Clin Immunol 2009 (Anand, Hatta)

ava i l ab l e a t www.sc i enced i rec t . com

C l i n i ca l Immuno logy

www.e l sev i e r. com/ l oca te /yc l im

Clinical Immunology (2009) xx, xxx–xxx

ARTICLE IN PRESS YCLIM-06425; No. of pages: 9; 4C: 7

Increased chitotriosidase activity in serum of leprosypatients: Association with bacillary leprosyAnand Iyer a,1, Marco van Eijk b,1, Eliane Silva c, Mochammad Hatta d,William Faber e, Johannes M.F.G. Aerts b, Pranab Kumar Das a,⁎

a Department of Pathology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlandsb Department of Medical Biochemistry, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlandsc Department of Immunology, Instituto Lauro de Souza Lima, Bauru, Brazild Department of Medical Microbiology, Hasanuddin University, Makassar, Indonesiae Department of Dermatology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands

Received 1 May 2008; accepted with revision 11 February 2009

drug therapy; NE, non-ENL BL/LL; NRBtuberculoid leprosy;M. leprae,Mycobac⁎ Corresponding author.E-mail address: [email protected]

1 Anand Iyer and Marco van Eijk cont

1521-6616/$ - see front matter © 200doi:10.1016/j.clim.2009.02.003

Please cite this article as: A. Iyer et aClin. Immunol. (2009), doi:10.1016/

Abstract Human phagocyte-specific chitotriosidase is associated with several diseases involvingmacrophage activation. Since macrophage activation plays an important role in the control ofMyco-bacterium leprae infection, we studied the association of chitotriosidase with leprosy both in serumand in situ in lesional skin biopsies from patients. Serum samples from 78 Indonesian leprosy patients(39 non-reactional and 39 reactional leprosy patients) and 36 healthy controls (HC) from the sameendemic region were investigated. The patients were classified as multibacillary (MB, n=69) orpaucibacillary (PB, n=9) based on the bacterial index in slit-skin smears. Thirty-six of the reactionalpatients had erythema nodosum leprosum (ENL), while only 3 had reversal reaction (RR). Follow-upserum samples after corticosteroid treatment were also obtained from 17 patients with ENL and onewith RR. Multibacillary (MB) patients showed increased chitotriosidase activity in serum as comparedto paucibacillary (PB) patients and healthy controls. Although no significant difference was observedbetween reactional and the corresponding non-reactional groups, ENL showed significantly higherchitotriosidase activity as compared to HC. Furthermore, corticosteroid treatment resulted insignificant decline of enzyme activity in ENL sera. Chitotriosidase activity correlated with levels ofneopterin, another macrophage activation marker, but not with IL-6, IFN-γ, TNF-α and IL-10.Immunohistochemical stainingof 6MB (LL=5, BL=1) lesional skin sections fromstoredmaterial showedpositive staining for chitotriosidase within lipid-laden macrophages suggesting that macrophages arethe source of the enzyme detected in serum. Thus, serum chitotriosidase activity is potentially usefulin distinguishing MB from PB leprosy and in monitoring response to therapy in ENL.© 2009 Elsevier Inc. All rights reserved.

KEYWORDSChitotriosidase;Leprosy;Macrophages;Neopterin

Abbreviations: BI, Bacteriological index; BB,mid-borderline; BL, borderline lepromatous; BT, borderline tuberculoid; CMI, cell-mediated immuneresponse; ENL, erythema nodosum leprosum; HC, healthy controls; LL, lepromatous leprosy; LAM, lipoarabinomannan; MB,multibacillary; MDT,multi-

, non-reactional borderline; PB, paucibacillary; PGL-I, phenolic glycolipid-I; RR, reversal reaction; TT,terium leprae.

l (P.K. Das).ributed equally to the manuscript.

9 Elsevier Inc. All rights reserved.

l., Increased chitotriosidase activity in serum of leprosy patients: Association with bacillary leprosy,j.clim.2009.02.003

mailto:[email protected]

http://dx.doi.org/10.1016/j.clim.2009.02.003


2 A. Iyer et al.

ARTICLE IN PRESS

Introduction

An important aspect of leprosy is its unique spectralpathology [1–3]. At one pole of the spectrum lepromatousleprosy (LL) presents with disseminated lesions character-ized histopathologically by a predominant macrophageinfiltrate with foamy changes and a high load of acid-fastMycobacterium (M.) leprae [3]. Tuberculoid (TT) leprosy, atthe other end of the spectrum, has a limited distribution,with lesions showing granulomas of epithelioid macrophagesand multi-nucleated giant cells surrounded by a cuff oflymphocytes, with no detectable M. leprae [3]. During thechronic course of leprosy, sudden increases in immuneactivity may occur, which are called reactions. These areeither type I (reversal reaction, RR), due to an acute increasein the cell-mediated immune response (CMI), or type II(erythema nodosum leprosum, ENL), described as an immunecomplex mediated disease with involvement of the CMI.

Early studies provided evidence for a defect in the CMI inLL using an in vitro lymphocyte transformation assay [4]. Astrong CMI response to M. leprae was however found in TTand contacts of leprosy patients (except those of LL patientstreated less than 6 months) [5]. Since the CMI response is theprimary line of defence against intracellular pathogens likemycobacteria, its defect in LL patients has importantconsequences for disease progression. In TT patients, thestrong CMI limits the spread of M. leprae, but sometimesresults in damage to nerves as a bystander effect [3].

