NDT Advance Access originally published online on November 24, 2006
Nephrology Dialysis Transplantation 2007 22(3):749-755; doi:10.1093/ndt/gfl688
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Lack of specific binding of Shiga-like toxin (verocytotoxin) and non-specific interaction of Shiga-like toxin 2 antibody with human polymorphonuclear leucocytes
1Department of Paediatric Nephrology and 2Department of Nuclear Medicine, Radboud University Nijmegen Medical Centre, PO Box 9101, 6500 HB Nijmegen, The Netherlands
Correspondence and offprint requests to: Leo A. H. Monnens, Radboud University Nijmegen Medical Centre, Department of Pediatric Nephrology, PO Box 9101, 6500 HB Nijmegen, The Netherlands. Email: l.monnens{at}cukz.umcn.nl
 |
Abstract |
|---|
 Top Abstract BackgroundMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Background. After gastrointestinal infection with Shiga-like toxin (Stx) producing Escherichia coli, the toxin is transported from the intestine to the renal microvascular endothelium. This is the main target for Stx in humans. Previous studies indicated that polymorphonuclear leucocytes (PMN) could serve as carriers for Stx in the systemic circulation. As at a later stage we could not confirm these data, we performed new studies.
Methods. The binding of Stx1 to PMN was determined in vitro (isolated human PMN and whole blood) and in vivo (injection in mice). The specificity of binding of an antibody against Stx2 to PMN from patients with haemolytic uraemic syndrome (HUS) was determined. This was compared with binding to PMN from healthy controls, and patients after haemodialysis (HD) or on peritoneal dialysis (PD). Furthermore, PMN were incubated with Stx to study possible activation.
Results. No specific binding of Stx1 to PMN could be detected. After intravenous injection of the toxin in mice, it was not associated with PMN. The binding of an antibody against Stx2 to PMN was detected in both patients with HUS and patients after HD, but not in patients on PD. Stx was not able to activate PMN.
Conclusions. PMN are not acting as transporter for Stx in the pathogenesis of HUS. The interaction of a Stx antibody with PMN from HUS patients is not specific as it can also be observed in patients after HD (possibly due to activation of the PMN). Therefore, binding of Stx antibody to PMN is not reliable as a diagnostic tool for HUS.
Keywords: binding experiments; haemolytic uraemic syndrome; polymorphonuclear leucocytes; Shiga-like toxin
|
Background |
|---|
|
TopAbstractBackgroundMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Shiga-like toxin (Stx, also called verocytotoxin) is an important mediator in the pathogenesis of the diarrhoea-associated haemolytic uraemic syndrome (HUS). Stx is produced by the pathogen Escherichia coli, of which O157:H7 is the serotype most commonly involved in HUS [1]. In an earlier study of 50 isolates of E. coli associated with HUS in The Netherlands, 45 were positive for the Stx2 gene and five for both Stx1 and Stx2 [2].
As the renal microvascular endothelium is the most important target of the toxin, a transport route from the intestine to the kidney is postulated. Stx was never found in sera from patients with HUS, but Stx2 could be detected in kidneys from HUS patients [3]. Furthermore, antibodies against Stx2 are found in serum from patients with HUS [4]. After reaching the renal endothelium, Stx will initiate cell damage with the formation of thrombi in the glomerular arterioles and capillaries [1]. Acute renal failure with thrombocytopenia and haemolytic anaemia will occur, which are the hallmark features of HUS.
Earlier work of te Loo et al. [5] suggested that circulating polymorphonuclear leucocytes (PMN) transported Stx2 in patients with HUS [5]. However, with a new batch of the same antibody, these results could not be reproduced. Therefore, binding studies with Stx to PMN in vitro were repeated. Also, additional binding experiments were performed in vivo in mice. The interaction of an antibody against Stx2 with PMN collected from blood of HUS patients was re-evaluated. It was hypothesized that the antibody could bind aspecifically to activated PMN in these patients. Consequently, the capacity of PMN to bind the antibody against Stx2 was compared between healthy controls, HUS patients, patients treated by haemodialysis (HD) and peritoneal dialysis (PD). It is known that PMN originating from patients on HD are highly activated [6].
