Parasitology International 62 (2013) 272–275

Contents lists available at SciVerse ScienceDirect

Parasitology International

j ourna l homepage: www.e lsev ie r .com/ locate /par in t

Incidence of toxoplasma retinochoroiditis in patients with ankylosing spondylitisafter using TNF-α blockers

Kelly Fernandes de Paula Rodrigues a, Tiago Eugênio Faria e Arantes a, Cristina Muccioli a,João Lins de Andrade Neto a, Marcelo M. Pinheiro b,⁎a Ophthalmology Department, Universidade Federal de São Paulo/Escola Paulista de Medicina (Unifesp/EPM), Brazilb Rheumatology Division, Universidade Federal de São Paulo/Escola Paulista de Medicina (Unifesp/EPM), Brazil

⁎ Corresponding author at: Av. Dr. Altino Arantes, 66São Paulo—SP CEP 04042-033, Brazil. Tel.: +55 11 5576 4

E-mail address: mpinheiro@uol.com.br (M.M. Pinhe

1383-5769/$ – see front matter © 2013 Elsevier Irelandhttp://dx.doi.org/10.1016/j.parint.2013.02.003

a b s t r a c t

a r t i c l e i n f o

Article history:Received 17 May 2012Received in revised form 15 February 2013Accepted 18 February 2013Available online 26 February 2013

Keywords:ToxoplasmaRetinochoroiditisAnkylosing spondylitisTNF-α blockersHLA-B27Prospective study with control group

TNF-α blockers are associated with reactivation of latent granulomatous infections and almost 6% of theworld population has some chorioretinitis (CR) caused by Toxoplasma gondii. Thus, the blockade of TNF-αcould reactivate a latent toxoplasmosis infection (LTxI). This study was conducted to evaluate the prevalenceand incidence of chronic and active CR related to T. gondii in patients with ankylosing spondylitis (AS). A totalof 74 eyes from 37 active AS outpatients starting TNFα blockers were compared with 35 AS patients, matchedto age and sex, under conventional therapy in a prospective and controlled trial. All patients underwent se-rological tests for T. gondii, as well as periodic ophthalmologic examination during 12 months. Active CR wasdefined if a white, focal retinochoroidal lesion with overlying vitreous inflammation had been found.Retinochoroidal lesions with sharp edges, hyperpigmented borders and atrophic center were defined as CRscars. At baseline, no patient had active CR. From the 144 eyes examined, almost 6% had CR scars and only2.1% had a typical toxoplasmic CR scar and all of them were negative for HLA-B27. During 12 months offollow-up, no recurrence or new CR were observed. AS patients using TNF-α blockers do not have a higherrisk of acute or chronic CR caused by T. gondii.

© 2013 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

The intracellular protozoan Toxoplasma gondii is capable to infect andreplicate in almost all human nucleated cells and cause a chronic asymp-tomatic infection in immunocompetent individuals [1,2]. However, in anyimmunosuppressant condition, this parasite is able to cause reactivationof latent ocular lesions, as also reactivation of other sites, which maylead to disseminated infection and life-threatening encephalitis [3].Worldwide, approximately 2 billion people are chronically infected withT. gondii. Ocular toxoplasmosis is the most common cause of posterioruveitis and it can be congenital or acquired after birth. Eye lesions usuallyare solitary and affect only one eye, emerging years after acute systemicinfection [1,2,4,5].

Tumor necrosis factor alpha (TNF-α) inhibitors have been used for thetreatment of chronic inflammatory arthritis, including rheumatoid arthri-tis (RA) and ankylosing spondylitis (AS) [6]. However, TNF-α blockers areassociatedwith a high rate of infections, particularly reactivation of latentgranulomatous diseases, such as tuberculosis, histoplasmosis and asper-gillosis [7–10]. Nonetheless, evidence of eye infection by Toxoplasmaafter the use of these agents is lacking.

9, apto 105, Vila Clementino,239; fax: +55 11 5579 6665.iro).

Ltd. All rights reserved.

