Cut, copy, move, delete: The study of human interferon genes reveal multiple mechanisms underlying their evolution in amniotes.

Interferons (IFNs) are rapidly evolving cytokines released when viral infections are detected in cells. Previous research suggests that genes encoding IFNs and their receptors duplicated extensively throughout vertebrate evolution. We present molecular genetic evidence that supports the use of nonallelic homologous recombination (NAHR) to expand select IFN genes during amniote evolution. The duplication of long regions of genome (encompassing at least one functional IFN gene) followed by the insertion of this genome fragment near its parent's location, is commonly observed in many amniote genomes. Duplicates inserted away from duplication hotspots are not as frequently perturbed with new duplicates, and tend to survive long periods of evolution, sometimes becoming new IFN subtypes. Although most duplicates are inserted parallel to and near the original sequence, the insertion of the Kelch-like 9 gene within the Type I IFN locus of placental mammals promoted antiparallel insertion of gene duplicates between the Kelch-like 9 and IFN-ε loci. Genetic exchange between highly similar Type I gene duplicates as well as between Type III IFN gene duplicates homogenized their diversification. Oddly, Type III IFN genes migrated long distances throughout the genome more frequently than did Type I IFN genes. The inter-chromosomal movement of Type I IFN genes in amniotes correlated with complete intron loss in their gene structure, and repeatedly occurred with occasional Type III IFN genes.

Analytical use of multi-protein Fluorescence Resonance Energy Transfer to demonstrate membrane-facilitated interactions within cytokine receptor complexes.

Experiments measuring Fluorescence Resonance Energy Transfer (FRET) between cytokine receptor chains and their associated proteins led to hypotheses describing their organization in intact cells. These interactions occur within a larger protein complex or within a given nano-environment. To illustrate this complexity empirically, we developed a protocol to analyze FRET among more than two fluorescent proteins (multi-FRET). In multi-FRET, we model FRET among more than two fluorophores as the sum of all possible pairwise interactions within the complex. We validated our assumption by demonstrating that FRET among pairs within a fluorescent triplet resembled FRET between each pair measured in the absence of the third fluorophore. FRET between two receptor chains increases with increasing FRET between the ligand-binding chain (e.g., IFN-γR1, IL-10R1 and IFN-λR1) and an acylated fluorescent protein that preferentially resides within subsections of the plasma membrane. The interaction of IL-10R2 with IFN-λR1 or IL-10R1 results in decreased FRET between IL-10R2 and the acylated fluorescent protein. Finally, we analyzed FRET among four fluorescent proteins to demonstrate that as FRET between IFN-γR1 and IFN-γR2 or between IFN-αR1 and IFN-αR2c increases, FRET among other pairs of proteins changes within each complex.

Improving the spectral analysis of Fluorescence Resonance Energy Transfer in live cells: application to interferon receptors and Janus kinases.

The observed Fluorescence Resonance Energy Transfer (FRET) between fluorescently labeled proteins varies in cells. To understand how this variation affects our interpretation of how proteins interact in cells, we developed a protocol that mathematically separates donor-independent and donor-dependent excitations of acceptor, determines the electromagnetic interaction of donors and acceptors, and quantifies the efficiency of the interaction of donors and acceptors. By analyzing large populations of cells, we found that misbalanced or insufficient expression of acceptor or donor as well as their inefficient or reversible interaction influenced FRET efficiency in vivo. Use of red-shifted donors and acceptors gave spectra with less endogenous fluorescence but produced lower FRET efficiency, possibly caused by reduced quenching of red-shifted fluorophores in cells. Additionally, cryptic interactions between jellyfish FPs artefactually increased the apparent FRET efficiency. Our protocol can distinguish specific and nonspecific protein interactions even within highly constrained environments as plasma membranes. Overall, accurate FRET estimations in cells or within complex environments can be obtained by a combination of proper data analysis, study of sufficient numbers of cells, and use of properly empirically developed fluorescent proteins.

Ligand-independent interaction of the type I interferon receptor complex is necessary to observe its biological activity.

