(CD14) on monocytes

4 downloads 0 Views 597KB Size Report
a rapid and transient monocytopenia during cupro- phane haemodialysis and a slow but persistent monocy- topenia during dialysis on AN69 membrane (data ...
Nephrol Dial Transplant (1996) 11: 657-662

Nephrology Dialysis Transplantation

Original Article

Cuprophane haemodialysis induces upregulation of LPS receptor (CD14) on monocytes: role of complement activation A. Marchant1, C. Tielemans2, C. Husson2, K. Gastaldello2, T. Schurmans2, D. De Groote3, J. Duchow1, J. L. Vanherweghem2 and M. Goldman1 Departments of 'Immunology and 2Nephrology, Dialysis and Transplantation, Cliniques Universitaires de Bruxelles, Hopital Erasme, Universite Libre de Bruxelles, Brussels; and 3Medgenix Group Research and Development Department, Fleurus, Belgium

Abstract

Key words: CD 14, complement activation, cuprophane membrane, haemodialysis, LPS receptor

Introduction

Besides activation of granulocytes, haemodialysis on cellulosic membranes also leads to significant alterations of monocytes. Indeed, patients dialysed on cuprophane membranes display an increased production of monocyte-derived cytokines such as interleukin-1 (IL-1), tumour necrosis factor-a (TNF-oc) and interleukin-6 (IL-6) [1-6, reviewed in 7]. Moreover, cuprophane haemodialysis induces an increased expression of various receptors at the surface of monocytes including leukocyte integrin MAC-1 (CDllb/CD18) [8], CD45 [8], complement receptor CR1 and HLA-DR [2]Several mechanisms might account for this monocyte activation. Firstly, it appears that the direct interaction of monocytes with dialyser polymer materials delivers activation signals [9]. Secondly, complement activation products such as C3a and C5a increase the expression of leukocyte integrins and are probably involved in the rapid upregulation of MAC-1 molecules observed at the surface of monocytes during cuprophane haemodialysis [8]. Complement activation products also prime monocytes to release TNF-a and IL-1 upon stimulation with LPS [10, 11]. Activation of cytokine gene transcription was shown to be one of the mechanisms involved in this priming effect [12]. We speculated that complement activation products might also act by modulating the expression of membrane receptors for LPS. Eventually, several reports suggest that bacterial lipopolysaccharide (LPS) or LPS-like substances could permeate through the cuprophane membrane and activate the leukocytes [13-15] but this issue remains largely controversial [16]. The CD 14 membrane antigen is a phosphatidylinosiCorrespondence and offprint requests to: Michel Goldman MD, Department of Immunology, Hopital Erasme, 808 Route de Lennik, tol-anchored protein expressed on monocytes, macrophages and to a lesser extent on granulocytes. It is a 1070 Brussels, Belgium.

© 1996 European Dialysis and Transplant Association-European Renal Association

Downloaded from ndt.oxfordjournals.org by guest on July 13, 2011

Background. The CD 14 molecule is a high-affinity receptor for the complex formed by lipopolysaccharide (LPS) and LPS-binding protein. Methods. We examined by flow cytometry the effect of in vitro and in vivo haemodialysis on cuprophane membrane and recombinant C5a on the expression of CD14 molecules at the surface of monocytes. Monocyte CD 14 expression was also studied during in vitro and in vivo haemodialysis on polyacrylonitrile AN69 membrane. Results. In vitro haemodialysis of whole blood from healthy volunteers on cuprophane membrane resulted within 30 min in upregulation of monocyte CD 14 expression. The reuse of the cuprophane membrane abolished both complement activation and CD 14 upregulation. Moreover, incubation of whole blood with recombinant C5a led to an increased monocyte CD 14 expression supporting a role for complement activation in the rapid cuprophane-induced CD14 upregulation. During AN69 dialysis which is not associated with complement activation in the blood phase, monocyte CD 14 expression did not change during the first 60 min but was significantly increased after 3 h of in vitro haemodialysis. This late increase might be related to the presence of complement activation products adsorbed on the membrane. In vivo dialysis on cuprophane membrane also resulted in early monocyte CD 14 upregulation as indicated by the higher CD 14 expression found after 60 min on monocytes obtained from the efferent as compared to the afferent line of the dialyser, a phenomenon that was not observed during haemodialysis on AN69 membrane. Conclusion. Haemodialysis on the complementactivating cuprophane membrane induces the rapid upregulation of the CD14 LPS-receptor on monocytes.

