Ashutosh Verma. Department of Microbiology, Immunology, and Molecular Genetics, University of Kentucky College of Medicine, MN 469 Chandler Medical Center, Lexington, KY 40536-0298, USA. Tel.: 1 859 257 9305; 1 859 257 9358; Fax: 1 859 257 8994; E-mail: email@example.com
Leptospirosis, caused by pathogenic species of genus Leptospira, is a highly prevalent zoonotic disease throughout many parts of the world, and an important emerging disease within the United States. Uveitis is a common complication of systemic infection in humans. A similar condition in horses is characterized by recurrent bouts of inflammation. In this article, we review advances in our understanding of leptospiral uveitis and its pathogenic mechanisms.
According to one estimate, more than one million cases of leptospirosis occur worldwide annually (Adler et al., 2011). The prevalence of leptospirosis in many parts of the world is due to chronic kidney infection of a wide variety of domestic, peridomestic and wild reservoir mammals, including rodents, cattle and dogs. Colonization of the renal tubules of carrier animals results in shedding of virulent leptospires in the urine. The bacteria then persist in fresh water until infection of a new host occurs via the conjunctiva, breaks in the skin or by invasion of mucous membranes in the respiratory or digestive tract.
Immune privilege is a special immunological status of an organ or tissue for which the threshold of tolerance has been increased to reduce chances of inflammatory injury. The immune privileged status of the eye was discovered by Medawar (1948) when he reported that allogenic skin grafts placed in the anterior chamber of rabbit’s eye survived for prolonged and sometimes indefinite periods of time (Medawar, 1948). In 1953, Billingham and Boswell provided evidence that the cornea is an immune privileged tissue. Subsequently, it was shown that the cornea, anterior chamber, vitreous cavity and subretinal space are immune privileged sites and that the cornea, lens, pigmented epithelium and retina are immune privileged tissues (Barker and Billinghamm, 1977; Streilein, 1987).
Although inflammation in general is required to ward off invading pathogens, an intense inflammatory reaction in an organ as delicate as the eye can distort the visual axis and thus impair vision. Ocular immune privilege helps the eye protect itself from pathogens while avoiding damage to its delicate components (Streilein et al., 2002). This is even more important in the case of corneal endothelium and neurosensory retina, as these are incapable of regeneration after injury (Matsubara and Tanishima, 1983; So and Aguayo, 1985). Immune privilege in the eye is achieved through several related mechanisms, viz., immunological ignorance, maintenance of an immunosuppressive environment and peripheral tolerance of eye-derived antigens (Streilein, 2003).
In initial studies on ocular immune privilege, the blood–eye barrier and absence of lymphatics in the eye were identified as possible reasons for this phenomenon (Medawar, 1948). Aside from these special anatomical features, additional mechanisms have been described that promote ‘immunological ignorance’ in the eye. For example, expression of MHC class I antigens is reduced in corneal endothelial cells. Moreover, normal cornea lacks MHC class-II-expressing antigen presenting cells (APCs) that usually activate alloreactive effector T cells (Streilein et al., 1979; Hamrah et al., 2003). In addition, normal healthy corneal epithelium prevents angiogenesis within the graft and its bed following corneal transplantation (Hori and Streilein, 2001).
Normal aqueous humor contains various immunomodulatory and anti-inflammatory factors. These factors suppress functions of molecules that mediate adaptive and innate immune inflammation and of cells that trigger intense inflammation, angiogenesis and bystander cell injury. The various soluble factors of aqueous humor form the immunosuppressive microenvironment in the eye, viz., transforming growth factor-β, α-melanocyte stimulating hormone, calcitonin gene-related peptide, vasoactive intestinal peptide, thrombospondin, macrophage migration inhibitory factor, interleukin-1Rα, CD46, CD55, CD59 and CD95L (Taylor, 1999; Streilein et al., 2002; Zamiri et al., 2005). In addition to these soluble immunomodulatory factors, cell-surface receptors in the eye interact with their co-receptors on inflammatory cells and inhibit effector functions and convert effector cells into regulators (Taylor, 1999; Streilein, 2003).
