PLASMA LIPOPOLYSACCHARIDE-BINDING PROTEIN AND BACTERICIDAL PERMEABILITY-INCREASING FACTOR IN CRF AND HD PATIENTS

The recent characterization of lipopolysaccharide binding protein (LBP ) and bactericidal/permeability increasing factor (BPI) have provided the opportunity to examine the natural factors that regulate cytokine production in response to endotoxin in patients on hemodialysis (HD). Whole blood was collected in EDTA from 28 undialyzed patients with chr onic renal failure (undialyzed CRF), 36 patients on chronic HD (HD) an d 15 healthy controls, and plasma levels of LBP and BPI were measured by a sandwich ELISA. Plasma LBP levels in undialyzed patients with CRF (P = 0.04) and patients on HD (P = 0.01) were significantly higher th an those in healthy controls, but not significantly different from eac h other. Plasma BPI levels in undialyzed patients with CRF and patient s on HD were not significantly different from those in healthy control s. There was no correlation between serum creatinine and plasma levels of either LBP or BPI. Peripheral blood mononuclear cells (PBMC) were harvested from healthy volunteers by Ficoll-Hypaque separation, and 0. 125 mL of 10 x 10(6)/mL suspensions were incubated with 0.125 mL of te st plasma (containing different LBP/BPI ratios) and 0.25 mL of RPMI, c ontaining 1 ng/mL of endotoxin, for 24 h at 37 degrees C. Samples were subjected to three freeze-thaw cycles, and total interleukin-1 recept or antagonist (IL-1Ra) or interleukin-1 alpha (IL-1 alpha) production was measured by a specific non-crossreactive RIA. The results of this study showed: (i) IL-1Ra production by endotoxin-stimulated PBMC incub ated with pooled plasma from HD patients with LBP/BPI ratios of 11 x 1 0(2), 167 x 10(2), 379 x 10(2), and 778 x 10(2), respectively was 1466 +/- 195 pg, 3105 +/- 462 pg, 8179 +/- 1020 pg, and 4770 +/- 1185 pg ( P < 0.001); (2) Paired plasma collected before dialysis (predialysis) and at 15 min after the start of dialysis (15 minute) with cellulose m embranes showed a negligible change in plasma LBP levels (-3 +/- 5%), but a 6681 +/- 1788% increase in plasma BPI levels. Consequently, comp ared with predialysis plasma, there was a 35 +/- 6% decrease in endoto xin-stimulated IL-la production by PBMC incubated with plasma drawn at 15 min (P = 0.001); (3) Compared with the PBMC incubated with predial ysis plasma from HD patients, there was a 39 +/- 5%, 53 +/- 5%, and 60 +/- 5% decrease in endotoxin-stimulated IL-1 alpha production in the presence of 1 ng/mL, 10 ng/mL, or 1 mu g/mL of recombinant BPI, respec tively (P < 0.003). These results suggest that the plasma LBP:BPI rati o could influence cytokine production in response to bacterial endotox in; the high LBP:BPI ratios observed in patients with chronic renal fa ilure probably imparts an increased susceptibility to endotoxin-stimul ated cytokine production; and natural or pharmacological increases in plasma BPI levels and the consequent decrease in LBP:BPI ratios could attenuate this susceptibility to endotoxin-stimulated cytokine product ion.