Harold Finkelstein

Macromolecular Components of Untreated and of Formolized Normal Chick Embryo Tissue.

Macromolecular substances have been found in healthy tissues of both plants and animals. 1 In experiments associated with studies of the equine encephalomyelitis virus protein 2 and the non-infectious immunizing principle 3 of formalin-treated virus-diseased chick embryos, 4 we have obtained such substances from normal chick tissue and from the tissue treated with formalin. Chick embryos incubated for 11 days were harvested, chilled, the heads removed and the bodies ground in a Ten Broeck grinder held in cracked ice. The resulting body mince was made to a 20% suspension and extracted as indicated. Large particles were removed with the angle centrifuge and the resulting extracts fractionated by alternate ultracentrifugal runs of 9,000 g for 5 minutes and 67,000 g for 1 to 1I/2 hours. The products have been studied by ultracentrifugal analysis. The results observed when extraction was made and the pellets were dissolved with 0.005 M ammonium hydroxide-ammonium chloride buffer pH 8.2 are shown in Figures 1 and 2. After 2 ultra-centrifugal cycles, double or diffuse boundaries were present whereas only one of slight diffuseness was seen after the third cycle. Here s20° = ca 70 × 10−13 cm sec−1 dynes−1. A similar material was obtained by extraction with glycerin in 50% concentration in the dilute buffer. When extraction was made in 0.9% sodium chloride solution alone or combined with glycerin and the pellets were dissolved in saline, indications of the same conponent were evident after the second cycle (Fig. 3) but further homogeneity was not attained by additional cycles (Fig. 4). The sharpest boundaries were obtained with material derived by extraction of tissue and solution of the pellets in distilled water (Fig. 9 and 10).

Rate of Inactivation of Equine Encephalomyelitis Virus (Eastern Strain) Relative to H ion Concentration.

Detailed studies of the pH stability range of infectivity of plant viruses, summarized recently by Stanley, 1 show that for each virus studied there is a well defined region where infectivity is relatively stable. On both sides of this region infectivity is lost in a regular way so that the curve plotted to relate infectivity with pH of the medium is smooth and trapezoidal. Certain of the animal viruses 2 , 3 appear to behave in a like manner. It was entirely unexpected, then, when conditions were encountered under which the virus of equine encephalomyelitis (Eastern Strain) seemed to behave in a different way. The experiments were made on virus propagated in chick embryos. 4 Virus-infected embryos were “harvested” when moribund and ground with sand in hormone broth to a 10% suspension. Centrifuged free of sand and gross tissue particles, the whole extract was mixed with composite buffer solution (0.05M) of various pHs. 2 , 3 Mixtures were maintained between 0°C and 5°C, and after various intervals, usually 1 hour, 1 day, and 1 week, portions were removed for test. The pH of each was readjusted approximately to neutrality and virus infectivity determined by titrations in decimal dilutions in mice. 5 pH was determined and frequently checked with the glass electrode. Four such experiments covering the range from pH 1.0 to pH 12.0 have shown essentially uniform results, and one is summarized in Fig. 1. The curve after 1 hour is similar in contour to those of other viruses, the region of greatest stability lying between pH 3.5 and pH 11.5. After 1 week, maximum stability is at pH 7.5 to pH 8.5, while a second region of relative stability is apparent between pH 3.5 to pH 5.0.

Macromolecular Components of Chick Embryo Tissue Diseased With the Virus of Equine Encephalomyelitis.

Chick embryo tissue diseased with virus of equine encephalomyelitis (Eastern strain) has been studied in relation to the homogeneous macromolecular substances recently obtained from normal tissue 1 and the results observed are described here. Eleven-day chick embryos were inoculated with virus and harvested when moribund. The heads were removed and the body tissue was ground without diluent in chilled Ten Broeck grinders. Extraction, ultracentrifugal fractionation and analysis were similar to those of the previous studies. 1 When the diseased tissue mince was extracted in 20% suspension with a mixture of equal parts 0.9% sodium chloride solution and glycerin at pH 8.5 and the pellets were dissolved in 0.9% saline (the sequence employed by Wyckoff 2 for isolation of the equine encephalomyelitis virus protein 3 ) the component with s20° = ca 250 × 10−13 cm sec−1 dynes −1 was observed in the second or third ultracentrifugal cycle (Fig. 1 and 2). This product was infectious to the order of 1014 mouse units per gram, and the result corroborated previous findings. 3 . On the other hand, when extraction was made with water, no evidence of this material was seen; instead, only boundaries with s20° = ca 70 × 10−13 cm sec−1 dynes −1 were obtained (Fig. 3) as from non-diseased tissue. 1 Infectivity to the order 1014 was likewise associated with this product. Fractionation of 0.9% saline extracts of virus-diseased tissue formolized to 0.4% yielded pellets which, on solution in saline, gave diffuse boundaries indicative of s20° = ca 70 × 10−13. Formolization and extraction with water at pH 7.0 gave a product with relatively sharp boundaries (Fig. 3). These derivatives of 0.9% saline and of water extraction were both highly antigenic, protecting guinea pigs against 1000 mouse infectious units of virus given intracerebrally.

