- Email: [email protected]
Role of Staphylococcus aureus in Chronic Allergic Conjunctivitis
Role of Staphylococcus aureus in Chronic Allergic Conjunctivitis Stephen]. Tuft, FRACS, I Meenakshi Ramakrishnan, MSC, 2 David V. Seal, MD,3 D. Michael Kemeny, PhD,4 Roger]. Buckley, FRCS 1
A
stud Y was undertaken to test the hypothesiS that Staphylococcus aureus colonization of the lid margins could determine the expression of allergic eye disease in atopic patients. The authors compared lid isolates of S. aureus from 23 adults who had both atopic dermatitis and chronic conjunctivitis and isolates from 14 patients who had atopic dermatitis but who lacked ocular disease. No significant difference was found in either the staphylococcal protein A or hemolytic toxin production by isolates from the two disease groups, and there was no difference between groups in the quantity of serum IgG nor IgE antibodies to staphylococcal ribitol-teichoic acid. In seven patients with chronic allergic conjunctivitis who were challenged with staphylococcal protein A or heat-killed S. aureus, delayed-type hypersensitivity was not enhanced. These results suggest that although S. aureus colonization of the lids is common in atopic patients, neither the pattern of toxin production nor humoral or cell-mediated immunity to S. aureus playa role in the expression of'chronic allergic conjunctivitis. Ophthalmology 1992; 99: 180-184
The skin of patients with atopic dermatitis is frequently colonized with Staphylococcus aureus, which may produce recurrent superficial infections. 1-9 There is an interdependence between the degree of staphylococcal colonization and the severity of dermatitis, in that the density of S. aureus skin carriage is reduced after treatment with corticosteroids9 and the dermatitis will usually improve when treated with antibiotics. 2,5-7 Chronic allergic eye disease most frequently occurs in highly atopic males. IO- 14 Affected individuals typically have eczema of the face and eyelids, which is associated with thickening of the lid margins, blepharitis, and conjunctival scarring. In some patients, there may be assoOriginally received: June 10, 1991. Revision accepted: October 7, 1991. I Moorfields Eye Hospital, London. 2 Institute of Ophthalmology, London. 3 Bacteriology Laboratory, Wolfson Centre, Glasgow. 4 Department of Allergy and Applied Respiratory Disorders, Division of Medicine, Guy's Hospital, London. Presented in part at the 22nd Ocular Microbiology and Immunology Group Meeting, Las Vegas, 1988. Correspondence to Stephen J. Tuft, FRACS, Moorfields Eye Hospital, City Road, London EClV 2PD.
180
ciated corneal di~, when the condition is termed atopic keratoconjunctivitis. 10 S. aureus can be isolated from the lid margins of the majority of patients with atopic keratoconjunctivitis, and, although delayed-type hypersensitivity to components of the S. aureus cell wall may contribute to marginal keratitis 16 and ulcerative blepharitis, 17-19 a mechanism whereby bacterial colonization of the lid margins could exacerbate chronic allergic conjunctivitis has not been examined. To determine if characteristics of the colonizing S. aureus could influence the expression of ocular disease in atopic patients we conducted a case-control study of serum and bacterial isolates from 23 atopic patients with chronic conjunctivitis and 14 patients with atopic dermatitis who lacked ocular disease. To determine if the properties of S. aureus colonizing non-atopes differed from those colonizing atopes, we also compared isolates from patients with blepharitis, suppurative keratitis, and non-atopic controls without lid disease. We have investigated known indices of pathogenicity of S. aureus that mediate inflammation. The protein A and hemolytic toxin production by each S. aureus isolate was investigated. Serum IgG antibodies to ribitol-teichoic acid of the S. aureus cell wall and 19B antibody binding to the ribitol-teichoic acid of S. aureus and Staphylococcus epidermidis were measured. Finally, to estimate in vivo
Tuft et al . Staphylococcus aureus in Chronic Allergic Conjunctivitis reactivity, we have measured delayed-type hypersensitivity to protein A and killed S. aureus cells in 7 patients with atopic keratoconjunctivitis.
Patients and Methods We investigated S. aureus isolates from 23 atopic patients who had had conjunctivitis for at least 2 years, who were receiving treatment for atopic dermatitis, had large papillae (> 1 mm) over the upper tarsal conjunctiva, and had a positive skin prick test to common airborne allergens. 15 This group was classified as having chronic allergic conjunctivitis, and no attempt was made to distinguish a supposed staphylococcal component to their ocular disease. Samples were also collected from 14 patients with atopic dermatitis who had a positive skin prick test but no history of chronic ocular irritation and who were without signs of conjunctival inflammation. A comparison was also made with S. aureus isolates from 3 groups of patients who had no history of atopic disease, comprising 12 with suppurative corneal ulcers, 17 with blepharitis, and 12 volunteers who did not have conjunctival or lid margin disease. These groups were not matched for age or sex.
Total IgE Estimation Total IgE in serum and tears was measured by a micropaper radioimmunosorbent test. 20 The normal range of serum IgE was considered to be 2 to 100 IUIml.
Bacteriologic Sampling Lids of both eyes were sampled by wiping once along the upper and lower margins with a cotton wool swab moistened with tryptic digest broth. These were plated onto 7% horse blood agar and incubated aerobically at 37° C for 18 hours, at which time the numbers of colonies of staphylococci were estimated on a scale of 0 to 3, the latter representing confluent growth. Colonies of S. aureus were identified by the tube coagulase test.
Protein A Estimation Protein A was quantified by nonspecific fluorescein isothiocyanate-labeled IgG binding by the Fc terminal to
protein A-containing S. aureus cell walls. 21 Each culture was grown in nutrient broth for 18 hours at 37° C. One drop was mixed with a 1I 100 dilution of the fluoresceinlabeled IgG and examined under ultraviolet microscopy. Positive cultures showed a fluorescent ring around each cell, which was graded from 1 to 3 for its intensity by 3 independent microscopists and the average recorded. S. aureus Wood 46 22 was used as a negative control as it lacks protein A in its cell wall.