Despite the above-mentioned anatomical and immune specializations, foreign entities placed in the eye come under the radar of ocular immune apparatus (Streilein, 2003). During the acute phase of leptospiral infection, the most common clinical sign involving the eye is non-specific ocular congestion with or without chemosis and rarely with subconjunctival haemorrhage. But as the adaptive immune response clears organisms from circulation, leptospires continue to persist in immune sequestered sites of the body such as the eye. The time between systemic infection and appearance of inflammatory changes in the eye may vary from a few weeks to years. Uveitis, inflammation of the uveal tract (iris, ciliary body and choroid) of the eye is a common complication of systemic leptospiral infection in humans and horses, and occasionally in dogs and cattle (Hoag and Bell, 1954; Rathinam et al., 1997; Chu et al., 1998; Martins et al., 1998; Faine et al., 1999; Mancel et al., 1999; Rathinam, 2002; Townsend et al., 2006; Pappachan et al., 2007).
Leptospiral Uveitis in Humans
Uveitis as a sequel to systemic human leptospirosis was first described by Weil in 1886, when one of four reported cases developed iridocyclitis 6 weeks after the initial infection (Weil, 1886). Uveitis develops 2 weeks to a few years after the systemic infection but in most cases the time of onset is about 6 months following a systemic infection (Rathinam, 2002). The common clinical signs include iritis, iridocyclitis, hypopyon (layering out of white blood cells in the anterior chamber), vitreous reaction, papillitis and retinal vasculitis (Fig. 1) (Sturman et al., 1949). Leptospiral uveitis is more common in young and middle-aged men, most likely due to occupational hazard associated with agricultural farming (Rathinam et al., 1997). The intraocular inflammation is usually non-granulomatous, may be acute or recurrent and anterior or diffuse. Panuveitis is often more severe and relapsing in nature. In endemic regions such as India, infection with Leptospira spp. accounts for approximately 10% of total uveitis and one-third of infectious uveitis cases in people (Rathinam and Namperumalsamy, 2007). With the emergence of leptospirosis in the United States, it is tactically important to address leptospiral uveitis as an impending threat.
The prolonged symptom-free period between leptospiral infection and onset of an ocular condition makes a definitive diagnosis difficult for the ophthalmologist. Due to current lack of any specific diagnostic assay, presumptive diagnosis is made on the basis of a past exposure to a potentially contaminated environment, the exclusion of other aetiologies and a positive microscopic agglutination test (MAT) (Rathinam, 2005). As the sensitivity of MAT is low and in some cases there is a prolonged gap between active infection and uveitis, a set of clinical diagnostic predictors were identified by Dr. S. R. Rathinam (Aravind Eye Hospital, Madurai, India) using samples from more than 500 seropositive leptospiral uveitis and 4800 nonleptospiral uveitis cases by multiple logistic regression analysis (Rathinam, 2006). According to these analyses, acute, anterior or pan, nongranulomatous uveitis with hypopyon, disc oedema, vasculitis and vitreous exudates should be taken as inclusion criteria (Rathinam et al., 1997; Dana, 2002; Rathinam, 2006). In addition, idiopathic uveitis and other entities associated with HLA-B27, leprosy, sarcoidosis, syphilis and tuberculosis should be carefully ruled out.
The severity of the disease is variable, but, in general, leptospiral uveitis in humans has a good prognosis if accurately diagnosed and timely treated (Rathinam et al., 1997). Anterior uveitis, which is the milder form, is either self-limiting or requires the use of topical corticosteroids and mydriatics. Patients with diffuse uveitis require periocular or systemic corticosteroids.
