cytokine pattern and endothelium damage markers in henoch-schönlein purpura
TRANSCRIPT
Immunopharmacology and Immunotoxicology, 30:623–629, 2008Copyright © Informa UK, Ltd.ISSN: 0892-3973 print / 1532-2513 onlineDOI: 10.1080/08923970801973646
LIPI0892-39731532-2513Immunopharmacology and Immunotoxicology, Vol. 00, No. 0, April 2008: pp. 1–15Immunopharmacology and ImmunotoxicologyCytokine Pattern and Endothelium Damage Markers in Henoch-Schönlein PurpuraMarkers in Henoch-Schonlein PurpuraD.V.G. Carlo et al.
Del Vecchio Giovanni Carlo, Penza Rosa, Altomare Maria, Piacente Laura, Aceto Gabriella, Lassandro Giuseppe, De Mattia Domenico, and Giordano Paola
Department of Biomedicine of Development Age, University of Bari, Bari, Italy
In a longitudinal cohort study our aim was to evaluate the cytokine pattern of childrenaffected by Henoch-Schönlein purpura (HSP) and to correlate this pattern to vascularendothelium damage and to nephropathy.
The following parameters were monitored at the onset of the disease (T0) and after6 months of follow-up (T1): clinical scores, serum levels of tumor necrosis factorα (TNF-α), interleukin 2 (IL-2), soluble IL-2 receptor (IL-2sRα), fibrinogen, vonWillebrand factor antigen (vWf:Ag) and soluble thrombomodulin (TMD) levels.
A total of 24 children (9 M, 15 F), affected by HSP, aged between 3–14 years(median 6 years), were enrolled into the study.
IL-2 serum levels were significantly increased at the onset of the disease comparedto control group and T1. The same pattern was observed for IL-2sRα and TNF-α.Fibrinogen and vWf:Ag concentrations were significantly higher at the onset of diseasethan t1 and in control group. TMD levels resulted constantly within the normal range.
Concerning the analyzed parameters, no significant difference resulted to be insubjects with and without renal involvement (hematuria and/or proteinuria). Finally,raised serum TNF-α concentration, related to vascular endothelium damage as shownby increased vWf:Ag levels, occurred invariably in children affected by HSP both withand without renal involvement.
Keywords Children, Vasculitis, Nephropathy, Coagulation, Cytokines.
Address correspondence to Professor De Mattia Domenico, Department of Biomedicineof Developmental Age, University of Bari Policlinico, P.zza G. Cesare 11, 70124 Bari,Italy; E-mail: [email protected]
Imm
unop
harm
acol
ogy
and
Imm
unot
oxic
olog
y D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y D
icle
Uni
v. o
n 11
/08/
14Fo
r pe
rson
al u
se o
nly.
624 D.V.G. Carlo et al.
INTRODUCTION
Henoch-Schönlein purpura (HSP) is a systemic vasculitis, characterized bycutaneous purpura, abdominal pain and bleeding, and arthritis.(1) Renalinvolvement has been reported to occur in 30% to 80% of the patients withHSP.(2) The mechanism of the renal involvement in HSP and the chronicity ofthe renal abnormality have not been completely elucidated, even if same riskfactors such as significant proteinuria have been pointed out.(3, 4)
The exact cause of HSP is unknown and several theories concerning thepathogenesis have been proposed. Recently the role of proinflammatory cytok-ines, such as tumor necrosis factor alfa (TNF-α),(5) as well as interleukin (IL)-5,IL-6, LTB4,(6,7) has been suggested in subjects affected by HSP.
Cytokines exert overlapping biological functions and form a complicatednetwork, even including various chemokines, and a number of ILs. Theobserved redundancy among the different cytokines and other mediators ofinflammation generally guarantees a substitution or complementation ofindividual components that may have been inactivated under pathologicalconditions.
Evidence has been provided that circulating pro-inflammatory cytokinesmay play an aggravating role and, as the role of especially TNF-α is nowstrongly suspected in the induction or aggravation of renal damage in variousglomerulonephritides, and as HSP has been described to be associated withthe aberrant expression of IL-2 or IL-2 receptors. In this framework we inves-tigated in HSP patients the time course of these serum cytokines togetherwith their soluble receptors.
