The Egyptian Rousette Genome Reveals Unexpected Features of Bat Antiviral Immunity

Bats harbor many viruses asymptomatically, including several notorious for causing extreme virulence in humans. To identify differences between antiviral mechanisms in humans and bats, we sequenced, assembled, and analyzed the genome of Rousettus aegyptiacus, a natural reservoir of Marburg virus and the only known reservoir for any filovirus. We found an expanded and diversified KLRC/KLRD family of natural killer cell receptors, MHC class I genes, and type I interferons, which dramatically differ from their functional counterparts in other mammals. Such concerted evolution of key components of bat immunity is strongly suggestive of novel modes of antiviral defense. An evaluation of the theoretical function of these genes suggests that an inhibitory immune state may exist in bats. Based on our findings, we hypothesize that tolerance of viral infection, rather than enhanced potency of antiviral defenses, may be a key mechanism by which bats asymptomatically host viruses that are pathogenic in humans.

Teleost fish interferons and their role in immunity

Interferons (IFNs) are the hallmark of the vertebrate antiviral system. Two of the three IFN families identified in higher vertebrates are now known to be important for antiviral defence in teleost fish. Based on the cysteine patterns, the fish type I IFN family can be divided into two subfamilies, which possibly interact with distinct receptors for signalling. The fish type II IFN family consists of two members, IFN-γ with similar functions to mammalian IFN-γ and a teleost specific IFN-γ related (IFN-γrel) molecule whose functions are not fully elucidated. These two type II IFNs also appear to bind to distinct receptors to exert their functions. It has become clear that fish IFN responses are mediated by the host pattern recognition receptors and an array of transcription factors including the IFN regulatory factors, the Jak/Stat proteins and the suppressor of cytokine signalling (SOCS) molecules.

Identification of a second group of type I IFNs in fish sheds light on IFN evolution in vertebrates

In this report, three type I IFN genes were identified in rainbow trout (rt) Oncorhynchus mykiss and are classified into two groups based on their primary protein sequences: group I containing two cysteine residues; and group II containing four cysteines residues. The group I rtIFNs were induced in fibroblasts (RTG-2 cells), macrophages (RTS-11 cells), and head kidney leukocytes when stimulated with polyinosinic:polycytidylic acid, whereas group II IFN was up-regulated in head kidney leukocytes but not in RTG-2 and RTS-11 cells. Recombinant group I rtIFNs were potent at inducing Mx expression and eliciting antiviral responses, whereas recombinant group II rtIFN was poor in these activities. That two subgroups of type I IFN exist in trout prompted a survey of the genomes of several fish species, including zebrafish, medaka, threespine stickleback and fugu, the amphibian Xenopus tropicalis, the monotreme platypus and the marsupial opossum, to gain further insight into possible IFN evolution. Analysis of the sequences confirmed that the new IFN subgroup found in trout (group II IFN) exists in other fish species but was not universally present in fish. The IFN genes in amphibians were shown for the first time to contain introns and to conserve the four cysteine structure found in all type I IFNs except IFN-betaepsilon and fish group I IFN. The data overall support the concept that different vertebrate groups have independently expanded their IFN types, with deletion of different pairs of cysteines apparent in fish group I IFN and IFN-betaepsilon of mammals.

Modification of TLR-induced activation of human dendritic cells by type I IFN: synergistic interaction with TLR4 but not TLR3 agonists

Upon detection of direct and indirect signs of infection, dendritic cells (DC) undergo functional changes that modify their ability to elicit immune responses. Type I interferon (IFN-alpha/beta), which includes a large family of closely related infection-inducible cytokines, represents one indirect signal that can act as a DC stimulus. We have investigated the ability of IFN-alpha/beta subtypes to affect DC function and to influence DC responses to Toll-like receptor (TLR) agonists (i.e., direct infection-associated signals). Subtle differences were observed among 15 subtypes of IFN-alpha/beta in the ability to stimulate expression of maturation markers and chemokines by human monocyte-derived DC, with IFN-omega being the most unique in its effects. Pre-treatment with IFN-alpha/beta did not alter the ability of DC to mature in response to subsequent contact with TLR agonists, but did modulate their secretion of chemokines. Conversely, IFN-alpha/beta was shown to act synergistically with TLR4 but not TLR3 agonists for the induction of maturation and chemokine production when DC were exposed to IFN-alpha/beta and TLR ligands simultaneously. Taken together, these results indicate a complex role for IFN-alpha/beta in regulating DC function during the course an infection, which varies according to IFN-alpha/beta subtype and the timing of exposure to other stimuli.