Hemolytic Toxin Production Estimation of a, {3, and 0 toxin production was carried out as described elsewhere. 23 This involved confluent culture for 20 hours at 37° C in 4% CO2 on a base agar medium (CM271, Oxoid) that contained sodium chloride (5 gfl) but which lacked blood and glucose. Each plate was flooded with 10 ml phosphate-buffered saline, which was left in situ for 90 minutes to elute the toxins from the agar. Each elution was tested for hemolysin production against 4% washed rabbit, sheep, human, and horse red blood cells that had been preserved in Alsever solution.
IgG Antibody to Ribitol-teichoic Acid Serum samples were tested for IgG antibodies to ribitolteichoic acid extracted from the cell wall of S. aureus strain Wood 46 by an enzyme-linked immunosorbent assay (ELISA) method. 24 Ribitol-teichoic acid was diluted in alkaline carbonate buffer and coated onto micro-ELISA plates. After washing, dilutions of serum from 1: 10 to 1:80 in phosphate-buffered saline (PBS}/Tween 20 (0.05%) were added and incubated at 22° C for 2 hours. After washing, peroxidase-labeled rabbit anti-human IgG was added to each well at a concentration of 1: 500 (Dako Ltd, High Wicombe, UK). The plate was incubated, washed again, and orthophenylene diamine substrate added. Values in control subjects using the same reagents have been estimated previously.24
IgE Antibodies to Staphylococcal Ribitol-teichoic Acid The presence of IgE antibodies to ribitol-teichoic acid of S. aureus Wood 46 or S. epidermidis were assayed by the radioallergosorbent test using cyanogen bromide-activated sepharose 4B (Pharmacia LKB, Milton Keynes, UK). This
Table 1. Estimation of Protein A in the Cell Walls of the Staphylococcus aureus Isolated from the Lids of Control Volunteers and Four Patient Groups Estimate of Protein A in S. aureua Cell Walls (Percentage in each Group) Patient Group Controls Chronic allergic conjunctivitis Atopic dermatitis Blepharitis Suppurative keratitis
No.
+++
12
0(%)
23
8 8
14 17 12
21
8
++
+
75(%) 35 57 32 50
25 (%) 35 21 37 25
Absent
0(%) 22
14 10
17
181
Ophthalmology
Volume 99, Number 2, February 1992
Table 2. Toxin Production by Staphylococcus aureus Isolated from the Lids of Control Patients and Four Patient Groups Production of S. aureus Toxins (Percentage of Isolates) Patient Group Controls Chronic allergic conjunctivitis Atopic dermatitis Blepharitis Suppurative keratitis
No.
a Toxin Alone
a+(3 Toxins
12 23
0(%) 13
92(%)
14
14
17
12
12
8
was coated with 1 mg of each extract per 0.5 g of sepharose and prepared as recommended by the manufacturer. The teichoic-acid-coated sepharose was incubated with serum from the study group (0.5 mg sepharose/0.5 ml PBS + 0.5% horse serum/50 ttl test serum sample) overnight at 22 0 C with mixing. The particles were washed 3 times with 1.5 ml PBS/Tween 20 and incubated with 50 ttl of iodine 125 radiolabeled anti-lgE, prepared as described previously,20 overnight at 22 0 C with mixing. The particles were washed as before and counted for 1 minute in a 1280 ultragammacounter (Pharmacia LKB, Milton Keynes, UK).
Investigation of Delayed,type Hypersensitivity After obtaining informed consent, 7 patients with chronic allergic conjunctivitis were challenged with 2.5 ng of purified protein A (Pharmacia Ltd, Milton Keynes, UK) in 0.05 ml of physiologic saline and 0.1 ml of heat-killed S. aureus (5 X 108 cells/ml) preserved in phenol saline (Bencard Ltd, UK), injected at separate intradermal sites on both forearms. The degree of swelling and induration was measured after 48 hours. Values in control subjects and blepharitic patients had been assessed previously using the same reagents. 24 As the patient groups were not controlled for age or sex, differences in serum or tear IgE were assessed using multiple linear logistic regression analysis. Comparisons between grouped data were made using chi-square analysis. Statistical significance was reached at P < 0.05.
Results The mean IgE levels in the serum of atopic patients with chronic allergic conjunctivitis (mean age, 35.5 years;
a+(3+o Toxins 8(%) 30
57 43 41 25
43 47 67
range, 14.9 to 60.3 years) and atopic patients without ocular signs (mean age, 32.3 years) was 17,075 (standard deviation, 33,774) and 37,189 (standard deviation, 59,535) IU/ml, respectively. The mean tear 19E levels were 850 (standard deviation 2281) and 609 (standard deviation 1490) IU /ml. These differences were not statistically significant. Fifty-two percent of the 23 patients with chronic allergic conjunctivitis and 57% of the eczematous patients without ocular disease colonized with S. aureus yielded a confluent growth on direct agar plating equivalent to 1 X 106 colonies per milliliter solubilized swab. The results of estimating the quantity of protein A in S. aureus cell walls from isolates from control subjects and the 4 patient groups are presented in Table 1. In 21 % of the isolates, protein A was absent altogether, and there is no statistical difference between the groups. As expected, all S. aureus isolates produced a-lysin, often in conjunction with ~-lysin. 25 There is a trend toward o-lysin production in pathogenic isolates, but this was not statistically significant (Table 2). Sixteen of the 23 patients with chronic allergic conjunctivitis had a keratopathy (atopic keratoconjunctivitis), but none of the tests distinguished this group. The quantity of IgG antibodies to S. aureus ribitolteichoic acid was similar in serum from eczematous patients with or without chronic allergic conjunctivitis (Table 3), and there was no evidence of specific serum 19E binding against ribitol-teichoic acid of either S. aureus or S. epidermidis in atopic patients with or without ocular disease (Table 4). Delayed-type hypersensitivity to protein A or heat-killed S. aureus was not observed in any of the seven patients tested, although two produced a flare; no patient developed delayed swelling after 48 hours.