Serological evidence indicates that 60% of the Brazilian population hasantibodies against T. gondii and 6% of them have some ocular lesions [4].In human dendritic cells, TNF-α plays an important role in controllingthe infection caused by this protozoan [3]. Thus, the blockade of TNF-αcould cause reactivation of latent toxoplasmosis infection (LTxI) as wellas increase the risk of acute toxoplasmosis.

This study aimed at evaluating the prevalence of chronic ocular andactive chorioretinitis lesions caused by Toxoplasma gondii in AS patientstaking TNF-α antagonists. Additionally, the incidence of new lesionswas also verified during 12 months of follow-up.

2. Material and methods

From January 2009 toApril 2010, a total of 74 eyes from37outpatientswith active AS [11] were evaluated. All of them had been taking TNF-αblockers (ADA—adalimumab, ETN—etanercept or IFX—infliximab) in thelast 12 months due to inadequate response to NSAIDs (non-steroidalanti-inflammatory drugs) and conventional DMARDs (disease-modifyingantirheumatic drugs), including methotrexate. Thirty-five AS patients,matched to age and sex, under conventional therapy (CT) with DMARDsandNSAIDswere used as control group. Individualswith uni- or bilater-al cataract were excluded from the study, as well as patients who werepregnant, with cancer or demyelinating disease or using pulsotherapywith methylprednisolone. Previous infections caused by hepatitisB or C, HIV or CMV were also excluded. All participants answered

Table 1Clinical characteristics of AS patients taking TNFα blockers and under conventionaltherapy (CT) with NSAIDs and DMARDs.

Taking TNF-blockers(N = 37)

CT(N = 35)

P

Age (years) 42.6 ± 11.5 43.7 ± 10.8 0.29Time since diagnosis (years) 13.4 ± 7.6 13.9 ± 8.1 0.43Male sex, N (%) 27 (73%) 25 (71.4%) 0.71Peripheral arthritis, N (%) 15 (40.5%) 13 (37.1%) 0.25Enthesitis, N (%) 13 (35.1%) 10 (28.6%) 0.12Previous IAU, N (%) 12 (32.4%) 11 (31.4%) 0.68HLA-B27, N (%) 30 (81.1%) 26 (74.3%) 0.08BASDAI 4.3 ± 1.9 3.3 ± 2.0 0.03BASFI 5.7 ± 2.1 5.9 ± 2.3 0.33BASMI 5.3 ± 2.6 5.5 ± 2.5 0.50ASDAS-CRP 3.3 ± 1.1 1.8 ± 1.5 0.01Regular and current NSAIDs, N (%) 31 (83.8%) 24 (68.7%) 0.02Current MTX and SSZ, N (%) 21 (56.7%) 18 (51.7%) 0.23TNF-α blockers, N (%)

IFX 22 (59.4%) – –

ADA 8 (21.6%) – –

ETN 7 (18.9%) – –

NSAIDs: non-steroidal anti-inflammatory drugs; CT: conventional therapy with DMARDsand NSAIDs; IAU: inflammatory anterior uveitis; BASDAI: Bath Ankylosing SpondylitisDisease Activity Index; BASFI: Bath Ankylosing Spondylitis Function Index; BASMI: BathAnkylosing SpondylitisMobility Index; ASDAS-CRP: Ankylosing Spondylitis Disease Activ-ity Score C-reactive protein;MTX:methotrexate; SSZ: sulfasalazine; IFX: infliximab; ADA:adalimumab; ETN: etanercept.

Table 2Serological lab tests (IgM and IgG) against toxoplasmosis in AS patients at baseline.

Toxoplasmosis lab tests(positivity)

AS patients taking TNFαblockers(N = 37)

AS patientsunder CT(N = 35)

P

IgG antibodies 23 (62.2%) 20 (57.1%) 0.67

CT: conventional therapy with DMARDs and/or NSAIDs.

273K.F. de Paula Rodrigues et al. / Parasitology International 62 (2013) 272–275

a questionnaire, which included details about demographic andanthropometrical data, medical history and concomitant medications.TheHLA-B27was studied in themajority of AS patients, but no significantstatistically difference between biologic and conventional groups.