In addition, if no binding of Stx to PMN occurs, it can be proposed that Stx has no effect on PMN function. For this reason, the expression of degranulation markers and cell adhesion molecules (CD66b, CD63, L-selectin and CD11/CD18) on PMN were studied after incubation with Stx2.
|
Methods |
|---|
|
TopAbstractBackgroundMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Materials
Purified Stx1, used for the binding experiments, was kindly provided by Dr M. A. Karmali (Public Health Agency of Canada, Ontario, Canada). The 125I-Stx1 B-subunit was a gift from Dr L. Johannes (Institut Curie, Paris, France). The unlabelled Stx1 B-subunit was a gift from Dr L. Brunton (University of Toronto, Canada). For the stimulation of PMN, Stx2 was purchased from Toxin Technology (Sarasota, United States). Ethylenediaminetetraacetic acid (EDTA) tubes were ordered from BD Vacutainer (Alphen a/d Rijn, The Netherlands). Ficoll-PaqueTM PLUS was purchased from Amersham Biosciences (Uppsala, Sweden). Human serum albumin (HSA; Cealb) from Sanquin (Amsterdam, The Netherlands) was used. Hanks’ balanced salt solution (HBSS) was ordered from MP Biomedicals (Eschwege, Germany).
Monoclonal antibodies against CD66b [fluorescein isothiocyanate (FITC)-labelled] and CD63 (PE-labelled) were obtained from Coulter/Immunotech (Marseille, France). Antibodies against L-selectin, CD11b/CD18 (both FITC-labelled) and CD13 (PE-labelled) were purchased from DAKO (Heverlee, Belgium), as was the secondary goat-anti-mouse FITC and goat serum. HBSS without phenol red was obtained from Life Technologies (Paisley, Scotland), luminol and 12-myristate 13-acetate and horseradish peroxidase were purchased from Sigma (St Louis, USA).
The antibody against Stx2 was a kind gift from Dr A. O’Brien (Uniformed Services University of Health Sciences, Bethesda, USA). This is a well-characterized, humanized antibody as described previously [7]. The secondary antibody (goat-anti-human FITC) for this antibody was purchased from Jackson Immunoreseach (Cambridgeshire, UK). The Fc-receptor blocker was ordered from Miltenyi Biotec (Utrecht, The Netherlands). Mouse serum was obtained from the Central Animal Laboratory (Nijmegen, The Netherlands). Lipopolysaccharide (LPS) was collected from Sigma (St Louis, USA).
Isolation of polymorphonuclear leucocytes
Fresh venous blood was drawn and collected in EDTA tubes. Blood was diluted two times with phosphate-buffered saline (PBS) and centrifuged (30 min, 400g at 4°C) over Ficoll-PaqueTM Plus. After centrifugation, the pellet containing PMN and erythrocytes was collected. Erythrocytes were lysed using ammonium chloride. This procedure routinely revealed a PMN population of over 95%.
Binding of Shiga-like toxin to isolated human polymorphonuclear leucocytes
After isolation of PMN from healthy donors (n=7), cells were washed with PBS and re-suspended in HBSS containing 1% HSA at 0°C. During a period of 3 h, 5x105 PMN were incubated with 125I-Stx1. Incubations were performed with different concentrations of Stx1 ranging from 0.3 up till 70 nmol/l. Stx1 was radio-iodinated according to the Iodogen method. To determine non-specific binding, the same incubation was carried out in the presence of a 25-fold molar excess of unlabelled Stx1. After the incubation, free 125I-Stx1 was separated from the PMN using Ficoll-Paque PLUS, and cells were washed. Cell-associated 125I-Stx1 was determined in a gamma counter. The binding of 125I-Stx1 to monocytes showed an adequate Kd value in the Scatchard plot analysis. In the glycolipids extracted from human monocytes, 125I-Stx is bound to Gb3.
Binding of Shiga-like toxin in whole blood
EDTA-blood from seven healthy human donors was obtained. Whole blood (2–2.5 ml) was incubated with 125I-Stx1 B-subunit (0.75 µg, 3 µCi) during 2.5 h at 4°C.
