PLATE VTHE haemolytic activity of Escherichia coli has been investigated by only a few authors. Although there are many reports in the literature about haemolytic E. coli, most authors have merely noted that haemolytic strains were isolated. Lovell and Rees (1960) and Smith (1963) first demonstrated the production of an extracellular haemolysin (designated a-haemolysin by Smith) in an alkaline meat-extract broth. Smith (1963) showed also that a cell-associated haemolysin (designated ,6-haemolysin) was produced by some strains. Other investigators (Snyder and Koch, 1966; Snyder and Zwadyk, 1969; Short and Kurtz, 1971) observed extracellular haemolytic activity when a strain of serogroup 0 6 was grown in chemically defined medium.There is little information about the purification of E. coli a-haemolysin. Zwadyk and Snyder (1971), like Lovell and Rees (1960), were unable to recover haemolytic activity after dialysis of ammonium-sulphate precipitates. Short and Kurtz (1971) found that two peaks of haemolytic activity appeared after chromatography of culture filtrates on Sephadex G-200 and Sepharose 6B. However, they gave no indication of the extent of purification and the activity of their starting material was low.Only one author, Smith (1963), has investigated the biological properties of a-haemolysin. He observed that the toxin was active against a range of erythrocyte species, that large amounts of haemolysin were required to kill mice and rabbits after intravenous injection, and that haemolytic culture filtrates did not cause dermonecrosis in the skin of rabbits or guinea-pigs. The haemolysin was unreactive after intragastric administration to rabbits.In the present study we have attempted to clarify some of the deficiencies of previous work. Our main objectives were as follows: (a) to produce large amounts of high-titre a-haemolysin, (b) to purify and characterise this agent, and (c) to investigate the biological properties of a-haemolysin in preparations with high specific activities. In an accompanying paper, the kinetics of erythrocyte lysis by a-haemolysin are reported (Rennie, Freer and Arbuthnott, 1974).

PLATE VITECHNIQUES for studying the mode of action of bacterial toxins at the molecular level are fast becoming valuable and necessary tools for the determination of the role of these agents as virulence factors in diseases of man and animals. The mechanism of action of Escherichia coli haemolysin is poorly understood. Although enough evidence now exists to conclude that calcium ions are required for haemolytic activity, only two groups of investigators have attempted to study the kinetics of erythrocyte lysis by E. coli a-haemolysin.The findings of Zwadyk and Snyder (1971) and Short and Kurtz (1971) indicated that lysis of sheep erythrocytes was dependent on haemolysin concentration, pH and temperature. Less haemolysis was observed as the concentration of red cells (RBC) was increased. It has been suggested that a complex consisting of haemolysin, calcium ions and RBC must be maintained until the lytic event (Short and Kurtz, 1971). These authors found that the addition of ethylenediaminetetraacetic acid (EDTA) at any time during the haemolytic reaction prevented subsequent haemolysis. They also noted that neither reducing agents nor lecithin or cholesterol affected a-haemolysin.Both groups of workers used methods that required centrifugation of reaction mixtures at different times, followed by spectrophotometric estimation of haemoglobin. In these processes there are unavoidable delays during which haemolysin-affected cells may be lysed.In this study, the turbidity of reaction mixtures was monitored continuously at 650 nm in a controlled-temperature spectrophotometer. By this technique it was possible to make rate measurements on haemolysis curves. MATERIALS AND METHODSHaemolysin production and purification. The preparations of E. coli u-haemolysin used in this study were produced in glucose-nutrient-broth medium and were purified by the three-stage method of Rennie and Arbuthnott (1974). Because it was more readily available, stage-I1 u-haemolysin was used for most assays. However, in each experiment, purified stage-III a-haemolysin was tested to confirm the results obtained with less pure preparations. (Rennie and Arbuthnott, 1974). Kinetic studies were carried out with silica cells of 1-cm light path in a controlled-temperature Unicam SP800 spectrophotometer, to which was attached a Unicam SP22 chart recorder (Pye-Unicam, Cambridge, England) set at a five-times multiplication factor. The spectrophotometer was adjusted so that extinction (E) was monitored at a constant wavelength of 650 nm.A 0.7% suspension of sheep erythrocytes (SRBC) was made in Veronal buffer, pH 7.3 (Cruickshank, 1969), containing lOmM CaCl2 (VC buffer). The spectrophotometer and chart recorder were then calibrated on a linear scale with the SRBC suspension. The contents of a 1-cm cell, containing 1-4 ml of VC buffer and 0.1 ml of 0.7% SRBC, were mixed by inversion and read at E 6 5 0 . The zero controls of the spectrophotometer and chart recorder were adjusted if necessary such that the E650 was 0.40. This represented 100% erythrocyte ...

The production of hemolysin by Escherichia coli after adding various carbohydrates to starved cultures was investigated.

A system for monitoring, recording and storing arterial pressure and heart rate during continuous extradural analgesia has been developed using an Apple II microcomputer and a Dinamap 1846 non-invasive arterial pressure monitor. The administration of local anaesthetic (or vasopressor) was recorded using a light pen. The computer was programmed to recognize this, and to initiate automatically more frequent measurements of heart rate and arterial pressure. The results of a study using this equipment are reported. The commands for computer control of the Dinamap 1846 are described.

There is a growing interest in the potential benefits to be obtained from closed-loop control of drug infusion [1-5]. The most widely available infusion pump which is specifically designed for control by computer is the 929 volumetric infusion pump produced by Imed. The increasing availability of microcomputers suggests that they will be used more frequently to control such pumps. However, difficulties can arise when the attempt is made to establish communication between the computer and the pump [6] and a method employed for this purpose is described in this article. The aim was to provide a simple method of conducting the required communication between the computer and the pump using a high-level language. The commands specified in the high-level language are passed to a machine-code subroutine which organizes transmission to and receives data from the pump.