Fibronectin in the tear film.

PURPOSE: Fibronectin plays an important role in corneal healing and has been detected previously in the tear film. To investigate the levels of fibronectin in normal human tears, the authors measured and compared fibronectin concentration in open-eye, closed-eye, and reflex tear fluid. The origin of fibronectin in the tear film was investigated by comparing fibronectin concentration in sequentially collected reflex tear samples with the concentrations of total protein and albumin in the same samples. METHODS: Open-eye and closed-eye tears were collected from 11 noncontact lens wearers. From 7 subjects, 20 uninterrupted reflex tear samples (10 microliters each) subsequently were collected, using the sneeze reflex method of stimulation, followed by an additional 10 nonstimulated tear samples (3 microliters each) immediately after cessation of stimulus. Enzyme-linked immunosorbent assays were used to determine fibronectin and albumin concentrations, and bicinchoninic acid protein assays were used to determine total protein concentration in each sample. RESULTS: Fibronectin concentration in open-eye tears (19 +/- 24 eta g/ml, range 3 to 78 eta g/ml) was significantly different (P = 0.004) from that in closed-eyes tears (4127 +/- 3222 eta g/ml, range 1177 to 11384 eta g/ml). In the first 50 microliters of reflex tears, fibronectin concentrations were low (10 +/- 23 eta g/ml), but they increased significantly (P = 0.028) after 100 microliters of reflex tears had been collected (220 +/- 126 eta g/ml). There was a further marked transient increase (767 +/- 946 eta g/ml) after cessation of stimulus. Total protein concentration in the same samples decreased significantly during reflex tear collection compared to open-eye tears, and it increased gradually after cessation of stimulus. Albumin concentration in the same samples, analyzed for two subjects only, showed a pattern similar to that for fibronectin. Dilation of conjunctival blood vessels was noted in all subjects after reflex tear collection. Administration of a topical vasoconstrictor in two subjects eliminated the increase in fibronectin concentration during reflex tearing but did not affect total protein concentration. Under reducing conditions, the molecular mass of fibronectin in open-eye and reflex tears was 240 kDa, identical to commercially available purified plasma fibronectin, whereas fibronectin in closed-eye tears was degraded into small molecular mass fragments. CONCLUSIONS: These findings suggest that fibronectin in tear fluid is derived from plasma and that the increase in concentration in closed-eye and reflex tear fluid is caused by leakage from dilated conjunctival blood vessels.

Dry eye among males infected with the human immunodeficiency virus.

BACKGROUND: Keratoconjunctivitis sicca (KCS) has been reported as a potential ocular manifestation of human immunodeficiency virus (HIV) infection. This study was designed to estimate prevalence of dry eye among HIV-infected males and correlate subjective and objective findings with disease severity. METHODS: All subjects and controls completed the McMonnies Dry Eye Questionnaire (MDEQ) and underwent diagnostic testing for dry eye including biomicroscopic assessment of the anterior segment, lacrimation kinetics, sodium fluorescein break up time, and vital staining with sodium fluorescein and rose bengal. RESULTS: A 38.8 percent prevalence of dry eye was found in the study group. Certain risk factors may be more likely to produce KCS than others and may be a better indication of potential dry eye than disease severity. CONCLUSIONS: Patient symptoms are not adequate predictors of dry eye, indicating the need for KCS diagnostic testing as part of a comprehensive ocular examination for HIV-infected individuals.

Tear protein profiles vs. clinical characteristics of untreated and cyclosporine-treated canine KCS.

BACKGROUND: Topically administered cyclosporine A (CsA) decreases ocular surface inflammation in canine keratoconjunctivitis sicca (KCS) and exerts lacrimomimetic effects. This study was performed to find correlations between clinical signs and tear protein levels in untreated and CsA-treated canine KCS. METHODS: Clinical profiles were scored in 16 KCS-affected dogs before and 6 weeks after commencing treatment with 0.2% topical CsA emulsion. Tear samples were also collected using polished micropipettes for specific protein assay by high-performance liquid chromatography (HPLC) and enzyme-linked immunosorbent assay (ELISA). RESULTS: Tear levels of serum proteins correlated with conjunctival clinical signs. Levels of lacrimal gland proteins in tears correlated most often with corneal clinical signs. CONCLUSIONS: The inflammatory features of KCS appear to link conjunctival signs to serum proteins in tears, while corneal signs are linked to lacrimal gland proteins.