In order to up-regulate a possible receptor, blood from four donors was pre-incubated with LPS (1 µg/ml) during 2 or 16 h at 37°C, before adding the 125I-Stx1 B-subunit.
To investigate the amount of binding of Stx1 B-subunit to PMN in the circulation, nine normal SE mice (female, age 10 weeks) were injected intravenously with 125I-Stx1 B-subunit. A total amount of 3.5 µg protein (1 µCi) was administered intravenously through the tail vein. In five mice, all blood was collected after 2.5 min. Two of these mice were co-injected with an excess of unlabelled Stx1 B-subunit (100 µg/mouse), to determine the specificity of the binding. The amount of unlabelled Stx1 B-subunit will saturate the specific Stx1 B-subunit binding sites in vivo. In four mice, blood was collected after 15 min. One of these mice received 100 µg unlabelled B-subunit. Blood was collected from the orbital plexus.
Blood from both mice and men, was centrifuged over Ficoll-Paque PLUS (30 min, 400g at 20°C). The different layers (plasma, interphase, Ficoll-Paque PLUS and the pellet containing PMN and erythrocytes) were separated. The interphase was centrifuged twice to remove platelets (10 min, 200g). Supernatants were also collected. From each layer, a fraction was counted in a gamma counter for the presence of 125I-Stx1 B-subunit. Subsequently, the fraction containing erythrocytes and PMN was lysed with ammonium chloride and washed. By this means, the presence of 125I-Stx1 B-subunit on the PMN could be measured. The 125I activity in each fraction was calculated.
Analysis of binding antibody Shiga-like toxin to polymorphonuclear leucocytes
Blood was collected from healthy controls (n=6), patients with HUS (Stx present in faeces) (n=3), patients on PD (n=3) and patients on HD (n=14). After isolation of PMN, cells were incubated with Fc-receptor blocker (15 µl for 15 min) and blocked with 10% goat- and mouse serum. Subsequently, PMN were incubated with the antibody against Stx2 followed by goat-anti-human FITC. After the incubations, the cells were suspended in paraformaldehyde 0.5% and analysed with FACS analyser.
Three other antibodies against Stx2 (two commercial and one non-commercial) reacted in a variable degree with normal PMN.
Stimulation of human polymorphonuclear leucocytes with Shiga-like toxin
Whole blood (100 µl) from four healthy volunteers was incubated with either LPS 1 µg/ml, Stx2 0.1 nM, Stx2 10 nM alone or LPS in combination with Stx2 (0.1 or 10 nM) for 2 h at room temperature. Subsequently, erythrocytes were lysed using ammonium chloride. PMN were identified by flow cytometry (Coulter® Epics® XL-MCL) using their morphological criteria and using CD13-PE as specific marker. Monoclonal antibodies against CD63 and CD66b were used as markers of degranulation. Also, monoclonal antibodies against L-selectin and CD11b/CD18 were used to study the expression of adhesion molecules. Antigen expression was measured as mean fluorescence intensity of 5000 cells by flow cytometry. Data acquisition and analysis were performed using XL-2 software (Coulter).
Furthermore, the superoxide production in PMN of eight healthy donors was measured after stimulation with Stx2 0.1 nM, Stx2 10 nM, LPS 1 µM alone or LPS in combination with Stx2 0.1 and 10 nM for 10 min, 1 h and 4 h, respectively. Phorbol 12-Myristate 13-Acetate (PMA) was used as a positive control. Immediately after the incubation, luminol-enhanced chemilumniscence of proteose-peptone-elicited PMN was measured on a Victor 1420 counter (Wallac, Turku, Finland) at 37°C. The measurements were performed in 96-well microplates as described previously [8]. Each well contained 2x105 PMN, 50 µM luminal and 4.5 U/ml horseradish peroxidase in 200 µl of HBSS without phenol red, supplemented with 0.25% HSA. Wilcoxon signed rank test was used to test significance of within-group differences. Differences were considered significant at P<0.05.
|
Results |
|---|
|
TopAbstractBackgroundMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Binding of Shiga-like toxin to isolated polymorphonuclear leucocytes
PMN from healthy donors were isolated and incubated with 125I-Stx1. To test the specificity of the binding, simultaneous incubations were done in the presence of an excess of unlabelled Stx1. In none of the performed experiments (n=7) was specific interaction of 125I-Stx1 with PMN found.