Diversity and Relatedness Among the Type I Interferons

Type I interferons (IFNs) include the IFN-alpha family of subtypes, IFN-beta, IFN-omega, IFN-tau, IFN-kappa, IFN-lambda, and IFN-zeta. IFN genes lack introns and encode secretory signal peptide sequences that are proteolytically cleaved prior to secretion from the cell. In contrast to the approximately 50% amino acid sequence identity among the human IFN-alpha subtypes, human IFN-alphas share approximately 22% identity with human IFN-beta and 37% identity with human IFN-omega. Many of the conserved residues among the type I IFNs are implicated in receptor recognition and structural integrity. This report provides an update on the gene annotations for the mouse and human IFN gene clusters on chromosome 4 and 9, respectively, with accompanying amino acid sequence alignments. Based on sequence identities, a phylogenic tree analysis for the different mammalian Type I IFNs is also presented, showing the high degree of relatedness among these IFNs. Notably, sequence alignment of the different human and mouse IFN promoter regions reveals different signature patterns for transcription factor binding sites, implying different inducers might differentially activate the transcription of the different IFNs.

Identification of interferon-tau isoforms expressed by the peri-implantation goat (Capra hircus) conceptus

Interferon-tau (IFN-tau) is the maternal recognition of pregnancy factor in pecoran ruminants. The aims of this study were to identify the various IFN-tau transcripts in the peri-implantation caprine (ca) conceptus and to compare these nucleotide sequences phylogenetically with established mRNA sequences from the goat. Conceptuses (n = 5) were collected from Boer crossbred and Angora female goats by laparotomy at days 17 and 18 of pregnancy. Total cellular RNA was extracted and RT-PCR was performed by standard procedures using a DNA polymerase with proofreading activity and gene-specific primers complimentary to non-coding regions of the published caIFN-tau sequence. Nine distinct nucleotide sequences were isolated that encode five distinct caIFN-tau proteins. These caIFN-tau have greater sequence homology with ovine IFN-tau (94-96% nucleotide identity; 90-93% amino acid identity) than with bovine IFN-tau (<92% nucleotide identity; <85% amino acid identity). The novel caIFN-tau isoforms contained pronounced nucleotide and amino acid sequence identity with one another (97-99% nucleotide identity; 94-99% amino acid identity) but only moderate sequence identity with the previously identified caIFN-tau (94-96% nucleotide identity; 87-90% amino acid identity). In conclusion, multiple caIFN-tau mRNA species are expressed during peri-implantation conceptus development and distinct clusters of caIFN-tau genes appear to have evolved in this species.

Transient transfection of mouse fibroblasts with type I interferon transgenes provides various degrees of protection against herpes simplex virus infection

Type I interferons (IFN) constitute one of the initial and most potent components of the innate immune response against viral infections. While there is only one IFN-beta gene, there are several IFN-alpha genes whose products act through the same receptor calling into question the role of these gene products against viral infection. The focus of the present study was to compare the anti-viral state of cells transiently transfected with different murine type I IFN transgenes including IFN-alpha1, -alpha4, -alpha5, -alpha6, -alpha9, and IFN-beta. Transfected cells produced biologically active IFN ranging from 6 to 46 units/ml. L929 and 3T12.3 cells transfected with the IFN-beta transgene consistently showed a 2-4 fold reduction in herpes simplex virus type 1 (HSV-1) and HSV-2 viral titers compared with cells transfected with the IFN-alpha transgenes which were much less consistent based on HSV species and cell type. Parallel with the reduction in viral titers, cells transfected with the IFN-beta transgene showed the complete absence or significant reduction in viral immediate early, early, and late gene expression. Collectively, the results suggest that the IFN-beta transgene is superior to IFN-alpha transgenes against HSV infection in vitro in part due to a reduction in viral gene expression. These results indicate events downstream of the type I IFN receptor distinguish between the subtypes of IFN-alpha species relative to the activation of genes ultimately responsible for the establishment of the anti-HSV state.

