Analysis of the Free Fatty Acid Component of Meibomian Secretions in Chronic Blepharitis

Size: px

Start display at page:

Download "Analysis of the Free Fatty Acid Component of Meibomian Secretions in Chronic Blepharitis"

Jean Ferguson
5 years ago
Views:

1 Analysis of the Free Fatty Acid Component of Meibomian Secretions in Chronic Blepharitis Joel M. Dougherty and James P. McCulley Meibomian secretions were collected from 43 patients with chronic blepharitis and 8 normal controls. Patients were divided into six clinically distinct groups of chronic blepharitis. Individual secretions were weighed and separated into specific lipid classes by thin-layer chromatography. The free fatty acid (FFA) fraction was recovered, methylated, and analyzed by gas-liquid chromatography. Quantitation was achieved through the use of an internal standard, and qualitative analyses were aided by the use of commercial external standards. Carbon numbers were expressed in terms of their equivalent chain lengths (E.C.L.). For statistical comparisons, specific acid weights were expressed as nanograms per milligram of secretion. Data from individual subjects were tabulated by group and analyzed by a nonparametric analysis of variance. The FFA portion made up from.21% to 1.3% of the total meibomian secretion. Acids ranged in length from 12 to 29 carbon atoms. Iso-branched and anteisobranched carbon chains made up approximately 33% of the FFA fraction. E.C.L.'s corresponding to Ci 6:, Ci 8:, and Ci8:i together made up a major portion of the total FFA fraction (mean = 49%). When compared to normals, a significantly decreased amount of Cj 2: o was seen in the mixed seborrheic/staphylococcal group and the meibomian seborrhea group. A significantly decreased amount of anteiso-branched C 15: was seen in the mixed seborrheic/staphylococcal group. Significantly decreased amounts of anteisobranched C 2 3:o were seen in all of the seborrheic blepharitides. A significantly increased amount of isobranched C 2 2:o was seen in the meibomian keratoconjunctivitis group. No significant differences were seen in the staphylococcal group. Invest Ophthalmol Vis Sci 27:52-56, 1986 Chronic blepharitis is a common, complex, frustrating disease. It is certainly one of the most common conditions seen in the ophthalmologist's office, accounting for approximately 59, patient visits in It is a complex condition that manifests several different and sometimes overlapping arrays of signs and symptoms. It is not easily diagnosed and is sometimes seen in association with other underlying disorders (sicca), and bacterial infections (staphylococci). It is often seen in association with sebaceous gland dysfunction (seborrhea, rosacea), 2 but the etiology of many forms of chronic blepharitis is still a mystery. Because of this lack of understanding, the disease remains one of the most frustrating for the physician and patient to deal with. From the Department of Ophthalmology, University of Texas Southwestern Medical School. Presented in part at the Annual Meeting of the Association for Research in Vision and Ophthalmology, Sarasota, Florida, April 3- May4, Supported by NIH Grant EY-232 and EY-365. This research was also supported by an unrestricted development grant from Research to Prevent Blindness, Inc., New York, New York. Submitted for publication: January 29, Reprint requests: Joel M. Dougherty, MS, Department of Ophthalmology, University of Texas Health Science Center at Dallas, 5323 Harry Hines Boulevard, Dallas, TX Six clinically distinct groups of chronic blepharitis can be recognized. 2 A bacterial pathogen can be cultured from only two of these groups (staphylococcal, mixed seborrheic/staphylococcal). 3 All seborrheic groups and meibomian keratoconjunctivitis have shown a high degree of association with sebaceous gland dysfunction. Inflammation of the meibomian glands is present in secondary (spotty) meibomitis and meibomian keratoconjunctivitis (MKC). In addition, MKC patients present plugging of the meibomian glands and an unstable tear film. Although there are differences, the meibomian glands are often thought of as specialized sebaceous glands. The consistent finding of generalized sebaceous gland dysfunction in the seborrheic blepharitides and MKC, and the apparent meibomian gland involvement in MKC, has brought added relevance to the study of meibomian secretions. Alterations in the lipid composition of the meibomian secretions could result in unfortunate consequences and contribute to the pathophysiology of some forms of chronic blepharitis. 2-4 Increases in free fatty acids could cause irritation to ocular tissues and destabilize the tear film. Shifts to lipid moieties with higher melting points could result in gland plugging and produce a less dynamic, stagnate tear film. For example, it has been shown that temporary stabilization of the tear film could be achieved 52 Downloaded From: on 9/13/218

No. 1 FREE FATTY ACIDS IN CHRONIC BLEPHARITIS / Dougherty and McCulley 53 in MKC by expressing meibomian secretions from deep within the gland ductule onto the tear film.

