During this presentation we will review techniques and recommendations for extracting endotoxin from cotton dust. Although several of the techniques have been used on numerous samples for years, a few of the recommendations do not appear to have been tested. One of the most unifying extraction methods, the standard extraction procedure, was developed after a special session at the 198 Cotton Dust Research Conference (Table 1)(9). This procedure was distributed along with a series of standard cotton dust samples by Cotton Incorporated in 1982-1984. The purpose of this study was to provide researchers with a single dust source and recommended extraction method so that comparisons in chemical, biological, and physiological properties could be made among the various laboratories. Six laboratories reported endotoxin contents for the <38 µm and 38-75 µm fractions of the standard cotton dust (see Table )(9). Endotoxin contents were determined by the Limulus amebocyte lysate (LAL) assay. Lab-to-lab variations were surprisingly small. Most of the researchers reported twice as much endotoxin in the fraction containing smaller particles (< 38 µm).

Fischer performed washing experiments on two 1982 standard cotton dust samples: stock (unsieved) dust and the 38-75 µm sieved fraction (5). Washing experiments were conducted by suspending 0.5 g dust in 50 ml of pyrogen-free water containing 0.1% Tween 80 for 20 minutes at 25°C and 65°C. Ten percent or less of the endotoxin (or lipopolysaccharide (LPS)) was retained in each dust sample regardless of wash temperature. The amount of LPS retained was lowest, and weight retained highest, for the 38-75 µm fraction washed at 65°C (Table 3). The amount of weight retained was not a simple function of temperature or dust type.

Experiments in our labs have failed to establish a mass balance for endotoxin extraction from cotton dust. Sequential extractions of cotton dust have provided more cumulative endotoxin in the extracts than is available from cotton dust by a single aqueous extraction for the LAL assay. Although mass balances for endotoxin removal from cotton dust have not been observed with aqueous extractions, they must exist. The lack of a mass balance indicates that the extraction methods for endotoxin analysis of the control dust are inadequate. According to studies by Milton and coworkers, endotoxin extraction is complicated by endotoxin adsorption to physical surfaces (14).