The effects of heparin fractions of defined chain length on human neutrophil functions in vitro R. Lever,1 W.T. Lo,2 R.A. Brown,2J. Gallagher3 & C.P. Page.2 1Department of Pharmacology, School of Pharmacy, University of London. WC1N 1AX. 2Sackler Institute of Pulmonary Pharmacology, King's College London. SE1 9RT. 3Christie Hospital NHS Trust, Manchester. M20 4BX.

Print abstract Search PubMed for: Lever R Lo WT Brown RA Gallagher J Page CP

Heparin has long been known to possess anti-inflammatory properties that are unrelated to its anticoagulant activity. However, whereas the structural requirements for the effects of heparin on haemostasis are well characterized, those involved in other pharmacological actions are, at best, poorly understood.

Previously, we have described the ability of unfractionated heparin (UH) and certain chemical analogues to modulate the adhesion of human neutrophils to cultured endothelial cells (Lever et al., 2000) and to inhibit neutrophil activation and degranulation in response to a variety of stimuli (Brown et al., 2003).

In the present study, we have examined the effects of a range of heparin fractions, prepared from the parent molecule by three different chemical or enzymatic methods, on neutrophil adhesion and activation in vitro, in order to determine whether heparin chains of specific lengths may possess differential effects in our assays.

Low molecular weight heparins (LMWH), prepared from UH by nitrous acid depolymerisation, heparinase digestion or beta-eliminative cleavage, were further fractionated by these methods, respectively, to yield heparin populations of four, six and fourteen saccharides in length (4s, 6s and 14s). These molecules were compared for their effects against a) neutrophil adhesion to interleukin-1ß (IL-1ß) stimulated HUVECs and b) neutrophil elastase release in response to f-met-leu-phe (fMLP), both with and without prior priming with tumour necrosis factor-(TNF-).

Neutrophils, isolated from the venous blood of healthy donors (n=6) were a) ⁵¹Cr-labelled and applied to IL-1ß-stimulated HUVECs (6h, 10 U ml^-1) at a density of 10^6, cells ml^-1, in the absence and presence of test compounds, non-adherent cells removed after 30 minutes and adherent cells quantified by -counting or, b) stimulated with fMLP (10^-7 M), with or without pre-treatment with TNF-(30 minutes; 100 U ml^-1), in the absence and presence of test compounds and elastase release quantified colorimetrically after 45 minutes. Data were analysed by ANOVA followed by Dunnett's test and were considered significant if P < 0.05.

The three LMWH inhibited adhesion weakly (max inhibition 29.8 ± 3.5%, 100 µg ml^-1, LMWH prepared by beta elimination) whereas the further fractionated materials were without effect. Elastase release was inhibited significantly by the three LMWH (1-100 µg ml^-1), the 14s heparins (10-100 µg ml^-1) and the 6s and 4s heparins (100 µg ml^-1) under both conditions of cell stimulation. There were no differences in the effects of the starting materials, nor between subfractions of equal chain length prepared by the different methods. Interestingly, the 14s heparins had a greater maximum effect on elastase release than the respective LMWH starting materials (eg 91.1 ± 3.4% inhibition vs 66.7 ± 2.1% inhibition, respectively, 100 µg ml^-1, materials prepared by nitrous acid depolymerisation, neutrophils stimulated with fMLP only).

Whereas the effects of UH on neutrophil-endothelial adhesion appear to be diminished or removed by fractionation, in contrast, inhibition of neutrophil degranulation is retained by smaller fractions of the parent molecule. However, our data possibly indicate that an optimum chain length of heparin exists for this effect. Further studies are required to examine this in more detail, the results of which may prove useful in the design of novel drugs with specific anti-inflammatory actions.

Brown et al. (2003). Br. J. Pharmacol. (in press).
Lever et al. (2000). Br. J. Pharmacol. 129, 533-540.