Table 3. Detection of IgG Antibody to Staphylococcus aureus Ribitol-teichoic Acid in Serum of Atopic Patients with or without Conjunctivitis Titers (Percentage of Isolates) Patient Group Chronic allergic conjunctivitis Atopic dermatitis
182
D
1/10
1/40
d/8D
23
0(%) 0
13(%) 0
35(%)
14
52(%) 86
No.
14
Tuft et al . Staphylococcus aureus in Chronic Allergic Conjunctivitis
Table 4. Detection of IgE Antibodies to Staphylococcal Ribitol-teichoic Acid in Serum of Atopic Patients with or without Conjunctivitis Counts per Minute (1 SD) No. Chronic allergic conjunctivitis Atopic dermatitis Non-atopic control
23
14 12
Discussion The mechanism of S. aureus colonization in atopy and its relationship to high IgE titers is unclear, but once colonization is established, S. aureus probably exacerbates dermatitis via a variety of enzymes and toxic products introduced into the dermis by scratching. Protein A is a component of the S. aureus cell wall, which can combine nonspecifically with the Fe portion of IgG to activate complement26 or produce histamine release when bound to the surface of basophils. 27 It can also bind with the Fe and F(ab) portions ofIgE. 28 Protein A produces an acute inflammatory response if applied to the skin after the stratum corneum has been removed,29 and induces a wheal and flare reaction similar to immediate hypersensitivity when injected intracutaneously.30 We found no statistically significant difference in protein A expression between any of the disease groups. Hemolytic toxin production by S. aureus is not believed to contribute to blepharitis,23 and we observed a similar pattern of hemolysin production by isolates from both atopic groups. However, an apparent increase in a-lysin production by S. aureus in clinically infected lesions has been previously noted,31 and we observed a-lysin production more frequently from isolates from suppurative corneal infections (67%) than from either the lids of atopes (35%) or control subjects (8%). Staphylococcal toxins have been shown to stimulate T cells, whose major histocompatibility complex receptor possess v-beta receptors; this could theoretically lead to local delayed hypersensitivity enhancement in predisposed individuals. 32 The role of immunodeficiency in atopic dermatitis is speculative. Although 40% of patients with chronic blepharitis have a delayed hypersensitivity reaction to protein A,24 none of the 7 patients with chronic allergic conjunctivitis were enhanced when challenged with protein A and killed S. aureus cells. This may be the result of a reduced delayed-type hypersensitivity response in atopic patients. The reason for the poor delayed-type hypersensitivity response in atopic patients is uncertain, but the granulocytes of atopic individuals are known to have a defect for engulfing S. aureus cells,33 and patients with atopic dermatitis may also have a significant reduction in their total numbers of T cells. 34 It has also been reported that polymorphonuclear leukocyte chemotaxis is reduced in patients with atopy during exacerbations of disease. 5,35 In an extreme form of atopic dermatitis, the hyperimmunoglobulin E syndrome, children have recurrent
Staphylococcus aureus
Staphylococcus epidermidis
527 (125) 525 (89) 533 (215)
1665 (559) 1686 (403) 1699 (336)
boils, sinopulmonary infections, and 19E levels that are as much as 10 times greater than the upper limit of normal. 36 These patients may have a mean serum anti-staphylococcal IgE level that represented up to 21 % of their total 19E, although it is unclear whether this is a direct result of staphylococcal colonization. 37 The elevation of serum anti-staphylococcal IgE is less marked in patients with atopic dermatitis who do not have a history of invasive staphylococcal infections. 38 Anti-staphylococcal ribitol-teichoic acid antibodies were not raised in the serum of our patients. We were unable to obtain a sufficient volume of tears for our assays, but expect that if there is local production of these antibodies in the eye, the measurement of tear concentrations may give a better correlation with disease. S. aureus is found in similar quantities on the lids of atopic individuals with or without chronic allergic conjunctivitis. ls We have previously reported a carriage rate of67% on the lids of patients with atopic keratoconjunctivitis, which compares with an incidence of between 5% and 20% in non-atopes.39 Although S. aureus colonization may exacerbate the pruritus of atopic dermatitis, our findings suggest that S. aureus is a passenger rather than a pathogen on the lids of patients with chronic allergic conjunctivitis. Thus, the presence of S. aureus may exacerbate lid disease if the eye with allergic conjunctivitis is rubbed, but we were unable to detect differences between isolates of S. aureus from the two different atopic groups that could explain the development of conjunctivitis in only some atopic individuals. Finally, as heavy S. aureus colonization may be present in the absence of clinical disease, the risk of infection should be considered in all atopes before surgery.
References 1. Andersen EK, Heilesen B. On the occurrence and typing of Staphylococcus pyogenes in weeping eczema (and the role played by these bacteria in the pathogenesis and course of eczema). Acta Derm Venereol1951; 31:679-703. 2. Leyden JJ, Marples RR, Kligman AM. Staphylococcus aureus in the lesions of atopic dermatitis. Br J Dermatol 1974; 90:525-30. 3. Aly R, Maibach HI, Shinefield HR. Microbial flora of atopic dermatitis. Arch Dermatol 1977; 113:780-2. 4. Bibel DJ, Greenberg JH, Cook JL. Staphylococcus aureus and the microbial ecology of atopic dermatitis. Can J Microbiol 1977; 23: 1062-8.
