Immune mechanisms of dry-eye disease

Clinical background

Key symptoms and signs

The symptoms of dry-eye disease are discussed in detail in Chapter 14 . A point that bears emphasis is that patient presentation is extremely variable. Patients may experience troublesome symptoms but present none of the standard clinical signs and even have increased fluid production; may report symptoms and show ocular surface pathology but have normal fluid production; or may show decreased fluid production but deny symptoms.

Historical development

Papers describing fibrotic and atrophic changes in the lacrimal glands of aged subjects began appearing in the late nineteenth century. In 1903 Schirmer noted that fluid production varied widely among normal individuals but tended to decrease with increasing age and to decrease more severely in women. Later investigators substantiated Schirmer’s conclusions. In the 1930s, Sjögren introduced the terms “keratoconjunctivitis sicca” (KCS) and “sicca complex,” and reported findings from patients with the sicca complex and arthritis, including documentation of inflammatory changes in glands obtained from several of the affected individuals.

The concept of autoimmune diseases emerged in the subsequent decade, along with the discovery that patients’ sera frequently contained antibodies directed against certain tissues or intracellular structures. Bloch and Bunim showed that the sicca complex and glandular histopathology occurred in patients with other autoimmune diseases in addition to rheumatoid arthritis. Bloch et al reported that the sicca complex and autoantibody titers also occurred in patients with no sign of autoimmune disease affecting other tissues. The distinction between primary Sjögren’s syndrome and secondary Sjögren’s syndrome was established by the mid-1970s.

The nature of lacrimal gland atrophy and dysfunction outside the setting of Sjögren’s syndrome and other inflammatory diseases has been somewhat controversial. Examining lacrimal gland and salivary gland histology in autopsy subjects, Waterhouse found at least slight adenitis in the lacrimal glands of between 8% and 22% of men in different age groups, with no indication of an age-associated increase. In contrast, the frequency of adenitis, which he interpreted as an autoimmune phenomenon, increased in women, from 22% in women younger than 44 years to 65% in women 75 and older. Whaley et al determined the frequency of decreased Schirmer test scores and increased rose Bengal staining in inpatients hospitalized for various indications but explicitly excluding autoimmune diseases. Because they found no correlation between the ocular surface findings and various autoantibodies, they concluded that dry-eye disease in their subjects was due to atrophic changes, rather than autoimmune phenomena. Subsequent postmortem and biopsy studies, discussed in detail in the section on pathology below, documented the frequency of age-associated fibrosis and parenchymal atrophy, and they generally also demonstrated associations with increased lymphocytic infiltration.

A development in oral pathology is of interest in this context. Daniels and Whitcher described histopathological features of labial salivary glands from patients who could and could not be diagnosed as having Sjögren’s syndrome on the basis of serum autoantibodies and clinical diagnoses of xerostomia and dry eye. Their conclusion might seem paradoxical to those who have been taught that the pathophysiology of Sjögren’s syndrome is autoimmune-mediated destruction of the secretory parenchyma: While parenchyma was replaced by lymphocyte aggregates, Sjögren’s syndrome cases were distinguished by an absence of acinar atrophy or ductal dilatation, even in parenchymal areas immediately adjacent to large aggregates.


This topic is reviewed in Chapter 14 .

Genetics and risk factors

There is a significant genetic influence on the incidence of Sjögren’s syndrome, since having a first-degree relative with an autoimmune disease increases the risk sevenfold. An association between human leukocyte antigen (HLA) DR alleles and autoantibodies is recognized, but other reported genetic associations have been controversial. Having delivered a baby doubles the risk for developing Sjögren’s syndrome 2.1-fold.

There appear to be no reports of genetic factors influencing dry-eye disease not associated with autoimmune diseases. Other risk factors are discussed in Chapter 14 .

That most dry-eye patients, and the large majority of patients with Sjögren’s syndrome, are women prompted studies of sex steroid actions in animal models ( Box 15.1 ).