It is ironical that the macrophage, which is an importantcomponent of the CMI response, is also one of the preferredhost cells for M. leprae [6]. Different survival mechanismshave been described forM. leprae, either through a defectivehost macrophage or an active subversion of the host defences[7–9]. Indirect evidence, however, suggests that macrophageactivation is not completely defective in lepromatous leprosypatients since elevated levels of neopterin, a marker ofmacrophage activation, are observed in sera of LL–borderlinelepromatous (BL) patients and leprosy reactions, as com-pared to healthy controls [10–12].

Recent studies have shown that human phagocyte-specific chitotriosidase, the first discovered mammalianchitinase, is an important component of the innate immuneresponse against fungal pathogens [13]. Chitotriosidase is anendoglucosaminidase belonging to family 18 of glycosylhy-drolases and cleaves chitin [14]. The enzyme was discoveredin sera of Gaucher patients, who lack the lysosomal enzymeacid β-glucocerebrosidase and therefore are unable todegrade the glycosphingolipid glucosylceramide within the

Table 1 Characteristics of the 6 MB patients used for immunohis

Patient Classification Reaction B.I.

A1 LL ENL n.a.A2 LL ENL n.a.BR15 LL ENL 4BR39 BL – 4BR43⁎ LL – 5BR47 LL ENL 5

Abbreviations used are LL — lepromatous leprosy; BL — borderline lepravailable. The ⁎ refers to the representative immunostainings depicted

Please cite this article as: A. Iyer et al., Increased chitotriosidase activClin. Immunol. (2009), doi:10.1016/j.clim.2009.02.003

lysosomes [14–16]. Chitotriosidase serves as a crucialmacrophage-derived biomarker to monitor disease onset,progression and therapeutic response in Gaucher disease[17]. Increased chitotriosidase activity has been found withinatherosclerotic lesions, in which cholesterol-laden foam cellsreside [18]. Elevated serum chitotriosidase activity has beenreported in malaria and sarcoidosis as well, but not inpulmonary tuberculosis, a mycobacterial disease [19–21].However, chitotriosidase activity has been reported intuberculous pleural effusions, suggesting that chitotriosi-dase production maybe local in these patients [22].

Since macrophage activation plays an important role inthe control ofM. leprae infection, we studied the associationof chitotriosidase with leprosy both in serum and in situ inlesional skin biopsies from patients.

Materials and methods

Patients and controls

The study included 78 leprosy patients of which 72attended the leprosy clinic at the Hasanuddin Universityhospital in Makassar, Indonesia. Thirty-six normal healthycontrols (HC) residing in the same area as the patientswere also included in the study. The study was approved bythe ethical committee of the Hasanuddin University andinformed consent was obtained from the patients and HC.In addition, serum and skin biopsy specimen were alsoavailable from 6 patients from archival material at theAcademic Medical Center, Amsterdam, The Netherlands,which were used for concomitant immunohistochemistry tostudy the localization of chitotriosidase within the immunecells and chitotriosidase activity within serum (Table 1).The median age of the patients was 29 years (range: 14–80 years) and included 51 males and 27 females. Themedian age of the HC was 28 years (range: 19–41 years)and included 28 males and 8 females.

Every patient was clinically assessed by detailed history,medical and dermatological examinations. Bacteriologicalexamination of slit-skin smears was carried out to deter-mine the bacteriological index (BI). The patients wereclassified according to Ridley and Jopling's five sub-groupclassification as 23 lepromatous (LL), 43 borderline lepro-matous (BL), 3 mid-borderline (BB), 2 borderline tubercu-loid (BT) and 7 tuberculoid (TT) patients [1]. Patients with aBIN0 were further grouped as multibacillary (MB, n=69)and included all the LL, BL and BB patients, whereas those

tochemical analysis.

Chitotriosidase (nmol/ml* h) Immunostaining

396 +115.3 +531.3 +136.2 +251.4 +456 +

omatous; ENL — erythema nodosum leprosum and; n.a. — notin Figure 3.

ity in serum of leprosy patients: Association with bacillary leprosy,


3Chitotriosidase activity in leprosy serum

ARTICLE IN PRESS

with BI=0 were grouped as paucibacillary (PB, n=9) andincluded the BT and TT patients. Thirty-nine of thepatients were diagnosed with reactions of which 36 hadtype II/erythema nodosum leprosum (ENL; LL=17, BL=19)and 3 had type I/reversal reaction (RR; BB=2, BT=1). ENLwas diagnosed by the acute appearance of nodular skinlesions, accompanied by fever with or without peripheralnerve pain and nerve dysfunction. RR reactions typicallypresented as an acute inflammation of pre-existing lesionsand/or onset of new erythematous skin lesions. For thepurpose of comparisons the BL and LL patients without ENLreactions were grouped together as NE (non-ENL BL/LL,n=34) since this group of patients is prone to ENL. Similarly,BL, BB and BT patients without RR were grouped as NRB(non-reactional borderline, n=25) since this group is proneto RR. Thus the BL patients were common to both NE and NRBgroups since potentially they might develop either ENL or RR[10]. The serum profiles of these groups were compared withENL and RR patients respectively.

Leprosy was treated with MDT according to World HealthOrganization guidelines [23]. The study group thus consistedof 30 untreated and 48 treated patients. Reactions weretreated using prednisolone starting at 40 mg/day andgradually tapering off over a period of 12 weeks [23]. Inthe absence of biopsy samples in the Indonesian patients, theinitial clinical assessment as well as assessment of improve-ment was done by the clinician-in-charge. Clinical improve-ment of reactions was defined as complete subsidence of allreactional symptoms. Follow-up samples at the end ofcorticosteroid treatment were obtained from 17 patientswith ENL and 1 patient with RR.

