Interferon Production Linked to Toxicity of Polyriboinosinic Acid-Polyribocytidylic Acid

Different procedures have been used in attempts to increase the production of interferon by polyriboinosinic acid-polyribocytidylic acid (poly [rI].poly [rC]) in mice: simultaneous injection of lead acetate, cycloheximide, or actinomycin D and prior injection of Freund's adjuvant, chlorite-oxidized oxyamylose (COAM), endotoxin, or Brucella abortus. In the experimental conditions tested, lead acetate, cycloheximide, Freund's adjuvant, and COAM brought about a parallel increase in interferon production and toxicity (lethality) of poly (rI).poly (rC); actinomycin D, endotoxin, and B. abortus increased the lethality of poly (rI).poly (rC) without a concomitant raise of its interferon-inducing capacity. Our results indicate that no significant increase in interferon production (or antiviral activity, as far as the antiviral activity is accounted for by interferon production) without an accompanying increase in toxicity can be achieved with poly (rI).poly (rC) and that it might be impossible to increase its therapeutic ratio (ratio of maximum tolerated dose to minimum effective dose).     

Effect of Separate and Combined Injections of Poly rI: Poly rC and Endotoxin on Reticulo-endothelial Activity, Interferon, and Antibody Production in the Mouse

The effect of repeated stimulation on both the interferon and antibody systems and on reticuloendothelial activity was studied by injection of endotoxin 2 days before injection of (poly rI):(poly rC). A single injection of endotoxin or (poly rI):(poly rC) increased or decreased the response in each system depending on the time of administration. If the injection of (poly rI):(poly rC) was preceded by an injection of endotoxin 2 days before, its activity was markedly reduced in all of the three systems studied. Although different doses of endotoxin were required to induce a state of hyporeactivity or tolerance to the effects of (poly rI):(poly rC) in either system, it is possible that a common mechanism underlies the hyporeactivity in all systems.     

Studies on the Mechanism of the Priming Effect of Interferon on Interferon Production by Cell Cultures Exposed to Poly(rI) · Poly(rC)

Interferon induction by poly(rI).poly(rC) in primary rabbit kidney and mouse L-929 cell cultures was markedly increased if the cells were previously treated with homologous interferon. This priming effect has been established with different times of exposure of the cells to poly(rI).poly(rC), and was most pronounced for short pulses of contact of the polynucleotide with the cells (10 s, 1 min). Treatment of the cells with pancreatic ribonuclease immediately after their exposure to poly(rI).poly(rC) brought about a relatively greater reduction of the interferon response in interferon-primed cells than it did in unprimed cell cultures. Priming of the cells with interferon did not increase cell-binding of poly(rI).poly(rC), whether this cell-binding was measured quantitatively (by radioactivity, upon exposure of the cells to radiolabeled polymer) or qualitatively (by antiviral activity, by assaying the cell extract for virus plaque reduction). Similarly, interferon priming did not alter the sensitivity of cell-associated poly(rI).poly(rC) to extraneous ribonuclease treatment. Finally, priming with interferon did not decrease the rate of degradation of cell-bound poly(rI).poly(rC) by cellular nucleases nor did it increase the anti-nuclease potency of the cells. The exact mechanism by which previous exposure of the cells to interferon enhances subsequent interferon production, induced by either synthetic polynucleotides or viruses, has not yet been resolved.     

The effect of combination of different inducers on the refractory state in interferon production.

A second injection of 100 mug poly (rI) poly (rC) per mouse at 6 and 24 hours after the first injection stimulated additional peaks of interferon production. The dynamics of the process of accumulation and disappearance of interferon was similar to that after a single injection of poly (rI). poly (rC). Injection of the above dose 12 hours after the first injection induced no interferon production as it apparently coincided with the refractory state in interferon production. After pretreatment of poly (rI) poly (rC) with DEAE-dextran, the refractory phase occurred in 6 hours. Inoculation of Venezuelan equine encephalomyelitis virus as a second interferon inducer resulted in a repeated stimulation of interferon production both in animals and in tissue culture; however, interferon titres in this case were low. The use of an inactivated virus as a second interferon inducer stimulated interferon production to higher titres (5120 IU/ml) than a single injection of DEAE-dextran-treated poly (rI). poly (rC). It is possible that a combined use of poly (rI). poly (rC) and noninfectious virus as a second interferon inducer eliminates the development of the refractory state.     

