Transactions of the Royal Society of Tropical Medicine and Hygiene (2009) 103, 575—580

avai lab le at www.sc iencedi rec t .com

journa l homepage: www.e lsev ierhea l th .com/ journa ls / t rs t

�1-acid glycoprotein and �1-antitrypsin as earlymarkers of treatment response in patients receivingthe intensive phase of tuberculosis therapy

Maria Luiza Doria Almeidaa, Marco Antonio Barbierib, Ricardo Q. Gurgela,Saddiq Tsimiri Abdurrahmanc, Uthman Alhaji Babac, C. Anthony Hartd,1,Alan Shenkind, Ângela Maria Silvaa, Luiz de Souzab, Luis Eduardo Cuevasc,∗

a Federal University of Sergipe, Departamento de Medicina, Campus da Saude, Rua Claudio Batista, s/n Sanatorio, Aracaju,Sergipe, Brazilb Sao Paulo University, Departamento de Puericultura e Pediatria, Ribeirao Preto, Sao Paulo, Brazilc Liverpool School of Tropical Medicine, Pembroke Place, Liverpool L3 5QA, UKd School of Infection and Host Defence, University of Liverpool, Duncan Building, Daulby Street, Liverpool,Merseyside L69 3GA, UK

Received 22 May 2007; received in revised form 21 November 2008; accepted 24 November 2008Available online 10 February 2009

KEYWORDSTuberculosis;�1-acid glycoprotein;�1-antitrypsin;C-reactive protein;Treatment response;Brazil

Summary The identification of early markers that predict the response to anti-tuberculosistreatment would facilitate evaluation of new drugs and improve patient management. Thisstudy aimed to determine whether selected acute phase proteins and micronutrients mea-sured at the time of diagnosis and during the first weeks of treatment could predict treatmentresponses during the 2-month standard intensive phase of therapy. For this purpose, �1-antitrypsin, �1-acid glycoprotein, �2-macroglobulin, C-reactive protein, C3, C4, zinc, copperand selenium concentrations were measured in Brazilian patients with smear-positive tuber-culosis at the time of diagnosis and 1, 3, 5 and 8 weeks after initiation of therapy. Patientswere classified into fast (n = 29), intermediate (n = 18) and slow responders (n = 10) if they weresmear-negative at 3, 5 or 8 weeks of treatment. �1-acid glycoprotein on enrolment and 1 weekof treatment, �1-antitrypsin at week 1 and C-reactive protein and C3 after 3 weeks of therapywere higher in slow responders than in fast responders. �1-antitrypsin and �1-acid glycoprotein

may be helpful in predicting treatment response at the time of initiation of therapy, and couldbe used as early markers to identify patients with an increased likelihood of treatment failure.© 2008 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rightsreserved.

∗ Corresponding author. Tel.: +44 151 708 3219; fax: +44 151 708 5233.E-mail address: lcuevas@liv.ac.uk (L.E. Cuevas).

1 Deceased.

0035-9203/$ — see front matter © 2008 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved.doi:10.1016/j.trstmh.2008.11.024

5

1

TfrwTapTtsprabttmt1ra

pdpwwc

mgpsira

2

TaePrmstpgpOwe3cwm

d

ePmsamfipret−

usm1rAamaes

uSpataptibp

oMC

wpfwitAs

3

Apmc

76

. Introduction

uberculosis (TB) is a major cause of morbidity and the mostrequent cause of adult death worldwide.1 Its treatmentequires combinations of antimicrobials for 6—8 months, andeeks, if not months, are needed for clinical improvement.his protracted recovery during therapy limits clinicians’bility to differentiate between patients with adequate andoor treatment response. The International Union againstuberculosis and Lung Diseases recommends monitoringhe patient’s response by obtaining sputum samples formear microscopy at the end of the 2-month intensivehase of treatment, as a significant proportion of patientsemain smear-positive at this stage.2 Recent studies havettempted to identify early clinical, radiological, micro-iological and immunological markers that could predictreatment response. Such markers would facilitate clinicalrials for the evaluation of new drugs and improve patientanagement. Detection of most potential markers iden-

ified to date, however, such as IFN-�-inducible protein0 and pentraxin 3,3 require technologies not available inesource-limited countries; more easily measured markersre required for high incidence areas.

