The activation of murine macrophages and natural killer cells by the partially thiolated double stranded RNA poly(I)-mercapto poly(C).

Partially thiolated analogs of the biological response modifier poly I.poly C (pI.pC) were synthesized. Each of these analogs (pI.MPC) contained a partially thiolated polycytidylate (MPC) strand containing either 1.2%, 4.6%, or 17% 5-mercaptocytosine (%SH) randomly introduced throughout the polynucleotide. The ability of these double stranded RNAs (dsRNAs) to activate murine peritoneal macrophages in vitro and augment natural killer (NK) cell activity in mice following intraperitoneal (i.p.) administration was determined. Macrophages were treated in vitro for 24 hours with pI.pC, or pI.MPC, washed, and then overlayed with exponentially growing L1210 leukemia cells at an effector to target (E:T) ratio of 25:1. The cytostatic effect of the macrophages on the L1210 cells was determined by 3H.thymidine pulse labelling. The rank order of potency for macrophage activation was determined to be: pI.pC>pI.MPC(1.2% SH)>pI.MPC(4.6% SH)pI.MPC(17% SH). Twenty hours following i.p. administration of 5 mg/kg of each pI.MPC analog, splenic NK cell activity was assessed in a standard 51Cr release assay using the murine tumor target cell line YAC- 1. The rank order of potency observed for NK cell activation was determined to be; pI.pCpI MPC(1.2% SH)>pI.MPC(4.6% SH)>pI.MPC(17% SH). These dsRNAs activated NK cells in a dose dependent manner. The efficacy and time course for NK cell activation following i.p. administration of pI.MPC (1.2% SH) at a dose of 10 mg/kg was directly compared to an equivalent 10 mg/kg i.p. dose of pI.pC. NK cell activation took place within three hours following treatment with pI-pC whereas the onset of NK cell activation by pI.MPC (1.2%) occurred between 8 and 20 hours post treatment. NK cell activity steadily declined from 24 to 50 hours post treatment at which time the NK activity in both treatment groups was similar. There was a significant correlation between the immunostimulatory potency of these dsRNAs and their experimentally determined melting temperatures (r2 = 0.88) and percent hyperchromicity upon thermal denaturation (r2 = 0.99). At the lower %SH, pI.MPC retains most of the immunostimulatory activities of p1.pC and may serve as a useful and potent biological response modifier.

Poly I-Mercapto Poly C: antiviral, anticellular, and pharmacologic effects.

Thiolation of position 5 of some of the cytosine bases in polycytidylic acid results in the formation of mercaptopolycytidylic acid (MPC). Annealing of MPC to polyinosinic acid (Poly I) results in the formation of double-stranded Poly I-MPC. In this study we investigated the interferon inducing ability, in vivo toxic effects, effect on DNA synthesis, and the effects in human tumor cell lines of Poly I-MPC. Poly I-MPC was capable of inducing human alpha, beta and gamma interferons in the appropriate cell systems. In vivo toxicity was measured in mice, guinea pigs, and rabbits according to FDA guidelines. Weight loss and lethal and pyrogenic effects were markedly lower in Poly I-MPC treated animals than in those that received unmodified Poly I-Poly C. In contrast to the lack of an effect of Poly I-Poly C in human lymphocytes, Poly I-MPC inhibited DNA synthesis. It also inhibited colony formation and was cytotoxic in several human tumor cell lines. Poly I-MPC's ability to induce human alpha, beta and gamma interferons, to inhibit DNA synthesis and its effects in human tumor cell lines demonstrate the potential of this drug for future clinical studies, both as an antiviral and antitumor agent.

The B reversible Z transition of poly(dI-br5dC).poly(dI-br5dC). A quantitative description of the Z form dynamic structure.

The study of poly(dI-br5dC).poly(dI-br5dC) films by infrared spectroscopy shows that in low salt concentration, the conformation of this polynucleotide belongs to the B-family and in high salt concentration to the Z-family. 31P nuclear magnetic resonance and circular dichroism confirm the existence of these two forms. By circular dichroism and ultraviolet absorption, it is shown that the equilibrium constant of the B reversible Z transition depends upon temperature. The deuteration rates of exchangeable protons involved in hydrogen bonds between base pairs were deduced from the changes in absorbance near 1700 cm-1. In the B-form, one class of protons is measured with an exchange half-time of 20 minutes. In the Z-form, two classes of protons are measured with very different exchange half-times, the exchange half-time of the slow protons being of the order of 850 minutes. By comparison of these results with those previously obtained for poly(dG-dC).poly(dG-dC), these very slow protons of these two Z-polynucleotides are identified as the cytosine amino protons. A quantitative description of the dynamic structure of the Z-form is presented.