A classification for the interferon-tau

An attempt has been made to provide a rational organization for the many interferon-tau (IFN-tau) sequences entered in GenBank based on phylogenetic analysis and common amino acid substitutions, which might form the basis for a universal nomenclature scheme. Over the 13 years since these genes were first discovered, large numbers of cDNA and gene sequences have been reported, and there is reason to suspect that representatives of all the major ovine and bovine forms have now been described. The data are consistent with the presence of many genes and also allelic variants in sheep and cattle analogous to what has been observed for the IFN-alpha in the human. Future variants should be easily accommodated into the scheme outlined here. A flexible system of nomenclature, based on that used for HuIFN, is needed to provide a common base for comparison between research done in different laboratories and to assign relative biologic potencies to these molecules.

The cross-species antiviral activities of different IFN-tau subtypes on bovine, murine, and human cells: contradictory evidence for therapeutic potential

It is claimed that interferon-tau (IFN-tau) has broad cross-species reactivity and less cytotoxicity than other type I IFN when used at high concentration either in vitro or in living animals. It can also amelioriate the development of experimental allergic encephalomyelitis (EAE) without the usual side effects of IFN therapy in mice autoimmunized with myelin basic protein. For these reasons, IFN-tau may have therapeutic potential in humans. Here, the antiviral (AV) activities of eight different recombinant IFN-tau were compared with those of several bovine, human, and murine type I IFN on bovine MDBK cells, murine L929 cells, and human WISH cells. The data show that only one of the IFN-tau, OvIFN-tau4, has broad cross-species reactivity. It was comparable in this respect to HuIFN-omega1 and HuIFN-alpha1. The other IFN-tau, including the variant form (OvIFN-tau1mod) tested by others in cytotoxicity experiments and for its ability to protect mice against EAE, had relatively weak AV activity on mouse and human cells. It is possibly because this particular bioengineered form of IFN-tau binds the common type I receptor of these two species with such low affinity that it lacks cytotoxic effects. The basis for its potent anti-EAE activity is unclear, but it seems possible that it does not involve the type I IFN receptor.

Differential expression patterns of type I interferon subtypes in mouse embryo fibroblasts: influence of genotype and viral inducer

Primary mouse embryo fibroblasts from 4 strains of mice (BALB/c, C57Bl/6, B6.C-H-28c and CBA) were infected with either Newcastle disease virus or murine cytomegalovirus. The time course of the total type I interferon response was assessed and the presence of individual subtypes determined. The total type I interferon produced was titrated using the cytopathic effect reduction assay and the relative levels of type I interferon subtypes expressed (alpha 1, alpha 4, alpha 5, alpha 6 and beta) were evaluated using a reverse transcription-polymerase chain reaction-based technique. In general, the patterns of type I interferon subtypes expressed appeared to be determined by the strain of mouse cells used rather than the inducing virus. However, the overall titre of type I interferons produced in response to a given virus was quite uniform across the strains of mice from which the mouse embryo fibroblasts were derived regardless of the subtype expression pattern. The latter observation fits the proposition that "cross-talk" or feedback between the type I interferon genes and their products is is occurring and that the inducer determines the level of response.

Biological activities of recombinant murine interferons alpha 1 and alpha 4: large difference in antiproliferative effect

The mature forms of two recombinant murine interferons alpha, alpha 1 and alpha 4, have been expressed in vitro using an established transcription and translation system. The relative specific antiviral activity, antiproliferative activity and the natural killer cell stimulating activity of both subtypes were compared in vitro. While the antiviral and natural killer cell stimulating activities of the 2 subtypes were similar, the relative antiproliferative activities varied markedly. On the basis of equal molar inputs, MuIFN-alpha 1 had less than 8% of the antiproliferative activity of MuIFN-alpha 4. This data shows that a large functional difference exists between these two subtypes which are known to be expressed at different levels in mouse L-cells in vitro.