Moreover, other authors reported an activation of coagulation secondaryto the endothelial damage as a typical feature of HSP.(8,9) In particular, TNF-αseems to be responsible for many alterations of the endothelium. In fact itinhibits anticoagulatory mechanism and promotes thrombotic processes and,therefore, plays an important role in pathological processes such as venousthrombosis, vasculitis. Of note, the expression of membrane thrombomodulin(TMD) is decreased by TNF-α.
In view of the interaction between cytokines and coagulation system(10, 11)
the present study was also undertaken to correlate cytokine pattern to vascu-lar endothelium damage and to nephropathy, in HSP.
PATIENTS AND METHODS
Design: Longitudinal cohort study.Subjects: 24 children (9 M, 15 F), aged between 3–14 years (median 6
years), affected by HSP were enrolled into the study. They were all patientsaffected by HSP, observed at University Paediatric Department of Biomedicinadell’Età Evolutiva, in Bari, Italy, between May 2002 and May 2003. Patients
Imm
unop
harm
acol
ogy
and
Imm
unot
oxic
olog
y D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y D
icle
Uni
v. o
n 11
/08/
14Fo
r pe
rson
al u
se o
nly.
Markers in Henoch-Schönlein Purpura 625
at the onset of their disease and complying with the diagnostic criteria forHSP(8) were enrolled.
Inclusion criteria: Subjects with a diagnosis of HSP based on purpuricrash on the buttocks and lower extremities, combined with one or more of thepeculiar manifestation of the syndrome as previously reported . (8).
Exclusion criteria: History of coagulation disease or any other chronicpathological condition, recent assumption of drugs able to interfere with coag-ulation (i.e., antiplatelet agent) and/or endothelium (i.e., steroids).
Twenty-four age- and sex-matched healthy children were selected as con-trol group, among the general pediatric out patient clinic, in the absence ofkidney or inflammatory disease.
None of the patients and controls was undertreatment at the time ofstudy.
The department ethics committee approved the study. Parental informedconsent was obtained for all children before participation in the study.
ParametersJoint, abdominal and renal symptoms were scored from 0 to 3 as previ-
ously reported(8) The following parameters and symptoms were consideredand scored: joint pain and/or swelling, abdominal pain and occult blood in thestool or melena, hematuria and proteinuria.
The following parameters were monitored at onset of the disease (T0) andafter 6 months of follow-up (T1): clinical scores, serum levels of TNF-α, IL-2,soluble IL-2 receptor (IL-2sRα), fibrinogen, von Willebrand factor antigen(vWf:Ag) and soluble TMD levels.
Blood specimens for coagulation studies were obtained using 0.129 Mconcentration trisodium citrate anticoagulant. Plasma was prepared shortlyafter blood collection by centrifugation at 4° C (2000 g for 10 min) and testedwithin 1 hour for fibrinogen assay. Plasma for vWf:Ag and TMD assays, wasfrozen at – 80° C in small aliquots and thawed once immediately before assay.
Serum levels of TNF-α, IL-2, IL-2sRα, were measured using quantitativeimmunoenzymatic assay (R & D Systems, Minneapolis, Minnesota, USA) follow-ing the manufacturer’s instructions for sample collection and storage and assayprocedure. The detection limits of these ELISAs were 4.4 pg/ml for TNF-α, 7 pg/ml for IL-2, 6 pg/ml for IL-2sRα. Optical density values obtained in two determi-nations were converted into pg/ml by the Bio Rad ELISA data analysis software.Fibrinogen was assayed by Clauss method (12) ; vWf:Ag and TMD by quantitativeimmunoenzymatic method (Boheringer Mannheim, Mannheim, Germany).
Statistical MethodsData were expressed as mean ± standard deviation, median, and evalu-
ated by paired and unpaired Student’s t-test, Mann-Whitney U test, Wilcoxon
Imm
unop
harm
acol
ogy
and
Imm
unot
oxic
olog
y D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y D
icle
Uni
v. o
n 11
/08/
14Fo
r pe
rson
al u
se o
nly.