183
Ophthalmology
Volume 99, Number 2, February 1992
5. Hanifin JM, Rogge JL. Staphylococcal infections in patients with atopic dermatitis. Arch Dermatol 1977; 113: 1383-6. 6. Wachs GN, Maibach HI. Co-operative double-blind trial of an antibiotic/corticoid combination in impetiginized atopic dermatitis. Br J Dermatol 1976; 95:323-8. 7. Dahl MY. Staphylococcus aureus and atopic dermatitis. Arch Dermatol1983; 119:840-6. 8. Hauser C, Wuethrich B, Matter L, et al. Staphylococcus aureus skin colonization in atopic dermatitis patients. Dermatologica 1985; 170:35-9. 9. White MI, Noble WC. Consequences of colonization and infection by Staphylococcus aureus in atopic dermatitis. ain Exp Dermatol 1986; 11:34-40. 10. Hogan MJ. Atopic keratoconjunctivitis. Am J Ophthalmol 1953; 36:937-47. 11. Jay JL. Clinical features and diagnosis of adult atopic keratoconjunctivitis and the effect of treatment with sodium cromoglycate. Br J Ophthalmol 1981; 65:335-40. 12. Buckley RJ. Atopic disease of the cornea. In: Cavanagh DH, ed. The Cornea: Transactions ofthe World Congress on the Cornea III. New York: Raven Press, 1988; chap. 79. 13. Tuft SJ, Dart JKG, Kemeny M. Limbal vernal keratoconjunctivitis: clinical characteristics and immunoglobulin E expression compared with palpebral vernal. Eye 1989; 3: 420-7. 14. Foster CS, Calonge M. Atopic keratoconjunctivitis. Ophthalmology 1990; 97:992-1000. 15. Tuft SJ, Kemeny DM, Dart JKG, Buckley RJ. Clinical features of atopic keratoconjunctivitis. Ophthalmology 1991; 98:150-8. 16. Mondino BJ. Inflammatory diseases of the peripheral cornea. Ophthalmology 1988; 95:463-72. 17. Mondino BJ, Caster AI, Dethlefs B. A rabbit model of staphylococcal blepharitis. Arch Ophthalmol1987; 105:40912. 18. Mondino BJ, Kowalski RP. Phlyctenulae and catarrhal infiltrates: occurrence in rabbits immunized with staphylococcal cell walls. Arch Ophthalmol 1982; 100:1968-71. 19. Mondino BJ, Dethlefs B. Occurrence ofphlyctenules after immunization with ribitol teichoic acid of Staphylococcus aureus. Arch Ophthalmol 1984; 102:461-3. 20. Kemeny DM, Richards D. Increased speed and sensitivity, and reduced sample size of a micro-radioallergosorbent test (MRAST). J Immunol Methods 1988; 108:105-13. 21. Lachica RYF, Genigeorgis CA, Hoeprich PD. Occurrence of protein A in Staphylococcus aureus and closely related Staphylococcus species. J Clin Microbiol 1979; 10:752-3. 22. Haukenes G. Serological typing of Staphylococcus aureus. 6. Antibodies of immune sera to strains cowan I, cowan II, cowan III, Wood 46, 670, 830, 1015,5687, and 6376. Acta Path Microbiol Scand 1964; 61:415-26.
184
23. Seal D, Ficker L, Ramakrishnan M, Wright P. Role of staphylococcal toxin production in blepharitis. Ophthalmology 1990; 97:1684-8. 24. Ficker L, Ramakrishnan M, Seal D, Wright P. Role of cellmediated immunity to staphylococci in blepharitis. Am J Ophthalmol 1991; 111:473-9. 25. Freer JH, Arbuthnott JP. Toxins of Staphylococcus aureus. Pharmacol Ther 1983; 19:55-106. 26. Sjoquist J, Stalenheim G. Protein A from Staphylococcus aureus. IX. Complement-fixing activity of protein A-IgG complexes. J Immunol 1969; 103:467-73. 27. Petersson BA, Stalenheim G. Induction of histamine release and desensitization in human leukocytes. Effect of anti-IgE or protein A from Staphylococcus aureus. Scand J Immunol 1975; 4:103-12. 28. Langone JJ. Protein A of Staphylococcus aureus and related immunoglobulin receptors produced by streptococci and pneumonococci. Adv Immunol 1982; 32:157-252. 29. White MI, Noble WC. Skin response to protein A. Proc R Soc Edin 1980; 79B:43-6. 30. White MI, Noble WC. The cutaneous reaction to staphylococcal protein A in normal subjects and patients with atopic dermatitis or psoriasis. Br J Dermatol 1985; 113: 179-83. 31 . Elek SD, Levy E. Distribution ofhaemolysins in pathogenic and non-pathogenic staphylococci. JPathol Bacteriol 1950; 62:541-54. 32. Kappler J, Kotzin B, Herron L, et al. Y-~-specific stimulation of human T cells by staphylococcal toxins. Science 1989; 244:811-3. 33. Matusiewicz R, Grzybowski A. The ability of granulocytes to engulflatex particles and Staphylococcus aureus in healthy children of parents with atopic asthma. Ann Allergy 1989; 62:445-8. 34. Carapeto FJ, Winkelmann RK, Jordon RE. T and B lymphocytes in contact and atopic dermatitis. Arch Dermatol 1976; 112:1095-100. 35. Hill HR, Quie PG. Raised serum-IgE levels and defective neutrophil chemotaxis in three children with eczema and recurrent bacterial infection. Lancet 1974; 1:183-7. 36. Buckley RH, Wray BB, Belmaker EZ. Extreme hyperimmunoglobulinemia E and undue susceptibility to infection. Pediatrics 1972; 49:59-70. 37. Schopfer K, Baerlocher K, Price P, et al. Staphylococcal IgE antibodies, hyperimmunoglobulinemia E and Staphylococcus aureus infections. N Engl J Med 1979; 300:835-8. 38. Abramson JS, Dahl MY, Walsh G, et al. Antistaphylococca1 IgE in patients with atopic dermatitis. J Am Acad Dermatol 1982; 7:105-10. 39. Badiani D, Bron A, Elkington A, Lea SH. Use of a novel transport method for quantification of the normal flora of theextemal eye. Microbiol Ecol Health Dis 1988; 1:57-9.