A. Marchant et at.

658

high-affinity receptor for LPS complexed to LPSbinding-protein (LBP) [17,18]. The binding of LPSLBP complexes to monocyte CD 14 triggers the release of TNF-oc, IL-6 and IL-8 [17,19]. Moreover, monocytes from CD14-deficient patients display an impaired production of cytokines in response to LPS, indicating that CD 14 plays an important role in LPS-induced monocyte activation [20]. The present study was undertaken to examine the expression of CD 14 molecules on monocytes during in vitro and in vivo dialysis on cuprophane membrane as compared to polyacrylonitrile AN69 which is not associated with significant activation of complement in the blood phase. The expression of MAC-1, another surface receptor known to be upregulated by complement and cuprophane dialysis [8] was used as a control. Subjects and methods Reagents

Whole blood activation by rC5a Blood obtained from healthy volunteers was collected in heparinized (10 U/ml) syringes. One millilitre of whole blood was incubated for various periods of time at 37°C with 50 |il RPMI 1640 medium containing various concentrations of rC5a. As control, whole blood was incubated with RPMI without rC5a. In vitro haemodialysis In vitro HD sessions of 3 h were performed using closedloop miniaturized blood and dialysate circuits. The dialyser consisted in miniaturized new oxide-sterilized cuprophane (« = 6) or polyacrylonitrile AN69 (« = 7) hollow-fibre dialysers that were provided by Hospal (Meyzieu, France). Thirty-five millilitres of heparinized whole blood (10 U/ml) from healthy volunteers were recirculated in the blood compartment. The dialysate consisted of the bicarbonate dialysate used for clinical HD in our facility that had first been uitrafikered through a polyamide membrane (Ultrafilter U 2000. Gambro, Sweden) to remove endotoxin and other bacterial products that can induce monokine synthesis [16]. The HD circuits were assembled aseptically, and apyrogenicity of both blood and dialysate compartments was assessed by the Limulus assay both at the start and by the end of each session. Blood and dialysate were recirculated at 37:C at a flow of 10 ml/min without any net ultrafiltration. Whole

Subjects End-stage renal failure patients (« = 7) undergoing intermittent haemodialysis (3 x week) were enrolled in this study. In order to avoid alterations of monocyte phenotype related to chronic exposure to complement activation products [2], we studied patients under intermittent dialysis on a polyacrylonitrile AN69 membrane (Filtral 20, Hospal, France). They were first studied during an AN69 dialysis session and then switched on a cuprophane membrane of the same surface (GF18E, Gambro, Sweden). Again, the dialysate consisted of bicarbonate dialysate ultrafiltered on line through a polyamide membrane (Ultrafilter U 2000, Gambro, Sweden). Its apyrogenicity was assessed by the Limulus assay. Heparinized whole blood was collected from the afferent line before and after 30, 60 and 180 min of haemodialysis for evaluation of monocyte CD14 expression. At 60 min, blood was collected both from the afferent and the efferent lines. Leukocytosis and differential count were determined with an automated cell counter (Coulter STKS; Coulter Electronics, Hialeah, Fla.) and confirmed under light microscopy. Flow cy tome try analysis Monocyte CD14 and MAC-1 expression was studied as described previously [8, 23]. Briefly, 200 ul whole blood was incubated with FITC anti-CD14 (Leu-M3) and PE antiCDllb MAb or irrelevant IgG-matched MAb at 4°C for 30 min. After lysis of red blood cells (FACS lysing solution, Becton Dickinson), white blood cells were fixed in a 1% paraformaldehyde solution. Fluorescence was measured using a FACScan flow cytometer (Becton Dickinson). Monocytes were gated according to forward and side scatter properties. Experiments with fluorescent standard microbeads demonstrated that meanfluorescencemeasured related linearly to the mean number offluoresceinmolecules bound per cell. Results were expressed as mean equivalent of soluble fluorescence (MESF) or change in MESF as compared to prestimulation or predialytic values. Statistical analysis Monocyte CD 14 expression levels were compared using Wilcoxon rank sum test on paired samples.