The immunological processes that take place in theatient under treatment are complex and change rapidlyuring the first stages of anti-TB treatment, when acutehase proteins (APP) and micronutrients exhibit changesith varying duration. There are, however, limited data onhether the changes that occur during this intensive phaseould predict treatment response.

This study therefore aimed to determine whetherarkers of inflammation [�1-antitrypsin (AAT), �1-acid

lycoprotein (AAG), �2-macroglobulin (AMG), C-reactiverotein (CRP), C3 and C4] or changes in zinc, copper andelenium concentrations at the time of diagnosis and dur-ng the first weeks of treatment could predict treatmentesponses after the 2-month standard intensive phase ther-py.

. Methods

his prospective study of patients with pulmonary TBttending primary health care centres in Aracaju, in north-ast Brazil took place between April 2004 and June 2005.atients with cough of more than 3 weeks duration wereequested to provide sputum samples for direct smearicroscopy and culture, and all patients with positive

mear microscopy were invited to participate. All study par-icipants gave informed consent. After an interview andhysical examination to obtain baseline clinical and back-round information, patients began the 2-month intensivehase of treatment following national treatment guidelines.nly newly diagnosed patients were enrolled and patientsith other chronic diseases such as diabetes and HIV werexcluded. All patients were followed during home visits 1,, 5 and 8 weeks after initiating treatment to assess their

linical progress. In addition, 30 healthy adult blood donorsere enrolled to obtain normal APP background levels andicronutrient concentrations.Sputum samples for smear microscopy were collected

uring each home visit and specimens collected on

fmhfia

M.L.D. Almeida et al.

nrolment and at week 8 were cultured on solid media.atients were placed into three groups according their smearicroscopy, culture and outcome. Patients with negative

mear microscopy after 3 weeks of treatment were classifieds having a fast response (Group I); those with positive smearicroscopy at 3 weeks but negative by week 5 were classi-ed as intermediate responders (Group II); and patients withositive smear microscopy at week 8 were classified as slowesponders (Group III). Blood samples were collected duringach visit using zinc and rubber-free equipment, allowedo coagulate, centrifuged and stored in separate tubes at70 ◦C within 6 h of collection.

Inductive coupled plasma mass spectrometry (ICPMS) wassed to measure zinc, selenium and copper concentrationsimultaneously using a PQ ExCell instrument (Thermo Ele-ental, Winsford, Cheshire, UK). Specimens were diluted

:60 with a diluent containing an internal standard and aspi-ated at 1.10 ml/min into the instrument, using a CETACSX-500 (CETAC, Omaha, NE, USA) sample changer. Galliumnd indium were used as internal standards. One hundredicrolitres of the calibrant, control or sample were diluted

nd prepared in duplicate. Calibrants were run as tests aftervery 22 aspirations as drift checks. Controls were run at thetart and finish of the specimen run.

CRP concentrations were measured using a Roche Mod-lar Analyser and the Roche CRPLX kit (Roche Ltd, Eastussex, UK). This assay is based on the principle ofarticle-enhanced immunoturbidimetry intended for use onutomated clinical chemistry analysers. Buffer is added tohe sample, then mixed and an anti-CRP antibody-latex isdded to start the reaction. The anti-CRP antibodies cou-led to the latex microparticles react with the antigen inhe sample to form an antigen/antibody complex. Follow-ng agglutination, this is measured turbidimetrically. Withinatch precision of the assay is 3% and the between batchrecision is 5%.

C3, C4, AAG, AAT and AMG were measured simultane-usly by nephelometry assays (Beckman Immage, Ramsey,N, USA) in the Clinical Chemistry Laboratory of Alder Heyhildren’s Hospital, Liverpool, UK.