Box 15.1

Gender-related lacrimal gland dimorphisms

The rodent lacrimal gland is subject to gender-related dimorphisms. Androgens support :

  • Larger population of dimeric immunoglobulin A (dIgA)-producing plasmacytes

  • Greater parenchymal expression of the polymeric immunoglobulin receptor (pIgR)

  • Greater secretion of secretory IgA (sIgA) and secretory component (SC)

  • Lower basal rate of fluid production

Androgens clearly influence immune cell activity in the lacrimal glands and the status of the ocular surface. Rocha et al proposed that androgens exert their influences by controlling expression of immunomodulatory mediators by parenchymal cells; this important concept is discussed at length in the section on pathophysiology, below. However, much remains to be learned about mechanisms underlying the androgens’ influences. Gonadectomizing or hypophysectomizing experimental animals causes significant biochemical and functional changes but does not cause acinar atrophy or fibrosis on the scale of the changes that occur in the aging human lacrimal gland or normally aging rats and mice.

Other hormones also influence lacrimal gland and ocular surface cytophysiology and immunophysiology. Mathers et al found that lacrimal fluid production correlates positively with increasing serum testosterone only in premenopausal women. In contrast, in all groups studied, i.e., premenopausal, postmenopausal without hormone replacement therapy, and postmenopausal with hormone replacement therapy, all measures of lacrimal function correlate negatively with increasing levels of serum prolactin (PRL). Notably, all subjects in this study had PRL values within the normal range.

Findings on the actions of estrogens defy explanations based on a single hormone-regulated process. Women with premature ovarian failure present increased signs and symptoms but produce fluid at normal rates. On the other hand, estrogen replacement therapy is associated with exacerbation of dry-eye symptoms in older women.

Differential diagnosis, treatment, and prognosis

These topics are addressed in Chapter 14 . The extent to which dry-eye disease occurs in association with local and systemic inflammatory diseases should be noted.


Lacrimal gland

Most studies of the histopathology of the human lacrimal gland in normal aging have confirmed the earlier finding that aging is associated with increased fibrosis, ductal pathology, acinar atrophy, and infiltration of immune cells. Damato et al and Pepose et al described the presence of lymphoid aggregates or foci, and, occasionally secondary follicles, even in individuals without a history of autoimmune disease. Roen et al and Obata noted the frequent occurrence of ductal dilatation, and Obata also documented the frequency of fatty infiltration. In an analysis of postmortem lacrimal glands, Obata et al found that diffuse fibrosis, diffuse acinar atrophy, and periductal fibrosis were more frequent in the orbital lobes of elderly women, and in the palpebral lobes of aging men. They also noted that the frequency of lymphoid foci increased with age and, as Wieczorek et al had found, that most foci were located near intralobular or interlobular ducts, i.e., those within, but at the periphery of, a lobule.

Nasu et al compared lacrimal glands from subjects with autoimmune diseases and with no history of autoimmune diseases. They concluded that the entire population shared common histopathological features and differed only by degree. Although the incidence of lymphoid foci was highest in patients with Sjögren’s syndrome and other autoimmune diseases, only 36% of lacrimal glands from subjects without autoimmune diseases appeared to be free of infiltrates. In the remainder, the incidence and severity of infiltration were highest among those older than 40 years and the incidence was nearly identical in males (63.9%) and females (62.8%).

The immunohistopathology of the lacrimal glands in patients with Sjögren’s syndrome presents some diversity. Pflugfelder et al found that, of 6 patients, lymphocytic infiltration was diffuse in 4 and focal in 2. Tsubota et al compared the histopathological features of the lacrimal glands of subjects with Mikulicz’s disease and Sjögren’s syndrome, which share several features, including massive infiltration by essentially identical proportions of CD4 + , CD8 + , and CD21 + lymphocytes. Whereas fluid production is severely impaired in patients with Sjögren’s syndrome, patients with Mikulicz’s disease retain exocrine function, and their ocular surfaces appear normal.