Binding of Shiga-like toxin in whole blood
To establish whether PMN in whole blood can bind Stx, we incubated blood from healthy donors (n=7) with 125I-Stx1 B-subunit. The B-subunit of Stx is the binding part of the toxin, whereas the A-subunit will only stimulate the uptake of the toxin and does not affect binding [9].
The experiment was performed with and without previous stimulation of the blood with LPS. The major part of the B-subunit can be found in the plasma. Seventy-one percent of total activity was detected in plasma, whereas 0.09% was found in the fraction containing the PMN (n=3). Pre-incubation with LPS did not change the amount of bound toxin [0.09 and 0.02% of total activity after, respectively, 2 and 16 h pre-incubation (n=4)].
To investigate the binding of Stx to PMN in vivo, the 125I-labelled Stx1 B-subunit was injected in mice (n=9). After 2.5 and 15 min, blood from the mice was collected and the distribution of the B-subunit in the various blood fractions was determined. Stx1 B-subunit rapidly cleared from the blood as is shown in Figure 1. Subsequently, the different cell populations were separated and the 125I-Stx1 B-subunit activity in each fraction was determined. The major part of the 125I-Stx1 B-subunit was found in the plasma. Only a minute fraction of the activity was found in the PMN fraction. The addition of an excess of unlabelled Stx1 B-subunit did not reduce the activity (Table 1).
|
20% only was still present.
|
Binding of antibody Shiga-like toxin to polymorphonuclear leucocytes
The specificity of attachment of the antibody against Stx2 was tested with PMN of patients with HUS, patients on HD or PD and healthy controls.
In two out of three blood samples from HUS patients in active disease, the antibody did not show any staining of the PMN (Figure 2). Only the PMN of one patient showed an attachment of the antibody to the PMN during active disease, which disappeared during remission.
|
The healthy controls were negative (n=6). In patients on PD, no attachment of the Stx2 antibody to PMN was found. However, if the antibody was added to PMN from patients on HD, in 11 out of 16 experiments, positive staining of the PMN with the antibody was observed (Figure 3). In these patients, no symptoms of HUS could be detected.
|
Stimulation of polymorphonuclear leucocytes with Shiga-like toxin
Incubation of PMN with different concentrations of Stx2 did not cause any change in the expression of degranulation markers (CD66b and CD63) on PMN (Figure 4). Similar results were obtained with the adhesion molecules. Stx2 did not induce significant changes in the expression of CD11b/CD18 and shedding of L-selectin (Figure 4).
|
Incubation of PMN from healthy donors did not show an increase of superoxide production whereas LPS induced a small increase (data not shown).
|
Discussion |
|---|
|
TopAbstractBackgroundMethodsResultsDiscussionAcknowledgementsReferences |
|---|
In a previous study, we have demonstrated specific binding of 125I-Stx1 to freshly isolated, non-stimulated human PMN. Also, a rapid binding of FITC-labelled Stx1 to PMN was detected when incubated in whole blood [5]. In the current study we were unable to reproduce these results. Neither in isolated PMN nor in whole blood was a specific binding of Stx1 to PMN observed. Fernandez et al. [10] also studied the binding of Stx to PMN. PMN from healthy adults and children were incubated with different concentrations of Stx1 or Stx2 for different time intervals. The binding of Stx to PMN was determined by flow cytometry using specific anti-Stx antibodies, and by quantification of the remnant Stx, evaluating the cytotoxicity of the supernatants on verocells. Neither of the two approaches revealed a positive binding of Stx to PMN surface in any of the conditions assayed. We also evaluated a possible binding of 125I-Stx1 B-subunit after intravenous injection in mice. In vivo, no specific interaction of Stx1 with PMN could be detected. Some caveats should be noted. It is not allowed to extrapolate data obtained in mice to the human. It is also possible that labelled toxin is detached from cells during the experimental procedure (centrifugation on Ficoll). We have no experience with erythrocytes, but labelled Stx on the surface of monocytes at a constant temperature is not released by centrifugation.