Ectopic coexpression of the two chains of the Type I and Type III interferon (IFN) receptor complexes (IFN-αR1 and IFN-αR2c, or IFN-λR1 and IL-10R2) yielded sensitivity to IFN-alpha or IFN-lambda in only some cells. We found that IFN-αR1 and IFN-αR2c exhibit FRET only when expressed at equivalent and low levels. Expanded clonal cell lines expressing both IFN-αR1 and IFN-αR2c were sensitive to IFN-alpha only when IFN-αR1 and IFN-αR2c exhibited FRET in the absence of human IFN-alpha. Coexpression of RACK-1 or Jak1 enhanced the affinity of the interaction between IFN-αR1 and IFN-αR2c. Both IFN-αR1 and IFN-αR2c exhibited FRET with Jak1 and Tyk2. Together with data showing that disruption of the preassociation between the IFN-gamma receptor chains inhibited its biological activity, we propose that biologically active IFN receptors require ligand-independent juxtaposition of IFN receptor chains assisted by their associated cytosolic proteins.

A dance between interferon-alpha/beta and p53 demonstrates collaborations in tumor suppression and antiviral activities.

In the recent paper by Takaoka et al. (2003), the authors demonstrate that interferons-alpha and -beta stimulate p53 expression but not p53 activation. The increase in p53 expression translates into significant enhancement of apoptosis and reduction of chemotherapeutic dosages in vitro to destroy tumor cells. Furthermore, viral infections are also modulated by p53 in collaboration with interferons-alpha and -beta. These observations are significant and may lead to new paradigms for therapy if the high doses of interferon necessary to obtain the effects in vitro can be combined with more active interferons, interferons with minimal side effects, and/or novel delivery systems to target interferons directly to tumors.

Binding and activity of all human alpha interferon subtypes.

Vertebrates have multiple genes encoding Type I interferons (IFN), for reasons that are not fully understood. The Type I IFN appear to bind to the same heterodimeric receptor and the subtypes have been shown to have different potencies in various experimental systems. To put this concept on a quantitative basis, we have determined the binding affinities and rate constants of 12 human Alpha-IFN subtypes to isolated interferon receptor chains 1 and 2. Alpha-IFNs bind IFNAR1 and IFNAR2 at affinities of 0.5-5 µM and 0.4-5 nM respectively (except for IFN-alpha1 - 220 nM). Additionally we have examined the biological activity of these molecules in several antiviral and antiproliferative models. Particularly for antiproliferative potency, the binding affinity and activity correlate. However, the EC50 values differ significantly (1.5 nM versus 0.1 nM for IFN-alpha2 in WISH versus OVCAR cells). For antiviral potency, there are several instances where the relationship appears to be more complicated than simple binding. These results will serve as a point of reference for further understanding of this multiple ligand/receptor system.

Historical developments in the research of interferon receptors.

Interferons (IFNs) were discovered 50 years ago independently by Isaacs and Lindemann and by Nagata and Kojima. When it was later realized that IFNs are active at very low concentrations, research began to determine how their powerful effects were generated from such a small initial signal. It has since been established that interferons, as well as all other cytokines, employ cell surface receptors to translate their presence in the serum to a potent cellular response to a viral infection. These receptor complexes are composed of multiple distinct glycosylated transmembrane polypeptides, a number of protein tyrosine kinases, and interact transiently with a large variety of other proteins including transcription factors, phosphatases, signaling repressors, and adaptor proteins coupling the receptor to alternative signaling pathways. Three major receptor complexes exist that are exclusive to each of three major classes of interferon. Even though the effects of each major class of interferon vary physiologically, each receptor complex interacts with its ligand in similar ways and activates similar signaling cascades. In this mini-review, we take a historical perspective at the major events in the characterization of interferon receptors, discussing interesting results that still need to be explained.

Protein arginine methyltransferases: evolution and assessment of their pharmacological and therapeutic potential.