Conjunctiva and cornea

The corneal and conjunctival epithelia undergo marked morphological changes in dry-eye disease. The number of goblet cells in the conjunctival epithelia decreases; cells in the superficial conjunctival epithelial layers flatten, such that the epithelium thins even as the number of strata increases; cells in the most superficial layer lose most of their microvilli and separate from their normally close attachment to the penultimate layer; hyaline bodies, suggested to represent the residua of defunct goblet cells, appear in the epithelium; and vacuoles and other inclusions appear within the cytoplasm. The lamina propria underlying areas of affected epithelium becomes increasingly populated by lymphocytes and leukocytes. Subsequent studies have confirmed that increased numbers of lymphocytes are present within the conjunctiva of patients with dry-eye disease. Epithelial cells expressing HLA DR (human major histocompatibility complex (MHC) class II) molecules were present in conjunctival impression cytology specimens from 50% of patients and in brush cytology specimens from 66% of patients with idiopathic dry-eye disease.


Explicit concepts are emerging for the mechanisms by which environmental stresses, iatrogenic factors, allergy and infection, and endocrine changes can initiate dry-eye disease. Before presenting these concepts, it is appropriate to review physiological principles and cytophysiological mechanisms that influence disease development.


Nexus between the visual system and the mucosal immune system

The normal ocular surface fluid provides a microenvironment for the living epithelial cells exposed in the interpalpebral regions of the cornea and conjunctiva. Figure 15.1 illustrates the general wiring scheme by which perception of irritation or dryness in the cornea or conjunctiva elicits production of lacrimal fluid.

Figure 15.1

Wiring of a physiological servomechanism. A perception of irritation or dryness by sensory nerve endings in the cornea and conjunctiva elicits afferent signals, which travel through the trigeminal ganglion to reach a lacrimal center in the brainstem. (Note that sensory nerve endings are also present in the lacrimal gland.) Like sensory information from the viscera, signals from the ocular surface are processed and lead to the generation of efferent autonomic secretomotor signals, even when there is no conscious awareness that the status of the ocular surface has deviated from its homeostatic setpoint. The secretomotor signals reach the lacrimal glands by way of both sympathetic and parasympathetic nerves, which release their neurotransmitters in the general vicinity of, but do not form synapses with, parenchymal epithelial cells.

The epithelia of the lacrimal glands, ocular surface, and lacrimal drainage system form a topological continuum with the mucosae of the respiratory system and the gastrointestinal system. The ocular surface system, like the respiratory and gastrointestinal systems, contains organized inductive sites for adaptive mucosal immunity, and it also performs both innate and adaptive mucosal immune effector functions. Even as the lacrimal epithelia perform the exocrine functions associated with production of the ocular surface fluid, they devote much of their cytophysiology to accomplishing mucosal immune effector functions.

Cytophysiological apparatus

Lacrimal epithelial cells employ ion pumps, symporters, exchangers, and channels that are common to essentially all nucleated cells. They generate vectorial ion fluxes by using transport vesicles to insert specific ion transporters into the basal, lateral and apical domains of their plasma membranes. Figure 15.2 illustrates the disposition of the ion transporters as an apparatus for secreting Cl ions and K + ions through the cells and Na + ions through the paracellular pathway. This apparatus is distinct from the apparatus that secretes proteins.

Figure 15.2

The cytophysiological apparatus for exocrine secretion of Cl ions, Na + ions, and, in ductal cells, K + ions. Secretion of the ions creates the osmotic driving force that causes water to move from the interstitial space to the lumen of the acinus−duct system. Recent studies indicate that the Na + /H + exchanger (NHE) and the Cl /HCO 3 exchanger (anion exchanger, AE) work in concert and in parallel with the Na + K + 2Cl symporters (NKCC) to drive Cl ions from the interstitial fluid to the cytosol, against an unfavorable electrochemical potential difference. Secretagogue-mediated opening of apical Cl channels allows Cl ions to flow into the lumen. Flux of Na + ions through the paracellular pathway dissipates the lumen-negative transepithelial voltage difference that results from the transcellular flux of Cl ions. Apical K + channels and K + -Cl symporters are primarily found in ductal epithelial cells. (Not illustrated is the fact that the transporters spend 90% of their time in the intracellular compartments depicted in Figure 15.3 . Secretagogue stimulation recruits more Na,K-ATPase (NKA) pump units to the basal lateral plasma membrane and activates NHE exchangers in the basal lateral membrane.)