Poly(rI) more important than poly(rC) in the interferon induction process by poly(rI)-poly(rC)

A significantly greater interferon production has been obtained in primary rabbit kidney cell cultures exposed to poly(rI) followed by poly(rC) than in cell cultures exposed to poly(rC) followed by poly(rI). The interferon response in cell cultures exposed to poly(rI) followed by poly(rC) was markedly more resistant to poly-l-lysine and pancreatic ribonuclease treatment than was the interferon response in cell cultures exposed to poly(rC) followed by poly(rI). In addition, poly-l-lysine treatment removed a substantially greater proportion of cell-associated radioactivity from cells exposed to [³H]poly(rC) followed by poly(rI) than from cells exposed to poly(rI) followed by [³H]poly(rC). These findings suggest that the poly(rI) · poly(rC) complex is more tightly and efficiently bound to the cell (surface) when the homopolymers are added in the order poly(rI), poly(rC) than when they are added in the order poly(rC), poly(rI) and that it is more effectively attached to the cell receptor site by its poly(rI) strand rather than by its poly(rC) strand.     

Binding of polyriboinosinic-polyribocytidylic acid with cultured cells

Summary Homogenates prepared from polyriboinosinic-polyribocytidylic acid copolymer [poly(rI) · poly(rC)]-treated cells exhibited antiviral activity in chick embryo, L and rabbit kidney cells. The antiviral activity in the homogenate co-sedimented with cellular membrane material and was shown to be poly(rI)·poly(rC) by a hybridization competition test with immobilized polyribocytidylic acid. The results indicate that poly(rI)·poly(rC) binds firmly to cellular membrane. These studies, however, could not differentiate between specific binding leading to the interferon induction and non-specific binding possibly unrelated to the induction of interferon.With 6 Figures     

The antiviral activity of certain thiophosphate and 2'-chloro substituted polynucleotide homopolymer duplexes

Poly(rI)·poly(rsC) (polyriboinosinic acid) (polythiophosphate ribocytidylic acid) was found to be a slightly less potent interferon inducer than poly(rI)·poly(rC). The rate of degradation for poly(rI)·poly(rsC) and poly(rI)·poly(rC) by serum nucleases also proved to be very similar. In contrast, substitution of the 2′-hydroxyl by a chlorine atom in poly(rU) (polyribouridylic acid) and poly(rC) greatly increased their resistance to serum nucleases. Poly(rA)·poly(2′-ClU) (polyriboadenylic acid) (poly2′-chlorouridylic acid) and poly(rI)·poly(2′-ClC) were incapable of producing interferon although they possessed as great or greater thermal stability and much greater nucleolytic resistance than the unsubstituted interferon producing duplexes. This lack of interferon production by the 2′-Cl duplexes could not be explained by the failure of human foreskin fibroblasts (HFF) to take up these polymers. In addition, poly(rI)·poly(2′-ClC) did not compete with the interfering activity or the cellular association of poly(rI)·poly(rC). Substitution of the 2′OH by a 2′NH₂ in poly(rU) also produced a polymer incapable of producing interferon.     

Susceptibility of various cells treated with interferon to the toxic effect of poly(rI).poly(rC) treatment.

Various cells were treated with interferon and then exposed to polyriboinosinic-polyribocytidylic acid complex [poly(rI).poly(rC)]. With mouse L cells, there was a marked cytotoxic effect from low doses of interferon and poly(rI.poly(rC), whereas chick embryo cells showed an effect only after high doses. When primate cells (LLC.Mk2,BSC.B, Vero and human embryo cells) were treated with human or monkey interferon, poly(rI).poly(rC) was not cytotoxic.     