Modification of the antigenic structure of human interferon alpha-2 by PH 2 treatment: a further support for the antigenic relationship between alpha and beta interferons

We have recently reported that a unique antigenic structure, designated as Common epitope 1, was found to be shared by human recombinant IFN alpha-2 and the human fibroblast IFN beta. The Common epitope 1 was identified with the aid of a synthetic IFN alpha-2 fragment SH 132-137. Based on this observation, we proposed the hypothesis that an antigenic relationship should exist also between natural human leukocyte IFN alpha and natural human fibroblast IFN beta. However, we were not able to detect any Common epitope 1 in preparations of conventional human leukocyte IFN alpha. In the present study, we were looking for a possible explanation of absence of the Common epitope 1 in conventional leukocyte IFN alpha. First, we demonstrated its acid labile nature in the recombinant IFN alpha-2 molecule and second, we proposed that the pH 2 lability of this unique epitope might be responsible for the lack of antigenicity also in pH 2-treated (conventional) leukocyte IFN alpha preparations. Actually, when pH 2 non-treated leukocyte IFN alpha was examined, we succeeded in demonstration of the Common epitope 1 in IFN-preparation. Moreover, anti-serum against pH 2 non-treated IFN alpha was capable of neutralizing both the conventional i.e. pH 2-treated leukocyte IFN alpha and fibroblast IFN beta. It is concluded that the nomenclature distinguishing two classes (i.e. alpha and beta as class I and gamma as class II) of IFNs is more appropriate than the current official nomenclature distinguishing three antigenic classes of IFNs.

[Interferon production by primary culture of human endothelial cells]

Endothelial cells of human umbilical vein are capable of producing interferon upon induction with Newcastle disease virus, influenza virus, and poly I: poly C, but not staphylococcal enterotoxin A. All the interferons produced belonged to the alpha-type. After treatment with influenza virus the endothelial cells produce two subtypes of alpha-interferon: acid-labile and acid-stable.

Effect of storage and 60Co irradiation on the antiviral activity and subtype pattern of native human interferon-alpha preparations

The stability of native human leukocyte IFN-alpha was investigated after one year storage at different temperatures, or after sterilization with 60Co-irradiation by determining the total antiviral activity of samples. Inactivation of VSV and Aujeszky virions by 60Co-irradiation was directly related to the radiolysis of samples, indicating the uselessness of this procedure for sterilization of IFN-alpha preparations. The presence and proportion of subtypes in the stored or irradiated preparations (with 50 and 25-70% inactivation, respectively) was analysed by chromatofocusing, comparing their patterns with that of the untreated controls. A logarithmic correlation was found between the pI values and temperature/irradiation sensitivity of subtypes.

Production of subtype-specific antipeptide antibodies to human interferon-alpha 1 and -alpha 4

Antibodies that are specific to the human interferon (IFN)-alpha 1 and -alpha 4 subtypes have been produced by immunizing rabbits with two short synthetic peptides, corresponding to residues 99-111 of IFN-alpha 1 and residues 37-50 of IFN-alpha 4, respectively. The IFN-alpha 1 peptide has at least three closely clustered residues that are different from those in the other IFN-alpha subtypes, while the IFN-alpha 4 peptide has only two unique amino acid residues, separated by five common residues. The antibodies raised against the IFN-alpha 1 peptide react with recombinant human IFN-alpha 1 but do not cross-react with recombinant human IFN-alpha 4 or IFN-alpha 2. The antibodies raised against the IFN-alpha 4 peptide react with IFN-alpha 4, cross-react with IFN-alpha 1 but not with IFN-alpha 2; the affinity of the antibodies to IFN-alpha 1, however, is at least 10 times lower than their affinity to IFN-alpha 4.

Prevention of viral myocarditis with recombinant human leukocyte interferon alpha A/D in a murine model