A
stud Y was undertaken to test the hypothesiS that Staphylococcus aureus colonization of the lid margins could determine the expression of allergic eye disease in atopic patients. The authors compared lid isolates of S. aureus from 23 adults who had both atopic dermatitis and chronic conjunctivitis and isolates from 14 patients who had atopic dermatitis but who lacked ocular disease. No significant difference was found in either the staphylococcal protein A or hemolytic toxin production by isolates from the two disease groups, and there was no difference between groups in the quantity of serum IgG nor IgE antibodies to staphylococcal ribitol-teichoic acid. In seven patients with chronic allergic conjunctivitis who were challenged with staphylococcal protein A or heat-killed S. aureus, delayed-type hypersensitivity was not enhanced. These results suggest that although S. aureus colonization of the lids is common in atopic patients, neither the pattern of toxin production nor humoral or cell-mediated immunity to S. aureus playa role in the expression of'chronic allergic conjunctivitis. Ophthalmology 1992; 99: 180-184
The skin of patients with atopic dermatitis is frequently colonized with Staphylococcus aureus, which may produce recurrent superficial infections. 1-9 There is an interdependence between the degree of staphylococcal colonization and the severity of dermatitis, in that the density of S. aureus skin carriage is reduced after treatment with corticosteroids9 and the dermatitis will usually improve when treated with antibiotics. 2,5-7 Chronic allergic eye disease most frequently occurs in highly atopic males. IO- 14 Affected individuals typically have eczema of the face and eyelids, which is associated with thickening of the lid margins, blepharitis, and conjunctival scarring. In some patients, there may be assoOriginally received: June 10, 1991. Revision accepted: October 7, 1991. I Moorfields Eye Hospital, London. 2 Institute of Ophthalmology, London. 3 Bacteriology Laboratory, Wolfson Centre, Glasgow. 4 Department of Allergy and Applied Respiratory Disorders, Division of Medicine, Guy's Hospital, London. Presented in part at the 22nd Ocular Microbiology and Immunology Group Meeting, Las Vegas, 1988. Correspondence to Stephen J. Tuft, FRACS, Moorfields Eye Hospital, City Road, London EClV 2PD.
180
ciated corneal di~, when the condition is termed atopic keratoconjunctivitis. 10 S. aureus can be isolated from the lid margins of the majority of patients with atopic keratoconjunctivitis, and, although delayed-type hypersensitivity to components of the S. aureus cell wall may contribute to marginal keratitis 16 and ulcerative blepharitis, 17-19 a mechanism whereby bacterial colonization of the lid margins could exacerbate chronic allergic conjunctivitis has not been examined. To determine if characteristics of the colonizing S. aureus could influence the expression of ocular disease in atopic patients we conducted a case-control study of serum and bacterial isolates from 23 atopic patients with chronic conjunctivitis and 14 patients with atopic dermatitis who lacked ocular disease. To determine if the properties of S. aureus colonizing non-atopes differed from those colonizing atopes, we also compared isolates from patients with blepharitis, suppurative keratitis, and non-atopic controls without lid disease. We have investigated known indices of pathogenicity of S. aureus that mediate inflammation. The protein A and hemolytic toxin production by each S. aureus isolate was investigated. Serum IgG antibodies to ribitol-teichoic acid of the S. aureus cell wall and 19B antibody binding to the ribitol-teichoic acid of S. aureus and Staphylococcus epidermidis were measured. Finally, to estimate in vivo
Tuft et al . Staphylococcus aureus in Chronic Allergic Conjunctivitis reactivity, we have measured delayed-type hypersensitivity to protein A and killed S. aureus cells in 7 patients with atopic keratoconjunctivitis.
Patients and Methods We investigated S. aureus isolates from 23 atopic patients who had had conjunctivitis for at least 2 years, who were receiving treatment for atopic dermatitis, had large papillae (> 1 mm) over the upper tarsal conjunctiva, and had a positive skin prick test to common airborne allergens. 15 This group was classified as having chronic allergic conjunctivitis, and no attempt was made to distinguish a supposed staphylococcal component to their ocular disease. Samples were also collected from 14 patients with atopic dermatitis who had a positive skin prick test but no history of chronic ocular irritation and who were without signs of conjunctival inflammation. A comparison was also made with S. aureus isolates from 3 groups of patients who had no history of atopic disease, comprising 12 with suppurative corneal ulcers, 17 with blepharitis, and 12 volunteers who did not have conjunctival or lid margin disease. These groups were not matched for age or sex.
Total IgE Estimation Total IgE in serum and tears was measured by a micropaper radioimmunosorbent test. 20 The normal range of serum IgE was considered to be 2 to 100 IUIml.
Bacteriologic Sampling Lids of both eyes were sampled by wiping once along the upper and lower margins with a cotton wool swab moistened with tryptic digest broth. These were plated onto 7% horse blood agar and incubated aerobically at 37° C for 18 hours, at which time the numbers of colonies of staphylococci were estimated on a scale of 0 to 3, the latter representing confluent growth. Colonies of S. aureus were identified by the tube coagulase test.
Protein A Estimation Protein A was quantified by nonspecific fluorescein isothiocyanate-labeled IgG binding by the Fc terminal to
protein A-containing S. aureus cell walls. 21 Each culture was grown in nutrient broth for 18 hours at 37° C. One drop was mixed with a 1I 100 dilution of the fluoresceinlabeled IgG and examined under ultraviolet microscopy. Positive cultures showed a fluorescent ring around each cell, which was graded from 1 to 3 for its intensity by 3 independent microscopists and the average recorded. S. aureus Wood 46 22 was used as a negative control as it lacks protein A in its cell wall.