Enhancements to the conjunctival impression cytology technique and examples of applications in a clinico-biochemical study of dry eye.

The conjunctival impression cytology (CIC) technique was optimized for cell pick-up and resolution of cytologic detail. Use of surfactant-free smaller pore-size Millipore filters yielded consistently high cell pick-up and resolution of cell detail. Staining by the modified Papanicolaou technique in a free-floating format allowed quantitative evaluation of cytologic features. Air exposure of CIC specimens for up to 8 minutes before fixing (as occurs with Nelson's standardized technique) did not adversely affect cell appearances. The optimized method was applied to a group of 50 patients with dry eyes. CIC features were correlated with symptoms, clinical test results, and serum and tear protein levels of dry eye patients.

Clinical profiles of non dry eye patients and correlations with tear protein levels.

An accurate clinical description of the normal patient must be established before the dry eye patient can be accurately profiled. The current study was designed to (a) determine normal values for, and to seek correlations between, clinical tests used for the evaluation of non dry eye patients' tear film and ocular surface, and (b) compare clinical findings with previously reported levels of non-stimulated and stimulated tear proteins for the same group. Thirty non-contact lens wearing patients (age range 20-64 years) were determined to be free of dry eye based on the results of a series of clinical tests. McMonnie's dry eye questionnaire was used as an initial screening step. Clinical tests included lacrimation kinetics (using sealed, calibrated filter paper strips), non-invasive tear break-up time (NIBUT) measurement by Xeroscope, biomicroscopic evaluation of the anterior segment, assessment of tear prism height and regularity, and fluorescein and Rose Bengal staining. No significant correlations were found between age, questionnaire score, lacrimation kinetics final tear secretion rate or NIBUT. However, NIBUT results did vary by gender. Mean NIBUT for females was 46.3 +/- 16.6 (standard deviation) seconds and for males 59.3 +/- 2.1 seconds. some significant correlations were found between clinical test results and levels of non-stimulated and stimulated tear proteins. Overall, the results indicate that this battery of clinical tests will be appropriate for distinguishing between normal and dry eye patients. Combined with tear protein assay, this clinical approach may improve our current understanding of the different types of dry eye.

Purification of the isoforms of tear specific prealbumin.

A chromatographic method for separating tear specific prealbumin (TSP) into six isoelectric forms is described. Size exclusion high performance liquid chromatography (SE-HPLC) was used to isolate TSP from whole tears, followed by chromatofocusing fast protein liquid chromatography (FPLC) of the SE-TSP fraction on a Mono P column. This yielded six fractions varying in isoelectric point (pI) between 5.3 and 4.6. Subsequent anion exchange FPLC (Mono Q column) allowed a slight further purification of each Mono P fraction and removed Polybuffer from the Mono P fractions. Isoelectric focusing (IEF) of the TSP isoforms verified that the heterogeneity was based on pI, and confirmed that the chromatofocusing separation was in many respects the same as an IEF separation. Purity of TSP isoforms was determined by SDS-polyacrylamide gel electrophoresis (SDS-PAGE), IEF, enzyme-linked immunosorbent assay (ELISA) and immunoblotting of samples separated by SDS-PAGE and IEF. Amino acid analysis and N-terminal amino acid sequencing revealed subtle differences between the TSP isoforms. The entire purification procedure was conducted both with and without the addition of reducing agents and protease inhibitors to tear samples and all buffers used in the purification process. Relatively little difference was seen in the TSP isoform profile under these two sets of conditions. However, the tendency of isolated TSP to aggregate and precipitate was dramatically decreased under reducing conditions, resulting in significantly higher protein recoveries. This chromatographic purification procedure provides a basis for further study of the nature of the heterogeneity of TSP and characterization of the properties of this protein.

Changes in human tear protein levels with progressively increasing stimulus.