After informed consent, blood samples were collected byvenipuncture, the serum was separated, aliquoted andstored in liquid nitrogen at Makassar until transported tothe Netherlands for analysis. No concomitant biopsy speci-men could be obtained from the Indonesian patients.

Cell culture

Monocytes were isolated and cultured as described pre-viously [13]. Briefly, the cells were cultured for 7 days in RPMI1640 (BioWhittaker, Viviers, Belgium) supplemented with10% human serum (HS) (BioWhittaker) to allow for chito-triosidase induction in the presence or absence of dexa-methasone at the following pre-determined optimalconcentration of 2 nM (Sigma-Aldrich Chemie BV, TheNetherlands). CD163 cell surface expression was determinedby flow cytometry according to standard procedures on aFACS Calibur and data were analyzed using the CellQuest Prosoftware (BD Biosciences, San Jose, CA, USA).

Chitotriosidase assay

Chitotriosidase enzyme activity was determined in sera ofleprosy patients, or in cell lysates, using the fluorescentsubstrate 4-methyl umbeliferyl (4-MU) β-D-N,N′,N″-(MU)-triacylchitotriose [24]. Activities were calculated as nano-moles per millilitre per hour (nmol/ ml⁎ h). Due to theabsence of a cDNA source of the leprosy patients andcontrols we were not able to screen for chitotriosidasedeficiency and have decided to use actual enzyme


activities. To exclude chitotriosidase deficient individualsfrom our analysis we did not take into account enzymeactivities below 5 nmol/ml* h.

Cytokine and soluble serum marker assays

The measurement of cytokines IL-6, IL-10, IFN-γ TNF-α,cytokine receptor (sIL-6R), sCD27 and neopterin has beendescribed previously [10,25].

Anti-PGL-I IgM assay

Anti-PGL-I IgM antibodies were detected as described inBrett et al. 1986 using natural tri-saccharide linked to bovineserum albumin via a phenolic ring (NT-p-BSA) as a semi-synthetic analogue of PGL-I [26].

Immunohistological techniques

Formalin-fixed, paraffin embedded tissues (6 μm) werestained by the Haematoxylin-Eosin (HE) technique for alight microscopic examination of the tissue morphology.The immune infiltrate was characterized by immunohisto-chemical staining as described earlier [27]. Briefly, thesections were deparaffinized and preincubated with 3%H2O2 in 0.01% sodium azide to inhibit endogenous perox-idase activity. Antigen retrieval for CD68 staining wascarried out by treating the slides with 0.25% pepsin in0.01 M HCl for 10 min at 37 °C. The sections were blockedwith normal goat serum. The antibodies used for thestainings were CD68 (Dako, Denmark), mycobacteriallipoarabinomannan (LAM, clone F30-5, from Dr. A. Kolk,KIT Biomedical Center, The Netherlands), M. lepraespecific phenolic glycolipid-I (PGL-I, clone 47-21, from Dr.A. Kolk, KIT Biomedical Center, The Netherlands) andchitotriosidase (mouse monoclonal clone CH1, IgG2b).Primary mouse monoclonal antibodies were applied to thespecimen followed by incubation with biotinylated rabbitanti-mouse Ig and subsequently horse radish peroxidase(HRP) labelled streptavidin. The colour reaction wasdeveloped using 3 amino-9 ethyl carbazole (AEC) as asubstrate. The sections were counterstained with haema-toxylin. The percentage of the various cellular populationsin the lesions was determined by light microscopy in threerandomly selected fields of sequential sections.

The double stainings were carried out to study thelocalization of chitotriosidase in relation to the macro-phages. The staining methods were similar to that used byVerhagen et al. [27]. The antibodies used in the stainingwere chitotriosidase (mouse monoclonal clone CH1, IgG2b)and the HAM-56 clone (Dako, Denmark) to stain macro-phages (since staining for CD68 on paraffin sections requiredantigen retrieval, which might affect the staining forchitotriosidase).

Statistical analysis

Since the data did not follow Gaussian distribution, theMann–Whitney U test was performed to compare levels ofthe serum markers and chitotriosidase activity between MBand PB patients and HC. Moreover, since NE and NRB



4 A. Iyer et al.

ARTICLE IN PRESS

patient groups are prone to ENL and RR respectively, ENLwas compared with NE while RR was compared with NRBand the differences in levels of serum markers andchitotriosidase activity were analyzed using the Mann–Whitney U test. Chitotriosidase activity in sera of patientsbefore and after treatment of ENL was compared by theWilcoxon Rank Test for paired samples. Correlationsbetween different serum markers and chitotriosidaseactivity were analyzed using the Spearman's rank correla-tion coefficient. A correlation was assumed when rN0.4with Pb0.05.

Figure 1 Chitotriosidase activity in the sera of leprosy patientschitotriosidase activity than HC (▴) (A). No significant differenc[multibacillary, MB (n); paucibacillary, PB (▾)] and multi-drug (MDT)showed increased chitotriosidase activity as compared to HC (●) (compared to PB leprosy (▴) or HC (▾) (C). No significant difference ipatients. ENL showed significantly higher chitotriosidase activity cochitotriosidase activity than NRB (▾) patients, although the sample


Results

Increased chitotriosidase activity in serum ofleprosy patients

Serum chitotriosidase activity was significantly elevated inleprosy patients as compared to healthy controls from thesame leprosy endemic area (Pb0.0001; Fig. 1A). The patientswere classified as MB, PB, ENL, RR, NE and NRB as describedin detail in materials and methods. Previously we have shown

and healthy controls (HC). Leprosy patients (n) showed highere in chitotriosidase activity was observed between untreatedtreated [MB (▴); PB (O)] patients. Both untreated and treated MBB). MB leprosy (n) displayed higher chitotriosidase activity asn chitotriosidase activity was noted between ENL (▴) and NE (n)mpared to HC (●). However, RR (O) showed significantly lowersize for RR patients is too small to draw firm conclusions (D).