A lack of specific binding of Stx to PMN suggests the absence of a direct effect of Stx on PMN function. Stx2 did not stimulate superoxide production in PMN and was unable to induce the expression of degranulation and activation markers CD63, CD66b and CD11b/CD18. Similar results were reported by Aoki et al. and Holle et al. [11,12].
However, Stx2 also has an inhibitory effect on spontaneous neutrophil apoptosis. Surprisingly, this effect is only partially inhibited by anti-Stx2 antibody [13]. Could Stx2 interfere with triggering factors for spontaneous apoptosis?
How to reconcile these data with the flow cytometric detection of Stx on PMN from HUS patients by us and recently by Brigotti et al. [14]? Using flow cytometry, in 13 out of 20 children with HUS, they showed binding of anti-Stx mouse monoclonal antibody to PMN. In our last three investigated patients with a proven HUS, PMN from only one patient were clearly positive for the antibody against Stx2 and became negative at remission. A possible explanation is the exposure of altered plasma membrane structures on activated PMN. In D+HUS children, PMN have been found to adhere more avidly to the endothelium and induce endothelial injury, assessed morphologically by degradation of endothelial cell fibronectin [15]. Alpha-1 antitrypsin complexed elastase is raised in HUS patients reflecting PMN activation and degranulation [16]. Fernandez et al. showed a reduced degranulatory capacity in response to cytokines and a decreased intracellular granule content indicating a preceding process of activation [10]. To underpin our hypothesis, we have tested the interaction of the antibody against Stx2 with PMN activated by HD and compared it with PMN obtained from patients treated by chronic PD. Degranulation of PMN appears to be an ongoing process during HD, starting directly after the initiation [17]. In the majority of patients on HD, the activated PMN collected after an HD session showed a binding of the antibody against Stx2. This was not seen in PMN from patients treated by chronic PD.
From these experiments we hypothesize that in activated PMN, membrane changes take place that cause binding of the Stx2-antibody. The characteristics of these binding sites are still unknown. Therefore, a reliable diagnosis of HUS patients demonstrating the presence of Stx on the surface of PMN by anti-Stx2 is doubtful.
It remains an unsolved issue how Stx is transported in the circulation. Recently, the acute phase protein human serum amyloid P has been suggested by Kimura et al. [18] as a possible neutralizing factor. But whether it can also target Stx to the glomerular endothelium needs to be elucidated.
|
Acknowledgements |
|---|
|
TopAbstractBackgroundMethodsResultsDiscussionAcknowledgementsReferences |
|---|
This work was supported by a grant from the Dutch Kidney Foundation (PC 153). The authors like to thank L. van Titz and G. Grutters for their excellent technical assistance. Also they would like to thank Dr H. van Hamersvelt for his support.
Conflict of interest statement. The results presented in this article have not been published previously in whole or part, except in abstract form.
|
References |
|---|
|
TopAbstractBackgroundMethodsResultsDiscussionAcknowledgementsReferences |
|---|
- Tarr PI, Gordon CA, Chandler WL. (2005) Shiga-toxin-producing Escherichia coli and haemolytic uraemic syndrome. Lancet 365:1073–1086.[Web of Science][Medline]
- Heuvelink AE, van de Kar NCAJ, van der Velden TJAM, Chart H, Monnens L. (1999) Verocytotoxin-producing Escherichia coli infection in household members of children with hemolytic-uremic syndrome in the Netherlands. Pediatr Inf Dis J 18:709–714.[CrossRef][Web of Science][Medline]
- Watanabe T, Tapchaisri P, Chongsa-nguan M, Chaicumpa W. (2001) Localization of Shiga toxins of enterohaemorrhagic Escherichia coli in kidneys of paediatric and geriatric patients with fatal haemolytic uraemic syndrome. Microb Patho 31:59–67.
- Ludwig K, Karmali MA, Sarkim V, et al. (2001) Antibody response to Shiga toxins Stx2 and Stx1 in children with enteropathic hemolytic-uremic syndrome. J Clin Microbiol 39:2272–2279.
[Abstract/Free Full Text] - te Loo DM, Monnens LA, van Der Velden TJ, et al. (2000) Binding and transfer of verocytotoxin by polymorphonuclear leukocytes in hemolytic uremic syndrome. Blood 95:3396–3402.
