Mouse interferons of three size classes (A, 35,000 to 40,000 daltons; B, 26,000 to 33,000 daltons; and C, 20,000 daltons) were purified from Ehrlich ascites tumor cells infected with Newcastle disease virus. The sequences of the first 24 amino acids (No. 17 has not been identified) of interferons A and B are identical. The sequence of the first 20 amino acids of interferon C differs from that of A and B in 18 positions. There is partial homology in amino terminal sequence between mouse interferons A (or B) and a human fibroblast interferon and between mouse interferon C and a human lymphoblastoid interferon.

The N-terminal sequences of mouse Ehrlich ascites tumor cell IFN beta (35,000-40,000 daltons) and IFN alpha (20,000 daltons) differ in 18 out of 20 positions. Furthermore, these two IFN species show little immunological cross reactivity. We treated mouse L929 cells and human HeLa S3 cells with essentially pure mouse IFN alpha or IFN beta or both at various concentrations and for various lengths of time. From the treated cells we prepared extracts and compared in these the activities of (2'-5')(A)n synthetase, an enzyme that was earlier shown to be induced by partially purified IFN preparations. The effects of treatment of mouse L929 cells with pure IFN alpha or IFN beta on (2'-5') (A)n accumulation in the cell extracts were very similar both in respect to the dependence on the length of exposure of the cells to the IFNs and on IFN concentration. Treatment with both IFN alpha and IFN beta at concentrations resulting in only partial induction of the enzyme led to an additive rather than to a synergistic effect. The maximal level of enzyme induced was the same in cells treated with high concentrations of IFN alpha or IFN beta or both. Mouse IFN alpha was as active as IFN beta in inducing a double-stranded RNA-dependent protein kinase in mouse L cells. The treatment of HeLa S3 cells with IFN beta did not affect the accumulation of (2'-5') (A)n in their extracts whereas treatment with IFN alpha boosted the accumulation though to a lesser extent than in the case of mouse L cells. These results are in line with the finding that mouse IFN alpha can, but mouse IFN beta cannot convert HeLa S3 cells into the antiviral state and also with the more pronounced homology in N-terminal sequence between mouse IFN alpha and a human (lymphoblastoid) IFN chi than between mouse IFN beta and a human IFN beta.

The Epstein-Barr virus-transformed B cell line RPMI 1788 constitutively produces autocrine growth factors with molecular masses of 17 kDa, 24 kDa and 35 kDa. All three molecular forms were completely neutralized with anti-interleukin (IL) 1 alpha antiserum. Although IL 1 alpha and IL 1 beta mRNA were both equally detectable by Northern blotting, no IL 1 beta activity was found in partially purified RPMI 1788 supernatant. The growth of low density-seeded RPMI 1788 cells is specifically dependent on the presence of either IL 1 alpha or IL 1 beta. Since no other cytokine was found to be capable of sustaining proliferation, this cell line is suitable for the identification and quantification of IL 1, even in the presence of other cytokines.

We report a simple procedure for the isolation of a mouse interferon beta (beta IFN) from Ehrlich ascites tumor cells infected with Newcastle disease virus. The procedure consists of four chromatographic steps (two on controlled pore glass, one on CM-Sephadex, and one on phosphocellulose) with stepwise elution. The product purified is a 35-40 kD beta IFN with a specific activity of 10(9) units/mg protein.

That only 20 amino acids occur naturally accounting for the structural and functional diversity of proteins remains a mystery. We show here that this is a consequence of the action of a symmetry group, identify the presence of hydrogen bond donor groups, presence of sp3 hybridized [gamma] carbons, absence of [delta] carbons and linearity as properties central to side chain design and quantify the chemical logic of protein sequences.