Equine Recurrent Uveitis and Leptospiral Infection
In 1819 James Wardrop, in his work An assay on the diseases of the eye of the horse, described a ‘specific inflammation’ of uveal origin that was different from what he observed as a ‘simple or common inflammation’ of corneal origin (Cook and Harling, 1983; Paglia et al., 2004). This ‘specific inflammation’ is now known as equine recurrent uveitis (ERU; also known as moon blindness, iridocyclitis or periodic ophthalmia). Equine recurrent uveitis is a chronic, recurrent inflammatory disease of the uveal tract (Cook and Harling, 1983) with a prevalence of approximately 8% in the United States (Schwink, 1992). It is also the most common cause of vision loss in horses worldwide (Errington, 1941; Hartskeerl et al., 2004). The Appaloosa breed and horses with MHC class I haplotype ELA-A9 have been observed to be at increased risk of developing uveitis (Deeg et al., 2004; Dwyer and Gilger, 2005). As per a recent estimate, the financial impact of this disease on the US equine industry alone could be as high as 100–250 million dollars a year (Dwyer and Gilger, 2005). Several known and unknown ‘factors’ can incite uveitis that leads to a recurring or persistent uveitic syndrome called ERU (Dwyer and Gilger, 2005).
Three clinical presentations of ERU have been described (Gilger and Michau, 2004). The most common and classical form of ERU is characterized by episodes of severe ocular inflammation followed by periods of low inflammation. Inflammation of the iris, ciliary body and choroid is often accompanied by inflammation of cornea, anterior chamber, lens, retina and vitreous. The insidious form of ERU is marked by a low-grade chronic inflammation that never resolves and is very difficult to recognize until the development of a cataract or vision loss. In posterior ERU, there is inflammation of the choroid, vitreous and retina accompanied by mild or no anterior uveitis (Gilger and Michau, 2004).
Onset of the disease is usually acute with variable degrees and duration. The acute signs include blepharospasm (uncontrolled muscle contraction of eyelid), photophobia, lacrimation, eyelid oedema, chemosis (swollen conjunctiva) and corneal oedema due to injury to corneal endothelium. In addition, miosis (constriction of pupil) ensues as a response to prostaglandins and other inflammatory mediators. As the disease progresses, aqueous flare and hypopyon may also be seen. Animals in the acute phase of ERU respond well to therapy (as described in following pages) and the outcome depends a lot on how aggressively the animal is treated during this phase. The acute phase is followed by a quiescent phase of minimal inflammation (Cook and Harling, 1983). Subsequent recurrence of inflammation results in a more severe disease with poor prognosis for preserving vision. Anterior and posterior synechia (anterior or posterior attachment of iris), secondary cataract, lens luxation (lens dislocation), vitreous exudates and retinal detachment are common features of subsequent inflammatory episodes (Rebhun, 1979; Cook and Harling, 1983; Gilger et al., 1999; Gilger and Michau, 2004). Presence of the eosinophilic linear cytoplasmic inclusion bodies in the non-pigmented ciliary epithelial cells and a thick hyaline membrane adjacent to the posterior aspect of the iris are considered pathognomonic for ERU (Cooley et al., 1990; Dubielzig et al., 1997).
Diagnosis and Treatment of ERU
The clinical diagnosis of ERU is made when characteristic clinical signs are accompanied by a history of recurrence or persistence of uveitis (Gilger and Michau, 2004). Accurate identification of the primary inciting cause is of paramount importance in designing an effective therapy. The MAT is a gold standard for the diagnosis of systemic leptospirosis and has also been accepted as an indicator of exposure to leptospiral infection. There is no specific test developed exclusively for the laboratory diagnosis of Leptospira-associated ERU.
Good management practices that reduce the chances of ocular injury and minimize inflammatory stimuli have been proposed as an effective strategy to delay or decrease recurrence of uveitis (Gilger et al., 2000; Gilger and Michau, 2004). Mydriatic cycloplegics, such as topical atropine, are used to reduce discomfort, and anti-inflammatory agents are used to reduce inflammation. Oral or intravenous flunixin meglumine is an effective anti-inflammatory drug for the management of ERU. In severe cases, systemic steroids have been shown to be effective (Gilger et al., 2000; Gilger and Michau, 2004).