2 No. 1 FREE FATTY ACIDS IN CHRONIC BLEPHARITIS / Dougherty and McCulley 53 in MKC by expressing meibomian secretions from deep within the gland ductule onto the tear film. 4 The study of meibomian secretions is hampered by the paucity of material that can be collected from an individual. However, there have been several recent investigations into the nature of the secretions from both humans and animals. 5 "" An extremely well-detailed report of the composition of pooled human and bovine secretions has been compiled by Nicolaides, et al. 5 The composition of rabbit secretions has been thoroughly described by Tiffany and Marsden. 7 ' 8 The purpose of this investigation was to compare the free fatty acid fractions of individual meibomian secretions obtained from chronic blepharitis patients to those of normal individuals, the goal being not to identify all the molecular species present in a free fatty acid fraction, but one of elucidating differences in composition among patient groups and normals. This paper describes the methodology for sampling and fractionating individual secretions and the analysis of the free fatty acid component. Materials and Methods Patient Selection and Evaluation Patients with chronic blepharitis were given careful ophthalmic evaluation. Patients had symptoms for at least 6 months and had received no therapy for at least 2 wk. Signs and symptoms were graded and recorded, and the patients classified according to the type of chronic blepharitis as previously described. 2 Normal volunteers served as controls, underwent the same evaluation as the patients, and were selected only after it was determined that they were free of ocular disease. Informed consent was obtained from all subjects prior to their participation in the study. Meibomian secretions were obtained and processed from all individuals as described in the following sections. Collection of Meibomian Secretions After the instillation of one drop of lidocaine onto the ocular surface, a lid conformer was placed into either the inferior or superior cul-de-sac. A sterile cotton swab was then passed over the lid margin to remove excessive debris and tears. The tarsal plate containing the meibomian glands was then squeezed between the swab and conformer, thereby expressing the meibomian secretions out onto the lid margin. The secretions were harvested with a sterile degreased platinum spatula. The entire area of the upper and lower lids was expressed and the meibomian secretions harvested as indicated. Samples were washed off the spatula into fresh chloroform in solvent washed and pre-weighed 6 X 5 mm glass tubes. Samples were taken to dryness with prepurified nitrogen gas at room temperature. Tubes were re-weighed, the difference in weight being taken as the weight of the secretion. The tubes were sealed under nitrogen and stored at 7 C until needed. Lipid Analysis All solvents used were of high purity, suitable for lipid work, and purchased from Burdick and Jackson Laboratories, Inc.; Muskegon, Michigan. Solvents were checked for purity prior to use, by concentrating and subjecting the residue to analyses by thin-layer and gasliquid chromatography (TLC and GLC respectively). All glassware was washed with 5% sulfuric acid, deionized water, acetone and chloroform prior to use. Teflon-lined caps were employed on all screw capped glassware. Blanks of silica gel were carried through all procedures to assess the level of contaminates in the system. The frozen secretions were brought to room temperature and redissolved in approximately 1 ix\ of chloroform. This was spotted in small increments onto precoated Analtech silica gel H (.-mm thickness) thin-layer chromatography plates under Nitrogen. The plates were activated at 12 C for 3 min prior to use. The inside of the tube was washed twice with 5 n\ of chloroform, and this was spotted along with the initial sample. Appropriate standards were spotted alongside the sample and included cholesteryl oleate, behenyl oleate, triolein, oleic acid, diolein, cholesterol, cetyl alcohol, and mono-olein. The plates were developed in hexane; diethyl ether; acetic acid (75; :1) for 3 min. After development, the plates were allowed to dry, sprayed with.2% dichlorofluorescein (in 95% ethanol), and viewed when dry under ultraviolet light. Bands of gel corresponding to different lipid classes were delineated then scraped from the plate into individual 13 X 1 mm glass tubes with screw caps. To the silica gel band containing the free fatty acid fraction was added.5 ml of benzene, 1 ml of BF 3 (Supelco Inc.; Bellefonte, PA), and up to 2.5 /zg of heptadecanoic acid (Ci 7: ) as an internal standard. (One milliliter of chloroform was added to the tubes containing the other bands. These were capped under Nitrogen, sealed, and stored at 2 C for future evaluation.) The free fatty acid methylation mixture was capped under nitrogen, and heated in a heating block at 65 C for 3 min with agitation every 1 min. At the end of the incubation and after cooling to room temperature, 1 ml of H 2 O was added, and the fatty acid methyl esters (FAME) were extracted 3 times with 1 ml portions of hexane. The extracts were combined in 1 X 75-mm glass tubes and taken to dryness with Nitrogen. The sides of the Downloaded From: on 9/13/218

