Conjunctival Surface Changes in Patients with Sjögren’s Syndrome: A Transmission Electron Microscopy Study

Abstract

purpose. To demonstrate the ultrastructural appearance of the conjunctival surface epithelium in patients with Sjögren’s syndrome (SS) compared with normal subjects.

methods. Conjunctival tissue specimens from seven normal subjects and eight patients with SS were obtained by bulbar conjunctival biopsy and examined by transmission electron microscopy.

results. The average number of microvilli per 8.3 μm epithelial length was significantly lower in the SS group than that in the control group (19.6 ± 2.5 vs. 28.0 ± 3.4, P < 0.0001). The microvillus height (0.539 ± 0.151 μm) and height-width ratio (1.825 ± 0.549) in the conjunctival epithelium in the SS group were significantly lower than those (height: 0.946 ± 0.117 μm, P < 0.001; and height-width ratio: 3.717 ± 0.696, P < 0.0001) in normal individuals. The microvilli in the SS group were wider than those in the control group (P = 0.003). Furthermore, the average number of secretory vesicles (per 8.3 μm epithelial length) in the apical conjunctival epithelial cell was significantly reduced in the SS group (16.4 ± 6.8 vesicles), compared with the control group (34.7 ± 1.2 vesicles, P = 0.003). In addition, although the ocular surface glycocalyx (OSG) was always present in control subjects, it was not detectable in all but one of the SS conjunctival specimens.

conclusions. The ultrastructural morphology of the apical conjunctival epithelium is altered in patients with SS. The findings suggest that an intact OSG may play a key role in the maintenance of a healthy ocular surface, possibly by preventing abrasive influences on the apical epithelial cells.

Dry eye syndrome (DES) is defined as a tear film disorder caused by tear deficiency or extreme tear evaporation that leads to injury on the ocular surface. ¹ The reported prevalence of significant DES is approximately 10% to 20% in the adult population, although severe DES is less frequent. ² DES leads to ocular irritation and ocular surface disease, termed keratoconjunctivitis sicca (KCS), resulting in blurred and fluctuating vision and increasing risk of sight-threatening corneal infection and ulceration. ³  

Sjögren’s syndrome (SS), a common disease-causing DES, shows a large female predilection (male-female = 1:9) ⁴ and is characterized by an enhanced immunologic responsiveness associated with several abnormal serum antibodies such as anti-Ro and anti-La. The salivary and lacrimal glands of patients with SS exhibit dense lymphocytic infiltration. ⁵  

Characteristic histologic changes of the ocular surface epithelia have been demonstrated in KCS and in DES. These changes include abnormal proliferation and differentiation of the ocular surface epithelium, with decreased density of conjunctival goblet cells and decreased and abnormal production of mucus by the ocular surface epithelium. ⁶  

The human ocular surface epithelia are composed of the uppermost layers of the corneal and conjunctival epithelia. The conjunctival epithelial cells are essential for tear film stability because: the epithelial cells contribute to secretion of tear mucus, the glycocalyx of the apical epithelial cells promotes tear adherence, and the microvilli of the apical epithelial cells increase the surface for tear adherence. ^{7 8 9} Mucins produced by conjunctival epithelial cells are either membrane tethered on the tips of the microvilli and participate in the formation of the glycocalyx or are secreted into the mucous layer in the tear film. ¹⁰ Membrane-tethered mucins form a dense barrier in the glycocalyx at the epithelial tear film interface. This barrier not only prevents pathogen penetrance, but also provides a lubricating surface that allows lid epithelia to glide over the ocular surface epithelia without adherence. ¹¹ Secretory vesicles are located just below the surface of the conjunctival epithelial cells. ^{10 12} These vesicles contain mucins, but the specific types have not yet been identified. ¹⁰  

The purpose of this study was to demonstrate the ultrastructural morphology of the conjunctival surface epithelium in patients with Sjögren’s syndrome compared with normal subjects, by using transmission electron microscopy (TEM). Evaluation was focused on the conjunctival microvilli, secretory vesicles, and the ocular surface glycocalyx (OSG). To our knowledge, this is the first study to demonstrate the ultrastructural morphology of the OSG in humans. 