Median concentrations of the APP and micronutrientsere described for each group and compared using non-arametric tests (Kruskal-Wallis). The Dunn test was appliedor variables with significant P-values. The Friedman testas used to assess median concentration changes over time

n each group. Patients in Group III had incomplete micronu-rient measurements due to a shortage of plasma and onlyPP was measured for this group. A P-value <0.05 was con-idered to be statistically significant.

. Results

total of 65 patients with TB were enrolled. Eight of theatients were excluded because they had abandoned treat-ent before the end of the 2 months (four), had other

hronic illnesses (two) or extrapulmonary TB (two). Thirty-

our (60%) of the 57 patients available for analysis wereale; their mean age was 37 (range 18—79) years and most

ad poor education. Twenty-nine (50%) patients were classi-ed as having a fast treatment response (Group I), 18 (32%)s intermediate responders (Group II) and 10 (18%) as slow

trea

og

n

�1-acid glycoprotein and �1-antitrypsin as early markers of

responders (Group III). There were no statistical differencesin the clinical presentation of the three groups at the timeof enrolment with the exception of a higher frequency ofhaemoptysis in Group I (3/29) than in Groups II (7/18) and III

(6/10; �2 for trend, P < 0.001), and a lower body mass index(BMI) in Group III (medians 20.8. 18.7 and 18 among GroupsI, II and III, respectively; P = 0.034). Most clinical parametersimproved rapidly, although a cough was still reported by 80%

hlwZ

Table 1 Mean (SD) concentrations of acute phase proteins and m

Group Time (weeks) [m

0

Copper (�mol/l) I 27.8 (7.7)II 29.8 (5.3)III 29.9 (3.8)Controlsa 16.7 (2.9)

Zinc (�mol/l) I 13.2 (4.2)II 14.0 (4.1)III 15.9 (7.7)Controlsa 13.2 (1.9)

Selenium (�mol/l) I 0.9 (0.2)II 1.0 (0.2)III 0.8 (0.2)Controlsa 1.3 (0.2)

Copper:zinc ratio I 2.2 (0.9)II 2.3 (0.6)III 2.3 (1.2)Controlsa 1.3 (0.3)

�1-antitrypsin (g/l) I 2.5 (0.7)II 2.8 (0.7)III 2.7 (0.9)Controlsa 1.3 (0.3)

�1-acid glycoprotein (g/l) I 2.1 (0.7)b

II 2.5 (0.5)III 2.8 (0.4)Controlsa 0.7 (0.7)

�2-macroglobulin (g/l) I 1.8 (0.4)II 2.0 (0.3)III 1.9 (0.4)Controlsa 1.9 (0.5)

C3 (g/l) I 1.6 (0.4)II 1.6 (0.4)III 1.6 (0.5)Controlsa 1.2 (0.2)

C4 (g/l) I 0.4 (0.1)II 0.4 (0.1)III 0.3 (0.1)Controlsa 0.3 (0.1

C-reactive protein (mg/l) I 99.1 (62.9)II 98.2 (43.9)III 138.1 (75.4)Controlsa 3.6 (4.4)

a P < 0.001 for all measurements when compared to controls with the eb P < 0.05 when compared to Group III.

tment response in patients receiving the intensive 577

f patients by the end of the 2-month follow up in the threeroups.

Copper and copper/zinc ratios were higher and sele-ium concentrations were lower in patients with TB than in

ealthy controls, as shown in Table 1 (P < 0.01 for all). Simi-arly, AAT, AAG, C3, C4 and CRP concentrations, but not AMG,ere higher in patients with TB than in healthy controls.inc concentrations increased 1 week after initiation of

icronutrients by study group

ean (SD)]

1 3 5 8

27.2 (9.0) 23.4 (5.0) 23.6 (4.2) 23.2 (5.2)29.9 (7.3) 24.8 (5.4) 24.8 (4.0) 22.9 (4.6)31.3 (5.0) 29.0 (4.8) 27.2 (5.1) 29.3 (6.1)

15.3 (4.6) 13.3 (3.4) 13.2 (3.6) 12.7 (3.4)15.2 (3.7) 12.9 (3.7) 12.8 (2.4) 12.7 (2.9)16.5 (5.2) 14.9 (3.2) 15.5 (5.8) 14.5 (5.3)