Differential effect of poly rI.rC and Newcastle disease virus on the expression of interferon and cellular genes in mouse cells

The expression of type I murine interferon (MuIFN) genes and several other cellular genes was examined in poly rI.rC induced and Newcastle disease virus (NDV) infected mouse cells. Northern analysis of RNA from induced L cells revealed that the MuIFN-alpha s are expressed efficiently in NDV infected cells but only at low levels in poly rI.rC induced cells. MuIFN-beta 1, however, is expressed equally well in cells treated with poly rI.rC or infected with NDV. As shown by the use of a probe specific for poly rI.rC, interferon induction correlates with the cellular uptake of poly rI.rC into the cells. The relative levels of alpha and beta 1 mRNAs in the cells reached a maximum at 10 hr after the induction which indicates coordinate expression of alpha and beta 1 interferon genes. The effect of viral infection on the expression of two murine genes coinduced with interferon (pMIF20/11 and pMIF3/10) and several cellular genes was also examined. While pMIF20/11 is an inducible gene, the pMIF3/10 gene is expressed constitutively in mouse L cells. Viral infection, but not poly rI.rC treatment, enhanced the expression of the pMIF3/10 gene, as well as two other cellular genes; H-2 and c-myc, however, the expression of beta-actin gene was unaltered. These data indicate that enhancement of gene expression in virus infected cells in not limited to the interferon system.     

The toxic effect of double-stranded RNA for interferon-treated cells: evidence for a heterogeneous cellular response and the role of the cell nucleus.

When L929 cells were treated with interferon and subsequently with poly(rI).poly(rC), there was a pronounced toxic effect. Most of the cells lysed, but some survived and grew at the same rate as control cells to yield cells which were as sensitive to the effects of interferon and poly(rI).poly(rC) as the original population. The proportion of surviving cells did not vary with either the cell cycle or the cell density. The treated cells produced interferon and some of the interferon was produced by the resistant cells. Cells which had been X-irradiated before treatment with interferon and poly(rI).poly(rC) behaved similarly so cell division was not necessary for the development of toxicity. The toxic effect also developed when cells were enucleated with the aid of cytochalas in B after treatment with interferon, but not if they were enucleated before treatment. It is concluded that the nucleus is essential for interferon to exert its effect on the cells, but not for the development of cytotoxicity after the addition of poly(rI).poly(rC).     

Protection of mice against infection with wild-type Mengo virus and an interferon sensitive mutant (IS-1) by polynucleotides and interferons.

Single-stranded polynucleotide preparations [tRNA, poly(rI) plus poly(ho5C)-copolymer] which protect mice against picornavirus infections without inducing interferon, protected mice equally against infection with an interferon-sensitive mutant (IS-1) of Mengo virus and with wild-type virus (IS+). Poly(rI) . poly(rC), and mouse macrophage interferon [i.e. serum from mice treated with poly(rI) . poly(rC)] protected mice equally against infections with the two viruses, but fibroblast interferon protected better against infection with the interferon-sensitive mutant than with the wild-type virus. These and other results indicate that: Mengo virus had a genetic locus affecting sensitivity to fibroblast but not macrophage interferon; these two types of interferon have different mechanisms of action against Mengo virus infections in mice; Mengo virus genes controlling sensitivity to fibroblast interferon may modulate disease since infection in vivo induces only fibroblast interferon; the antiviral activity of the single-stranded polynucleotides is unlikely to be mediated by induction of either macrophage or fibroblast interferon.     

The preventive effects of incomplete Freund’s adjuvant and other vehicles on the development of adjuvant-induced arthritis in Lewis rats

The present study showed a novel finding that the development of adjuvant-induced arthritis (AA) in Lewis rats was completely prevented by incomplete Freund's adjuvant (IFA) injected 21 or 28 days before complete Freund's adjuvant (CFA) challenge. Hexadecane also completely prevented AA and squalane, methyl oleate and pristane moderately prevented AA, though pristane by itself induced mild arthritis in two out of five rats. Concanavalin A-stimulated lymph node cells (LNCs) isolated from AA rats were able to adoptively transfer the severe polyarthritis to all the naive recipients or even to the IFA pretreated recipients with earlier onset and more rapid progression than those of AA. The LNCs from the donors who had been pretreated with IFA and subsequently challenged with CFA could induce mild arthritis in only two out of eight naive recipients, whereas all the recipients who were challenged with CFA immediately after intravenous injection of these LNCs developed significantly less severe arthritis. However, the LNCs from IFA-pretreated donors failed to prevent AA. According to the T helper type 1 (Th1)/Th2 paradigm, it was suggested that the adjuvant-active vehicles such as IFA, hexadecane, squalane, methyl oleate and pristane, can affect and deviate the Th1/Th2 balance of immune responses in host. CFA could promote the propagation of Th2 cells rather than Th1 cells in these vehicle-pretreated rats through as yet undetermined mechanisms, eventually resulting in the prevention of AA. Finally, we discussed a regulatory role of adjuvant vehicles for induction and suppression of AA.     