3 54 INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / January 1986 Vol. 27 tubes were washed down with small volumes of chloroform. This was then dried off with Nitrogen, and the FAME'S were redissolved in 5 nl of chloroform and stored sealed under Nitrogen at -2 C until needed. All FAME analyses were performed isothermally (at both 175 and 2 C) in a Hewlett Packard 571A Gas Chromatograph (Hewlett Packard; 12 B, Palo Alto, CA) equipped with dual flame ionization detectors. Columns were stainless steel (W X 6') packed with acid washed Chromosorb W (1/12 mesh) and coated with 1% SP233 (Supelco Inc.). The carrier gas was Nitrogen at 3 ml/min. Injection ports were maintained at 2 C, the detectors at 3 C. Injections of.5 jil-2. n\ were made manually. Runs lasted 45 min, and columns were purged between runs with a temperature program to, and 1 min hold at, 24 C. Peak areas were determined by integration relative to the internal standard with a Hewlett Packard 339A Reporting Integrator. External standards consisted of quantitative commercial preparations of free fatty acids and fatty acid methyl esters. The free acid standards were carried through all analytical procedures that the samples were subjected to in order to check extraction and methylation efficiencies. Standards included homologous series of normal straight chain saturated and mono-unsaturated esters as well as iso- and anteisobranched members, from 12 to 3 carbon atoms in length. Identifications of sample peaks were based on these standards. A standard curve was prepared from a plot of the log of the retention times versus the carbon numbers of straight chain saturated FAME'S. Carbon numbers for sample peaks were then read from the curve using the observed retention times. Carbon numbers of unsaturated and branched chains yielded fractional carbon numbers. These were expressed in terms of their equivalent chain lengths (E.C.L.) 12 by interpolation from the carbon numbers of their equivalent saturated straight chain neighbors. The total weight of the FFA fraction was found by summation of the weights of all specific FFA's observed in a run. This total was divided by the weight of the secretion to find the percentage of the sample weight contributed by the FFA fraction. For statistical analyses, the weight of each specific acid was normalized by expressing it in terms of nanograms per milligram of sample. Data from individual subjects were tabulated by group and analyzed by a non-parametric one-way analysis of variance. Individual differences among groups were then isolated by Student Newman-Keul's test. A P-value of.1 was chosen because of the large number of analyses and multiple comparisons. Duncan's multiple comparison test was used to assess group differences caused by changes in free fatty acid chain configurations. Patient Classification Results Patients fell within six clinically distinct groups of blepharitis as described in detail elsewhere. 2 The number of patients from each group used in this study was as follows: staphylococcal, 8; seborrheic, 7; seborrheic/ staphylococcal mixed, 5; meibomian seborrhea, 8; secondary (spotty) meibomitis, 8; and meibomian keratoconjunctivitis, 7. Eight normal subjects served as controls. Collection of Meibomian Secretions The amount of meibomian secretion (henceforth referred to as "meibum" 5 ) that could be collected from each individual varied according to one or several factors: including the individual's tolerance to the discomfort caused by the procedure, the physical condition of the meibomian glands (eg, plugging), and the general sebaceous tone of the individual (oily or dry skin). The amount of material collected from the individuals in this study ranged from 1 ng to 5, ng (mean = 66 ^g)- No significant differences were seen among groups in the amount of meibum that could be collected. Lipid Analysis Thin-layer chromatography (TLC) separated the meibum into lipid classes which corresponded to wax and sterol esters, triglycerides, free fatty acids, free sterols, alcohols, and diglycerides, monoglycerides, and origin material. Sterol and wax esters were the most prominent component of the secretion. Table 1 shows the group means for most of the free fatty acids separated on GLC. The FFA fraction made up from.21% to 1.3% (2,1 ng/mg to 13, ng/ mg) of an individual's total secretion. Although there were individual differences in the qualitative pattern of specific acids, the group means for the total FFA fraction were remarkably consistent and averaged.48% (4, ng/mg to 6,1 ng/mg). Normals averaged.52% (5,2 ng/mg). Acids ranged in length from 12 to 29 carbon atoms. Many of the acids resolved had E.C.L.'s corresponding to even-numbered iso-branched and odd-numbered anteiso-branched carbon chains. Together, these acids made up approximately 33% of the free acid fraction of a normal secretion. No significant difference in the amount of branched chains was seen among groups. E.C.L.'s corresponding to C (6 :' C 18; ' and C 18:) together made up a large portion of the total FFA fraction (mean = 49%; range = 4% to 54%). No statistically significant differences were seen among groups with these major components. Downloaded From: on 9/13/218