The levels of 13 proteins were measured in six tear samples collected atraumatically at progressively increasing flow rate from nonstimulated (less than 0.5 microliter/min) to highly stimulated (greater than 50 microliters/min) in ten subjects. Tears were fractionated initially by size-exclusion high-performance liquid chromatography (SE-HPLC). Enzyme-linked immunosorbent assays and kinetic assays were then applied to relevant SE-HPLC fractions to determine specific protein levels. Nine of the 13 proteins assayed showed significantly higher concentrations in nonstimulated tears than in any other tear sample. Immunoglobulin (Ig) M, secretory IgA, polymeric IgA1, and polymeric IgA2 all decreased progressively in concentration from nonstimulated tears to the higher flow-rate stimulated samples. The level of IgG, albumin, and transferrin showed a large drop in concentration between nonstimulated tears and the first (lowest flow-rate) stimulated sample, with relatively little decrease for any subsequent sample. Levels of lactoferrin, tear-specific prealbumin, lysozyme, and peroxidase were relatively constant throughout the series of tear samples. These results indicate that the mechanisms responsible for changes in concentration of constitutive, serum-derived, and regulated tear proteins with stimulus can be studied successfully using noninvasive methods to collect human tears. They also show that simply distinguishing between nonstimulated and stimulated tears is not sufficient to completely characterize the effect of stimulus conditions on tear protein composition.

Protein levels in nonstimulated and stimulated tears of normal human subjects.

Atraumatically collected nonstimulated (less than 1 microliter/min) and stimulated (greater than 50 microliters/min) tears from 30 clinically normal subjects were fractionated by size exclusion high-performance liquid chromatography (SE-HPLC). Enzyme-linked immunosorbent assay (ELISA) and kinetic assays were applied to relevant HPLC fractions to quantitatively identify 12 tear proteins. Secretory IgA levels were much higher in nonstimulated than in stimulated tears, and a similar disparity was seen also with IgA1 and IgA2 in the HPLC fraction containing secretory IgA. IgM levels were also higher in nonstimulated tears. Levels of the primary lacrimal gland proteins, lactoferrin, tear specific prealbumin, and lysozyme were similar in both types of tears. Significantly higher concentrations of the major serum proteins, IgG, transferrin, and serum albumin were measured in nonstimulated tears. Overall, 8 of the 12 proteins assayed were present at significantly higher concentrations in nonstimulated tears. These results show that tear flow rate strongly influences the protein profile obtained. Therefore, to allow valid comparisons of tear protein profiles within and between studies that use atraumatic collection procedures, an indication of flow rate during collection should be reported.

Cell sloughing with proparacaine.

The corneal epithelium sloughs cells directly into the pre-corneal tears. By irrigating tears from the surface of the cornea under different circumstances it is possible to compare the rates at which cells are sloughed. In this experiment the rate of appearance of cells following one drop of topical anesthetic was compared with the same eye without the drop. Initially the anesthetized cornea sloughed less, but later more cells were sloughed. This effect lasted for at least 6 hours, and showed that the return of normal sensation is only the initial part of a recovery process which takes much longer.

Identification of proteins in small tear volumes with and without size exclusion HPLC fractionation.

A protocol for quantitative assay of several proteins in small tear volumes is described. Assays included sandwich ELISA's for IgM, IgA, secretory IgA, IgG, lactoferrin and "serum" albumin, and kinetic assays for lysozyme and peroxidase. All assays were applied to whole stimulated tears and size exclusion (SE) HPLC-separated fractions of stimulated tears. Comparison of results revealed several sources of error in whole tear assays. Tear IgG and albumin levels were considerably lower than reported by others. IgM was routinely detected at very low levels. Secretory IgA levels greatly exceeded "serum" IgA. SE-HPLC-isolated tear specific prealbumin was separated into five sub-fractions by anion exchange HPLC. Due to its ability to identify proteins in very small tear volumes, the protocol will be most useful for assay of non-stimulated tears.

Investigation of sloughed corneal epithelial cells collected by non-invasive irrigation of the corneal surface.