Protein arginine N-methylation is a post-translational modification whose influence on cell function is becoming widely appreciated. Protein arginine methyltransferases (PRMT) catalyze the methylation of terminal nitrogen atoms of guanidinium side chains within arginine residues of proteins. Recently, several new members of the PRMT family have been cloned and their catalytic function determined. In this report, we present a review and phylogenetic analysis of the PRMT found so far in genomes. PRMT are found in nearly all groups of eukaryotes. Many human PRMT originated early in eukaryote evolution. Homologs of PRMT1 and PRMT5 are found in nearly every eukaryote studied. The gene structure of PRMT vary: most introns appear to be inserted randomly into the open reading frame. The change in catalytic specificity of some PRMT occurred with changes in the arginine binding pocket within the active site. Because of the high degree of conservation of sequence among the family throughout evolution, creation of specific PRMT inhibitors in pathogenic organisms may be difficult, but could be very effective if developed. Furthermore, because of the intricate involvement of several PRMT in cellular physiology, their inhibition may be fraught with unwanted side effects. Nevertheless, development of pharmaceutical agents to control PRMT functions could lead to significant new targets.

Site-specific phosphorylation of raf in cells containing oncogenic ras-p21 is likely mediated by jun-N-terminal kinase.

In a study of interactions between the raf-MEK-MAPK (ERK) and JNK-jun pathways, we found previously that JNK can induce phosphorylation of raf but not vice versa. In this study, we investigate the nature of the JNK-induced phosphorylation of raf. In in vitro experiments in which immunobead-bound raf is phosphorylated by activated JNK, we find strong phosphorylation signals at raf-Ser259 and Ser338. The Ser259 phosphorylation is surprising since it is associated with inhibition of migration of raf to the cell membrane where it can interact with ras-p21. We also find that in oocytes induced to mature with oncogenic ras-p21, which induces high levels of phosphorylated JNK and MAPK, the same pattern of phosphorylation of raf occurs. In contrast, in oocytes induced to mature with insulin, which requires activation of wild-type ras-p21, phosphorylation of raf-Ser338 but not raf-Ser259 occurs. In oncogenic ras-transformed human pancreatic cancer MIA-PaCa-2 cells, phosphorylation of both raf serines occurs. Treatment of these cells with the ras peptide, PNC-2 attached to a penetrating sequence that blocks JNK and MAPK phosphorylation and induces tumor cell necrosis, results in a marked decrease in phosphorylation of raf-Ser259, but not that of raf-Ser338. These results suggest that oncogenic ras-p21 induces phosphorylation of both raf-Ser259 and Ser338 and that raf-Ser 259 phosphorylation may be effected by activated JNK.

Elevated expression of ISG15 in tumor cells interferes with the ubiquitin/26S proteasome pathway.

IFN-stimulatory gene factor 15 (ISG15) is a ubiquitin-like protein, which is conjugated to many cellular proteins. However, its role in protein degradation is unclear. Here, we show that ISG15 is highly elevated and extensively conjugated to cellular proteins in many tumors and tumor cell lines. The increased levels of ISG15 in tumor cells were found to be associated with decreased levels of polyubiquitinated proteins. Specific knockdown of ISG15 expression using ISG15-specific small interfering RNA (siRNA) was shown to increase the levels of polyubiquitinated proteins, suggesting an antagonistic role of ISG15 in regulating ubiquitin-mediated protein turnover. Moreover, siRNA-mediated down-regulation of the major E2 for ISG15 (UbcH8), which blocked the formation of ISG15 protein conjugates, also increased the levels of polyubiquitinated proteins. Together, our results suggest that the ISG15 pathway, which is deregulated during tumorigenesis, negatively regulates the ubiquitin/proteasome pathway by interfering with protein polyubiquitination/degradation.

The dual-specificity kinases, TOPK and DYRK1A, are critical for oocyte maturation induced by wild-type--but not by oncogenic--ras-p21 protein.