At baseline, serological tests for Toxoplasma gondii (IgM andIgG antibodies) were performed in all patients by a commercialimmunofluorimetric method in microparticles. Non-reagent IgG andIgM results were considered if values below 1.6 IU/mL and 0.6 IU/mL,respectively. Reagent IgG and IgM tests if values above 6 IU/mL and1.0 IU/mL, respectively, according to manufacturer's manual.

The patients underwent ophthalmologic examination by a specializedophthalmology physician, including binocular indirect ophthalmoscopyundermydriasiswith a 20 diopter lens. During the exam, the toxoplasmicretinochoroidal lesions were classified into one of two main typesaccording to the clinical characteristics. Active chorioretinitis was definedif a white, focal, retinochoroidal lesion with overlying vitreous inflamma-tion had been found. On the other hand, hyperpigmented lesions withwell-delimited edgeswere used todefine retinochoroidal scars or inactivetoxoplasmosis [1,2]. Moreover, periodic ophthalmologic re-evaluationwas performed in order tofind any sign of reactivation during 12 months.

All participants gave written informed consent prior to entry inthe study and the Unifesp/EPM's Ethics and Research Committeeapproved the protocol (N° 1478/09).

2.1. Statistical analysis

Descriptive analysis was performed according to demographic,anthropometric andmedical history data, as well as eye examinationand lab tests. The data are expressed as percentage (%) and mean ±standard deviation (SD). Variables achieving statistical significancein the initial univariate analysis or chi-squared association testwere included in the logistic regression models. The dependent vari-ables were active chorioretinitis and retinochoroidal scars in order toknow if any risk factor could be significantly related to these lesions.The Statistical Package for Social Science software (SPSS, version 15)was used for all statistical analyses. Statistical significance was set asp b 0.05.

3. Results

Themain clinical characteristics of the AS patients are listed in Table 1.Although none of them had previous history of toxoplasmosis infection,the majority had reagent IgG antibodies against T. gondii. However,none of them had positive IgM antibodies and 45% was negative for IgMand IgG antibodies (Table 2). The first serological profile IgG (+) IgM(−) demonstrates immunity or chronic infection. The latter profile char-acterizes a susceptible population, since it has no protective antibodies.

At baseline ophthalmologic examination, no patient had activeuveitis. In biologic group (or taking TNF-α blockers), from the 74 eyesexamined, 6.8% had chronic lesions (retinal or retinochoroidal scars),of which only one patient (1.3%) had a pathognomonic finding ofretinochoroidal scar caused by toxoplasmosis. Similarly, in CT group,from 70 eyes evaluated, only 2 (2.9%) had typical chronic toxoplasmicscars. These 3 (4.2%) patients with typical lesions caused by T. gondiiwere negative for HLA-B27. Moreover, they had some visual impairmentand positivity for antibodies (IgG) against the protozoan.

During 12 months of follow-up, no recurrence of toxoplasmosisuveitis or new eye lesions were observed. Thus, the ocular lesionsremained stable throughout the follow-up period and no reactivationwas observed, regardless of age, disease duration, baseline BASDAI orASDAS-CRP, serological tests for toxoplasmosis, anti-TNF-α agent orconventional treatment group and previous eye injury (Table 3). Noneof them had positivity for IgM antibodies. No clinical variable was associ-ated with eye lesions or LTxI. On the other hand, the positivity forHLA-B27 plays a protective role for ocular chronic lesions (retina orretinochoroidal scars).

No clinical findings suggestive of systemic acute infection or Jarisch–Herxheimer's reaction, due to apoptosis of T. gondii tachyzoites, wereverified in our patients.

