosen for study as previous work had shown that these were significantly increased in the tears of patients with keratoconus compared to normals.[19]

Protease activities

The collagenase activities in tear samples of individual subjects before and after rubbing were examined using the Enzchek® Gelatinase/Collagenase Assay Kit (Invitrogen Australia Pty, Melbourne, Victoria, Australia). The protocol for the assay was modified for use with tear samples.[19]

Statistical methods

A sample size with a minimum of 10 patients in each group was required to determine a statistically significant difference with 80 per cent power at 95 per cent confidence, based on previous reports.[19, 25] Paired Student's t-tests were performed to determine significant differences in the levels of total tear protein, MMP-13, IL-6, TNF-α, and collagenase activities in BR and AR tear samples. SPSS predictive analytics software, version 18 (SPSS Inc., Chicago, IL, USA) was used for all the analyses and a p-value less than 0.05 was considered significant.

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Corneal curvature

The central corneal power (K-value) and infero-superior asymmetry (I-S) of the subjects were 42.7 ± 1.2 D and 0.20 ± 0.30 D, respectively.

Total tear protein concentration

The total protein levels in tear samples before and after rubbing are shown in Table 2. The pre-rubbing tear samples had a total protein concentration not significantly (p = 0.68) different from after rubbing samples.

Table 2. Levels of total protein, inflammatory mediators and proteolytic activity in tears
GroupsBefore rubbingAfter rubbingp-value
  1. Results are expressed as mean ± standard deviation. FIU: fluorescent intensity units, IL: interleukin, MMP: matrix metalloproteinase, TNF: tumour necrosis factor

Total protein (mg/ml)7.74 ± 1.528.03 ± 2.150.68
MMP-13 (pg/ml)51.86 ± 34.2963 ± 36.800.006
Collagenase activity (FIU/mg total protein)5.02 ± 3.007.50 ± 3.900.14
IL-6 (pg/ml)1.25 ± 0.982.02 ± 1.520.004
TNF-α (pg/ml)1.16 ± 0.741.45 ± 0.660.003

Tear levels of MMP-13, IL-6 and TNF-α

The concentrations of MMP-13, IL-6 and TNF-α in BR and AR tear samples are shown in Table 2. The concentration of MMP-13, IL-6 and TNF-α increased significantly (p < 0.05) in the tears after rubbing (Table 2).

Tear proteolytic activities

The activities of collagenases in the tear samples of individual subjects before and after rubbing were examined and the results are shown in Table 2. Collagenolytic activities in tear samples before rubbing were not significantly (p > 0.05) different compared with the after rubbing samples (Table 2).

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The expressions of MMP-13,[18] protease activity,[20-22] IL-6 and TNF-α[25, 37] have been examined previously to establish their involvement in keratoconus. The tear film in keratoconus showed increased levels of MMP-13, IL-6 and TNF-α compared to normal controls.[19] To establish a link between keratoconus and eye rubbing, this study was designed to investigate the influence of eye rubbing on the tear levels of MMP-13, IL-6, TNF-α and protease activity in normal subjects. The present study appears to be the first experimental evidence to demonstrate significantly elevated levels of MMP-13, IL-6 and TNF-α in tear film after eye rubbing, thereby providing evidence for the association between keratoconus and eye rubbing.

Various case reports have examined the involvement of eye rubbing in keratoconus.[38-41] Other case studies have linked eye rubbing in keratoconus to nasolacrimal duct obstruction,[42] punctual agenesis[43] and Tourette's syndrome.[44] These associations between eye rubbing and keratoconus have meant that clinicians often advise people of the risk of progression of keratoconus, if they indulge in eye rubbing. Although many patients with keratoconus do not report or acknowledge abnormal eye rubbing, it is not uncommon for parents or family members to contradict a patient's statement about rubbing history.[4]

The ocular environment changes dynamically during eye rubbing. There is increased friction between the palpebral conjunctiva of the upper eyelid and the ocular surface during closed eye rubbing. Animal model studies[3] of eye rubbing have illustrated significant infiltration of inflammatory cells, degranulation of mast cells and surface irregularity of the upper tarsal conjunctiva.