Results In vitro haemodialysis on cuprophane membrane induces rapid upregulation of monocyte CD14 In vitro dialysis was performed using whole blood from healthy volunteers and endotoxin-free dialysate.

Downloaded from ndt.oxfordjournals.org by guest on July 13, 2011

RPMI 1640 medium was obtained from Grand Island Biological Corporation (Grand Island, NY). Heparin was purchased from Novo (Novo Industry A/S, Bagsvaerd, Denmark). Recombinant C5a (rC5a) was obtained from Sigma Chemical Co. (St Louis, MO). The endotoxin content of RPMI medium and of heparin was less than 2 pg/ml, endotoxin content of rC5a was below 200 pg/ng protein as determined by the Limulus assay (LAL-QCL-1000, Whittaker Bioproducts, Walkersville, MD). Fluoresceinated anti-CD 14 (Leu-M3) and phycoerythrin anti-CDllb monoclonal antibody (MAb) were purchased from Becton Dickinson (Mountain View, CA). IgG isotypic control was obtained from DAKO (DAKO A/S, Glostrup, Denmark).

blood was collected at 0, 30, 60 and 180 min for evaluation of monocyte CD14 and MAC-1 expression. In some experiments, the same miniaturized cuprophane haemodialyser was repeatedly used after formaldehyde sterilization in order to decrease its complement activating properties [22]. After every HD session, the circuit was washed with apyrogenic saline, filled with a 4% formol solution, incubated overnight and finally washed with saline immediately before the HD session. Cuprophane membrane was treated four times by this reuse procedure to completely prevent complement activation as assessed by plasma levels of C3a desarg measured before and after 30 and 180 min of HD, using a commercially available radioimmunoassay (Amersham, Buckinghamshire, UK).

659

CD 14 upregulation by cuprophane haemodialysis

Io

1.e+4 -

2

Oe+0 -

§ i

i

i

i

i

i

i

0

30

60

90

120

150

180

UH

(75 J

1.2e+6 -

Reuse of cuprophane dialyser abolishes its ability to upregulate monocyte CD14 expression To further evaluate the role of complement activation in the upregulation of monocyte CD 14 expression during cuprophane dialysis, we compared the changes in CD 14 expression during in vitro HD with either new

c 'Z

10e+6 -

£

8.0e+5 H

35000 -

CJ ""

30000 -

6.0e+5 -

25000 -

4.0e+5 -

3 o o ^

2.0e+5 -

|

20000 -

2

15000 -

5

0.0e+0 -

( >

-

i

i

i

i

0

30

60

90

i

i

i

10000 5000 -

120 150 180 0 -

Time (min) Fig. 1. Modulation of monocyte CD14 and MAC-1 expression during in-vitro haemodialysis. Whole blood from healthy volunteers was circulated in a closed loop miniaturized haemodialyser system as described in Methods. Blood samples were collected before and after 30, 60 and 180 min haemodialysis. Monocytes were stained with FITC-anti-CD14 (Leu-M3) or PE-anti-CDl lb MAb and mean fluorescence intensity (MESF) was measured by flow cytometry. The figures show the change in monocyte CD14 (upper panel) and MAC-1 (lower panel) expression as compared to predialytic values during dialysis on cuprophane (closed circles, mean + SEM of four (MAC-1) to six (CD 14) different experiments) or AN69 membrane (open circles, n = 5 (MAC-1) to 7 (CD14) different experiments). *P