Effects of recombinant human leukocyte interferon alpha A/D on experimental myocarditis due to encephalomyocarditis virus were investigated. Plaque reduction assays revealed that 50% of plaque formation in vitro in human amnion (FL) cells was inhibited by interferon alpha A/D (9.7 U/ml) when it was administered 24 hours before infection with the encephalomyocarditis virus. Four week old male DBA/2 mice were inoculated intraperitoneally with 10 plaque-forming units (pfu) of encephalomyocarditis virus. Interferon alpha A/D was administered subcutaneously (10(2) U/g body weight per day in Group 1, 10(3) U/g per day in Group 2 and 10(4) U/g per day in Group 3) starting 1 day before infection. It was also administered starting the same day in Group 4 and 1 day after virus inoculation in Group 5 (10(4) U/g per day in both groups). Control mice were injected with saline solution. Each group consisted of 10 mice; they were killed on day 4 for evaluation. Myocardial virus titers were significantly lower in Group 3 (8.2 +/- 25.2 X 10(2) pfu/mg, p less than 0.05) and Group 4 (3.0 +/- 5.5 X 10(3) pfu/mg, p less than 0.05) than in control mice (5.6 +/- 4.1 X 10(4) pfu/mg). Histologic examination showed extensive myocardial necrosis and cellular infiltration in all control mice, but no myocardial necrosis or cellular infiltration in Group 3 and less severe necrosis and infiltration in Group 4. There were no significant differences in myocardial virus titers or histologic changes between control mice and Group 1, 2 or 5.(ABSTRACT TRUNCATED AT 250 WORDS)

Anomalous behavior of human leukocyte interferon subtypes on polyacrylamide gel electrophoresis in the presence of dodecyl sulfate

35S-labeled human leukocyte interferon (IFN) subtypes produced in a cell-free system derived from Escherichia coli were analyzed by polyacrylamide gel electrophoresis in the presence of SDS (SDS-PAGE). Some IFN subtypes anomalously showed lower electrophoretic mobilities than those expected from their formula molecular masses. The results with hybrid IFNs and esterification suggest that this anomaly of IFN subtypes on SDS-PAGE is due to the introduction of one or two negative charges in the middle of the molecule.

[Formation of an acid-labile interferon induced in human leukocytes by the influenza virus]

Predominantly acid-labile interferon was produced by treatment with influenza virus (inducer) of a portion of leukocytes followed by their addition to the main cell pool. This interferon could be detected by treatment of the material with antiserum to the inducer virus and ultracentrifugation. This treatment completely eliminated the activity of the virus-inducer present in the material. Typing of the acid-labile interferon with the employment of antisera to alpha and gamma interferons demonstrated its belonging to the alpha type.

Regulatory effects on macrophages of human recombinant interferons-alpha

The regulatory effects of human recombinant and hybrid interferons-alpha (IFN-alpha) on macrophage-mediated tumoricidal activity were examined. Recombinant hybrid IFN-alpha-A/D suppressed the capacity of murine interferon-gamma (IFN-gamma) to activate mouse peritoneal macrophages to a tumorilytic state, and blocked the killing of syngeneic syngeneic melanoma target cells by macrophages previously committed to the cytotoxic phenotype with a 4-h pretreatment with IFN-gamma. This suppressive activity was limited to IFN-alpha-A/D, as IFN-alpha-A and IFN-alpha-D were not effective. In contrast, IFN-alpha-A, -D, and -A/D were all capable of activating human peripheral blood monocytes to lyse human tumor cells. When encapsulated in liposomes, only IFN-alpha-A/D maintained its monocyte activating efficacy. These findings suggest that the immunomodulatory effects of IFN-alpha subtypes and hybrid molecules are dependent on species of monocytes/macrophages, subtype, and nature of presentation to effector cells.

Human monocytes and lymphocytes produce different mixtures of alpha-interferon subtypes

alpha-Interferon (IFN-alpha) was produced by either peripheral blood lymphocytes or by monocytes and purified by an anti-IFN-alpha affinity column. When these preparations were analyzed by reversed-phase HPLC, a difference in the distribution of IFN-alpha subtypes from the two cell types was found. While the two major subtypes of IFN from induced lymphocytes had apparent molecular weights of 20K and 21K, monocytes produce an additional subtype of molecular weight 26K in large quantities (50%). This subtype had greater activity on human cells than on bovine cells in comparison to other IFN-alpha subtypes.