Immunization with Non-Infectious Formalin Derivative of Purified Equine Encephalomyelitis Virus Protein.∗

Previous ultracentrifugal studies 1 of crude equine encephalomyelitis chick vaccines 2 have failed to yield definite evidence of the character of the immunizing principle. Information concerning its probable nature has been sought in the present work by study of the purified virus protein treated with formaldehyde. Purified protein (Eastern strain), 3 dissolved in Ringers solution, pH 8-9, 2.0 mg per cc, was treated with various concentrations of CH2O. Before exposure to CH2O, one m.i.u. 4 of protein was 10-13.5 g. When the concentration of CH2O was less than 0.01 M, inactivation was not always complete in 2 weeks, and tests for immunizing capacity were not made. With other concentrations of CH2O used, inactivation was complete in the time shown in Table I, as judged by the failure of 10-5.5 g of protein to infect mice. The effect of CH2O on the protein molecules was somewhat similar to its action on tobacco mosaic virus protein. 5 With CH2O 0.01 M or less, the protein remained in solution but lost slightly in homogeneity, showing a slightly diffuse boundary with a sedimentation pattern similar to that of the untreated protein 3 persisting with little change for more than 3 weeks. In 0.02 M and higher concentrations of CH2O, the protein rapidly became quite inhomogeneous and increasingly insoluble, losing the definite boundary in 2-4 days. In no instance was there evidence of molecular disruption to small soluble protein fragments of uniform size, nor was there evidence of the component S20° = ca 60 × 10-13 cm sec-1 dynes-1 previously described. 6 The non-infectious, soluble though slightly inhomogeneous protein derivative with 0.01 M CH2O has immunized all guinea pigs receiving it, as shown in the typical experiment in Table I.

Macromolecular Component of Chick Embryo Tissue Diseased with Western Strain Equine Encephalomyelitis Virus.

A macromolecular component has been isolated 1 by ultracentrifu-gation from chick embryo tissue diseased with the Eastern strain of equine encephalomyelitis virus (E.S.). This protein, within the limits of tests thus far made, behaves as the virus and is specific 2 to the virus-diseased embryo tissue, though it is separable with difficulty from the lighter normal tissue component with s20° = ca 70 × 10−13 cm sec−1 dynes−1. 3 In the present paper are described ultracentri-fugal studies of extracts of chick embryo tissue diseased with Western strain virus (W.S.) from which a macromolecular protein possessing the properties of this strain has been isolated. In a typical experiment, diseased tissue was extracted 24 hours at about 5°C in 4 times its volume of 0.15 M NaCl solution made to pH 8.5 with NHiOH. Cleared of tissue debris by angle centrifugation, the extract was ultracentrifuged in 8 15-cc tubes at 67,000 g for 30 minutes. The 8 pellets were taken up in 60 cc water, pH 8.5 with NH4OH, and the resulting suspension was spun in 4 tubes at 6000 g for 5 minutes. The supernatant fluid was then spun in 4 tubes at 17,000 g for 30 minutes. A specimen from the 4 pellets was dissolved in 0.2 M NaCl solution adjusted to pH 9.0 with NH4OH for examination in the analytical ultracentrifuge (Fig. 1), and the remainder was taken up again in 60 cc water for repetition of the cycle of 6000 g and 17,000 g. The final pellets were dissolved in 0.2 M NaCl solution for ultracentrifugal analysis (Fig. 2). The sedimentation diagram after the second cycle showed the Indistinct, diffuse boundary (Fig. 1, a) of persisting s20° — ca 70 × 10−13 and the more prominent, somewhat diffuse boundary (Fig. 1, b) of a heavier material.

Improved Sequence for Rapid Consistent Purification of Equine Encephalomyelitis Virus Protein.

The procedure described here for purification of equine encephalomyelitis virus protein 1 yields consistently a product of high molecular homogeneity. Advantage is taken chiefly of the following principles: (1) prolonged extraction to aid in eliminating the normal chick tissue component; 2 (2) extraction, fractionation and solution of the protein in a balanced salt solution, mammalian Ringer, 3 instead of 0.9% NaCl or buffer salt solutions; (3) filtration of crude extracts with celite to remove mucoid and colloid materials; and (4) aggregation or partial precipitation of the protein in slightly acid medium prior to the first ultracentrifugal cycle. Diseased embryo tissue ground in the cold in the usual way, 2 is suspended in 4 times its volume of normal Ringers solution made to pH 9.0 with NH4OH without buffer salts. After extraction at 5°C for 72 to 96 hours, gross material is eliminated in the angle centrifuge. To each 100 cc of the turbid supernatant fluid, 5 g of No. 512 Celite Filter Aid (Johns-Manville Co., N. Y.) are added and the suspension is filtered with suction through a 1 to 2 mm mat of No. 503 Celite. Standard celite then added to the filtrate, 2 g per 100 cc, is filtered off through a mat also of standard celite. Sometimes the latter step is repeated to obtain an entirely clear filtrate. The filtrate is acidified to pH 6.5 with 0.2 N HC1, and 120 cc of it distributed immediately in 8 collodion tubes is spun at 17,000 g for 45 minutes. The pellets are taken up in 30 cc of Ringers solution. In 2 tubes, the solution is spun at 17,000 g for 5 minutes. The supernatant fluid diluted to 60 cc is ultracentrifuged 30 minutes at 67,000 g and the resulting pellets are taken up in 15.0 cc of Ringer fluid.