ARTICLE IN PRESS

that multi-drug therapy (MDT) status did not significantlychange levels of serum cytokines and receptors [10].Similarly, in the present study, statistical analysis revealedthat the MDT status did not significantly affect chitotriosi-dase activity in the patients (Fig. 1B). Hence, phenotypicpatient groups were assigned irrespective of their MDTstatusto have statistically significant numbers of patients withineach group. Activity of serum chitotriosidase was signifi-cantly higher in MB (mean: 207.3±371.5) as compared to HC(mean: 18±12.7) (Pb0.0001), but no difference was foundbetween PB (mean: 27.9±26.1) and HC. A significantdifference in chitotriosidase activity was also observedbetween MB and PB patients (P=0.011) (Fig. 1C). Whereasserum chitotriosidase activity was elevated in ENL (mean:274.5±484.4) (Pb0.0001) as compared to HC, no significantdifference was noted between RR (mean: 12±10.6) and HC.On the other hand, while RR showed significantly lowerchitotriosidase activity as compared to NRB (mean: 143.5±151.6) (P=0.031), no difference was noted between NE(mean: 134.5±139) and ENL (Fig. 1D). The latter observationshould be interpreted with caution however, because thenumber of RR patients in the study is low.

Serum samples from 17 patients with ENL and 1 with RRwere available at the onset of reaction and at the end ofcorticosteroid treatment. Chitotriosidase activity wasobserved to decline significantly following corticosteroidtreatment in ENL (mean: 68.4±105.7) (P=0.002) (Fig. 2A)and in the single RR patient (data not shown) and was notsignificantly different from the activity observed in HC.Furthermore, we found that in vitro polarization of humanmonocytes towards macrophages with the corticosteroiddexamethasone prevented chitotriosidase induction (Fig.

Figure 2 Effect of corticosteroid treatment on chitotriosidase actreatment with prednisolone in comparison with HC (▾). Chitotriosidobserved in HC (A). Chitotriosidase activity during dexamethasone-inChitotriosidase activity was suppressed (B) concomitantly with induc


2B). The dexamethasone-driven polarization of the macro-phages was confirmed, for instance, by analysis of theinduction of CD163 at the cell surface (Fig. 2C).

Chitotriosidase activity correlates with neopterin inleprosy serum

The results of a cross-sectional analysis of serum focussing oncytokines and soluble cellular markers have been describedelsewhere and have been summarized in Table 2 [10]. Here,we analyzed whether serum chitotriosidase activity corre-lated with any of the soluble markers of the above mentionedcross-sectional analysis. Serum chitotriosidase activity cor-related well with another macrophage activation product,namely neopterin (r=0.48, P=0.0003). Chitotriosidase activ-ity did not correlate with the cytokines IL-6 (r=0.02), IL-10(r=0.19), IFN-γ (r=0.16), TNF-α (r=−0.10), the solublecytokine receptor sIL-6R (r=0.21) or sCD27 (r=0.10). NeitherBI (r=0.36) nor levels of anti-PGL-I IgM antibodies (r=0.11)correlated with chitotriosidase activity in sera of thepatients.

Immunohistochemical analysis of chitotriosidase intissues of MB and PB patients

As chitotriosidase is expressed by human macrophages andalso stored in specific granules of polymorphonuclearneutrophils, immunohistochemical staining was performedto identify the source of the observed increase in chito-triosidase activity in leprosy patients [13]. Positive stainingfor chitotriosidase protein, which appeared to be localized

tivity. Chitotriosidase activity in ENL before (n) and after (▴)ase activity declined on prednisolone treatment of ENL to levelsduced polarization of human monocytes towards macrophages.tion of CD163 (C) at the macrophage cell surface.



Table2

Mea

nleve

lsof

cytokine

s,cytokine

rece

ptor,ce

llular

activa

tion

prod

ucts,PG

L-Ia

ndch

itotriosidaseac

ross

thediseasegrou

ps.

Classification

IL-6

pg/m

lsIL-6R

ng/m

lIL-10pg

/ml

IFN-γ

pg/m

lTN

F-αpg

/ml

sCD27

U/m

lNeo

pterin

(nmol/l)

Chitotriosida

se(nmol/m

lh)

anti-PGL-IIgM

O.D.

at49

2nm

.

MB(n

=69

)96

0.1(±17

70)

43.4

(±26

.7)

43.5

(±84

.7)

6.4(±6.5)

82.7

(±13

5.8)

173.7(±10

9.9)

25.9

(±18

.7)

207.3(±37

1.5)

0.26

7(±0.49

0)PB

(n=9)

36(±48

.6)

46.7

(22.7)

6.6(±13

.1)

7.8(±5.2)

59(±12

4.5)

159.7(±36

.3)

12.6

(±7.2)

27.9

(±26

.1)

0.05

0(±0.07

1)NE(n

=34

)10

01(±19

62)

35.7

(±13

.7)

72.9

(±13

9.7)

3.5(±16

)77

.3(±14

5.6)

296.1(±17

6.6)

27.3

(±23

.4)

134.5(±13

9)0.22

7(±0.37

7)EN

L(n

=36

)82

1.4(±18

72)

47.8

(±26

.1)

42.4

(±76

.5)

5.8(±6.5)