We have previously found that oncogenic ras-p21 and insulin, which activates wild-type ras-21 protein, both induce Xenopus laevis oocyte maturation that is dependent on activation of raf. However, oncogenic ras-p21 utilizes raf-dependent activation of the two classic raf targets, MEK and MAP kinase (MAPK or ERK) while insulin-activated wild-type ras-p21 does not depend on activation of these two kinases. Utilizing a microarray containing the entire Xenopus genome, we discovered two dual specificity kinases, T-Cell Origin Protein Kinase (TOPK), known to bind to raf and the nuclear kinase, DYRK1A, that are expressed at much higher levels in insulin-matured oocytes. Using SiRNA's directed against expression of both of these proteins, we now show that each inhibits insulin-but not oncogenic ras-p21-induced oocyte maturation. Control siRNA's have no effect on either agent in induction of maturation. We find that each SiRNA "knocks down" expression of its target protein while not affecting expression of the other protein. These results suggest that both proteins are required for maturation induced by wild-type, but not oncogenic, ras-p21. They also suggest that oncogenic and wild-type ras-p21 utilize pathways that become divergent downstream of raf. On the basis of these findings, we propose a model for two signal transduction pathways by oncogenic and activated wild-type ras-p21 showing points of overlap and divergence.

MDA-7/IL-24: novel cancer growth suppressing and apoptosis inducing cytokine.

The melanoma differentiation-associated gene-7 (mda-7) was cloned by subtraction hybridization as a molecule whose expression is elevated in terminally differentiated human melanoma cells. Current information based on structural and sequence homology, has led to the recognition of MDA-7 as an IL-10 family cytokine member and its renaming as IL-24. Northern blot analysis revealed mda-7/IL-24 expression in human tissues associated with the immune system such as spleen, thymus, peripheral blood leukocytes and normal melanocytes. The MDA-7/IL-24 mouse counterpart, FISP, appears to be a Th2-specific protein and the rat counterpart, C49A/MOB-5, is associated with wound healing and is also induced as a consequence of ras-transformation. A notable property of MDA-7/IL-24 is its ability to induce apoptosis in a large spectrum of human cancer derived cell lines, in mouse xenografts and upon intratumoral injection in human tumors (phase I clinical trials). Various aspects of this intriguing molecule including its cytokine and anti-tumoral effects are described and discussed.

Two dual specificity kinases are preferentially induced by wild-type rather than by oncogenic RAS-P21 in Xenopus oocytes.

In prior studies, we have found that oncogenic ras-p21 protein induces oocyte maturation using pathways that differ from those activated by insulin-induced wild-type ras-p21. Both oncogenic and wild-type ras-p21 require interactions with raf, but unlike oncogenic ras-p21, insulin-activated wild-type ras-p21 does not depend completely on activation of MEK and MAP kinase (MAPK or ERK) on the raf kinase pathway. To determine what raf-dependent but MAPK-independent pathway is activated by wild-type ras-p21, we have analyzed gene expression in oocytes induced to mature either with oncogenic ras-p21 or with insulin using a newly available Xenopus gene array. We find a number of proteins that are preferentially expressed in one or the other system. Of these, two proteins, both dual function kinases, T-Cell Origin Protein Kinase (TOPK) and the nuclear kinase, DYRK1A, are preferentially expressed in the insulin system. Confirming this finding, blots of lysates of oocytes, induced to mature with oncogenic ras-p21 and insulin, with anti-TOPK and anti-DYRK1A show much higher protein expression in the lysates from the insulin-matured oocytes. Neither of these kinases activates or is activated by MAPK and is therefore an attractive candidate for being on a signal transduction pathway that is unique to insulin-activated wild-type ras-p21-induced oocyte maturation.

Evolution of the Class 2 cytokines and receptors, and discovery of new friends and relatives.