0.9 (0.2) 1.0 (0.2) 0.9 (0.2) 0.9 (0.1)1.0 (0.2) 1.0 (0.2) 1.0 (0.2) 1.0 (0.2)1.0 (0.1) 1.0 (0.2) 1.0 (0.2) 0.9 (0.1)

2.0 (1.0) 1.9 (0.7) 1.9 (0.7) 2.0 (0.8)2.0 (0.5) 2.1 (0.7) 2.0 (0.5) 1.9 (0.5)2.1 (0.9) 2.1 (0.7) 1.9 (0.7) 2.2 (0.7)

2.4 (0.7)b 2.1 (0.8) 1.9 (0.5)b 1.7 (0.5)b

2.6 (0.3) 2.4 (0.3) 2.1 (0.4) 2.0 (0.5)3.1 (0.3) 2.5 (0.2) 2.7 (0.6) 2.4 (0.7)

1.8 (0.7)b 1.4 (0.7) 0.9 (0.5)b 0.9 (0.4)b

2.1 (0.6) 1.4 (0.4) 1.2 (0.7) 1.1 (0.6)2.6 (0.3) 1.6 (0.3) 1.8 (0.6) 1.9 (0.8)

1.9 (0.5) 2.1 (0.5) 2.1 (0.5) 2.1 (0.5)2.0 (0.4) 1.9 (0.3) 2.0 (0.2) 2.1 (0.3)2.0 (0.4) 2.2 (0.3) 2.1 (0.3) 2.1 (0.3)

1.5 (0.4) 1.6 (0.4) 1.5 (0.4)b 1.4 (0.3)b

1.6 (0.4) 1.5 (0.3) 1.4 (0.3) 1.4 (0.3)1.8 (0.3) 1.8 (0.3) 1.8 (0.2) 1.9 (0.3)

0.3 (0.1) 0.3 (0.1) 0.3 (0.1) 0.3 (0.1)0.3 (0.1) 0.3 (0.1) 0.3 (0.1) 0.3 (0.1)0.3 (0.1) 0.3 (0.1) 0.3 (0.1) 0.3 (0.1)

47.9 (50.4) 35.8 (41.9) 14.5 (18.4)b 13.4 (17.9)b

58.1 (44.9) 29.7 (32.7) 33.9 (40.7) 25.3 (31.7)84.0 (63.4) 27.8 (20.4) 52.9 (39.8) 51.9 (36.5)

xception of zinc (P = 0.8) and AMG (P = 0.9).

5 M.L.D. Almeida et al.

t1dtow

aGItseCh

sGptc

Frb

Figure 2 Box plots of �1-antitrypsin (AAT) and �1-acid gly-coprotein (AAG) concentrations in patients with fast (Group I),

78

reatment and subsequently decreased. Copper increasedweek after initiation of treatment and subsequently

ecreased in Groups I and II. Selenium had a less defined pat-ern, with median concentrations increasing and decreasingver time in the three groups. None of these micronutrientsere predictors of treatment response.

C3 concentrations were similar among the study groupst the start of the study, and remained high over time inroup III but decreased with time in patients in Groups I and

I. CRP concentrations were high in all groups at the start ofreatment, decreased rapidly over the first 3 weeks and thentarted to increase again in Group III with significant differ-nces by weeks 5 and 8. AMG concentrations increased and4 concentrations decreased with time; the three groupsad similar concentrations at all time points.

AAT, AAG and CRP concentrations varied significantly by

tudy group (Figure 1). AAT concentrations were similar inroups I and II on enrolment and decreased over time, withatients in Group III having significantly higher concentra-ions than patients in Groups I and II at weeks 1, 5 and 8. AAGoncentrations were higher in Group III on enrolment and

igure 1 �1-antitrypsin (A), �1-acid glycoprotein (B) and C-eactive protein (C) concentrations on enrolment and follow upy study group. Error bars show 95% CI.

iaC

r1wAs

t16hs

4

AitttitAmToaoTMhagwmm

trph

ntermediate (Group II) and slow (Group III) treatment responsesfter 1 week of therapy. Each box plot illustrates the mean, 95%I and range.

emained significantly higher during the follow up (weeks, 5 and 8). Patients in Group III had increasing levels aftereek 3 while Groups I and II continued decreasing. AAT andGM values at week 1 for patients in Group I, II and III arehown in Figure 2.