This paper reports the development of a non-contact corneal irrigation chamber (NC-CIC) which enables non-invasive collection of epithelial cells from the corneal surface of human subjects. Cells were viewed by fluorescence microscopy following vital staining with acridine orange (AO). Staining characteristics revealed two corneal epithelial cell types: cells with (i) green and (ii) orange-red cytoplasmic staining. The green cytoplasmic stain appeared to indicate a more viable cell. Multi-cell aggregates were regularly collected from the corneal epithelial surface. Groups of up to seven epithelial cells were obtained. Quantitative studies of corneal epithelial cell sloughing, using isotonic NaCl (305 mOsm/kg) and isotonic "basic tear solution" (BTS, 305 mOsm/kg) as irrigating solutions, involving hourly irrigations between 8 a.m. and 4 p.m. were conducted. Consistently higher cell counts were obtained with NaCl. Using BTS, data scatter was reduced sufficiently to reveal significant differences in sloughing rate as a function of time of day. Instillation of one drop of 0.5% proparacaine caused a significant, but gradual, increase in epithelial cell sloughing rate over a period of hours, as indicated by subsequent BTS irrigations of the cornea. Since the NC-CIC technique is able to discriminate these effects, it may be an appropriate system for in vivo studies of the relationship between corneal epithelial cell mitosis and sloughing.

Use of tear enzyme activities to assess the corneal response to contact lens wear.

A non-invasive biochemical method for assessing the effects of contact lens wear on the in vivo corneal epithelium is described. A fluorometric technique is used to measure the activities of lactate dehydrogenase (LDH) and malate dehydrogenase (MDH) in human tear fluid. In view of the corneal epithelial origin of these enzymes, changes in their activity can indicate the severity of environmental stresses on the corneal epithelium. Following short-term wear of contact lenses, LDH and MDH activities are altered so that the tear LDH/MDH ratio is elevated. The magnitude and time course of this elevation are influenced by contact lens type, fit and duration of wear. The technique can yield more specific measures of the corneal response to contact lens wear than previous techniques.

Human tear enzyme changes as indicators of the corneal response to anterior hypoxia.

A non-invasive biochemical technique for quantifying the effects of anterior corneal hypoxia on the in vivo corneal epithelium of the human eye is described. Following short-term exposure of the cornea to low atmospheric oxygen pressures, lactate dehydrogenase (LDH) and malate dehydrogenase (MDH) activities in tears are altered so that the tear LDH/MDH ratio is elevated. The degree of elevation of the ratio and its timing are related to the severity of hypoxia. Possible explanations for the elevation of the tear LDH/MDH ratio include unbinding of intracellular LDH and increased cell membrane permeability. For severe hypoxia, de novo LDH synthesis may also contribute. These changes suggest that control mechanisms within the corneal epithelium may be responding to optimize the efficiency of anaerobic metabolism during conditions of environmental stress, and offer the prospect of a non-invasive technique for monitoring epithelial metabolic stress.

Diurnal variation in human tear enzymes.

A fluorometric technique for measuring the activities of lactate dehydrogenase (LDH) and malate dehydrogenase (MDH) in human tear fluid is described. Measurement of separate LDH and MDH activities during the day resulted in fluctuating values with no definite pattern. However, a consistent diurnal pattern emerged when the tear LDH/MDH ratio was measured for individual tear sample. The tear LDH/MDH ratio was highest upon waking, declining soon after to a stable level for the remainder of the day. Evidence for a corneal epithelial origin of tear LDH and MDH is discussed, as well as the ability of the technique to detect differences in the metabolic status of the epithelium between the open-eye and closed-eye environments. Possible causes of the elevation of the tear LDH/MDH ratio following lid closure are unbinding of intracellular M-type LDH within epithelial cells and increased cell membrane permeability, both processes occurring in response to reduced cellular energy content under the more hypoxic conditions of lid closure. Although the nature of the present experiment did not allow any conclusive statement, de novo LDH synthesis may also be contributing to the increased tear LDH/MDH ratio. Deliberate corneal epithelial surface disruption was found to have no quantifiable effect on the tear LDH/MDH ratio.