Production and characterization of poly(I):poly(C)-induced rat interferons in vitro

The synthetic polyribonucleotide polyinosinate:polycytidylate [poly(I):poly(C)] was used to induce interferon (IFN) production in a rat leiomyosarcoma cell line. Three separate IFN species were distinguished by antiviral activity on homologous and heterologous cells, interaction with Blue Sepharose CL-6B chromatography matrix, and antibody neutralization studies. Two IFN species, one with a weakly ionic and the other with a hydrophobic interaction with Blue Sepharose, exhibited heterologous antiviral activity, and were acid- and heat-labile. Both had a molecular weight of 30,000 daltons by HPLC-based molecular exclusion chromatography. The weakly ionic species was neutralized only by antiserum to L-cell IFN, whereas the hydrophobic species was neutralized by antiserum to L-cell IFN and antiserum to human leukocyte IFN. The remaining IFN species, with a weakly ionic interaction with Blue Sepharose, was acid- and heat-stable, exhibited no heterologous cell antiviral activity, and was not neutralized by any antisera to IFN. This species had a molecular weight of 39,000 daltons by HPLC-based molecular exclusion chromatography and 34,000 daltons by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. These rat fibroblast IFNs possess greater heterogeneity than that reported or observed for murine or human fibroblast IFNs. These IFNs could thus be classified as follows: one species of rat IFN-beta and two different species of unique acid-labile rat IFNs.

Interferon effects on herpes simplex virus in rabbit and human cell cultures

The effects of four subtypes of recombinant human alpha interferon (RIFN alpha), (A,B,D, and the hybrid A/D) were tested on six strains of herpes simplex virus (HSV). RIFN alpha -D was the most effective subtype in rabbit kidney cells, which is consistent with our previous in vivo results in the rabbit herpetic keratitis model. In human corneal cells, however, RIFN alpha -D was one of the least effective IFN subtypes tested. Conversely, RIFN alpha-A appeared to be relatively more effective in the human corneal cells than in the rabbit kidney cells, but RIFN alpha -B and RIFN alpha -A/D were the most effective interferon subtypes in human corneal cells. Different strains of HSV had different susceptibilities to the various IFN subtypes tested.

A novel class of human type I interferons

The screening of a cDNA library prepared from mRNA of Sendai virus induced Namalwa (human Burkitt's lymphoma) cells, using a human IFN-alpha 2 DNA probe under conditions of low stringency, identified two weakly hybridizing clones containing sequences related to, but discernably different from those of the IFN-alpha class. Sequence and hybridization analysis of these cDNAs as well as expression in E. coli provided evidence that they encode proteins which have the characteristics of IFN type I but which are sufficiently diverged in sequence from both IFN-alpha s and IFN-beta to suggest that they are representatives of a new and distinct class of interferons named interferon-omega. Hybridization of these sequences to genomic DNA reveals that this class contains at least four members.

Immunologic differentiation between E. coli and CHO cell-derived recombinant and natural human beta-interferons

The products of the human IFN-beta gene expressed in E. coli, Chinese hamster ovary (CHO) cells, and human fibroblasts appear similar when purified on a monoclonal antibody column and analyzed by reverse-phase HPLC, indicating little difference in their hydrophobic nature. SDS-PAGE differentiates E. coli-rHuIFN-beta ser (Mr = 17,000) from CHO-rHuIFN-beta and HuIFN-beta (Mr = 23,000), with glycosylation accounting for 26% of the apparent m.w. of the latter two proteins. CHO-rHuIFN-beta is preferentially neutralized by mouse monoclonal and monospecific rabbit polyclonal anti-HuIFN-beta antibodies, whereas E. coli-rHuIFN-beta ser is preferentially neutralized by goat polyclonal anti-E. coli-rHuIFN-beta antibodies. Adsorption measurements by a sensitive radioimmunoassay indicate that the binding of the three proteins to anti-HuIFN-beta antibodies is similar. The results show that all three molecules can be differentiated by the heteroclitic cross-reactivities of anti-HuIFN-beta and anti-E. coli-rHuIFN-beta antibodies to the antigens.

Isolation and characterization of subspecies of murine interferon alpha

Interferon produced by mouse L-929 cells by incubation with poly(rI).poly(rC) is known to be composed of a mixture of MuIFN-alpha and MuIFN-beta. The alpha component was separated from the bete species by affinity chromatography over a monoclonal anti-MuIFN-beta agarose column and partially purified by gel filtration. MuIFN-alpha, prepared by this method was separated into at least five subspecies by chromatofocusing. The approximate pI values of these components are greater than or equal to 7.5, 6.5, 6.2, 5.9 and 5.6, respectively. Component 3 (pI 6.2) was the most prominent subspecies present in our MuIFN-alpha preparations, representing 40 to 50% of the total antiviral activity. Component 1 (pI greater than or equal to 7.5) which accounted for about 5% of the antiviral activity on mouse cells, differed in some properties from the other interferon subspecies. It showed a relatively high antiviral activity on heterologous cells and it was eluted from a Sephadex column after the other alpha subspecies. Furthermore, it showed a diminished binding to heparin as compared to the other MuIFN-alpha subspecies, indicating a lower affinity for polynucleotides.