Patients in Group III were followed up 1 year after ini-iation of therapy to ascertain their outcome. Four out of0 patients had died; one continued to be smear positivemonths later and was declared a treatment failure; two

ad negative microscopy; two were declared cured withoutmear confirmation and one was lost.

. Discussion

cute phase proteins and micronutrients have been stud-ed for a considerable time in TB.4—6 However, as most ofhese studies were conducted before the advent of shorter-erm antimicrobial combinations, this is the first time thatheir combined patterns are reported in patients receiv-ng the modern intensive phase of treatment. Data fromhe 1970s and 1980s indicate that most micronutrients andPPs are modified by TB.7 It is, for example, well docu-ented that CRP concentrations are high in patients withB, and decrease with the resolution of infection8; it is onef the most commonly used indicators to monitor treatment,lthough it has rarely been described as an early markerf treatment efficacy.9 C3 also plays an important role inB. Active complement components enhance the binding ofycobacterium tuberculosis to receptors in the surface ofuman monocytes and macrophages10 enhancing adherencend uptake by mononuclear phagocytes.11 Our findings sug-est that both CRP and C3 are unable to identify patientsith slow treatment responses at the time of starting treat-ent, but that they have some prognostic value after 1onth of therapy.

�2-macroglobulin is a zinc-binding protein. Among mul-

iple roles, it regulates extracellular proteolytic activitiesesulting from inflammation and has specific activity againstroteinases of fungal or bacterial origin. Although AMGas been reported to play a pivotal role in immune effi-

trea

mcs

AMAmAmiMaLamo

AiHpm

FLC

C

EriS

R

Legk 2004:35—40.

�1-acid glycoprotein and �1-antitrypsin as early markers of

ciency, protecting against oxidative damage,12 it does notseem to have much potential as predictor of treatmentresponses. Similarly, although zinc, selenium and copperconcentrations were abnormal, their potential to differen-tiate patients with fast and slow treatment responses isminimal.

�1-acid glycoprotein and �1-antitrypsin seem to havegood potential as markers of treatment response. AAGis involved in the induction of non-specific resistance toinfection.13,14 Its synthesis and glycosylation are regulatedby cytokines and glucocorticoids, and its reactivity in vivovaries from non-reactive, weakly reactive to strongly reac-tive depending on the degree of glycosylation. Differentglycoforms have different effects on the induction and sup-pression of cytokines such as IL-1,15 IL-1�, IL-1 receptorantagonist, and enhance the effect of lipopolysaccharidesto induce IL-1�, IL-6 and TNF expression by peripheralblood monocytes and alveolar macrophages.16 AAG there-fore has pro- and anti-inflammatory effects, inhibits humanneutrophil aggregation and neutrophil activities, includingsuperoxide generation,17 inhibits lymphocyte proliferation18

and the antibody response of B cells with both humoral andT-cell suppressive functions.13 Abnormally high concentra-tions have previously been reported in patients with TB, withtwo peaks at 1 and 21 days post-infection, in association withgranuloma formation.19 TB infection results in an increasedglycosylation of AAG, which in turn has an increased bind-ing affinity to the antimicrobials used for treatment.14 Thushigh AAG levels may affect the efficacy of anti-TB therapy14

and is theoretically possible that the high AAG concentra-tion and affinity may result in slower treatment responses.In agreement with this hypothesis, AAG concentrations con-tinued to be higher in slow responders and this is the firsttime that AAG has been identified as an early marker of slowresponse to treatment. Indeed, it seems that this potentialis already present before the start of therapy.