[Abstract/Free Full Text] - Grooteman MPC, Bos JC, van Houte AJ, van Limbeek J, Schoorl M, Nube MJ. (1997) Mechanisms of intra-dialyser granulocyte activation: a sequential dialyser elution study. Nephrol Dial Transplant 12:492–499.
[Abstract/Free Full Text] - Perera LP, Marques LR, O'Brien AD. (1988) Isolation and characterization of monoclonal antibodies to Shiga-like toxin II of enterohemorrhagic Escherichia coli and use of the monoclonal antibodies in a colony enzyme-linked immunosorbent assay. J Clin Microbiol 26:2127–2131.
[Abstract/Free Full Text] - van Tits LJ, Hak-Lemmers HL, Demacker PN, Stalenhoef AF, Willems PH. (2000) Oxidized low-density lipoproteins induces calcium influx in polymorphonuclear leukocytes. Free Radic Biol Med 29:747–755.[CrossRef][Web of Science][Medline]
- Torgersen ML, Lauvrak SU, Sandvig K. (2005) The A-subunit of surface-bound Shiga toxin stimulates clathrin-dependent uptake of the toxin. FEBS Journal 272:4103–4113.
- Fernandez GC, Gomez SA, Rubel CJ, et al. (2005) Impaired neutrophils in children with the typical form of hemolytic uremic syndrome. Pediatr Nephrol 20:1306–1314.[CrossRef][Web of Science][Medline]
- Aoki Y and Takeda T. (2002) Effect of Shiga toxin on granulocyte function. Microb Pathog 32:279–285.[CrossRef][Web of Science][Medline]
- Holle JU, Williams JM, Harper L, Savage CO, Taylor CM. (2005) Effect of verocytotoxins (Shiga-like toxins) on human neutrophils in vitro. Pediatr Nephrol 20:1237–1244.[CrossRef][Web of Science][Medline]
- Liu J, Akahoshi T, Sasahana T, et al. (1999) Inhibition of neutrophil apoptosis by verotoxin 2 derived from Escherichia coli O157:H7. Infect Immun 67:6203–6205.
[Abstract/Free Full Text] - Brigotti M, Caprioli A, Tozzi AE, et al. (2006) Shiga toxins present in the gut and in the polymorphonuclear leukocytes circulating in the blood of children with hemolytic-uremic syndrome. J Clin Microbiol 44:313–317.
[Abstract/Free Full Text] - Forsyth KD, Simpson AC, Fitzpatrick MM, Barratt TM, Levinsky RJ. (1989) Neutrophil-mediated endothelial injury in haemolytic uraemic syndrome. Lancet 2:411–414.[Web of Science][Medline]
- Fitzpatrick MM, Shah V, Trompeter RS, Dillon MJ, Barratt TM. (1992) Interleukin-8 and polymorphoneutrophil leucocyte activation in hemolytic uremic syndrome of childhood. Kidney Int 42:951–956.[Web of Science][Medline]
- Gritters M, Grooteman MP, Schoorl M, et al. (2006) Citrate anticoagulation abolishes degranulation of polymorphonuclear cells and platelets and reduces oxidative stress during haemodialysis. Nephrol Dial Transplant 21:153–159.
[Abstract/Free Full Text] - Kimura T, Tani S, Matsumoto Yi Y, Takeda T. (2001) Serum amyloid P component is the Shiga toxin 2-neutralizing factor in human blood. J Biol Chem 276:41576–41579.
[Abstract/Free Full Text]
Accepted in revised form: 25.10.06
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Brigotti, D. Carnicelli, E. Ravanelli, S. Barbieri, F. Ricci, A. Bontadini, A. E. Tozzi, G. Scavia, A. Caprioli, and P. L. Tazzari Interactions between Shiga toxins and human polymorphonuclear leukocytes J. Leukoc. Biol., October 1, 2008; 84(4): 1019 - 1027. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. P. Griener, G. L. Mulvey, P. Marcato, and G. D. Armstrong Differential binding of Shiga toxin 2 to human and murine neutrophils J. Med. Microbiol., November 1, 2007; 56(11): 1423 - 1430. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||