Novel Human Leukocyte Interferon Subtype and Structural Comparison of Alpha Interferon Genes

In the cDNA library of virus-induced human leukocytes a novel subtype of IFN-alpha gene has been identified and sequenced, named IFN-alpha-N. A comparison of nucleotide sequences within the genes coding for different subtypes of human leukocyte interferon (IFN-alpha) has revealed the natural hybrid structure of individual alpha-IFNs-H,B,F, and N. Certain regions of the genes for IFN-alpha-H,B,F,N show a homology with one of the two structurally distinct groups of leukocyte interferons--either IFN-alpha-A,D or IFN-alpha-C,C1 utilized as reference standards.

High and low potency interferon-alpha subtypes induce (2'-5') oligoadenylate synthetase with similar efficiency

Three major subtypes of human interferon-alpha (IFN-alpha), isolated from virus-induced leukocytes, were compared for their antiviral and anticellular activities on one hand, and for their ability to induce (2'-5') oligoadenylate synthetase on the other hand. One subtype, IFN-alpha 1, was found to have low specific antiviral (6.10(6)-5.10(7) units/mg) and anticellular activities when measured on a variety of human cells. A second subtype, exhibiting an unusually high molecular weight (26,000) by SDS-polyacrylamide gel electrophoresis (IFN-alpha 26K), was found to have the highest known specific antiviral (8.10(8)-2.10(9) units/mg) and anticellular activities. Thus, these two subtypes of IFN-alpha differ by a factor of 330 and represent the two extremes in the antiviral scale on human cells. A third subtype, IFN-alpha 2, was tested as well and was found to have intermediate antiviral and anticellular activities. The ability of these three subtypes to induce (2'-5') oligoadenylate synthetase in human cells was then measured. It was found that on a weight basis, the three subtypes were equally effective in inducing the enzyme. Since the level of (2'-5') adenylate oligomers is affected also by the interferon-induced (2'-5') phosphodiesterase, the ability of these subtypes to induce this enzyme was compared as well and was found to be very similar. We therefore conclude that the differences in potency between these IFN-alpha subtypes are not related to their ability to induce (2'-5') oligoadenylate synthetase.

Different antiviral spectra of human macrophage interferon activities

The antiviral activity of alpha-interferon (IFN-alpha) produced by human leukocytes in response to viral infection has been considered to be independent of the virus which induced its production. Recently, however, IFN-alpha has been found to include at least eight different subtypes, as indicated by measurement of antigenic variability, DNA hybridization and amino acid sequencing. We considered the possibility that interferon heterogeneity may play a part in enhancing host antiviral defence and now present data suggesting that purified human macrophages, when exposed to different viruses, produce interferons having differing spectra of antiviral activity. These findings may provide a functional correlation for IFN-alpha heterogeneity.

Distinction of two subtypes of human leukocyte interferon (IFN-alpha) on B cell activation. B cell proliferation by two subtypes of IFN-alpha

We examined the effect of interferon (IFN), with particular emphasis on the effects of the two subtypes of IFN-alpha (IFN-alpha A and IFN-alpha B) on the B cell proliferation induced by Staphylococcus aureus Cowan I bacterium (SpA Col). An increase of SpA Col-induced proliferation was observed in the presence of 100 to 1000 U/ml of IFN-alpha, but a decrease of SpA Col-induced proliferation was observed in the presence of 1000 to 10,000 U/ml of IFN-beta. The two subtypes of IFN-alpha had different effects on cell proliferation; a significant enhancement was shown in the presence of 1000 to 10,000 U/ml of IFN-alpha A, but inhibition was shown in the presence of 1000 to 10,000 U/ml of IFN-alpha B. In the reconstitution test of the two subtypes of IFN-alpha, the boundary between enhancement and inhibition of SpA Col-induced proliferation was revealed when the proportion of IFN-alpha A and IFN-alpha B (IFN-alpha A:IFN-alpha B) ranged between 8:2 and 9:1. Toward the SpA Col-induced responses, the above IFN were all found to act on B cells directly, independent of the presence of T cells. Proliferative responses by IFN-alpha and IFN-alpha A, however, were shown to be slightly dependent on the presence of monocytes. The lymphocyte proliferation induced by other mitogens (phytohemagglutinin, concanavalin A, pokeweed mitogen, and protein A of S. aureus) were all inhibited by the above IFN.