626 D.V.G. Carlo et al.
signed rank test and Spearman rank correlation coefficient as appropriate byusing STAT view program (Abacus Concepts, Berkley, CA, USA). A p valuebelow 0.05 was considered to be statistically significant.
RESULTS
Patient’s clinical data are reported in Table 1. At the onset of disease 5/24children (21%) had renal involvement (hematuria and/or proteinuria); only 2of them displayed same alterations at 6 months of follow-up.
Cytokine serum levels and coagulation data are reported in Tables 2 and 3,respectively. IL-2 serum levels were significantly increased at the onset of thedisease (T0) respect to control group (p < 0.001) and T1 (p < 0.05) (Table 2). Thesame pattern was observed for IL-2sRα and TNF-α (Table 2). At T1 IL-2 wassignificantly increased compared to control group (p < 0.001), whereas IL-2sRα and TNF-α were falling within the normal range (p = NS).
Fibrinogen and vWf:Ag concentrations were significantly higher at theonset of disease (T0) than at T1 (p < 0.01, p < 0.01) and control group (p < 0.05,p < 0.05). TMD levels were consistently within the normal range (Table 3).
No significant differences in relation to the analyzed parameters resultedwhen subjects where divided into two subgroups, with and without renalinvolvement (hematuria and/or proteinuria) (data not shown).
Table 1: Joint, abdominal and kidney scores in the study population, median (minimum – maximum).
Score T0 T1 p
Joint 1 (0 – 3) 0 (0 – 0) < 0.001Abdnominal 0 (0 – 2) 0 (0 – 2) < 0.05Kidney 0 (0 – 3) 0 (0 – 3) NSTotal 1 (0 – 8) 0 (0 – 5) < 0.001
NS = not significant
Table 2: Cytokines (pg/ml) in cases and controls, median (minimum – maximum).
(no.) IL-2 IL-2sRa TNFa
Cases (24)T0 12.8 (6.5 – 81.9) 1603.5 (293.0 – 4316.0) 15.0 (6.5 – 88.0)T1 8.0 (0.5 – 25.0) 1199.5 (400.0 – 2106.0) 7.4 (5.0 – 12.0)
Controls (24) 1.0 (0.5 – 3.2) 1250.0 (420.0 2000.0) 7.2 (3.7 – 10.0)T0 vs C p < 0.001 p < 0.05 p < 0.05T1 vs C p < 0.001 p = NS p = NST0 vs T1 p < 0.05 p < 0.001 p < 0.001
NS = not significant.
Imm
unop
harm
acol
ogy
and
Imm
unot
oxic
olog
y D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y D
icle
Uni
v. o
n 11
/08/
14Fo
r pe
rson
al u
se o
nly.
Markers in Henoch-Schönlein Purpura 627
Finally, a significant positive correlation resulted to be among the follow-ing parameters: renal score and vWf:Ag (rho = 0.4 e p < 0.01); articular scoreand IL-2sRα (rho = 0.6 e p < 0.01); TNF-α and vWf:Ag (rho = 0.5 e p < 0.05);vWf:Ag and fibrinogen and TMD (rho = 0.6 e p < 0.01; rho = 0.6 e p<0.01).
DISCUSSION
The present study was undertaken to evaluate the cytokines’ pattern in chil-dren affected by HSP and to correlate it either to markers of endotheliumdamage or to renal involvement, revealed as hematuria and/or proteinuria. Inthis study, renal involvement (hematuria and/or proteinuria) was present atthe onset of disease in 21% of cases and in 8% of cases at 6 months. This prev-alence was similar to that previously reported regarding renal involvement inunselected children with HSP: 28%–33%.(4, 13)
As claimed in literature dysregulated cytokine production plays a crucialrole in HSP and probably in renal involvement.(5,14–.16) In fact, serum TNF-αlevels during HSP acute phase were significantly raised compared to remis-sion phase particularly in subjects with renal involvement.(5) Moreover renalinvolvement, unmasked as nephrotic syndrome and/or renal insufficiency, wasassociated with specific interleukin gene polymorphism.(15,16)
In our study, we demonstrated a pro-inflammatory cytokine pattern at theonset of the disease, with increased serum levels of IL-2, IL-2sRα and TNF-αthat was less significant at the follow-up evaluation. This pattern is analogousto that previously reported(14) in adult patients affected by IgA nephropathy, 5of whom were HSP patients and in 20 children affected by HSP(5) The impor-tance of pro-inflammatory cytokines, and, in particular of TNF-α, in thepathogenesis of HSP has been evidenced by an in vitro study dealing withenhanced levels of IgA antibodies to human umbilical vein endothelial cells inHSP children.(17) However, different from previously reported results(5) wewere not able to detect any difference in pro-inflammatory pattern betweenchildren with or without renal involvement.