Methods

Selection Criteria for Normal and SS Specimens

This study adhered to the tenets of the Declaration of Helsinki. Participation was voluntary, and informed consent was obtained after the subjects had been informed about the nature of the study. The study protocol was approved by the University of Thessaly and the University Hospital of Larissa Ethics Committee. Inclusion criteria for diagnosis of SS in the subjects were: symptoms of dry eye; symptoms of dry mouth; rose Bengal staining ≥4/9, according to Van Bijsterveld’s scoring system; Schirmer-1 test performed without anesthesia with results of ≤5 mm in 5 minutes in at least one eye; positive findings in a minor salivary gland biopsy; and positive serous autoantibodies ANA ≥ 1:160, RF ≥ 1:160, and positive SS-A (Ro) and/or SS-B (La). ¹³  

The age distribution was similar in the control (mean age, 50 ± 11.8 years) and SS (mean age, 54.4 ± 11.8 years, P = 0.49) groups. All normal subjects exhibited normal aqueous tear production (i.e., had Schirmer-1 test results of >15 mm in 5 minutes without anesthesia in both eyes, no rose Bengal staining, and negative serous autoantibodies). The information from the normal subjects and patients with primary and secondary SS from whom biopsy specimens were obtained is summarized in Table 1 . 

Conjunctival Tissue Collection

Superior bulbar conjunctival tissue specimens from four eyes of four patients with primary SS, four eyes of four patients with secondary SS, and seven eyes of seven randomly selected asymptomatic normal control subjects were obtained with topical anesthesia (Alcaine 0.5% drops; Alcon, Fort Worth, TX) under slit lamp biomicroscopy. 

Transmission Electron Microscopy

The specimens were fixed in 3% glutaraldehyde in phosphate buffer saline (PBS; pH 7.3) for 3 hours. Subsequently, they were postfixed in 2% osmium tetroxide in PBS for 1.5 hours. After being washed with PBS and double-distilled water, the specimens were dehydrated in a graded series of alcohol and were embedded in epoxy resins (SERVA Electrophoresis GmbH). Ultrathin sections (59–90 nm) were studied under a transmission electron microscope (model 2000 FX2; JEOL, Tokyo, Japan) after staining with uranyl acetate and lead citrate solutions. 

Data Analysis

Measurements were performed manually as follows: For each specimen, 10 to 12 images were acquired on high-quality negative film, which was processed according to manufacturer’s recommendations and printed using a magnification of ×24,000. All micrographs were examined, and of those, for each specimen, five representative micrographs were selected for quantitative analysis. They were scored in a masked fashion by two independent examiners who made measurements in millimeters with a surgical caliper. The measured values were converted into micrometers by simple arithmetic calculation, taking into account the magnification used. For measurements of the microvilli population and calculation of the number of secretory vesicles present in apical epithelial cells, a standardized ocular surface epithelial length of 8.3 μm was examined. 

Statistical analysis included descriptive statistics and Student’s t-test. P < 0.05 was considered as statistically significant. 

Results

Height, Width, and Number of Conjunctival Epithelial Microvilli

To obtain measurements of microvillus height, width, and number of conjunctival epithelia, electronmicrographs of conjunctival epithelia from normal subjects and patients with primary and secondary SS were obtained and scored in a masked fashion. In the SS group, the microvillus height, width, and ratio were measured as 0.539 ± 0.151, 0.299 ± 0.019, and 1.825 ± 0.549 μm, respectively. These values differ significantly from those measured in the control group (height: 0.946 ± 0.117 μm, P < 0.001; width: 0.259 ± 0.029 μm, P = 0.003; and height-width ratio: 3.717 ± 0.696, P < 0.0001). Furthermore, the average number (±SD) of microvilli per 8.3 μm epithelial length was significantly lower in the SS group (19.6 ± 2.5) than that in control subjects (28.0 ± 3.4, P < 0.0001). 

Although gender distribution differed between the control (four women, three men) and study (eight women) subjects, further analysis of the results showed height, width, and height-width ratio, as well as number of vesicles to be independent of gender distribution (P = 0.93, 0.24, 0.56, and 0.51, respectively). The results of the control group are shown in Table 2 . 

Structural complexity of the microvilli in the SS group, often with bifurcation and branching of the microvilli, was a common observation (Figs. 1A 1B) . 

Secretory Vesicles and OSG in the Apical Conjunctival Epithelia

Another important finding observed in the SS group, was a significant reduction in the number of secretory vesicles (per 8.3 μm epithelial length) in the apical conjunctival epithelial cells (mean, 16.4 ± 6.8 vesicles), compared with the number of vesicles in the control subjects (mean 34.7 vesicles ± 1.2, P = 0.003; Figs. 2A 2B 2C ). 