4. Discussion

Our data demonstrate that there is low prevalence of symptomaticprevious toxoplasmosis infection, ocular or systemic, in adults with AStaking TNF-blockers or under conventional treatment with DMARDsand NSAIDs. The prevalence of chronic and active eye lesions caused byToxoplasma gondii is not increased in these patients and it is similar tothe general population. Although 60% of these patients had previouslyhad contact with the protozoan at some point of their lives, there wasno case of ocular reactivation of LTxI after treatment with TNF-α inhibi-tors orMTX or SSZ. Moreover, taking these agents, the chance of contam-ination or re-exposition was also extremely low, at least during the first12 months of follow-up. However, it is important to note that approxi-mately 45% of the patients were susceptible since they did not have anyIgG or IgM antibodies against T. gondii.

These data are different of those from other granulomatous diseases,particularly tuberculosis [6,7], in which the formation and maintenanceof the granuloma are alsomediated through IFNγ and Th1-dependent re-sponse. The majority of active cases of tuberculosis related to anti-TNF-αtherapy are associatedwith reactivation of latent infection (LTbI). Usually,it occurs within the first six months, according to the agent used, beinglower in patients exposed to ETNwhen compared to users of monoclonalantibodies. The incidence of new LTbI cases was significantly reducedafter the introduction of guidelines before beginning these drugs, includ-ing epidemiologic data, tuberculin skin test, chest X-ray and isoniazide for6 months in positive cases [12,13].

Table 3Ophthalmologic acute and chronic lesions in AS patients taking TNF blockers and underconventional therapy with DMARDs and NSAIDs, according to time of evaluation.

Chronic eye injuries Baseline evaluation After 12 months P

AS patients taking TNFα blockers(N = 37)

5 (6.8%) 5 (6.8%) 1.0

AS patients under CT(N = 35)

2 (2.9%) 2 (2.9%) 1.0

ANOVA test.

274 K.F. de Paula Rodrigues et al. / Parasitology International 62 (2013) 272–275

Considering our results on reactivation of LTxI, the structure of thegranuloma may not have been significantly missed nor should ithave been increasing the protozoan load, showing the safety forusing TNF-α blockers in AS patients. According to these findings, wecould speculate that the TNF-α blockade may not be as relevant fortoxoplasma as it is for mycobacteria. Furthermore, the granuloma in-duced by T. gondii should be more stable than those from LTbI.

In a mice model, the endogenous protective role of TNF-α on infec-tion caused by T. gondii has been demonstrated [3,10,14]. Some geneticparticularities, the disequilibrium of innate and acquired immuneresponse, as well as peculiarities related to the diseases themselvesand to the pathogens could explain the differences among the studiesand the incidence rates of infections. Comparing T. gondii to hepatitisC virus (HCV), another intracellular microorganism, TNF-α has asecondary, but synergistic, role on its replication. In HCV patients,there is no higher rate of reactivation of subclinical infection afterusing TNF-α blockers [15]. On the other hand, for other virus, includingvaricella-zoster [16], hepatitis B virus [17], cytomegalovirus [18], andherpes simplex-1 [19], the TNF-α blockade can have devastating effects.

Another interesting aspect that could justify our results, lowprevalence of reactivation of LTxI in AS patients after TNF-α blockersis related to the human leukocyte antigen (HLA). Recent evidencehas shown the ability of HLA-B27 to confer resistance to the progressionof some infections, including HIV and HCV [20]. The co-occurrence ofthese protective and pathogenic features suggests a more pronouncedimmune/inflammatory response leading to an effective clearance ofsome pathogens on one side and to autoimmunity on the other. Thismight be due to the antigen presenting properties and/or to theco-inheritance of gene variants that contribute to an altered homeostasisin case of microbial infections or tissue injury [20]. This same group be-lieves that the Plasmodium falciparum, for instance, may have also playeda negative selection role on HLA-B27, since the geographic distribution ofHLA-B27 shows a latitude-related gradient inverse to that of malaria en-demic [21]. On the other hand, positive HLA-B27 patients have more dif-ficulty for the clearance of Chlamydia, another intracellular germ relatedto chronic urethritis or cervicitis, than negative individuals. Thus, theyhave higher risk for chronic arthritis [22,23].