55.2

(±75

.9)

266.7(±18

7.5)

22.5

(±13

.9)

274.5(±48

4.4)

0.31

7(±0.61

4)NRB

(n=25

)88

4.4(±20

12)

35.8

(±17

.5)

80.3

(±15

4.4)

4(±17

.4)

51(±10

9.1)

286.2(±17

2.2)

24.1

(±19

.6)

143.5(±15

1.6)

0.23

3(±0.35

6)RR

(n=3)

988.2(±13

00)

50.9

(±33

.7)

22.9

(±41

.6)

6.7(±6.5)

149.9(231

)20

5(±57

)30

.6(±24

.6)

12(±10

.6)

0.15

9(±0.22

1)HC(n

=36

)27

.8(±12

2.2)

61.1

(±40

.3)

9.7(±32

.5)

0(±2.5)

9.8(±82

.8)

27.8

(±12

2.1)

5.9(±1.9)

18(±12

.7)

0.01

(±0.00

9)

Dataarede

picted

asthemea

nof

thedifferen

tpa

rametersmea

suredwithS.D.indica

tedbe

low.Abb

reviations

used

aremultiba

cillary(M

B),pa

ucibac

illary(PB),no

n-EN

LBL

/LL(NE),erythe

ma

nodo

sum

leprosum

(ENL)

non-reac

tion

albo

rderlin

e(NRB

),reve

rsal

reac

tion

(RR)

andhe

althyco

ntrols(H

C).

6 A. Iyer et al.

ARTICLE IN PRESS

Please cite this article as: A. Iyer et al., Increased chitotriosidaseClin. Immunol. (2009), doi:10.1016/j.clim.2009.02.003

activ

within macrophages, was seen in lesional skin biopsies from 6MB patients (Figs. 3A and E). However, detailed analysis ofserially stained sections showed that only a sub-population ofthe CD68+ cells expressed of chitotriosidase protein, suggest-ing that chitotriosidase activity was only upregulated in asubset of macrophages. This positive staining in tissuesections corresponded to the high serum levels of theenzyme seen in these patients as depicted in Table 1.Immunohistochemical double staining of corresponding serialsections with antibodies to chitotriosidase and the macro-phage-specific antibody HAM56 showed co-localization,confirming the macrophage localization of the enzyme(Fig. 3B). This suggests that the macrophages, present inthe skin, are the source of the chitotriosidase activitydetected in serum of leprosy patients. To study thelocalization of M. leprae associated antigens in relation tochitotriosidase, serial tissue sections were also stained formycobacterial LAM (Fig. 3C) and the M. leprae specificglycolipid PGL-I (Fig. 3D). Positive staining for both antigenswas seen in foam cells, which were also positive for themacrophage marker CD68 in serial sections (Fig. 3E),suggesting that these antigens are also localized withinmacrophages.

Discussion

Human phagocyte-specific chitotriosidase activity is asso-ciated with several diseases involving macrophage activationand is a valuable tool for monitoring the efficacy of therapyin Gaucher disease [14,15,28,29]. The crucial role of themacrophage in leprosy prompted us to look at chitotriosidaseactivity in the sera of leprosy patients and healthy controlsmainly from leprosy endemic areas in Indonesia. The presentreport is, to our knowledge, the first one to describe anassociation of serum chitotriosidase activity with leprosy.

Serum chitotriosidase activity was elevated in MB ascompared to PB patients or healthy controls in the presentstudy. MB leprosy is characterized by an abundant macro-phage infiltrate with many macrophages showing foamymorphology depending on the chronicity of the disease [1,3].Representative immunohistochemical staining of skin biop-sies from 6 MB patients (LL/BL) with high serum chitotrio-sidase activity showed positivity for chitotriosidase, whichwas localized within the macrophages (Figs. 3A, B). Thissuggested that these macrophages are the active producersof chitotriosidase activity detected in MB serum. Detailedanalysis showed only a subset of the CD68+ macrophages tobe positive for chitotriosidase. Similar observations have alsobeen reported for foamy macrophages in atheroscleroticplaques, where only a sub-population of the cells are positivefor chitotriosidase [18]. Chitotriosidase activity was higher inENL as compared to HC although no significant differencewas observed between ENL and non-ENL MB, which suggeststhat chitotriosidase activity is independent of the reactionstate of the patient. However, surprisingly, RR patientsshowed a trend towards decreased chitotriosidase activity ascompared to NRB patients. This apparently contradictoryresult may reflect the inherently different nature of the tworeactions. RR is associated with an increased CMI whereasENL is thought to be an immune complex disease with someinvolvement of the CMI. However, this result needs to be



Figure 3 Immunohistochemical staining of lesional skin from a MB patient (BR43). Single staining for chitotriosidase (A); doublestaining for chitotriosidase (blue) and macrophages (HAM56) (red),→ shows double stained cells (B); single staining for LAM (C); PGL-I(D) and CD68 (E). Magnifications 20×.


ARTICLE IN PRESS

interpreted cautiously on account of the low number of RRpatients in the study. Interestingly, MDT-treated patientsshowed a bimodal distribution with respect to chitotriosidaseactivity in serum (Fig. 1B). Analysis of the clinical detailsshowed that patients with low chitotriosidase values had aBI≤2, suggesting an association of low chitotriosidaseactivity with low bacterial load in the patient. However,the opposite was not true, i.e. not all patients with BI≤2 hadlow chitotriosidase values in serum.