All control group specimens showed the presence of an OSG, characterized by a rather continuous, fine network of filamentous electron-dense material associated with the apical cell membrane. This filamentous layer was more prominent on the tips of the microvilli, extending among the microvilli, and forming an extrinsic cell surface coat (Figs. 1C 1D) . An interesting finding in the SS group was that no OSG was detectable by TEM, except in one specimen. However, in that particular specimen, the OSG morphologically differed from that in the control group. In this specimen, the OSG appeared irregular, in that it did not consist of a fine filamentous cell surface coat, but it appeared discontinuous, and the filaments were shorter and thicker, clustered onto the epithelial surface of the entire microvillus (Fig. 2D) . 

Discussion

The most striking changes in the ultrastructure of the conjunctival epithelium observed in the SS group were the decreased absolute number of microvilli per epithelial length, along with a reduced height-width ratio of the microvilli. The microvilli in SS conjunctivas were shorter and wider than those in control subjects. Furthermore, the OSG was not detectable in specimens except one from the SS group. A significant reduction in the number of secretory vesicles in the apical conjunctival cells was also a consistent finding in the SS group. Together, these findings, compared with the control, definitely represent at least a deviation from the appearance of the ocular surface epithelium in control, non-SS subjects. 

We hypothesize that one of the primary reasons for the decreased absolute number of microvilli per surface area, along with the reduced height-width ratio of the microvilli in the SS group, may be the loss of the membrane-tethered mucins from the conjunctival apical epithelial cell (Figs. 1A 1B 2A 2B 2C) . Membrane-tethered mucins that contribute to the formation of the glycocalyx (MUC1, -4, -13, -15, -16, and -17) ^{14 15 16 17 18} and secretory mucins found in the tears (MUC5AC and -2), ^{15 17} are juxtaposed at the boundary between the ocular surface and tear film, but do not firmly adhere to each other. ¹⁵ Adherence of membrane-tethered and soluble mucins is thought to be prevented by hydrogen-bonding of water molecules on polar regions in mucin molecules, which, by competition, effectively prevent mucin molecules from binding to one another. ¹⁹ Shielding of the mucin molecules by absorbed water molecules promotes diffusion of the tear film, reducing friction during blinking and preventing microtrauma on the ocular surface. ²⁰ Our findings suggest that the conjunctival epithelium in SS lacks the lubricating surface of the membrane-tethered mucins that allows lid epithelia to glide over the ocular surface epithelia without adherence. Therefore, we speculate that mechanical forces exerted by lid epithelia onto a dry ocular surface may play a role in the flattening and the structural complexity of the conjunctival microvilli in SS. Dilly ¹² hypothesized that the secretory vesicles fuse with the cell membrane, providing a considerable amount of extra cell membrane to the cell, and thus the microvilli are formed. Provided that the number of the secretory vesicles is reduced in the SS group, this hypothesis could explain the reduced number of microvilli, but not their reduced size. Our findings suggest that the secretory vesicles are too small to be the source of the microvilli, so that there is no morphologic link between them and the microvilli. 

The reduced number of secretory vesicles observed in the SS group could explain the observed absence of the membrane-tethered mucins from the conjunctival apical epithelial cells. Many investigators have shown that these vesicles contain mucin, but the type has not been identified yet. ^{10 17 21} We hypothesize that the membrane-tethered mucins MUC1, -4, and MUC16, which have been identified in the conjunctiva, ¹⁰ could comprise some of the mucins of the secretory vesicles of the conjunctival epithelial cells. This hypothesis is supported by the observation of material morphologically identical with the OSG on the inner wall of the vesicles’ membrane (Fig. 2D) . The secretory vesicles and their contents are produced by the endoplasmic reticulum and modified by the Golgi complex before they reach the subsurface site, according to the principles of exocrine secretion. The array of cytoskeletal elements in the cytoplasm, observed just below the exposed surface of the epithelial cells may guide these vesicles to their position beneath the membrane and finally assist their fusion with the cell membrane. ¹² In this way, the inner wall of the vesicle’s membrane and the mucins attached to it may become incorporated into the cell surface and form the glycocalyx. 

Argüeso et al. ¹⁸ showed a decrease in the distribution of a carbohydrate epitope, known to be carried on the membrane-tethered mucin MUC16 of the apical conjunctival epithelial cells from patients with non-Sjögren’s dry eye. These data suggest that the glycosylation of MUC16, the expression of the gene itself, or the rate of shedding of the mucin from the cell surface is altered. ¹¹ The same investigator showed that keratinization of the ocular surface epithelia is accompanied by changes in the pattern of expression of glycosyltransferases that initiate O-glycosylation of mucins, ²² which may lead to alterations in carbohydrate structures of the mucins. ^{11 22} These data are consistent with our finding that the OSG is dramatically altered in patients with SS and could explain the reduction in the number of secretory vesicles observed in the SS group, as they serve as vehicles for mucins trafficking. It is likely that under the influence of a multisystemic disease like Sjögren’s syndrome, protein synthesis in the conjunctival epithelial cell is altered. Subsequently, the phenotype and function of the epithelial cells are altered, leading to abnormal mucin synthesis and mucin release from the secretory vesicles. 