Although highly prevalent worldwide, the clinical association be-tween recurrent inflammatory chronic uveitis, related to HLA-B27, andtoxoplasmosis posterior uveitis is not observed. Sampaio-Barros et al.also did not find significant association between toxoplasmosis and pos-terior uveitis after analyzing 350 patients with spondyloarthritis [24].Nevertheless, it is important to emphasize that four cases of reactivationof LTxI (2 cerebral and 2 retinochoroiditis) were reported in RA patientsafter using TNF-α antagonists [8–10] and one case of acute toxoplasmosisinfection in an AS patient treated with adalimumab [25]. Additionally,some murine studies have shown reactivation of cerebral toxoplasmosisafter treatment with dexamethasone or etanercept [26,27].

These findings highlight a possible protective mechanism of thisHLA on some infections, such as malaria and toxoplasmosis. Both thePlasmodium spp. and the Toxoplasma gondii are obligate intracellularprotozoanparasites of the Apicomplexa group. Both of themuse specificproteolytic enzymes that are essential for both parasite survival andpropagation inside host cells [28–30]. In addition, they share complexN-glycan structures that stress the endoplasmic reticulum prior to

threading into the apicoplast [31]. Similarly, in AS patients, the HLA-B27 canmisfold within the environment of the endoplasmatic reticulumand lead to a “gain of toxicity function” by inducing cellular stressresponses, according to the misfolding hypothesis [32].

To our best knowledge, this is the first prospective study addressed toinvestigate the relationship between previous or acute ocular lesions,LTxI, and TNF-α blockers in rheumatic patients under some immunosup-pressant and immunobiological treatment. Thus, these findings could linksome pathogens and the innate immunity, via Toll-like receptors, forexample, to the pathophysiology of the ankylosing spondylitis, particular-ly on role of the HLA-B27 in the clearance of microorganisms.

Nonetheless, this study has some limitations. We did not performserological tests ormolecular analysis in the puncturefluid of the aqueoushumor, another strategy to confirm this infection, nor we did IgG aviditytests.

Thus, based on data and current literature, there is no evidence thatTNF-α inhibitors cause reactivation of LTxI in AS patients. Additionally,according to our results, we do not suggest any lab test or ophthalmo-logic evaluation addressed to LTxI before beginning TNF-α therapy inAS patients, since there is no approach able to prevent the recurrenttoxoplasma retinochoroiditis nor any therapeutic intervention is re-quired, including chemoprophylaxis. Furthermore, its clinical identifi-cation is not easy. The ocular lesions are often asymptomatic and arenot associated with systemic symptoms [2]. Large prospective studiesare needed to confirm our findings, including the absence of causalitybetween TNF-α blockers and recurrence of toxoplasma chorioretinitis.In addition, in other chronic inflammatory arthritis, not ankylosingspondylitis, need to be evaluated to the risk LTxI.

In conclusion, patients with active ankylosing spondylitis do nothave higher risk of acute or chronic chorioretinitis caused by Toxoplas-ma gondii after 12 months of taking TNF-α blockers, regardless of age,disease duration, disease activity index, lifestyle habits, concomitantmedication and previous exposition.

Acknowledgments

Wewould like to thank the Spondylarthritis Section – RheumatologyDivision and Uvea/AIDS Section – Ophthalmology Department,Universidade Federal de São Paulo/Escola Paulista de Medicina fortheir support in the recruitment of patients, as well as performingall measurements.

References

[1] Wu L, Evans T, García RA. Toxoplasmosis. http://emedicine.medscape.com. [Availableat: eMedicine Specialties > Ophthalmology > Infections Disease. Accessed: 12.23.10.].

[2] Commodaro AG, Belfort RN, Rizzo LV, Muccioli C, Silveira C, Burnier Jr MN, et al.Ocular toxoplasmosis—an update and review of the literature. Memórias doInstituto Oswaldo Cruz 2009;104(2):345–50.

[3] Giese A, Stuhlsatz S, DaubenerW,Mackenzie CR. Inhibition of the growth of Toxoplasmagondii in immature human dendritic cells is dependent on the expression of TNF-alphareceptor 2. Journal of Immunology 2004;173(5):3366–74.