Minimal interferon induction in dogs, using a modified polyriboinosinic-polyribocytidylic acid complex.

Adult Beagles failed to respond to high concentrations of interferon (IF) when they were injected with a nuclease-resistant complex poly I:C with poly-L-lysine and carboxymethylcellulose (PICLC), by the IV or intrathecal route. An IV dose of 1 mg of PICLC/kg of body weight was lethal to 1 of 3 adult dogs, but induced IF in only 2 dogs. Smaller doses were less toxic, but also were less effective. The injection of a high dose of a known IF inducer (3 X 10(8) egg LD50 of Newcastle disease virus) also failed to induce IF in Beagles. Interferon could not be induced in vitro when primary cultures of neonatal dog lung or kidney were treated with cultures of neonatal dog lung or kidney were treated with PICLC. When these primary cell cultures were compared with the cell line Madin-Darby canine kidney in an IF assay, no difference in sensitivity to IF-induced protection from infection with vesicular stomatitis virus could be shown. This indicated that the sensitivity of the Madin-Darby cell line was not the only factor in determining the lack of IF response in dogs and indicates that the dogs are poor responders to IF induction.

Binding of polynucleotides to fibroblasts. Effects of complex formation with vinyl analogs of nucleic acids.

The binding of polynucleotides or of their vinyl analogs to human fibroblasts is changed when complexes are formed between these compounds. The following polymers have been studied: poly(1-vinylcytosine), poly(1-vinyluracil, poly(9-vinyladenine), polyuridylate, polyadenylate and polyinosinate. Only that complex formation (between poly(1-vinylcytosine) and polyinosinate) which is accompanied by aggregation leads to a considerable increase (30 fold) in binding to cells; all the other complex formations have only moderate effects (0.2-3 fold). Furthermore, a comparison of unordered complexes containing polyinosinate whows that enhanced binding to cells is paralleled by an increased cellular resistance to viral infection.

Treatment of a murine cytomegalovirus infection with exogenous interferon, polyinosinic-polycytidylic acid, and polyinosinic-polycytidylic acid-poly-L-lysine complex.

Treatment of a cytomegalovirus infection of mice with exogenous murine interferon did not alter final mortality or mean day of death. Pretreatment with two interferon inducers significantly reduced mortality, but treatment initiated after infection was not effective.

Recognition of polynucleotides by antibodies to poly(I), poly(C).

The binding of anti poly(I). poly (C) Fab fragments to double or triple stranded polynucletides has been studied by fluorescence. Association constants were deduced from competition experiments. The comparison of the association constants leads to the conclusion that several atoms of the base residues do not interact with the amino acid residues of the binding site of Fab fragment while the hydroxyl groups of furanose rings interact. These results suggest that the Fab fragments do not bind to the major groove of the double stranded polynucleotides. An interaction between the C(2)O group of pyrimidine residues and Fab fragments cannot be excluded. Circular dichroism of poly(I). poly(C) or poly(I). poly(br5C)-Fab fragments complexes are very different from the circular dichroism of free polynucleotides which suggests a deformation of the polynucleotides bound to the Fab fragments.

Action of partially thiolated polynucleotides on the DNA polymerase alpha from regenerating rat liver.

The effects of partially thiolated polynucleotides on the DNA polymerase alpha from regenerating rat liver were investigated. The enzyme was isolated from the nuclear fraction essentially according to the method of Baril et al.; it was characterized as the alpha polymerase on the basis of its response to synthetic templates and its inhibition with N-ethylmaleimide. Although polycytidylic acid had no effect on the DNA polymerase alpha either as a template or as an inhibitor, partially thiolated polycytidylic acid (MPC) was found to be a potent inhibitor, its activity being directly related to its extent of thiolation (percentage of 5-mercaptocytidylate units in the polymer). In comparison, the DNA polymerase beta which was purified from normal rat liver nuclear fraction, was much less sensitive to inhibition by MPC. Analysis of the inhibition of the alpha polymerase by the method of Lineweaver and Burk showed that the inhibitory action of MPC was competitively reversible with the DNA template, but the binding of the 7.2%-thiolated MPC to the enzyme was much stronger than that of the template (Ki/Km less than 0.03). Polyuridylic acid as such showed some inhibitory activity which increased on partial thiolation, but the 8.4%-thiolated polyuridylic acid was less active than the 7.2% MPC. When MPC was annealed with polyinosinic acid, it lost 80% of its inhibitory activity in the double-stranded configuration. However, 1 to 2%-thiolated DNA isolates were significantly more potent inhibitors than were comparable (1.2%-thiolated) MPC and showed competitive reversibility with the unmodified (but "activated") DNA template. These results indicate that the inhibitory activities of partially thiolated polynucleotides depend not only on the percentage of 5-mercapto groups but also on the configuration, base composition, and other specific structural properties.