Hemolytic Toxin Production Estimation of a, {3, and 0 toxin production was carried out as described elsewhere. 23 This involved confluent culture for 20 hours at 37° C in 4% CO2 on a base agar medium (CM271, Oxoid) that contained sodium chloride (5 gfl) but which lacked blood and glucose. Each plate was flooded with 10 ml phosphate-buffered saline, which was left in situ for 90 minutes to elute the toxins from the agar. Each elution was tested for hemolysin production against 4% washed rabbit, sheep, human, and horse red blood cells that had been preserved in Alsever solution.
IgG Antibody to Ribitol-teichoic Acid Serum samples were tested for IgG antibodies to ribitolteichoic acid extracted from the cell wall of S. aureus strain Wood 46 by an enzyme-linked immunosorbent assay (ELISA) method. 24 Ribitol-teichoic acid was diluted in alkaline carbonate buffer and coated onto micro-ELISA plates. After washing, dilutions of serum from 1: 10 to 1:80 in phosphate-buffered saline (PBS}/Tween 20 (0.05%) were added and incubated at 22° C for 2 hours. After washing, peroxidase-labeled rabbit anti-human IgG was added to each well at a concentration of 1: 500 (Dako Ltd, High Wicombe, UK). The plate was incubated, washed again, and orthophenylene diamine substrate added. Values in control subjects using the same reagents have been estimated previously.24
IgE Antibodies to Staphylococcal Ribitol-teichoic Acid The presence of IgE antibodies to ribitol-teichoic acid of S. aureus Wood 46 or S. epidermidis were assayed by the radioallergosorbent test using cyanogen bromide-activated sepharose 4B (Pharmacia LKB, Milton Keynes, UK). This
Table 1. Estimation of Protein A in the Cell Walls of the Staphylococcus aureus Isolated from the Lids of Control Volunteers and Four Patient Groups Estimate of Protein A in S. aureua Cell Walls (Percentage in each Group) Patient Group Controls Chronic allergic conjunctivitis Atopic dermatitis Blepharitis Suppurative keratitis
No.
+++
12
0(%)
23
8 8
14 17 12
21
8
++
+
75(%) 35 57 32 50
25 (%) 35 21 37 25
Absent
0(%) 22
14 10
17
181
Ophthalmology
Volume 99, Number 2, February 1992
Table 2. Toxin Production by Staphylococcus aureus Isolated from the Lids of Control Patients and Four Patient Groups Production of S. aureus Toxins (Percentage of Isolates) Patient Group Controls Chronic allergic conjunctivitis Atopic dermatitis Blepharitis Suppurative keratitis
No.
a Toxin Alone
a+(3 Toxins
12 23
0(%) 13
92(%)
14
14
17
12
12
8
was coated with 1 mg of each extract per 0.5 g of sepharose and prepared as recommended by the manufacturer. The teichoic-acid-coated sepharose was incubated with serum from the study group (0.5 mg sepharose/0.5 ml PBS + 0.5% horse serum/50 ttl test serum sample) overnight at 22 0 C with mixing. The particles were washed 3 times with 1.5 ml PBS/Tween 20 and incubated with 50 ttl of iodine 125 radiolabeled anti-lgE, prepared as described previously,20 overnight at 22 0 C with mixing. The particles were washed as before and counted for 1 minute in a 1280 ultragammacounter (Pharmacia LKB, Milton Keynes, UK).
Investigation of Delayed,type Hypersensitivity After obtaining informed consent, 7 patients with chronic allergic conjunctivitis were challenged with 2.5 ng of purified protein A (Pharmacia Ltd, Milton Keynes, UK) in 0.05 ml of physiologic saline and 0.1 ml of heat-killed S. aureus (5 X 108 cells/ml) preserved in phenol saline (Bencard Ltd, UK), injected at separate intradermal sites on both forearms. The degree of swelling and induration was measured after 48 hours. Values in control subjects and blepharitic patients had been assessed previously using the same reagents. 24 As the patient groups were not controlled for age or sex, differences in serum or tear IgE were assessed using multiple linear logistic regression analysis. Comparisons between grouped data were made using chi-square analysis. Statistical significance was reached at P < 0.05.
Results The mean IgE levels in the serum of atopic patients with chronic allergic conjunctivitis (mean age, 35.5 years;
a+(3+o Toxins 8(%) 30
57 43 41 25
43 47 67
range, 14.9 to 60.3 years) and atopic patients without ocular signs (mean age, 32.3 years) was 17,075 (standard deviation, 33,774) and 37,189 (standard deviation, 59,535) IU/ml, respectively. The mean tear 19E levels were 850 (standard deviation 2281) and 609 (standard deviation 1490) IU /ml. These differences were not statistically significant. Fifty-two percent of the 23 patients with chronic allergic conjunctivitis and 57% of the eczematous patients without ocular disease colonized with S. aureus yielded a confluent growth on direct agar plating equivalent to 1 X 106 colonies per milliliter solubilized swab. The results of estimating the quantity of protein A in S. aureus cell walls from isolates from control subjects and the 4 patient groups are presented in Table 1. In 21 % of the isolates, protein A was absent altogether, and there is no statistical difference between the groups. As expected, all S. aureus isolates produced a-lysin, often in conjunction with ~-lysin. 25 There is a trend toward o-lysin production in pathogenic isolates, but this was not statistically significant (Table 2). Sixteen of the 23 patients with chronic allergic conjunctivitis had a keratopathy (atopic keratoconjunctivitis), but none of the tests distinguished this group. The quantity of IgG antibodies to S. aureus ribitolteichoic acid was similar in serum from eczematous patients with or without chronic allergic conjunctivitis (Table 3), and there was no evidence of specific serum 19E binding against ribitol-teichoic acid of either S. aureus or S. epidermidis in atopic patients with or without ocular disease (Table 4). Delayed-type hypersensitivity to protein A or heat-killed S. aureus was not observed in any of the seven patients tested, although two produced a flare; no patient developed delayed swelling after 48 hours.