Molecular species of interferon induced in mouse L cells by Newcastle disease virus and polyriboinosinic-polyribocytidylic acid.

Interferons were stimulated in mouse L cells by Newcastle disease virus (NDV) or by polyriboinosinic-polyribocytidylic acid poly(rI).poly(rC). These were fractionated by sequential affinity chromatography on bovine plasma albumin (BPA)-Sepharose and on omega-carboxypentyl (CH)-Sepharose. Based on their interaction with CH-Sepharose, interferon induced by NDV was resolved into three major bands of activity (L/NDV-1,2,3) and poly(rI).poly(rC)-interferon into two (L/rI:rC-1,2). These interferon components were purified to a specific activity of 3 X 10(7) to 4 X 10(7) units/mg protein by antibody affinity chromatography and examined by electrophoresis in SDS-polyacrylamide gels. A total of five molecular species was thus identified for NDV-induced interferon and three for poly(rI).poly(rC) induced interferon, as summarized in Table 1. We conclude from our observations that mouse interferons can be produced by L cells in multiple forms with specific physiochemical properties and in proportions determined by the type of agent employed for induction.     

The Adjuvant Effect of Silicone-Gel on Antibody Formation in Rats

The extent of immunological adjuvancy of silicone-gel, from mammary implants, up to now, has not been determined definitively. This study compares the immune potentiation effects of silicone-gel with that of Freund's adjuvant, using bovine serum albumin (BSA) as the test antigen in rats. Sixty, 250 gr., male Sprague Dawley rats were divided into six groups: I- phosphate buffered saline (PBS) only, II- silicone oil (Dow Corning Medical Grade 360 liquid silicone), III- 50% silicone-gel (McGhan Medical Corp.- mammary implant) in silicone oil, IV- complete Freund's adjuvant (CFA), V- incomplete Freund's adjuvant (IFA), and VI- 50% silicone oil in IFA. Each adjuvant was mixed or emulsified with an equal volume of 50 μg of BSA in 150 μl of PBS. Each immunization was given intramuscularly in a single injection. Cardiac puncture test bleeds were taken at 12, 22, 40 and 56 days post immunization and the serum anti-BSA-antibody was measured by ELISA. The results indicate that silicone-gel is a potent immunological adjuvant, compared to both CFA and IFA. Silicone oil alone is not as potent an adjuvant and seems to inhibit the immune response when mixed with IFA. There thus appears to be a distinct possibility that silicone-gel may also be able to mediate an auto-immune reaction.     

A limited role of IL-1 in immune-enhancement by adjuvants.

Generation of interleukin-1 (IL-1) in the draining lymph nodes after injection of complete Freund's adjuvant (CFA), incomplete Freund's adjuvant (IFA) and alum was studied in line with IL-1 mRNA expression (cytoplasmic slot blot analysis) and IL-1 beta antigen detection (ELISA and immunohistochemistry) in rabbits. The expression of IL-1 beta mRNA was marked from 6 to 96 hr, with a maximum at around 24 hr post-injection of CFA, while injection of the other two adjuvants elicited only a moderate or negligible response. On the other hand, IL-1 alpha mRNA expression was almost negligible during the entire 8-week observation period after injection of the above three adjuvants. Generation of IL-1 beta antigen in the draining lymph nodes after CFA injection paralleled the expression of IL-1 beta mRNA. Immunohistochemistry revealed that cells containing IL-1 beta resided in the medullary sinuses, marginal sinuses and para-cortical area, but not in the follicles. Despite marked generation of IL-1 beta in CFA-treated draining lymph nodes, the primary antibody response (IgG) to ovalbumin differed only slightly between the three animal groups that were immunized with the antigen incorporated in CFA, IFA and alum. Further, rIL-1 beta did not significantly enhance the immune response when it was entrapped together with the antigen in IFA and alum. IL-1 beta enhanced the immune response only when it was injected with antigen without adjuvant. Thus, IL-1 seemed to play only a limited role, if any, in the augmentation of the primary immune response by the above-mentioned adjuvants.     