Dependence of interferon induction on nucleic acid conformation.

UV and circular dichroism characteristics of duplex analogs belonging to the (A)n-(U)n and (I)n-(C)n series were determined to assign qualitatively the nature of conformational differences caused by 5-pyrimidine and c7 purine substitutions in such duplexes. Evidence is presented which shows that 5-pyrimidine substitution by bromine or methyl changes the duplex conformation of both series in a similar way, if at all. A c7 substitution in the purine ring affects the duplex conformation of the members in the same series similarly, but the conformational change appears to be different for the two series. To wit, it is proposed that in the duplexes the effect of the change (A)n leads to (c7A)n is an increase of the positive base tilt, whereas the change (I)n leads to (c7I)n causes a decrease where (c7A)n is poly (7-deazaadenylic acid) and (c7I)n is poly(7-deazainosinic acid), respectively. Poly(5-bromocytidylic acid) (br5C)n proved to be useful as a sensor strand for the intepretation of the spectroscopic data. The circular dichroism findings correlate well with observations made earlier on the interferon inducing ability for such duplexes, namely, duplexes based on the (c7A)n are inactive as interferon inducers, whereas duplexes based on (c7I)n are potent inducers. Furthermore, a 5-pyrimidine substitution does not substantially affect the interferon inducing ability, unless the thermal stability of the analog becomes critical, as in the case of (A)n-(br5U)n. Thus, this study provides the first evidence to link the interferon-inducing ability of a nucleic acid to a defined physical parameter of double helix, and reinforces the concept that interferon induction is dependent on the recognition of a particular spatial and steric organization of a double-stranded RNA.

Modification of chronic hepatitis-B virus infection in chimpanzees by administration of an interferon inducer.

Chimpanzees chronically infected with hepatitis-B virus showed transient changes in several markers of infection when treated with the interferon inducer polyriboinosinic-polyribocytidylic acid-poly-l-lysine carboxymethyl cellulose. Serum Dane-particle-associated D.N.A. polymerase, e antigen and hepatitis-B surface antigen, and intrahepatic hepatitis-B surface and core antigens diminished during treatment. Defective (D.N.A.-polymerase-negative) Dane particles increased in titre transiently during treatment; these may play a role in the modulation of hepatitis-B virus infection. Humoral immune responses in chronic hepatitis-B carrier chimps were unaffected. Interferon inducers (or exogenous interferon) may be useful for the treatment of chronic hepatitis-B virus infection.

Polynucleotide aggregates enhance the transport of protein at the surface of cultured mammalian cells.

It was known that polycationic polymers enhance the entry of macromolecules into cells. We now show that polynucleotides may have similar effects, when used as large aggregates. Poly(1-vinylcytosine):polyinosinic acid, an inducer of interferon production in human cells, can cause at 40 mug/ml a 75-fold enhancement of albumin uptake by sarcoma cells in culture. Most of this activity (85%) is related to the presence of aggregates retained by 0.65 mu millipore membranes. The prior finding that enhancers of albumin transport have increasing effects with increasing molecular sizes may thus extend to complexes of supramolecular sizes.

Structural features of double-stranded polyribonucleotides required for immunological specificity and interferon induction.

Purified antibody to poly(adenylic acid)-poly(uridylic acid) was used in quantitative microcomplement fixation assays to detect conformational variations among several double-helical polyribonucleotide analogs of poly(adenylic acid)-poly(uridylic acid) or poly(inosinic acid)-poly(cytidylic acid) that had been previously evaluated for their ability to induce interferon. Modification at the furanose 2'-position of one or both strands resulted in a dramatic decrease in serological reactivity. Most modifications of the bases caused smaller serological changes, and no base modification caused complete loss of reactivity. The reaction patterns support the conclusion that the structure of the furanose and the overall conformation of the helix are critical in the formation of antigenic determinants. The backbones of both strands appear to be involved in forming a single antigenic site, and base modifications may alter the steric relationship between the backbones. In addition, the same structural changes that substantially alter recognition by antibody also lead to large changes in the interferon-inducing ability of the nucleic acid.