The organelles that lacrimal epithelial cells use to secrete proteins are also common to most cell types. However, by using transport vesicles to transfer products between specific organelles, they organize the organelles into apparatus to perform an exocrine function. Figure 15.3 illustrates this dynamic apparatus. The regulated exocrine apparatus exocytoses proteins, including signaling mediators, into the nascent ocular surface fluid. The transcytotic apparatus takes dimeric IgA (dIgA) up from the stromal fluid and exocytoses secretory IgA (sIgA) at the apical membrane; it also functions as a paracrine secretory apparatus, which exocytoses extracellular matrix components and signaling mediators into the stromal space ( Box 15.2 ). A novel paracrine apparatus, discussed below, is induced under certain physiological and pathophysiological conditions. The autophagic-lysosomal apparatus mediates the catabolic turnover of glycoproteins and phospholipids; it communicates with the transcytotic paracrine apparatus at the late endosome.

Figure 15.3

Cytophysiological apparatus for exocrine secretion of glycoproteins, transcytotic secretion of secretory component and secretory IgA, paracrine secretion of signaling mediators, and catabolism of cellular proteins, lipids, and carbohydrates. The apparatus are essentially the same in acinar cells and duct cells, but the exocrine and paracrine secretory products differ.

The glycoproteins that each epithelial cell secretes and the proteins it uses in its ion-transporting apparatus are assembled in the common biosynthetic apparatus, which consists of the endoplasmic reticulum (ER) and the Golgi complex. The products then traffic to the trans -Golgi network (TGN), which is the most complex of the cells’ several sorting nexuses. The TGN sorts specific products into distinct microdomains, and in these microdomains the transport vesicles form that will traffic to the other organelles.

The immature secretory vesicle is the entry compartment of the classic exocrine secretory apparatus for glycoproteins. The late endosome and the isolation membrane are the entry compartments of the autophagic-lysosomal apparatus. The recycling endosome and early endosome comprise both a transcytotic secretory apparatus and, simultaneously, a paracrine secretory apparatus. Under certain circumstances, both physiological and pathophysiological, the cell is induced to express a second paracrine secretory apparatus.

Autoantigens that are present in the cytosol, as well as autoantigens that are embedded in the lipid bilayers of membrane-bound organelles, enter the luminal spaces of the membrane traffic apparatus through the process of autophagy and through the formation of multivesicular bodies. Both processes are of interest. Formation of multivesicular bodies in lacrimal gland epithelial cells has not been studied, but it is a fundamental cytophysiological process with potentially important implications for autoantigen exposure. Like proteins that are released to the fluid phase, the proteins in microvesicles that may be released from the cell can be taken up, processed, and presented by antigen-presenting cells.

Box 15.2

Acute and long-term regulation of lacrimal epithelial cells

Secretagogues acutely activate:

  • The ion- and water-secreting apparatus

  • The exocrine apparatus for secreting proteins

  • Transcytotic apparatus for secreting secretory component and secretory immunoglobulin A

Hormones, cytokines, and inflammatory mediators likely influence:

  • Expression of specific components of each apparatus

  • Expression of secretagogue receptors

  • Expression of intracellular signaling cascades activated by secretagogue receptors

The lacrimal gland’s innate mucosal immune effector function is to secrete bacteriostatic and bactericidal effector molecules. These include lactoferrin, lactoperoxidase, and lytic enzymes, such as lysozyme and other glycohydrolases, proteases, and phospholipases. Its adaptive mucosal immune function is to deliver sIgA into the ocular surface fluid.

The transcytotic paracrine apparatus that delivers IgA to the ocular surface fluid is described in detail in Box 15.3 . The authors have proposed that, because this apparatus communicates with the autophagic lysosomal apparatus at the late endosome, it also mediates the secretion of autoantigens to the underlying stromal space. Moreover, the unique strategy lacrimal epithelial cells use to regulate secretion of sIgA and secretory component acutely may give them a propensity to generate and secrete autoantigen fragments that contain otherwise cryptic epitopes.