Biological activity of poly rI:poly rC: effect on poxvirus-specific functions

Poxvirus-specific functions have been studied during the “early” period of the replication cycle in chick embryo fibroblasts pretreated with poly rI:Poly rC. “Early” viral protein synthesis and formation of DNA was found to be strongly reduced in a similar fashion to that of cells exposed to homologous interferon preparations. “Early” viral RNA of cowpox or vaccinia (WR), on the other hand, was synthesized at the rate expected for cells in which protein synthesis is inhibited. Even at high concentrations, (100 μg/ml), poly rI:poly rC had no inhibitory effect on “early” viral RNA transcribed in vitro by the RNA-polymerase contained in the virion. The results are discussed with respect to the mechanism of action of interferon and poly rI: poly rC.     

Interferon production by human-mouse hybrid cells: dominant mouse control of superinduction and priming

We have examined the production of interferon by a number of human-mouse hybrid clones in response to polyriboinosinic acid:polyribocytidylic acid copolymer [poly(rI).poly(rC)] all of which produced both human and mouse interferons when stimulated with a virus. Their capacity to be superinduced and primed for interferon production in response to poly(rI).poly(rC) was compared to that of the parental human and mouse cells. It was found that the hybrids responded in a way similar to their mouse cell parents, indicating dominant mouse control of both the priming and superinduction phenomena.     

Induction of Viral Interference: Effects of Poly rI·rC and Diethylaminoethyl-Dextran on the Activity of the Antiviral Protein

The activity of the antiviral protein induced by various ratios of poly rI.rC and diethylaminoethyl (DEAE)-dextran was studied. It was found that, when large doses of poly rI.rC were used, very little viral interference was observed. This effect was initially attributed to the cells being refractory for production of antiviral protein. Subsequent experiments offered alternative explanations suggesting that, at any given dosage of poly rI.rC, an excess of DEAE-dextran is necessary for the production of viral interference. It is suggested that DEAE-dextran acts by exposing a cell receptor site for poly rI.rC.     

Regulation of the interferon system: evidence that Vero cells have a genetic defect in interferon production.

A clone of Vero cells was isolated and shown to be totally unable to synthesize interferon and insensitive to the toxic effect of poly(rI).poly(rC) treatment. Cells of this clone and mouse L cells were fused by treatment with polyethylene glycol or Sendai virus. Hybrid cell clones were isolated following selection in medium containing hypoxanthine, thymidine and ouabain. The hybrids were sensitive to the antiviral effect of poly(rI).poly(rC) and synthesized mouse, but not primate, interferon. It is proposed that in Vero cells, the gene for interferon synthesis is defective or absent.     

Identification of a cell membrane receptor for interferon induction by poly rI:rC.

PR-RK, a cell line derived from rabbit kidney cells (RK-13), was insensitive to the cytotoxic effect and interferon (IFN) inducing activity of the copolymer of riboinosinic and ribocytidylic acid (poly rI:rC). However, PR-RK was sensitive to the cytotoxic effect of the copolymer of riboadenylic and ribouridylic acid (poly rA:rU). Comparison of PR-RK cells and RK-13 cells by cytofluorometric analysis revealed that the binding of poly rI:rC was considerably reduced on PR-RK cells. These results suggested that the receptor for poly rI:rC might be different from the receptor for poly rA:rU, and this difference could provide a basis for the identification of the dsRNA receptor on cell surface. Western blot analysis of the components of cell membrane fraction prepared from RK-13 cells was performed by using a monoclonal antibody, which binds to cell membrane of RK-13 cells but not to PR-RK cells, and which blocks IFN induction by poly rI:rC in RK-13 cells. The 60K protein was identified as one of the poly rI:rC receptor protein.