The sequencing of a wide variety of genomes and their transcripts has allowed researchers to determine how proteins or protein families evolved and how strongly during evolution a protein has been conserved. In this report, we analyze the evolution of the Class 2 ligands and their cognate receptors by analyzing Class 2 ligand and receptor chain gene sequences from a variety of DNA sequence databases. Both the Class 2 cytokines and receptor chains appear to have developed during the evolution of the chordate phyla: distant homologues of type I interferon (IFN) receptors are the only Class 2 cytokine receptors identified in the Ciona genomes, while a wide variety of Class 2 ligands and receptor chains are encoded in the currently available genomes of bony vertebrates (teleost fish, amphibians, reptiles, birds, mammals). Phylogenetic trees of ligands and ligand-binding receptor chains demonstrate that proteins involved in conferring antiviral activity diverged before those involved in adaptive immunity. Genes encoding IFNs and IFN receptors duplicated multiple times during chordate evolution, suggesting that duplication of genes encoding IFN activity conveyed an evolutionary advantage. Altogether, these data support a model whereby the original Class 2 cytokines and receptors evolved and duplicated during the evolution of the chordate innate immune response system; new receptor and ligand duplications evolved into signaling molecules to fulfill communication requirements of a highly specialized and differentiated vertebrate immune system. In addition, the genomic analysis led to the discovery of some new members of this family.

Modulation of the activation of Stat1 by the interferon-gamma receptor complex.

The activation of Stat1 by the interferon-gamma (IFN-gamma) receptor complex is responsible for the transcription of a significant portion of IFN-gamma induced genes. Many of these genes are responsible for the induction of an apoptotic state in response to IFN-gamma. In the absence of Stat1 activation, IFN-gamma instead induces a proliferative response. Modifying Stat1 activation by IFN-gamma may have pharmacological benefits. We report that the rate of activation of Stat1 can be altered in HeLa cells by overexpressing either the IFN-gammaR1 chain or the IFN-gammaR2 chain. These alterations occur in hematopoietic cell lines: Raji cells and monocytic cell lines, which have average and above-average IFN-gammaR2 surface expression, activate Stat1 similarly to HeLa cells and HeLa cells overexpressing IFNgammaR2, respectively. The rapid Stat1 activation seen in HeLa cells can be inhibited by overexpressing a chimeric IFN-gammaR2 chain that does not bind Jak2 or (when high concentrations of IFN-gamma are used) by overexpressing IFN-gammaR1. These data are consistent with a model in which the recruitment of additional Jak2 activity to a signaling complex accelerates the rate of Stat1 activation. We conclude that the rate of activation of Stat1 in cells by IFN-gamma can be modified by regulating either receptor chain and speculate that pharmacological agents which modify receptor chain expression may alter IFN-gamma receptor signal transduction.

Preassembly and ligand-induced restructuring of the chains of the IFN-gamma receptor complex: the roles of Jak kinases, Stat1 and the receptor chains.

We previously demonstrated using noninvasive technologies that the interferon-gamma (IFN-gamma) receptor complex is preassembled (1). In this report we determined how the receptor complex is preassembled and how the ligand-mediated conformational changes occur. The interaction of Stat1 with IFN-gammaR1 results in a conformational change localized to IFN-gammaR1. Jak1 but not Jak2 is required for the two chains of the IFN-gamma receptor complex (IFN-gammaR1 and IFN-gammaR2) to interact; however, the presence of both Jak1 and Jak2 is required to see any ligand-dependant conformational change. Two IFN-gammaR2 chains interact through species-specific determinants in their extracellular domains. Finally, these determinants also participate in the interaction of IFN-gammaR2 with IFN-gammaR1. These results agree with a detailed model of the IFN-gamma receptor that requires the receptor chains to be pre-associated constitutively for the receptor to be active.

Expression analysis and genomic characterization of human melanoma differentiation associated gene-5, mda-5: a novel type I interferon-responsive apoptosis-inducing gene.