More is known about �1-antitrypsin than AAG regardingits role in pulmonary pathology. AAT is a glycoprotein thatblocks proteolytic enzymes such as trypsin, chymotrypsinand elastase, and correlates with poor lung function.20 Itsdeficiency has been associated with increased susceptibilityto lung disease21 and increased concentrations have beenreported in Russian children with pulmonary TB.5 Increasedlevels in patients with severe pulmonary TB at the end oftreatment are inversely correlated with decreased forcedexpiratory volumes,6 and adolescents with persistently highconcentrations at the end of 6 months of therapy had aless favourable prognosis than adolescents with decreasingconcentrations.4 Rao et al.22 reported that the AAT con-centrations of 20 adults with pulmonary TB who becamesmear negative after 1 month of treatment decreased moresharply than in 15 patients who were still smear posi-tive at this stage, suggesting that this marker could haveprognostic value. In our patients, AAT concentrations werehigher in slow responders 1 week after initiation of treat-ment and this is thus a potential early marker of treatmentresponse.

In conclusion, AAG and AAT could be used at the time of

diagnosis or the first week of treatment to identify patientswith an increased likelihood of treatment failure. Thesemarkers are easy to measure, use readily available tech-nologies and could be used singly or in combination to

tment response in patients receiving the intensive 579

onitor treatment responses in patients in high TB incidenceountries. Larger studies with longer follow-up are neces-ary for the confirmation of these findings.

uthors’ contributions: LEC and CAH conceived the study;LDA, STA, UAB and RQG collected the field data; CAH andS screened blood samples for acute phase proteins andicronutrients, respectively, and revised the manuscript;MS and LdS identified patients with TB attending the pri-ary health care centres and interpreted the laboratory

nvestigations before referring the patients to the study;AB supervised MLDA’s fieldwork and provided statisticaldvice for the final analysis and interpretation of the data;EC prepared the study protocols and together with MLDAnalysed and interpreted the data and prepared the initialanuscript. All authors read and approved the final version

f the manuscript. LEC is guarantor of the study.

cknowledgements: We are grateful to Mr Keith Bestwickn the Clinical Chemistry Laboratory of Alder Hey Children’sospital, Liverpool, UK, for the measurement of the acutehase proteins and to Eddi Roberts for measurement oficronutrients.

unding: MLDA received funding for a visiting fellowship toiverpool from the Conselho Nacional de Desenvolvimentoientifico e Tecnologico (CNPq), Brazil.

onflicts of interest: None declared.

thical approval: The study protocol was approved by theesearch ethics committees of the Liverpool School of Trop-cal Medicine, Liverpool, UK, and the Federal University ofergipe, Aracaju, Sergipe, Brazil.

eferences

1. Keeler E, Perkins MD, Small P, Hanson C, Reed S, CunninghamJ, Aledort JE, Hillborne L, Rafael ME, Girosi F, Dye C. Reducingthe global burden of tuberculosis: the contribution of improveddiagnostics. Nature 2006;444(Suppl 1):49—57.

2. Brahmbhatt S, Black GF, Carroll NM, Beyers N, Salker F, Kidd M,Lukey PT, Duncan K, van Helden P, Walzl G. Immune markersmeasured before treatment predict outcome of intensive phasetuberculosis therapy. Clin Exp Immunol 2006;146:243—52.

3. Azzurri A, Sow OY, Amedei A, Bah B, Diallo S, Peri G, BenagianoM, D’Elios MM, Mantovani A, Del Prete G. IFN-gamma-inducibleprotein 10 and pentraxin 3 plasma levels are tools for mon-itoring inflammation and disease activity in Mycobacteriumtuberculosis infection. Microbes Infect 2005;7:1—8.

4. Kaminskaia GO, Firsova VA, Ovsiankina ES, Grigor’eva ZP,Grigor’eva EV. Relationship of tuberculosis course in adoles-cents with acute phase blood proteins and genetic types ofhaptoglobin [in Russian]. Probl Tuberk 1997:36—40.