[4] Muccioli C, Silveira C, Belfort Jr R. Ocular toxoplasmosis. Contemporary Ophthalmology2007;6:1–8.

[5] Glasner PD, Silveira C, Kruszon-Moran D, Martins MC, Burnier Júnior M, Silveira S,et al. An unusual high prevalence of ocular toxoplasmosis in southern Brazil.American Journal of Ophthalmology 1992;114:136–44.

[6] Furst DE, Breedveld FC, Kalden JR, Smolen JS, Burmester GR, Biljlsma JW, et al.Updated consensus statement on biological agents, specifically TNF-alphablocking agents and interleukin-1 receptor antagonist (IL-r1a) for the treatmentof rheumatic diseases. Annals of the Rheumatic Diseases 2005;64:Siv2–Siv14.

[7] Giles GT, Bathon JM. Serious infections associated with anticytokine therapies inthe rheumatic diseases. Journal of Intensive Care Medicine 2004;19:320–34.

[8] Young JD, McGuire BS. Infliximab and reactivation of cerebral toxoplasmosis. TheNew England Journal of Medicine 2005;353:1530–1.

[9] Lassoued S, Zabraniecki L,Marin F. Toxoplasmic chorioretinitis and antitumor necro-sis factor treatment in rheumatoid arthritis. Seminars in Arthritis and Rheumatism2007;36:262–3.

[10] Gonzalez-Vincent M, Diaz MA, Sevilla J, Madero L. Cerebral toxoplasmosis followingetanercept treatment for idiopathic pneumonia syndrome after autologous peripheralblood progenitor cell transplantation (PBPCT). Annals of Hematology 2003;82:649–53.

[11] Zochling J, Van der Heijde D, Dougados M, Braun J. Current evidence for the manage-ment of ankylosing spondylitis: a systematic literature review for the ASAS/EULAR

275K.F. de Paula Rodrigues et al. / Parasitology International 62 (2013) 272–275

management recommendations in ankylosing spondylitis. Annals of the RheumaticDiseases 2006;65(4):423–32.

[12] Gomez-Reino J, Carmona L, Descalzo MA. Risk of tuberculosis in patients treatedwith tumor necrosis factor antagonists due to incomplete prevention ofreactivation of latent infection. Arthritis and Rheumatism 2007;57(5):756–61.

[13] Dixon WG, Hyrich KL, Watson KD, Lunt M, Galloway J, Ustianowski A, et al.Drug-specific risk of tuberculosis in patients with rheumatoid arthritis treatedwith anti-TNF therapy: results from the British Society for Rheumatology Bio-logics Register (BSRBR). Annals of the Rheumatic Diseases 2010;69:522–8.

[14] Derouich-Guergour D, Aldebert D, Vigan I, Jouvin-Marche E, Marche PN, Aubert D, et al.Toxoplasma gondii infection can regulate the expression of tumor necrosis factor-alphareceptors on human cells in vitro. Parasite Immunology 2002;24(5):271–9.

[15] Ferri C, Ferraccioli G, Ferrari D, Galeazzi M, Lapadula G, Montecucco C, et al. Safety ofanti-tumor necrosis factor-alpha therapy in patients with rheumatoid arthritis andchronic hepatitis C virus infection. Journal of Rheumatology 2008;35(10):1944–9.

[16] Salmon-Ceron D, Tubach F, Lortholary O, Chosidow O, Bretagne S, Nicolas N, et al.Drug-specific risk of non-tuberculosis opportunistic infections in patients receiv-ing anti-TNF therapy reported to the 3-year prospective French RATIO registry.Annals of the Rheumatic Diseases 2011;70(4):616–23.

[17] Garcia-Vidal C, Rodríguez-Fernández S, Teijón S, Esteve M, Rodríguez-CarballeiraM, Lacasa JM, et al. Risk factors for opportunistic infections in infliximab-treatedpatients: the importance of screening in prevention. European Journal of ClinicalMicrobiology and Infectious Diseases 2009;28(4):331–7.