Melanoma differentiation associated gene-5 (mda-5) was identified by subtraction hybridization as a novel upregulated gene in HO-1 human melanoma cells induced to terminally differentiate by treatment with IFN-beta+MEZ. Considering its unique structure, consisting of a caspase recruitment domain (CARD) and an RNA helicase domain, it was hypothesized that mda-5 contributes to apoptosis occurring during terminal differentiation. We have currently examined the expression pattern of mda-5 in normal tissues, during induction of terminal differentiation and after treatment with type I IFNs. In addition, we have defined its genomic structure and chromosomal location. IFN-beta, a type I IFN, induces mda-5 expression in a biphasic and dose-dependent manner. Based on its temporal kinetics of induction and lack of requirement for prior protein synthesis mda-5 is an early type I IFN-responsive gene. The level of mda-5 mRNA is in low abundance in normal tissues, whereas expression is induced in a spectrum of normal and cancer cells by IFN-beta. Expression of mda-5 by means of a replication incompetent adenovirus, Ad.mda-5, induces apoptosis in HO-1 cells as confirmed by morphologic, biochemical and molecular assays. Additionally, the combination of Ad.mda-5+MEZ further augments apoptosis as observed in Ad.null or uninfected HO-1 cells induced to terminally differentiate by treatment with IFN-beta+MEZ. The mda-5 gene is located on human chromosome 2q24 and consists of 16 exons, without pseudogenes, and is conserved in the mouse genome. Present data documents that mda-5 is a novel type I IFN-inducible gene, which may contribute to apoptosis induction during terminal differentiation and during IFN treatment. The conserved genomic and protein structure of mda-5 in human and mouse will permit analysis of the evolution and developmental aspects of this gene.

The IL-10R2 binding hot spot on IL-22 is located on the N-terminal helix and is dependent on N-linked glycosylation.

IL-22 is a class 2 alpha-helical cytokine involved in the generation of inflammatory responses. These activities require IL-22 to engage the cell surface receptors IL-22R1 and the low-affinity signaling molecule IL-10R2. IL-10R2 also interacts with five other class 2 cytokines: IL-10, IL-26, and the interferon-like cytokines IL-28A, IL-28B, and IL-29. Here, we define the IL-10R2 binding site on IL-22 using surface plasmon resonance (SPR) and site-directed mutagenesis. Surprisingly, the binding hot spot on IL-22 includes asparagine 54 (N54), which is post-translationally modified by N-linked glycosylation. Further characterization of the glycosylation reveals that only a single fucosylated N-acetyl glucosamine on N54 is required for maximal IL-10R2 binding. Biological responses of IL-22 mutants measured in cell-based luciferase assays correlate with the in vitro SPR studies. Together, these data suggest that IL-22 activity may be modulated via changes in the glycosylation state of the ligand during inflammation.

Increased interleukin-10 mRNA stability in melanoma cells is associated with decreased levels of A + U-rich element binding factor AUF1.

Abnormal production of interleukin-10 (IL-10) is observed in some pathologic conditions. For example, compared with normal melanocytes, IL-10 expression is elevated in melanoma cells. IL-10 overexpression could inhibit both immune surveillance and tumor rejection. We investigated a potential posttranscriptional mechanism for IL-10 overexpression in melanoma cells. In normal melanocytes, the half-life of IL-10 mRNA is 7 min, whereas in the melanoma cell line MNT1, the half-life is 75 min. This 10-fold difference could account, at least in part, for IL-10 overexpression in MNT1 cells. Examination of the 3'-untranslated region (3'-UTR) of IL-10 mRNA revealed a suspected A + U-rich element (ARE) that might target the mRNA for rapid degradation. Transfection experiments confirmed that these sequences promote rapid degradation when inserted into a normally stable mRNA, indicating ARE functionality. As AREs act via their interactions with ARE-binding proteins, we examined cytoplasmic proteins from normal melanocytes and MNT1 cells for IL-10 ARE-binding activity. Compared with cytoplasmic extracts of normal melanocytes, cytoplasmic extracts of MNT1 cells possess substantially less ARE-binding activity, consistent with the extended half-life of IL-10 mRNA in MNT1 cells. Finally, we find that the ARE-binding protein AUF1 comprises the major ARE-binding activity in cytoplasmic extracts of normal melanocytes. By contrast, AUF1 is not detectable in cytoplasmic extracts of MNT1 cells but appears restricted to the nuclear fraction. Together, these data suggest a mechanism whereby reduced cytoplasmic levels of AUF1 in MNT1 melanoma cells may lead to IL-10 overexpression, with deleterious consequences for tumor surveillance and rejection.