Ocular cyclosporine A device implantation and pars plana vitrectomy have been shown to be effective in long term control of the disease. Pars plana vitrectomy has been shown to be effective in reducing recurrent episodes, but surgical complications such as cataract formation have been seen in a large number of operated cases (Fruhauf et al., 1998; Gilger and Michau, 2004).
Pathogenesis of ERU
The pathogenesis of ERU is currently under investigation, and several possible mechanisms have been proposed (Parma et al., 1985, 1987, 1997; Gilger et al., 1999; Deeg et al., 2001, 2002b, 2004, 2006b; Hartskeerl et al., 2004; Verma et al., 2005; Brandes et al., 2007b). Affected eyes exhibit infiltration of lymphocytes, plasma cells and macrophages into the ciliary body and the iris, thereby constituting morphological evidence of breach of immune privilege. The most abundant infiltrating cells are CD4+ T lymphocytes in the anterior uveal tract (Romeike et al., 1998). The T-lymphocyte response in such horses has a Th1 bias based on quantitative reverse transcription-PCR (RT-PCR) analyses showing significantly greater interleukin-2 (IL-2)/gamma interferon- than IL-4-specific mRNA (Gilger et al., 1999). Moreover, peripheral blood leucocytes of chronically uveitic horses do not exhibit a Th1 response, consistent with an independent local response (Gilger et al., 1999). Kalsow et al. (1994) demonstrated infiltration of T and B cells and MHC class II-expressing APCs to the anterior uvea and pineal gland of ERU horses. They also reported that in active leptospiral ERU, unlike quiescent ERU cases, the T- and B-cell areas are highly organized as in germinal centres. They hypothesized that this organizational structure may explain intraocular antibody production and greater antibody reactivity to leptospiral antigen in aqueous humor as compared to serum (Kalsow et al., 1994).
Although direct Leptospira-mediated injury to the eye structure may be a possible pathogenic mechanism of ERU, there is a growing body of evidence that autoimmune responses to ocular tissue components play a significant role in pathogenesis (Parma et al., 1985, 1987, 1997; Gilger et al., 1999; Deeg et al., 2001, 2002a,b, 2004, 2006b; Verma et al., 2005). Parma et al. (1985) demonstrated an antigenic relationship between Leptospira spp. and equine cornea by showing the reactivity of anti-equine cornea antibodies with Leptospira and binding of anti-Leptospira and anti-equine cornea antibodies to equine cornea (Parma et al., 1985). Subsequently, an antigenic relationship between equine lens and leptospires was proposed by the same group (Parma et al., 1987). Electron microscopic studies revealed that the antigenic protein of L. interrogans, which shares epitopes with equine cornea and lens is not exposed on the outer surface of leptospires (Parma et al., 1997), although neither the leptospiral nor the ocular proteins were identified in those studies.
