A simple method of concentrating and purifying bacteriophage has been described. The procedure consisted essentially in collecting the active agent on a reinforced collodion membrane of a porosity that would just retain all the active agent and permit extraneous material to pass through. Advantage was taken of the fact that B. coli will proliferate and regenerate bacteriophage in a completely diffusible synthetic medium with ammonia as the only source of nitrogen, which permitted the purification of the bacteriophage by copious washing.
The material thus obtained was concentrated by suction and after thorough washing possessed all the activity of the original filtrate. It was labile, losing its activity in a few days on standing, and was quickly and completely inactivated upon drying. This material contained approximately 15 per cent of nitrogen and with 2 or 3 mg. samples of inactive dry residue it was possible to obtain positive protein color tests.
The concentrated and purified bacteriophage has about 10–14 mg. of nitrogen, or 6 x 10–17 gm. of protein per unit of lytic activity. Assuming that each unit of activity represents a molecule, the calculated maximum average molecular weight would be approximately 36,000,000, and on the assumption of a spherical shape of particles and a density of 1.3, the calculated radius would be about 22 millimicra.
By measurement of the diffusion rate, the average radius of particle of the fraction of the purified bacteriophage which diffuses most readily through a porous plate was found to be of the order of magnitude of 9 millimicra, or of a calculated molecular weight of 2,250,000. Furthermore, when this purified bacteriophage was fractionated by forcing it through a thin collodion membrane, which permits the passage of only the smaller particles, it was possible to demonstrate in the ultrafiltrate active particles of about 2 millimicra in radius, and of a calculated molecular weight of 25,000.
It was of interest to apply this method of purification to a staphylococcus bacteriophage. Since this organism does not readily grow in synthetic medium, a diffusate of yeast extract medium was employed. The better of two preparations contained about 10–12 mg. of nitrogen per unit of lytic activity. Although this is about one hundred times the amount of nitrogen found in an active unit of B. coli bacteriophage, nevertheless, the diffusion rate experiments gave results which paralleled those obtained with the coliphage. The diffusible particles of the crude staphylococcus bacteriophage had a radius of about 7 millimicra, and a calculated molecular weight of about 1,000,000, while the particles of the same phage which appeared in the ultrafiltrate through a thin collodion membrane had a radius of about 2.4 millimicra and a calculated molecular weight of about 45,000.
It appears, therefore, that the active principle is distributed as particles of widely different sizes. However, since the smaller particles have all the properties of bacteriophage, the larger particles probably do not represent free molecules, but either are aggregates, or more likely, inactive colloids to which the active agent is adsorbed.
The protein isolated, which bears the phage activity, is capable of stimulating the production of antilytic antibodies on parenteral injection into rabbits or guinea pigs. It retains its specific antigenicity when inactivated by formalin, but not when inactivated by drying.