[18] Haerter G,Manfras BJ, de Jong-Hesse Y,Wilts H, Mertens T, Kern P, et al. Cytomegalovi-rus retinitis in a patient treatedwith anti-tumor necrosis factor alpha antibody therapyfor rheumatoid arthritis. Clinical Infectious Diseases 2004;39(9):e88–94.

[19] Justice EA, Khan SY, Logan S, Jobanputra P. Disseminated cutaneous herpes sim-plex virus-1 in a women with rheumatoid arthritis receiving infliximab: a case re-port. Journal of Medical Case Reports 2008;2:282.

[20] Mathieu A, Paladini F, Vacca A, Cauli A, FiorilloMT, Sorrentino R. The interplay betweenthe geographic distribution of HLA-B27 alleles and their role in infectious and autoim-mune diseases: a unifying hypothesis. Autoimmunity Reviews 2009;8(5):420–5.

[21] Mathieu A, Cauli A, Fiorillo MT, Sorrentino R. HLA-B27 and ankylosing spondylitisgeographic distribution as the result of a genetic selection induced by malaria

endemic? A review supporting the hypothesis. Autoimmunity Reviews 2008;7(5):398–403.

[22] Carter JD, Gérard HC, Espinoza LR, Ricca LR, Valeriano J, Snelgrove J, et al. Chla-mydiae as etiologic agents in chronic undifferentiated spondylarthritis. Arthritisand Rheumatism 2009;60(5):1311–6.

[23] AppelH, KuonW,KuhneM,WuP, Kuhlmann S, Kollnberger S, et al. Use ofHLA-B27 tet-ramers to identify low-frequency antigen-specific T cells in chlamydia-triggered reac-tive arthritis. Arthritis Research & Therapy 2004;6(6):R521–34.

[24] Sampaio-Barros PD, Conde RA, Bonfiglioli R, Bértolo MB, Samara AM. Characteri-zation and outcome of uveitis in 350 patients with spondyloarthropathies. Rheu-matology International 2006;26(12):1143–6.

[25] Azevedo VF, Pietrovski CF, de Almeida Santos Jr M. Acute toxoplasmosis infectionsin a patient with ankylosing spondylitis treated with adalimumab: a case report.Reumatismo 2010;62(4):283–5.

[26] Prandota J. The importance of Toxoplasma gondii infection in diseases presenting withheadaches. Headaches and aseptic meningitis may be manifestations of the Jarisch–Herxheimer reaction. International Journal of Neuroscience 2009;119(12):2144–82.

[27] Gazzinelli RT, Eltoum I, Wynn TA, Sher A. Acute cerebral toxoplasmosis is inducedby in vivo neutralization of TNF-alpha and correlates with the down-regulatedexpression of inducible nitric oxide synthase and other markers of macrophageactivation. Journal of Immunology 1993;151:3672–81.

[28] Li H, Child MA, Bogyo M. Proteases as regulators of pathogenesis: examples from theapicomplexa. Biochimica et Biophysica Acta Jun 14 2011 [Epub ahead of print].

[29] Tuda J, Mongan AE, Tolba ME, Imada M, Yamagishi J, Xuan X, et al. Full-parasites:database of full-length cDNAs of apicomplexa parasites, 2010 update. NucleicAcids Research 2011;39:D625–31 [Database issue].

[30] López-Estraño C. Comparative analysis of stage specific gene regulation ofapicomplexan parasites: Plasmodium falciparum and Toxoplasma gondii. InfectiousDisorders Drug Targets 2010;10(4):240–1.

[31] Bushkin GG, Ratner DM, Cui J, Banerjee S, Duraisingh MT, Jennings CV, et al. Suggestiveevidence for Darwinian selection against asparagine-linked glycans of Plasmodiumfalciparum and Toxoplasma gondii. Eukaryotic Cell 2010;9(2):228–41.

[32] Antoniou AN, Lenart I, Guiliano DB. Pathogenicity of misfolded and dimericHLA-B27 molecules. International Journal of Rheumatology 2011;2011:486856.