Recently, we identified two leptospiral proteins, LruA and LruB, which are expressed in the eyes of uveitic horses. In uveitic eyes, there is possibly a deviant LruA- and LruB-directed immune response, as LruA- and LruB-specific IgG and IgA antibodies in the eye fluids of uveitic horses were significantly higher than in the companion sera. This observation was in agreement with earlier reports indicating local production of antibodies in the eye and possibly a role in the pathogenesis. We also demonstrated cross-reactivity of LruA- and LruB-specific antiserum with equine eye lenticular and retinal extracts (Verma et al., 2005). Subsequently, we identified the lens proteins cross-reacting with LruA antiserum to be α-crystallin B and vimentin. Similarly, the retinal protein cross-reacting with LruB-antiserum was found to be β-crystallin B2. Purified recombinant human α-crystallin B and vimentin were recognized by LruA-directed antiserum, but not by control pre-immune serum. Recombinant β-crystallin B2 was likewise recognized by LruB-directed antiserum, but not by pre-immune serum. Moreover, uveitic eye fluids contained significantly higher levels of antibodies that recognized α-crystallin B, β-crystallin B2 and vimentin than did normal eye fluids. Antibody levels were found to be significantly elevated in uveitic compared to normal eye fluids. The presence of antibodies recognizing α-crystallin B, vimentin and β-crystallin B2 in uveitic, but not normal eye fluids, strongly suggests a role for these antibodies in Leptospira-associated recurrent uveitis (Verma et al., 2010). In the immune privileged ocular environment, it is likely that the early phase of leptospiral infection involves a non-inflammatory immune response specific for LruA and LruB. Resulting antibodies may interact with cross-reacting proteins in lens and retinal tissues and may therefore initiate a process of desequestration of these ocular antigens, and possibly other components. These studies indicate that LruA and LruB share immuno-relevant epitopes with α-crystallin B, vimentin and β-crystallin B2 of lens and retina, suggesting that cross-reactive antibody interactions with these ocular proteins may contribute to immunopathogenesis of Leptospira-associated recurrent uveitis (Verma et al., 2010).
In another study, LruA and LruB antibodies were found in sera of more than 65% of leptospiral uveitis patients (Verma et al., 2008). Intriguingly, LruA and LruB were recognized by antibodies from Behçet’s and Fuchs uveitis patients, without any evidence of those patients having been exposed to a leptospiral infection. In the same study, we also observed an association in humans between high levels of antibodies recognizing LruA and LruB and the presence of cataract. The high levels of antibodies cross-reactive with LruA and LruB in patients with Fuchs or Behçet’s uveitis, and the strong association of LruA and LruB antibodies with cataract could be due to increased levels of antibodies to the common autoantigens, α-crystallin B, vimentin and β-crystallin B2 in those diseases. Moreover, elevated levels of LruA- and LruB-antibodies in sera of human patients with leptospiral uveitis and reactivity of LruA- and LruB-antiserum with human alpha-crystallin B and β-crystallin B2 suggest a similar phenomenon in human leptospiral uveitis.
In spontaneous ERU cases, a role for retinal S-antigen (S-Ag) in autoimmunity was proposed initially by Hines and Halliwell (1991). Antibodies and T lymphocytes specific for retinal S-Ag and interphotoreceptor retinoid binding protein (IRBP) were observed in the eyes of uveitic horses (Deeg et al., 2001). Moreover, injection of IRBP with complete Freund’s adjuvant induced a disease similar to ERU (Deeg et al., 2002b). The uveitogenic potential of retinal S-Ag was investigated and although T- and B-cell responses specific for the S-Ag were observed in all five immunized horses, only one developed uveitis (Deeg et al., 2002b). The authors concluded that as compared to IRBP, S-Ag is a weak autoantigen in horses. Recently, a novel retinal autoantigen, cellular retinaldehyde-binding protein (c-RALBP), was identified by screening of two dimensional porcine-retinal proteome with autoantibodies from spontaneous ERU sera (Deeg et al., 2006b). Cellular retinaldehyde-binding protein was evaluated for its role in the induction of uveitis in rats and horses and produced uveitis in both the species (Deeg et al., 2006b). Deeg et al. (2006a) found inter- and intramolecular epitope spreading to S-Ag and IRBP derived epitopes in a spontaneous model of ERU and proposed that the shifts in immunoreactivity could explain the alternate phases of active inflammation and quiescence (Deeg et al., 2006a).
In conclusion, leptospiral uveitis is an important uveitic entity of humans and equines. The lack of a specialized diagnostic test makes an accurate estimate of the scope of the problem difficult. Although some progress has been made in understanding the pathogenesis of leptospiral uveitis, a lot of gaps still exist. Unravelling of the molecular mechanisms underlying the disease process will be a key to better management of this ocular pathology.
We thank Claire Adams, Catherine Brissette, Alicia Chenail, Brandon Jutras, S. R. Rathinam and John Timoney for their helpful comments.