Recently, it has been suggested that chitotriosidase RNAexpression levels, derived from Kupffer cells in the liver fromindividuals suffering from non-alcoholic steatohepatitis andsimple steatosis, correlate amongst others with TNF-α RNAexpression [30]. In the present study, serum chitotriosidaseactivity did not correlate at all with serum cytokines (IL-6, IL-


10, IFN-γ, TNF-α), the IL-6 receptor, or the T-cell activationmarker CD27, in leprosy patients. The lack of associationbetween chitotriosidase and other indicators of inflamma-tion suggested that chitotriosidase and cytokine responsepathways could be regulated differently. Possibly, chitotrio-sidase induction is a response to lipid accumulationspecifically within the macrophages [31] irrespective of thecytokine environment in the surrounding tissue. A positivecorrelation was observed, however, between chitotriosidaseactivity and serum levels of neopterin, another macrophageactivation associated molecule [32]. Elevated levels of serumneopterin were previously reported in leprosy patients ingeneral and in particular in reactions [10–12]. As observedfor chitotriosidase, elevated levels of serum neopterin werealso reported in diseases associated with macrophage



8 A. Iyer et al.

ARTICLE IN PRESS

activation like malaria, tuberculosis and coronary arterydisease [17–20,33–36]. Serum neopterin levels were sig-nificantly higher in sarcoidosis patients with progressivedisease (Stage II) as compared to patients with no indicationsfor corticosteroid therapy (Stage 0 or Stage I) [37]. Similarly,increased chitotriosidase activity was observed in Stage II andStage III sarcoidosis patients in a separate study [20]. Serumchitotriosidase activity also correlated with neopterin levelsin Gaucher patients although it was weaker as compared tothe correlation with angiotensin converting enzyme (ACE),adenosine deaminase (ADA) and β-hexosaminidase respec-tively [38]. Interestingly, in the present study, serumchitotriosidase activity did not correlate with either anti-PGL-I IgM levels or with the BI of the patients. This suggestsalso that the chitotriosidase activity does not reflect theactual bacterial load, but is a measure of accumulation of thelipid antigens such as LAM and PGL-I (Figs. 3C, D) in themacrophages. This may in turn explain why chitotriosidaseactivity persists even after completion of MDT in the patients,since M. leprae antigens such as LAM and PGL-I are known topersist even after completion of MDT in the patients [27].However, in the absence of functional data, these suggestionsremain speculative.

Chitotriosidase activity declined in patients with ENL ontreatment with prednisolone. This was supported by in vitrodata showing lack of induction of chitotriosidase duringdexamethasone-induced polarization of monocytes tomacrophages. Our previous study with the same patientgroup showed a decline in serum IFN-γ, TNF-α and sIL-6R butnot neopterin on corticosteroid treatment of ENL [10]. Thissuggests that although chitotriosidase and neopterin areboth products of macrophage activation, regulation of theirexpression may be different as also reported elsewhere [38].In this regard, neopterin has been proposed as a marker of anactive CMI response caused by IFN-γ produced by T-cells [32].This will induce a classically activated M1 macrophagephenotype. On the other hand, chitotriosidase is elevatedin response to lipid accumulation within alternativelyactivated M2 macrophages as is seen in Gaucher disease [31].

We have emphasised in an earlier report that a majorlimitation of analysis of cytokines and activation markers isthat serum measurements may not adequately reflect thetissue immune response [10,27]. However, in the presentreport, high serum levels of chitotriosidase corresponded topositive staining for the enzyme within foamy macrophagesin the MB leprosy patients. Cytokines and other solublemarkers studied in the present report are a reflection of amore general immune-mediated condition, involving variouscell types [10]. In contrast, both neopterin and chitotriosi-dase are macrophage-derived and specifically elevated indiseases involving macrophage activation and lipid accumu-lation such as leprosy [15,17–20,33,34]. Nevertheless, theassociation of these markers with disease should be inter-preted with caution. Still, neopterin and chitotriosidase areuseful in distinguishing MB from PB leprosy patients fortreatment purposes. Moreover, chitotriosidase, but notneopterin, may be a useful prognostic marker to monitorthe response to corticosteroid therapy in ENL.

The present study does not shed any light on the role ofchitotriosidase in leprosy pathology. In this respect, van Eijket al. previously demonstrated the anti-fungal activity ofhuman chitotriosidase in vitro and in vivo in mouse models of


systemic Candidiasis and systemic Aspergillosis [13]. Anotherreport observed that deficiency in chitotriosidase was alsoassociated with infections with Wuchereria bancrofti, thecausative agent of filariasis [39]. This led to the suggestionthat human chitotriosidase is a component of the innateimmunity involved in protection against chitin containingpathogens. However,M. leprae is not known to contain chitinin its cell wall/membrane fraction [40]. It is interesting tonote that variants of chitotriosidase affecting its activity areassociated with Gram-negative bacteremia in childrenundergoing therapy for acute myeloid leukemia, suggestingmore pleiotropic effects of chitotriosidase [41]. Alterna-tively, increased chitotriosidase in leprosy may be an indirectphenomenon related to lipid overloading of lysosomes inmacrophages. However, in the context of the present study,these suggestions remain purely speculative and need to bestudied further at a functional level.

In conclusion, serum chitotriosidase activity was observedto be associated with multibacillary leprosy and may bepotentially useful in monitoring response to therapy in ENLreactions.

Acknowledgments

George Gussenhoven is acknowledged for performing theanti-PGL-I antibody ELISAs. Dr. Michael Tank is gratefullyacknowledged for his help with the statistical analysis. TheRoyal Netherlands Academy of Sciences (KNAW), the Q.M.Gastmann Wichers Stichting and the Netherlands are kindlyacknowledged for their maintenance support for Anand Iyer.The study will form a part of the Ph.D. thesis of Anand Iyer atthe University of Amsterdam, the Netherlands.