The present study is unique in that the ultrastructural morphology of the OSG is demonstrated in humans for the first time, both in normal subjects and in patients with SS. Taken together, our findings, coupled with those in previous studies, ^{11 18 22} suggest that the membrane-tethered mucins, MUC1, -4, and -16, which form the OSG of the conjunctival apical epithelial cell, may be a part of the mucins contained in the secretory vesicles of the stratified squamous epithelial cells of the conjunctiva. 

Further studies will facilitate the identification of the mucins contained in the secretory vesicles and the regulation of their secretion and trafficking. 

 Presented in part at the annual meeting of the Association for Research in Vision and Ophthalmology, Fort Lauderdale, Florida, May 2005.

 Submitted for publication June 24, 2005; revised August 31 and October 6, 2005; accepted December 15, 2005.

 Disclosure: D.I. Koufakis , None; C.H. Karabatsas, None; L.I. Sakkas, None; A. Alvanou, None; A.K. Manthos, None; D.Z. Chatzoulis, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Donor Information for Biopsy Specimens

Donor	Age	Gender	Eye	Diagnosis
1	49	F	OD	Normal Subject
2	65	F	OS	Normal Subject
3	56	F	OD	Normal Subject
4	33	M	OS	Normal Subject
5	62	M	OS	Normal Subject
6	39	M	OS	Normal Subject
7	46	F	OD	Normal Subject
8	60	F	OD	Primary SS
9	60	F	OS	Secondary SS
10	67	F	OD	Primary SS
11	56	F	OD	Secondary SS
12	36	F	OS	Secondary SS
13	53	F	OD	Primary SS
14	37	F	OS	Secondary SS
15	66	F	OS	Primary SS

Table 2.

 

Measurements of Microvilli and Number of Secretory Vesicles in Apical Conjunctival Epithelial Cells in Control Subjects

Gender	MV Height (μm)	MV Width (μm)	Height-Width Ratio	MV Number per 8.3 μm	SV Number per 8.3 μm
Male	0.952 ± 0.152	0.277 ± 0.034	3.503 ± 0.869	28.0 ± 3.60	35.25 ± 1.06
Female	0.942 ± 0.109	0.244 ± 0.017	3.877 ± 0.620	28.0 ± 3.74	34.50 ± 1.29

 Data are expressed as the mean ± SD. MV, microvilli; SV, secretory vesicles.

Figure 1.

 

(A, B) Electronmicrographs of conjunctival apical epithelial cells from a 53-year-old woman with primary Sjögren’s syndrome show an example of the structural complexity (chevrons) of the microvilli (MV) observed in the Sjögren’s group, exhibiting extensive branching and bifurcation. (C, D) Electronmicrographs of conjunctival apical epithelial cells from two normal subjects, a 56-year-old (C) and a 46-year-old (D) woman, shown for comparison. The microvilli appear single, elongated and do not branch. Also, note the numerous secretory vesicles (V). The OSG appears as a rather continuous, fine network of filamentous electron-dense material, forming an extrinsic cell surface coat (arrows). Magnification: (A, B) ×28,000; (C, D) ×20,000.

Figure 2.

 

Electronmicrographs of conjunctival apical epithelial cells from patients with Sjögren’s syndrome (SS). The microvilli (MV) appear either deformed (A, B) or flattened (C, D). The secretory vesicles (V) are reduced in number. The OSG is not detectable in (A), (B) or (C). In (D), the OSG is different morphologically from that in the control group, appearing discontinuous with shorter and thicker filaments, clustered onto the epithelial surface of the entire microvillus (small arrows). (D, inset), higher magnification (×60,000) of the boxed area showing material morphologically identical with the OSG observed on the inner wall of the vesicle membrane (large arrows). (A, B) A 37-year-old woman with secondary SS. N, nucleus. (C) A 67-year-old woman with primary SS; (D) a 60-year-old woman with primary SS. Magnification: (A–D) ×24,000; (D, inset) ×60,000.

The authors thank IOVS volunteer editor Dongli Yang, MD, PhD, and Dimitris A. Velonis, DDS, PhD, for assistance in editing the manuscript.