Table 3. Detection of IgG Antibody to Staphylococcus aureus Ribitol-teichoic Acid in Serum of Atopic Patients with or without Conjunctivitis Titers (Percentage of Isolates) Patient Group Chronic allergic conjunctivitis Atopic dermatitis
182
D
1/10
1/40
d/8D
23
0(%) 0
13(%) 0
35(%)
14
52(%) 86
No.
14
Tuft et al . Staphylococcus aureus in Chronic Allergic Conjunctivitis
Table 4. Detection of IgE Antibodies to Staphylococcal Ribitol-teichoic Acid in Serum of Atopic Patients with or without Conjunctivitis Counts per Minute (1 SD) No. Chronic allergic conjunctivitis Atopic dermatitis Non-atopic control
23
14 12
Discussion The mechanism of S. aureus colonization in atopy and its relationship to high IgE titers is unclear, but once colonization is established, S. aureus probably exacerbates dermatitis via a variety of enzymes and toxic products introduced into the dermis by scratching. Protein A is a component of the S. aureus cell wall, which can combine nonspecifically with the Fe portion of IgG to activate complement26 or produce histamine release when bound to the surface of basophils. 27 It can also bind with the Fe and F(ab) portions ofIgE. 28 Protein A produces an acute inflammatory response if applied to the skin after the stratum corneum has been removed,29 and induces a wheal and flare reaction similar to immediate hypersensitivity when injected intracutaneously.30 We found no statistically significant difference in protein A expression between any of the disease groups. Hemolytic toxin production by S. aureus is not believed to contribute to blepharitis,23 and we observed a similar pattern of hemolysin production by isolates from both atopic groups. However, an apparent increase in a-lysin production by S. aureus in clinically infected lesions has been previously noted,31 and we observed a-lysin production more frequently from isolates from suppurative corneal infections (67%) than from either the lids of atopes (35%) or control subjects (8%). Staphylococcal toxins have been shown to stimulate T cells, whose major histocompatibility complex receptor possess v-beta receptors; this could theoretically lead to local delayed hypersensitivity enhancement in predisposed individuals. 32 The role of immunodeficiency in atopic dermatitis is speculative. Although 40% of patients with chronic blepharitis have a delayed hypersensitivity reaction to protein A,24 none of the 7 patients with chronic allergic conjunctivitis were enhanced when challenged with protein A and killed S. aureus cells. This may be the result of a reduced delayed-type hypersensitivity response in atopic patients. The reason for the poor delayed-type hypersensitivity response in atopic patients is uncertain, but the granulocytes of atopic individuals are known to have a defect for engulfing S. aureus cells,33 and patients with atopic dermatitis may also have a significant reduction in their total numbers of T cells. 34 It has also been reported that polymorphonuclear leukocyte chemotaxis is reduced in patients with atopy during exacerbations of disease. 5,35 In an extreme form of atopic dermatitis, the hyperimmunoglobulin E syndrome, children have recurrent
Staphylococcus aureus
Staphylococcus epidermidis
527 (125) 525 (89) 533 (215)
1665 (559) 1686 (403) 1699 (336)
boils, sinopulmonary infections, and 19E levels that are as much as 10 times greater than the upper limit of normal. 36 These patients may have a mean serum anti-staphylococcal IgE level that represented up to 21 % of their total 19E, although it is unclear whether this is a direct result of staphylococcal colonization. 37 The elevation of serum anti-staphylococcal IgE is less marked in patients with atopic dermatitis who do not have a history of invasive staphylococcal infections. 38 Anti-staphylococcal ribitol-teichoic acid antibodies were not raised in the serum of our patients. We were unable to obtain a sufficient volume of tears for our assays, but expect that if there is local production of these antibodies in the eye, the measurement of tear concentrations may give a better correlation with disease. S. aureus is found in similar quantities on the lids of atopic individuals with or without chronic allergic conjunctivitis. ls We have previously reported a carriage rate of67% on the lids of patients with atopic keratoconjunctivitis, which compares with an incidence of between 5% and 20% in non-atopes.39 Although S. aureus colonization may exacerbate the pruritus of atopic dermatitis, our findings suggest that S. aureus is a passenger rather than a pathogen on the lids of patients with chronic allergic conjunctivitis. Thus, the presence of S. aureus may exacerbate lid disease if the eye with allergic conjunctivitis is rubbed, but we were unable to detect differences between isolates of S. aureus from the two different atopic groups that could explain the development of conjunctivitis in only some atopic individuals. Finally, as heavy S. aureus colonization may be present in the absence of clinical disease, the risk of infection should be considered in all atopes before surgery.
References 1. Andersen EK, Heilesen B. On the occurrence and typing of Staphylococcus pyogenes in weeping eczema (and the role played by these bacteria in the pathogenesis and course of eczema). Acta Derm Venereol1951; 31:679-703. 2. Leyden JJ, Marples RR, Kligman AM. Staphylococcus aureus in the lesions of atopic dermatitis. Br J Dermatol 1974; 90:525-30. 3. Aly R, Maibach HI, Shinefield HR. Microbial flora of atopic dermatitis. Arch Dermatol 1977; 113:780-2. 4. Bibel DJ, Greenberg JH, Cook JL. Staphylococcus aureus and the microbial ecology of atopic dermatitis. Can J Microbiol 1977; 23: 1062-8.