Table 3: Fibrinogen, vW:Ag and trombomodulin in cases and controls (M ± SD).
(no.) Fibrinogen (mg/dl) vW:Ag (%) TMD (ng/ml)
Cases (24)T0 328.20 ± 95.92 102.61 ± 21.55 22.40 ± 7.55T1 253.33 ± 35.33 87.71 ± 19.97 22.71 ± 6.94
Controls (24) 274.37 ± 57.27 88.45 ± 18.90 24.45 ± 5.90T0 vs C p < 0.05 p < 0.05 NST1 vs C NS NS NST0 vs T1 p < 0.01 p < 0.01 NS
NS = not significant.
Imm
unop
harm
acol
ogy
and
Imm
unot
oxic
olog
y D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y D
icle
Uni
v. o
n 11
/08/
14Fo
r pe
rson
al u
se o
nly.
628 D.V.G. Carlo et al.
Moreover, in our study we evaluated vWf:Ag and TMD levels as markersof endothelial cell injury, as previously described(8,18,19) We confirmed aninvolvement of vWf:Ag in HSP as in other vasculitis(8) although its patternwas not useful in differentiating children with or without renal involvement.Interestingly, in our study a positive correlation was evident between vWf:Agand TNF-α. Such a correlation could be explained considering the emerginginteraction between cytokines and coagulation(10,11) and it fits well in a patho-genetic model in which TNF-α seems to play a pivotal role.(17)
TMD levels resulted consistently within the normal range in our studywhile others(19) reported increased levels in subjects with HSP and nephritis,this suggesting a more severe and more prolonged damage to the endothelialcells in patients with nephritis than in those without nephritis. This differ-ence between our and others’ study(19) could be due either to the smaller num-ber of our children with renal involvement (5/24) compared to the other one(15/25)(19) or to a different degree of renal involvement.
In conclusion, children affected by HSP show evidence of a documentedvascular endothelium damage related to elevated serum levels of TNF-α withno difference between subjects with and without renal involvement.
REFERENCES
1. Roberts, P.F.; Waller, T.A.; Brinker, T.M.; Riffe, I.Z.; Sayre, J.W.; Bratton, R.L.Henoch-Schonlein Purpura: A review article. South Med. .J 2007, 100, 821–824.
2. Shenoy, M.; Bradbdury, M.G.; Lewis, M.A.; Webb, N.J. Outcome of Henoch-Schön-lein purpura nephritis treated with long-term immunosuppression. Pediatr. Neph-rol. 2007, 22, 1717–22.
3. Sano, H; Izumida, M; Shimizu, H; Ogawa, Y. Risk factors of renal involvement andsignificant proteinuria in Henoch-Schonlein purpura. Eur. J. Pediatr. 2002, 161,196–201.
4. Kaku, Y; Nohara, K; Honda, S. Renal involvement in Henoch-Schonlein purpura: Amultivariate analysis of prognostic factors. Kidney Int. 1998, 53, 1755–1759.
5. Besbas, N.; Saacti, U.; Ruacan, S.; Ozeu, S.; Sungur, A.; Bakkaloglu, A.; Eluabas,A.M. The role of cytokines in Henoch-Schonlein purpura. Scand. J. Rheumatol.1997, 26, 456–460.
6. Liao, P.Y.; Wu, S.H. Serum levels of IL-5 and LTB-4 in children with Henoch-Schon-lein purpura. Zhonggou Dang Dai Er Keza Zhi. 2006, 8, 198–200.