References

[1] D.S. Ridley, W.H. Jopling, Classification of leprosy according toimmunity. A five-group system, Int. J. Lepr. Other Mycobact.Dis. 34 (1966) 255–273.

[2] S.L. Walker, D.N. Lockwood, Leprosy, Clin. Dermatol. 25 (2007)165–172.

[3] S.L. Walker, D.N. Lockwood, The clinical and immunologicalfeatures of leprosy, Br. Med. Bull. 77–78 (2006) 103–121.

[4] T. Godal, B. Myrvang, S.S. Froland, J. Shao, G. Melaku, Evidencethat themechanism of immunological tolerance (“central failure”)is operative in the lack of host resistance in lepromatous leprosy,Scand. J. Immunol. 1 (1972) 311–321.

[5] T. Godal, K. Negassi, Subclinical infection in leprosy, Br. Med. J.3 (1973) 557–559.

[6] T.J. Birdi, N.H. Antia, The macrophage in leprosy: a review onthe current status, Int. J. Lepr. Other Mycobact. Dis. 57 (1989)511–525.

[7] L.D. Sibley, J.L. Krahenbuhl, Defective activation of granulomamacrophages from Mycobacterium leprae-infected nude mice,J. Leukoc. Biol. 43 (1988) 60–66.

[8] E. Caron, A. Hall, Identification of two distinct mechanisms ofphagocytosis controlled by different Rho GTPases, Science 282(1998) 1717–1721.

[9] A.C. Moura, M. Mariano, Lipids from Mycobacterium leprae cellwall suppress T-cell activation in vivo and in vitro, Immunology92 (1997) 429–436.

[10] A. Iyer, M. Hatta, R. Usman, S. Luiten, L. Oskam, W. Faber, A.Geluk, P. Das, Serum levels of interferon-gamma, tumournecrosis factor-alpha, soluble interleukin-6R and soluble cell




ARTICLE IN PRESS

activation markers for monitoring response to treatment ofleprosy reactions, Clin. Exp. Immunol. 150 (2007) 210–216.

[11] F.F. Hamerlinck, P.R. Klatser, D.S. Walsh, J.D. Bos, G.P. Walsh,W.R. Faber, Serum neopterin as a marker for reactional statesin leprosy, FEMS Immunol. Med. Microbiol. 24 (1999) 405–409.

[12] W.R. Faber, A.M. Iyer, T.T. Fajardo, T. Dekker, L.G. Villaher-mosa, R.M. Abalos, P.K. Das, Serial measurement of serumcytokines, cytokine receptors and neopterin in leprosy patientswith reversal reactions, Lepr. Rev. 75 (2004) 274–281.

[13] M. van Eijk, C.P. van Roomen, G.H. Renkema, A.P. Bussink, L.Andrews, E.F. Blommaart, A. Sugar, A.J. Verhoeven, R.G. Boot,J.M. Aerts, Characterization of human phagocyte-derivedchitotriosidase, a component of innate immunity, Int. Immunol.17 (2005) 1505–1512.

[14] A.P. Bussink, D. Speijer, J.M. Aerts, R.G. Boot, Evolution ofmammalian chitinase(-like) members of family 18 glycosylhydrolases, Genetics 177 (2007) 959–970.

[15] C.E. Hollak, S. van Weely, M.H. van Oers, J.M. Aerts, Markedelevation of plasma chitotriosidase activity. A novel hallmark ofGaucher disease, J. Clin. Invest. 93 (1994) 1288–1292.

[16] A.P. Bussink, M. van Eijk, G.H. Renkema, J.M. Aerts, R.G. Boot,The biology of the Gaucher cell: the cradle of humanchitinases, Int. Rev. Cytol. 252 (2006) 71–128.

[17] J.M. Aerts, C.E. Hollak, M. van Breemen, M. Maas, J.E. Groener,R.G. Boot, Identification and use of biomarkers in Gaucherdisease and other lysosomal storage diseases, Acta Paediatr.,Suppl. 94 (2005) 43–46 discussion 37–8.

[18] R.G. Boot, T.A. van Achterberg, B.E. vanAken, G.H. Renkema,M.J.Jacobs, J.M. Aerts, C.J. de Vries, Strong induction of members ofthe chitinase family of proteins in atherosclerosis: chitotriosidaseand human cartilage gp-39 expressed in lesion macrophages,Arterioscler. Thromb. Vasc. Biol. 19 (1999) 687–694.

[19] R. Barone, J. Simpore, L. Malaguarnera, S. Pignatelli, S.Musumeci, Plasma chitotriosidase activity in acute Plasmo-dium falciparum malaria, Clin. Chim. Acta 331 (2003) 79–85.

[20] S. Grosso, M.A. Margollicci, E. Bargagli, Q.R. Buccoliero, A.Perrone, D. Galimberti, G. Morgese, P. Balestri, P. Rottoli,Serum levels of chitotriosidase as a marker of disease activityand clinical stage in sarcoidosis, Scand. J. Clin. Lab. Invest. 64(2004) 57–62.

[21] E. Bargagli, M. Margollicci, N. Nikiforakis, A. Luddi, A. Perrone,S. Grosso, P. Rottoli, Chitotriosidase activity in the serum ofpatients with sarcoidosis and pulmonary tuberculosis, Respira-tion 74 (2007) 548–552.

[22] L. Bouzas, E. San Jose, J.C. Tutor, Chitotriosidase activity inpleural effusions, Clin. Lab. 53 (2007) 449–452.