The increased levels of tear MMP-13, IL-6 and TNF-α after 60 seconds of closed eye, experimental eye rubbing on normal subjects seen in the present study could be due to the mechanical effects of rubbing the upper palpebral conjunctiva against the corneal surface and bulbar conjunctiva. These proteases and inflammatory molecules have the potential to induce apoptosis of the keratocytes,[15, 45] which is the major form of cell death in keratoconic corneas.[46, 47] MMP-13 or collagenase 3 is expressed by human corneal epithelial cells and is essential for the regulation of corneal wound healing and remodelling.[48] Increased levels of MMP-13 have also been reported in vernal keratoconjunctivitis (VKC)[49] and systemic immune disorders, such as rheumatoid arthritis[50] and osteoarthritis.[51] Cytokines such as IL-6 and TNF-α regulate immune responses and inflammation. Corneal[52, 53] and conjunctival epithelial cells[54] express IL-6 and TNF-α and the levels of these inflammatory mediators are increased in response to corneal wound healing.[55, 56] Increased levels of IL-6 and TNF-α have also been reported in dry eye syndromes,[57] allergic keratoconjunctivitis[58, 59] and vernal keratoconjunctivitis.[60]

Levels of MMP-13, IL-6 and TNF-α are also elevated in tears of atopic keratoconjunctivitis (AKC) and VKC.[49, 58-60] Immune-mediated mechanisms triggered by environmental allergens are actively involved in AKC- and VKC-related ocular rubbing.[61] Mast cells of the conjunctiva play a central role in causing the signs and symptoms of ocular itching.[62] The allergens bind to immunoglobin E (IgE) molecules on the surface of mast cells, stimulating the release of histamine, increasing the possibility of an itch stimulus for eye rubbing.[9] Along with the release of histamine, the mast cells also liberate cytokines leading to inflammatory changes in the cornea. Increased levels of serum IgE have also been reported in keratoconus;[63, 64] however, the type of itch-rub cycle seen in AKC or VKC[61] might be absent in keratoconus unrelated to atopy and associated itch provocations.

The temptation to rub the eyes in AKC and VKC is probably predominantly due to ocular itching. In addition to ocular itching, the motivating factors for eye rubbing in keratoconus could be improvement in vision or relief from burning and gritty sensations.[13] Nevertheless, AKC and VKC aggravate the progression of keratoconus,[65, 66] where a combination of both atopic-type and keratoconus-type rubbing might have a significant role in these conditions. The reports on keratoconus-type eye rubbing have only anecdotal status[13] and future experimental studies are necessary to clearly distinguish the eye rubbing in atopy and keratoconus.

Keratocytes produce IL-6 when exposed to TNF-α,[67] and MMP-13 levels are increased in human epithelial cell cultures when treated with TNF-α.[48] The active interplay between proteases and cytokines[19] might be aggravated during the persistent eye rubbing seen in keratoconus. McMonnies[68] hypothesised that rubbing can cause spikes in corneal temperature that might increase collagenase activity. The activity of tear collagenases was not statistically different in this study and future investigations with ample sample sizes are essential to substantiate the rubbing-related increase in protease activity.

In summary, the levels of MMP-13, IL-6 and TNF-α were over-expressed in normal eyes after experimental eye rubbing. In ectatic corneal conditions such as keratoconus, persistent eye rubbing might cause a greater increase in the levels and activity of these mediators, with the potential to contribute to the development or progression of the disease. Up-regulated proteases and inflammatory molecules may be causal links between eye rubbing and keratoconus. Elucidating the factors responsible for the habit of abnormal chronic central eye rubbing in some patients with keratoconus could help in the management of rubbing habits and in understanding the aetiology of the disease.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

This study was supported by a scholarship from the Brien Holden Vision Institute. The authors would like to thank Dr Thomas Naduvilath for his guidance in statistical analysis. The authors also thank Dr Judith Flanagan for her assistance in preparation of this manuscript.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References