Box 15.3

Secretory immunoglobulin A (sIgA)

  • Immobilizes microbes, preventing penetration of the epithelial barrier

  • Does not fix complement

  • Reduces the risk that inflammatory responses will be needed to fight infections

The task of delivering of sIgA to the ocular surface fluid requires that lacrimal parenchymal cells maintain the stromal space as a niche for dIgA-secreting plasmacytes. They do so by secreting paracrine mediators that induce plasmablasts to undergo terminal differentiation and additional mediators that support the survival of mature plasmacytes. While both acinar and ductal epithelial cells secrete sIgA and secretory component, they do not express the same spectra of paracrine mediators. More is known about mediators produced by duct cells, while more is known about cytophysiological mechanisms of acinar cells.

In mucosal immune effector sites transforming growth factor-β (TGF-β) conveys the signal for plasmablast differentiation. In the rabbit lacrimal gland, TGF-β expression is largely localized to interlobular duct epithelial cells. Normally, duct cells apportion TGF-β into both their exocrine secretory apparatus and their transcytotic paracrine apparatus. TGF-β is synthesized in a proform; proteolytically processed, mature TGF-β typically remains in a noncovalent complex with its latency-associated peptide. Latent TGF-β entering the stroma is thought to associate with the extracellular matrix proteoglycan, decorin, which appears to be produced by both ductal and acinar epithelial cells. Thus, decorin may create a diffusion barrier that keeps latent TGF-β concentrated in the spaces surrounding the interlobular ducts, where it would signal to plasmablasts entering the gland from venules that parallel the ducts.

TGF-β also conveys additional signals: proliferative signals to mesenchymal cells and some epithelia, and antiproliferative or proapoptotic signals to other epithelia and lymphocytes. The antiproliferative signals contribute to its immunosuppressive actions. Moreover, TGF-β can induce expression of its own mRNA in T lymphocytes and antigen-presenting cells. It appears that mucosal immune effector tissues are able to use TGF-β to induce plasmablast differentiation because they also secrete paracrine mediators that abrogate TGF-β’s proapoptotic signals. In the lamina propria of the small intestine, interleukin-6 (IL-6) appears to be a major plasmacyte survival factor. In the lactating mammary gland, plasmacyte survival is supported by PRL. Schechter et al demonstrated that PRL, like TGF-β, is concentrated in epithelial cells of the interlobular ducts in the rabbit lacrimal gland. Retention of TGF-β in the stromal space surrounding the interlobular ducts may minimize its deleterious actions. In contrast, PRL is thought free to diffuse from the periductal regions to the periacinar regions, where it would promote survival of plasmacytes and acinar cells.

Azzarolo and coworkers found that plasmacytes in the rabbit lacrimal gland undergo apoptosis within hours after animals have been ovariectomized. Apoptosis is prevented if either dihydrotestosterone or estradiol is administered prior to ovariectomy. These findings highlight the importance of survival signals for plasmacytes, and they suggest that the sex steroids either support expression of PRL or interact with PRL to generate the survival signal.

Studies of lacrimal glands of pregnant rabbits demonstrate that the influences of epithelium-derived paracrine mediator extend beyond plasmablast differentiation and plasmacyte survival.

As described in detail in Box 15.4 , expression of both TGF-β and PRL increases during pregnancy. The increase of TGF-β appears to be driven by estradiol (E2) and progesterone (PRG), while the increase for PRL is driven, at least in part, by pituitary PRL. PRL induces expression of the novel paracrine apparatus, amplifying delivery of both mediators to the stroma and decreasing delivery of both to the ocular surface fluid. These changes are associated with significant changes in lacrimal fluid production and with a redistribution of lymphocytes and plasmacytes from periductal perivenular spaces to periacinar spaces. Evidence now suggests that the counterpoise between TGF-β and PRL may be a key factor determining the lacrimal gland’s immunophysiological status.

Aug 26, 2019 | Posted by in OPHTHALMOLOGY | Comments Off on Immune mechanisms of dry-eye disease

Full access? Get Clinical Tree

Get Clinical Tree app for offline access