183
Ophthalmology
Volume 99, Number 2, February 1992
5. Hanifin JM, Rogge JL. Staphylococcal infections in patients with atopic dermatitis. Arch Dermatol 1977; 113: 1383-6. 6. Wachs GN, Maibach HI. Co-operative double-blind trial of an antibiotic/corticoid combination in impetiginized atopic dermatitis. Br J Dermatol 1976; 95:323-8. 7. Dahl MY. Staphylococcus aureus and atopic dermatitis. Arch Dermatol1983; 119:840-6. 8. Hauser C, Wuethrich B, Matter L, et al. Staphylococcus aureus skin colonization in atopic dermatitis patients. Dermatologica 1985; 170:35-9. 9. White MI, Noble WC. Consequences of colonization and infection by Staphylococcus aureus in atopic dermatitis. ain Exp Dermatol 1986; 11:34-40. 10. Hogan MJ. Atopic keratoconjunctivitis. Am J Ophthalmol 1953; 36:937-47. 11. Jay JL. Clinical features and diagnosis of adult atopic keratoconjunctivitis and the effect of treatment with sodium cromoglycate. Br J Ophthalmol 1981; 65:335-40. 12. Buckley RJ. Atopic disease of the cornea. In: Cavanagh DH, ed. The Cornea: Transactions ofthe World Congress on the Cornea III. New York: Raven Press, 1988; chap. 79. 13. Tuft SJ, Dart JKG, Kemeny M. Limbal vernal keratoconjunctivitis: clinical characteristics and immunoglobulin E expression compared with palpebral vernal. Eye 1989; 3: 420-7. 14. Foster CS, Calonge M. Atopic keratoconjunctivitis. Ophthalmology 1990; 97:992-1000. 15. Tuft SJ, Kemeny DM, Dart JKG, Buckley RJ. Clinical features of atopic keratoconjunctivitis. Ophthalmology 1991; 98:150-8. 16. Mondino BJ. Inflammatory diseases of the peripheral cornea. Ophthalmology 1988; 95:463-72. 17. Mondino BJ, Caster AI, Dethlefs B. A rabbit model of staphylococcal blepharitis. Arch Ophthalmol1987; 105:40912. 18. Mondino BJ, Kowalski RP. Phlyctenulae and catarrhal infiltrates: occurrence in rabbits immunized with staphylococcal cell walls. Arch Ophthalmol 1982; 100:1968-71. 19. Mondino BJ, Dethlefs B. Occurrence ofphlyctenules after immunization with ribitol teichoic acid of Staphylococcus aureus. Arch Ophthalmol 1984; 102:461-3. 20. Kemeny DM, Richards D. Increased speed and sensitivity, and reduced sample size of a micro-radioallergosorbent test (MRAST). J Immunol Methods 1988; 108:105-13. 21. Lachica RYF, Genigeorgis CA, Hoeprich PD. Occurrence of protein A in Staphylococcus aureus and closely related Staphylococcus species. J Clin Microbiol 1979; 10:752-3. 22. Haukenes G. Serological typing of Staphylococcus aureus. 6. Antibodies of immune sera to strains cowan I, cowan II, cowan III, Wood 46, 670, 830, 1015,5687, and 6376. Acta Path Microbiol Scand 1964; 61:415-26.
184
23. Seal D, Ficker L, Ramakrishnan M, Wright P. Role of staphylococcal toxin production in blepharitis. Ophthalmology 1990; 97:1684-8. 24. Ficker L, Ramakrishnan M, Seal D, Wright P. Role of cellmediated immunity to staphylococci in blepharitis. Am J Ophthalmol 1991; 111:473-9. 25. Freer JH, Arbuthnott JP. Toxins of Staphylococcus aureus. Pharmacol Ther 1983; 19:55-106. 26. Sjoquist J, Stalenheim G. Protein A from Staphylococcus aureus. IX. Complement-fixing activity of protein A-IgG complexes. J Immunol 1969; 103:467-73. 27. Petersson BA, Stalenheim G. Induction of histamine release and desensitization in human leukocytes. Effect of anti-IgE or protein A from Staphylococcus aureus. Scand J Immunol 1975; 4:103-12. 28. Langone JJ. Protein A of Staphylococcus aureus and related immunoglobulin receptors produced by streptococci and pneumonococci. Adv Immunol 1982; 32:157-252. 29. White MI, Noble WC. Skin response to protein A. Proc R Soc Edin 1980; 79B:43-6. 30. White MI, Noble WC. The cutaneous reaction to staphylococcal protein A in normal subjects and patients with atopic dermatitis or psoriasis. Br J Dermatol 1985; 113: 179-83. 31 . Elek SD, Levy E. Distribution ofhaemolysins in pathogenic and non-pathogenic staphylococci. JPathol Bacteriol 1950; 62:541-54. 32. Kappler J, Kotzin B, Herron L, et al. Y-~-specific stimulation of human T cells by staphylococcal toxins. Science 1989; 244:811-3. 33. Matusiewicz R, Grzybowski A. The ability of granulocytes to engulflatex particles and Staphylococcus aureus in healthy children of parents with atopic asthma. Ann Allergy 1989; 62:445-8. 34. Carapeto FJ, Winkelmann RK, Jordon RE. T and B lymphocytes in contact and atopic dermatitis. Arch Dermatol 1976; 112:1095-100. 35. Hill HR, Quie PG. Raised serum-IgE levels and defective neutrophil chemotaxis in three children with eczema and recurrent bacterial infection. Lancet 1974; 1:183-7. 36. Buckley RH, Wray BB, Belmaker EZ. Extreme hyperimmunoglobulinemia E and undue susceptibility to infection. Pediatrics 1972; 49:59-70. 37. Schopfer K, Baerlocher K, Price P, et al. Staphylococcal IgE antibodies, hyperimmunoglobulinemia E and Staphylococcus aureus infections. N Engl J Med 1979; 300:835-8. 38. Abramson JS, Dahl MY, Walsh G, et al. Antistaphylococca1 IgE in patients with atopic dermatitis. J Am Acad Dermatol 1982; 7:105-10. 39. Badiani D, Bron A, Elkington A, Lea SH. Use of a novel transport method for quantification of the normal flora of theextemal eye. Microbiol Ecol Health Dis 1988; 1:57-9.