The effect of tunicamycin on interferon-alpha production in a human lymphoblastoid cell line (Namalwa)

Tunicamycin alters the biological properties of human interferon-alpha (HuIFN-alpha) produced in its presence by changing the mixture of IFN-alpha subtypes produced. When IFN mRNA extracted from treated cells was microinjected into oocytes the product was also altered. However, treatment of the oocytes with tunicamycin did not alter the properties of the IFN-alpha produced.

Structure and expression of human IFN-alpha genes

Copy DNA (cDNA) was prepared from induced leucocyte poly(A) RNA and cloned in Escherichia coli. IFN-alpha cDNA clones were isolated by subculture cloning with the use of a translation hybridization assay. Definitive identification of the clones was based on the production of an interferon-like protein by the transformed bacteria. Different IFN-alpha cDNAs, with characteristic target cell specificities, were identified. The cloned cDNAs typically encode a mature polypeptide of 166 (or, for IFN-alpha 2, 165) amino acids and a signal sequence of 23 amino acids. A human chromosomal library was screened with IFN cDNA and 17 distinct IFN-alpha-related sequences were isolated and identified, of which 7 proved to be nonallelic authentic genes and 4 pseudogenes; 6 sequences remain to be elucidated. Taking into account the work of Goeddel and his colleagues, 13 non-allelic authentic genes and 6 pseudogenes can be distinguished. In addition, 9 genes believed to be allelic to the 13 authentic genes have been sequenced. The IFN-alpha genes may be classified into two major subfamilies, which diverged at least 33 Ma ago, but perhaps much earlier, if sequence rectification occurred. At least one IFN-alpha gene appears to have resulted by a recombinational event between members of the subfamily I and II. IFN-beta is distantly related to IFN-alpha's and may have diverged from a common ancestor at least 500 Ma ago. Both IFN-alpha and IFN-beta genes differ from most other genes of higher organisms by being devoid of introns. The mouse was found to possess an IFN-alpha gene family of a size similar to that of man; the murine genes also do not have introns. IFN-alpha genes devoid of their signal sequence were joined to prokaryotic promoters to produce the mature interferons in E. coli in high yield. IFN-alpha 2, purified to homogeneity, has been crystallized by T. Unge and B. Strandberg (Uppsala). Hybrid genes consisting of IFN-alpha 1 and IFN-alpha 2 segments were constructed and expressed in E. coli; the target cell specificities of such hybrids were dependent on the arrangement of the segments and were different from those of either parent. The chromosomal gene for HuIFN-alpha 1 was introduced into mouse L cells to study the mechanism of its expression. Correct transcription was only detected after induction (with Newcastle disease virus); expression was transient, with the same kinetics as those of the endogenous mouse IFN mRNA. Natural murine IFNs and human IFN-beta and IFN-gamma are glycosylated. Because E. coli cells transformed with the genes of eukaryotic glycoproteins are not expected to yield correctly glycosylated polypeptides, we prepared lines of hamster cells permanently transformed with hybrid plasmids, which contained an IFN gene linked to the SV40 early promoter, as well as dihydrofolate reductase as a selective marker. After intracellular amplification of the introduced genes, cell lines were obtained which constitutively produced IFN at about 40 000 units ml-1 and could be propagated for at least several months.

Antigenic correlations between components of C243 and L cell interferons

The three components of virus-induced interferon (IFN) from C243 cells, with molecular weights of 19,000, 24,000 and 30,000, were examined for neutralization by antisera against the alpha (24,000) and beta (36,000) types of virus-induced L cell IFN. The results indicated that the C243 19,000 component belongs to the alpha type, and the 24,000 and 30,000 components to the beta type. The 19,000 component was also indistinguishable from L cell IFN-alpha in the reaction with antisera against human IFN-alpha.