5. Kaminskaia GO, Abdullaev R, Elufimova VF, Mitinskaia LA, Iukhi-menko NV. The phenotypes of haptoglobin and the level ofacute-phasic proteins in the sera of children with local formof intrathoracic tuberculosis [in Russian]. Probl Tuberk Bolezn

6. Plit ML, Anderson R, Van Rensburg CE, Page-Shipp L, Blott JA,Fresen JL, Feldman C. Influence of antimicrobial chemother-apy on spirometric parameters and pro-inflammatory indices insevere pulmonary tuberculosis. Eur Respir J 1998;12:351—6.

5

chronic obstructive pulmonary disease. Am Rev Respir Dis 1975;111:587—93.

80

7. Wong CT, Saha N. Changes in serum proteins (albumin,immunoglobulins and acute phase proteins) in pulmonarytuberculosis during therapy. Tubercle 1990;71:193—7.

8. Koyanagi A, Kuffo D, Gresely L, Shenkin A, Cuevas LE. Relation-ships between serum concentrations of C-reactive protein andmicronutrients, in patients with tuberculosis. Ann Trop MedParasitol 2004;98:391—9.

9. Ostrowski SR, Ravn P, Hoyer-Hansen G, Ullum H, Andersen AB.Elevated levels of soluble urokinase receptor in serum frommycobacteria infected patients: still looking for a marker oftreatment efficacy. Scand J Infect Dis 2006;38:1028—32.

10. Schlesinger LS. Mycobacterium tuberculosis and the comple-ment system. Trends Microbiol 1998;6:47—9, discussion 9—50.

11. Mueller-Ortiz SL, Wanger AR, Norris SJ. Mycobacterial proteinHbhA binds human complement component C3. Infect Immun2001;69:7501—11.

12. Mocchegiani E, Costarelli L, Giacconi R, Cipriano C, MutiE, Malavolta M. Zinc-binding proteins (metallothionein andalpha-2 macroglobulin) and immunosenescence. Exp Gerontol2006;41:1094—107.

13. Hochepied T, Berger FG, Baumann H, Libert C. Alpha(1)-acidglycoprotein: an acute phase protein with inflammatory andimmunomodulating properties. Cytokine Growth Factor Rev2003;14:25—34.

14. Johnson DA, Smith KD. The efficacy of certain anti-tuberculosisdrugs is affected by binding to alpha-1-acid glycoprotein.Biomed Chromatogr 2006;20:551—60.

15. Singh VK, Fudenberg HH. Lymphocyte stimulation in vitro byorosomucoid glycoprotein. Immunol Lett 1986;14:9—13.

M.L.D. Almeida et al.

16. Boutten A, Dehoux M, Deschenes M, Rouzeau JD, BoriesPN, Durand G. Alpha 1-acid glycoprotein potentiateslipopolysaccharide-induced secretion of interleukin-1 beta,interleukin-6 and tumor necrosis factor-alpha by humanmonocytes and alveolar and peritoneal macrophages. Eur JImmunol 1992;22:2687—95.

17. Costello MJ, Gewurz H, Siegel JN. Inhibition of neutrophilactivation by alpha1-acid glycoprotein. Clin Exp Immunol1984;55:465—72.

18. Elg SA, Mayer AR, Carson LF, Twiggs LB, Hill RB, Ramakr-ishnan S. Alpha-1 acid glycoprotein is an immunosuppressivefactor found in ascites from ovaria carcinoma. Cancer 1997;80:1448—56.

19. Hernandez-Pando R, Arriaga AK, Panduro CA, Orozco EH,Larriva-Sahd J, Madrid-Marina V. The response of hepatic acutephase proteins during experimental pulmonary tuberculosis.Exp Mol Pathol 1998;65:25—36.

20. Levy H, Kalish LA, Huntington I, Weller N, Gerard C, SilvermanEK, Celedón JC, Pier GB, Weiss ST. Inflammatory markers of lungdisease in adult patients with cystic fibrosis. Pediatr Pulmonol2007;42:256—62.

21. Barnett TB, Gottovi D, Johnson AM. Protease inhibitors in

22. Rao S, Rau PVP, Sahoo RC. Decline in alpha1-antitripsin withtuberculosis treatment - correlation with bacteriological sta-tus. Tuber Lung Dis 1992;73:306—7.