4 No. 1 FREE FATTY-ACIDS IN CHRONIC BLEPHARITIS / Dougherry and McCulley 55 Table 1. Composition of free fatty acid component of meibum in chronic blepharitis" FFA] C 2: aic 3: Cl4: aic 15: Ci5: 'Ci6: C 6: C 6:l C 8:o C 8: C 8M lc2: C2: aic2l: '^: C: aic23: C24: C24: C24M aic :o C26: C26: aic 27;o ic28:o C28:l aic.29: NORM not STAPH SEB MIX 99 2 d SEB MEIB MKC * Acids listed are those identified and do not represent 1% of the total acids separated by GLC. t Free fatty acid identification based on ECL (see text). % Numbers represent group means (nanograms acid per milligram sample) Significantly different than normals, P <..5 (see text) Significant differences were seen among some minor components. When compared to normals, a significantly decreased amount of C 12: o was seen in the mixed seborrheic/staphylococcal group and the meibomian seborrhea group. A significantly decreased amount of anteiso-branched Ci 5: was seen in the mixed seborrheic/staphylococcal group. Significantly decreased amounts of anteiso-branched C 2 3 : o were seen in all of the seborrheic blepharitides. And a significantly increased amount of iso-branched C 2 2 : o was seen in the MKC group. All differences were significant at a P- value less than or equal to.5. No significant differences were seen in the staphylococcal group. These differences are noted in Table 1. Discussion There have been few investigations in which the lipid composition of meibum was used to assess meibomian gland function or was used as a marker for disease. Cory et al used thin-layer chromatography to compare meibum from normal adults and meibum from patients with ocular rosacea. 13 They were unable to detect differences using this relatively insensitive technique. Most recently, Nicolaides et al compared mei- bum of normal adults, acne rosacea and blepharitis patients using gas-chromatography/mass-spectrometry. 14 They found only minor insignificant differences among the blepharitis patients. However, they did not try to classify the blepharitis patients, and it is not clear whether the patients were in an acute or chronic state of disease. Tiffany found meibum composition so variable among six individuals, that he was moved to conclude that results from pooled material might not be a reliable indicator of lid disease. 6 We chose to compare the free fatty acid fraction of meibum from normal adults and patients with chronic blepharitis because, of all the lipid components present in meibum, the FFA fraction has been thought to have the greatest potential for causing ocular toxicity, irritation, and destabilization of the tear film. We chose not to pool individual secretions, since to do so would have compromised our statistical analyses and would not have allowed us to assess individual variation. Like Tiffany, we found a degree of individual variation. The variability was most noticeable among minor acids, whereas major components were relatively uniform among individuals. Within groups, individual variation tended to cancel out and a striking consistency of pattern emerged. Inter-group differences therefore; tended to be among minor acids as is the case for C 2, iso- Downloaded From: on 9/13/218

$56 INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / Jonuory 1986 Vol. 27 branched C 2 2, and anteiso-branched C 15: and C 2 3 : o. All of these acids represent minor components of the FFA fraction.$