7. Lin, C.Y.; Yang, Y.H.; Lee, C.C.; Huang, C.L.; Wang, L.C.; Chiang, B.L. Thrombopoi-etin and interleukin-6 in Henoch-Schonlein-purpura. J. Microbiol. Immunol. Infect.2006, 39, 476–482.
8. De Mattia, D.; Penza, R.; Giordano, P.; Del Vecchio, G.C.; Aceto, G.; Altomare,M.; Schettini, F. Von Willebrand factor and factor XIII in children with Schonleinpurpura. Pediatr. Nephrol. 1995, 9, 603–605.
9. Brendel-Muller, K.; Hahn A.; Schneppenheim, R.; Santer, R. Laboratory signs ofactive coagulation are common in Henoch-Schonlein purpura. Pediatr. Nephrol.2001, 16, 1048–8.
Imm
unop
harm
acol
ogy
and
Imm
unot
oxic
olog
y D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y D
icle
Uni
v. o
n 11
/08/
14Fo
r pe
rson
al u
se o
nly.
Markers in Henoch-Schönlein Purpura 629
10. Joseph, L.; Fink, L.M.; Hauer-Jensen, M. Cytokines in coagulation and thrombosis:A preclinical and clinical review. Blood Coagul. Fibrinolysis, 2002, 13, 105–116.
11. Levi, M.; Ten Cate, H.; Van der Poll, T. Endothelium: Interface between coagulationand inflammation. Crit. Care Med. 2002, 30 (suppl), S220–224.
12. Clauss, A. Gerimmungasphisiologische schnellmethoden zur bestimmung desfibrinogens. Acta Haematol. 1957, 17, 237–241.
13. Stewart, M.; Savage, J.M.; Bell, B.; MCCord, B. Long term renal prognosis ofHenoch-Schonlein purpura in an unselected childhood population. Eur. J. Pediatr.1988, 147, 113–115.
14. Rostoker, G.; Rymer, J.; Bagnard, G.; Petit-Phar, M.; Griuncelli, M.; Pilatte,Y. Imbalances in serum proinflammatory cytokines and their soluble receptors: Aputative role in the progression of idiopathic IgA nephropathy and HSP nephritisand a potential target of immunoglobulin therapy?. Clin. Exp. Immunol. 1998,114, 468–476.
15. Amoli, M.M.; Thomson, W.; Hajeer, A.H.; Calvino, MC; Garcia-porrua, C; Ollier, WE;Gonzales-Gay, M.A. Interleukin 8 gene polymorphism is associated with increasedrisk of nephritis in cutaneous vasculitis. J. Rheumatol. 2002, 29, 2367–2370.
16. Amoli, M.M.; Thomson, W.; Hajeer, A.H.; Calvino, M.C.; Garcia-porrua, C.; Ollier,W.E.; Gonzales-Gay, M.A. Interleukin 1 receptor antagonist gene polymorphismis associated with severe renal involvement and renal sequelae in Henoch-Schon-lein purpura. J. Rheumatol. 2002, 29, 1404–1407.
17. Yang,, Y.H.; Wang, S.J.; Chuang, Y.H.; Lin, Y.T.; Ching, B.L. The level of IgAantibodies to human umbilical vein endothelial cells can be enhanced by TNF-alpha treatment in children with Henoch-Schonlein purpura. Clin. Exp. Immunol.2002, 130, 352–357.
18. Endo, S.; Inada, K.; Nakae, H.; Takakuwa, T.; Kasai, T.; Yamada, Y.; Suzuki, T.;Taniguchi, S.; Yoshida, M. Blood levels of endotheiln-1 and thrombomodulin inpatients with disseminated intravascular coagulation and sepsis. Res. Commun.Mol. Pathol. Pharmacol. 1995,; 90, 277–288.
19. Fujieda, M.; Oishi, N.; Naruse, K.; Hashizume, M.; Nishiya, K.; Kurashige, T.; Ito,K. Soluble thrombomodulin and antibodies to bovine glomerular endothelial cellsin patients with Henoch-Schonlein purpura. Arch. Dis. Child, 1998, 78, 240–244.
Imm
unop
harm
acol
ogy
and
Imm
unot
oxic
olog
y D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y D
icle
Uni
v. o
n 11
/08/
14Fo
r pe
rson
al u
se o
nly.