[23] S.N. Marlowe, R.A. Hawksworth, C.R. Butlin, P.G. Nicholls, D.N.Lockwood, Clinical outcomes in a randomized controlled studycomparing azathioprine and prednisolone versus prednisolonealone in the treatment of severe leprosy type 1 reactions inNepal, Trans. R. Soc. Trop. Med. Hyg. 98 (2004) 602–609.

[24] G.H. Renkema, R.G. Boot, A.O. Muijsers, W.E. Donker-Koop-man, J.M. Aerts, Purification and characterization of humanchitotriosidase, a novel member of the chitinase family ofproteins, J. Biol. Chem. 270 (1995) 2198–2202.

[25] P. Mayersbach, R. Augustin, H. Schennach, D. Schonitzer, E.R.Werner, H. Wachter, G. Reibnegger, Commercial enzyme-linkedimmunosorbent assay for neopterin detection in blood donationscompared with RIA and HPLC, Clin. Chem. 40 (1994) 265–266.

[26] S.J. Brett, S.N. Payne, J. Gigg, P. Burgess, R. Gigg, Use ofsynthetic glycoconjugates containing the Mycobacterium


leprae specific and immunodominant epitope of phenolicglycolipid I in the serology of leprosy, Clin. Exp. Immunol. 64(1986) 476–483.

[27] C. Verhagen, W. Faber, P. Klatser, A. Buffing, B. Naafs, P. Das,Immunohistological analysis of in situ expression of mycobac-terial antigens in skin lesions of leprosy patients across thehistopathological spectrum. Association of mycobacteriallipoarabinomannan (LAM) and Mycobacterium leprae phenolicglycolipid-I (PGL-I) with leprosy reactions, Am. J. Pathol. 154(1999) 1793–1804.

[28] L. Malaguarnera, Chitotriosidase: the yin and yang, Cell. Mol.Life Sci. 63 (2006) 3018–3029.

[29] J. Kzhyshkowska, A. Gratchev, S. Goerdt, Human chitinases andchitinaselike proteins as indicators for inflammation andcancer, Biomarker Insights 2 (2007) 128–146.

[30] L. Malaguarnera, M. Di Rosa, A.M. Zambito, N. dell'Ombra, F.Nicoletti, M. Malaguarnera, Chitotriosidase gene expression inKupffer cells from patients with non-alcoholic fatty liverdisease, Gut 55 (2006) 1313–1320.

[31] L.A. Boven, M. van Meurs, R.G. Boot, A. Mehta, L. Boon, J.M.Aerts, J.D. Laman, Gaucher cells demonstrate a distinctmacrophage phenotype and resemble alternatively activatedmacrophages, Am. J. Clin. Pathol. 122 (2004) 359–369.

[32] C. Murr, B. Widner, B. Wirleitner, D. Fuchs, Neopterin as a markerfor immune systemactivation, Curr. DrugMetab. 3 (2002) 175–187.

[33] G. Reibnegger, V. Boonpucknavig, D. Fuchs, A. Hausen, E.Schmutzhard, H. Wachter, Urinary neopterin is elevated inpatients with malaria, Trans. R. Soc. Trop. Med. Hyg. 78 (1984)545–546.

[34] A.E. Brown, P. Teja-Isavadharm, H.K. Webster, Macrophageactivation in vivax malaria: fever is associated with increasedlevels of neopterin and interferon-gamma, Parasite Immunol.13 (1991) 673–679.

[35] C. Immanuel, R. Swamy, M. Kannapiran, S. Vijayalakshmi, V.Sundaram, K. Jagannath, C.N. Paramasivan, Neopterin as amarker for cell-mediated immunity in patients with pulmonarytuberculosis, Int. J. Tuberc. Lung Dis. 1 (1997) 175–180.

[36] V. Videm, R. Wiseth, S. Gunnes, H.O. Madsen, P. Garred,Multiple inflammatory markers in patients with significantcoronary artery disease, Int. J. Cardiol. 118 (2007) 81–87.

[37] M.W. Ziegenhagen, U.K. Benner, G. Zissel, P. Zabel, M. Schlaak,J. Muller-Quernheim, Sarcoidosis: TNF-alpha release fromalveolar macrophages and serum level of sIL-2R are prognosticmarkers, Am. J. Respir. Crit. Care Med. 156 (1997) 1586–1592.

[38] J.A. Casal, L. Lacerda, L.F. Perez, R.A. Pinto, M. Clara SaMiranda, J. Carlos Tutor, Relationships between serum markersof monocyte/macrophage activation in type 1 Gaucher'sdisease, Clin. Chem. Lab. Med. 40 (2002) 52–55.

[39] E.H. Choi, P.A. Zimmerman, C.B. Foster, S. Zhu, V. Kumar-aswami, T.B. Nutman, S.J. Chanock, Genetic polymorphisms inmolecules of innate immunity and susceptibility to infectionwith Wuchereria bancrofti in South India, Genes Immun. 2(2001) 248–253.

[40] P.J. Brennan, H. Nikaido, The envelope of mycobacteria, Annu.Rev. Biochem. 64 (1995) 29–63.

[41] T. Lehrnbecher, T. Bernig, M. Hanisch, U. Koehl, M. Behl, D.Reinhardt, U. Creutzig, T. Klingebiel, S.J. Chanock, D.Schwabe, Common genetic variants in the interleukin-6 andchitotriosidase genes are associated with the risk for seriousinfection in children undergoing therapy for acute myeloidleukemia, Leukemia 19 (2005) 1745–1750.



Documents

Clin Immunol 2009 (Anand, Hatta)