5 56 INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / Jonuory 1986 Vol. 27 branched C 2 2, and anteiso-branched C 15: and C 2 3 : o. All of these acids represent minor components of the FFA fraction. For instance Ci 2: o and anteiso-branched Ci 5: each make up approximately 2% of the total FFA present in a normal secretion. Iso-branched C 2 2:o and anteiso-branched C 23: each make up.1% and 3.4% respectively, (see Table 1). The effects that changes in these components would have on the total meibum, the tear film, or the ocular surface are unclear. Shifts from branched to straight chain configurations of the same carbon number would tend to raise melting points and could contribute to a gland plugging. Shifts from shorter to longer chains of the same configuration would have a similar effect. Such shifts were not detected in this study. It is interesting to note, however, that the staphylococcal group which presents a relatively normal sebaceous function (dermatologically normal and no apparent meibomian abnormality) also presents a normal free fatty acid fraction. The percentage of FFA in meibum observed in this study (.48%) is lower than that reported by Nicolaides (2.1%), 5 Tiffany (%-24%), 6 or Cory (1.4%). 13 This may be due to differences in sampling, methodology, or the population studied. As Nicolaides has pointed out, 5 it would not be appropriate for the meibomian gland to produce large amounts of FFA, but a small amount may be tolerated and could be necessary as a surfactant. We did not find a difference in the total amount of FFA present among groups. However, we were analyzing freshly extruded meibum, whereas the amount of FFA actually present in the tear film was not measured. It is possible that the large amount of FFA normally present on the skin surface could spill over the lid margin onto the tear film, contributing greatly to the total FFA present already from the secreted meibum. In a population given to sebaceous gland dysfunction, 2-4 sebaceous FFA might be a source of ocular irritation. Bacteria present on the lid margins and the conjunctivae could also contribute to tear film FFA, by producing extracellular Upases capable of hydrolyzing meibum lipids. We have found that the normal and blepharitic lid and conjunctival flora is made up largely of coagulase negative Staphylococcus spp., Propionibacterium acnes, and lipophillic Corynebacterium spp. 3 All of these organisms are capable of producing extracellular Upases, 15 which could greatly influence tear film FFA levels. In summary, the FFA composition of meibum from patients with all types of chronic seborrheic blepharitides and MKC is significantly different in minor components from normal individuals, and patients with chronic staphylococcal blepharitis. Although the ramifications of this finding remain unclear at this time, changes in meibum could contribute to, or arise from, the pathophysiologic mechanisms in chronic blepharitis. Key words: chronic blepharitis, meibomian glands, meibum, free fatty acids, meibomian lipids, thin layer chromatography, gas-liquid chromatography, meibomian keratoconjunctivitis Acknowledgments The authors thank Dale Meyer, Ph.D., for his help in data evaluation and Dorothy Taylor for typing the manuscript. References 1. National Disease and Therapeutics Index (NDTI), IMS America, Dec McCulley JP, Dougherty JM, and Deneau DG: Classification of chronic blepharitis. Ophthalmology 89:1173, Dougherty JM and McCulley JP: Comparative bacteriology of chronic blepharitis. Br J Ophthalmol 68:524, McCulley JP and Sciallis GF: Meibomian keratoconjunctivitis. Am J Ophthalmol 84:788, Nicolaides N, Kaitaranta JK, Rowdah TN, Macy JI, Boswell FM, and Smith RE: Meibomian gland studies: comparison of steer and human lipids. Invest Ophthalmol Vis Sci 2:5, Tiffany JM: Individual variations in human meibomian lipid composition. Exp Eye Res 27:289, Tiffany JM: The meibomian lipids of the rabbit. Exp Eye Res 29:195, Tiffany JM and Marsden RG: The meibomian lipids of the rabbit. II. Detailed composition of the principle esters. Exp Eye Res 34: 61, Nicolaides N and Ruth EC: Unusual fatty acids in the lipids of steer and human meibomian gland excreta. Curr Eye Res 2:, 1982/ Baron C and Blough HA: Composition of the neutral lipids of bovine meibomian secretions. J Lipid Res 17:3, Andrews JS: Human tear film lipids. I. Composition of the principle non-polar component. Exp Eye Res 1:3, Miwa TK, Mikolajczak KL, Earle FR, and Wolff IA: Gas chromatographic characterization of fatty acids. Anal Chem 32:1739, Cory CC, Hinks W, Burton JL, and Shuster S: Meibomian gland secretion in the red eyes of rosacea. Br J Dermatol 89:, Nicolaides N, Santos EC, Robin J, and Smith RE: Meibum lipids in rosacea, blepharitis and chalazia. ARVO Abstracts. Invest Ophthalmol Vis Sci 24(Suppl):78, Pablo G, Hammons A, Bradley S, and Fulton JE: Characteristics of the extracellular lipases from Corynebactehum acnes and Staphylococcus epidermidis. J Invest Dermatol 63:231, Downloaded From: on 9/13/218

Bacterial Upases and Chronic Blepharitis

Bacterial Upases and Chronic Blepharitis Joel M. Dougherty and Jomes P. McCulley Eyelids and conjunctivae of 36 normal individuals and 6 patients from six clinical groups of chronic blepharitis were cultured