Explore with me as we take a deeper look into the rich world of the IFN induced enzymes (2-5OAS, RNase L and PKR) dysfunctional in CFS. Even after reading CFS ABA there is still much to learn. This exploration is always informed by the question “What activates the p100 2-5OAS isoform that produces the dimers that bind to RNase L in such a way that it is easily fragmented?
I. IFN signalling and RNase L –overview
Our understanding of RNase L and PKR activity must start with the IFN signaling system believed to activate them in CFS patients. Interferons are produced in response to stressful stimuli (viruses, bacteria, toxins). The IFN signals do two things; they ‘wake up’ 2-50AS, PKR and RNase L (and other enzymes) in the cytoplasm and they activate interferon stimulated genes (ISG’s) in the nucleus. Over 300 ISG’s have been identified.
Because RNase L is already present in the cell in large amounts relatively little RNase L is produced (2x’s) in contrast to other enzymes (20-100x’s). (This is presumably why RNase L was given its latent (L) moniker. PKR and other enzymes are along with RNase L important parts of the antiviral response. Why then is it necessary to have RNase L always present in the cytoplasm? Why not just induce its production when the time comes? Perhaps because RNase L is alot more than an antiviral enzyme. See below).
The IFN System (Stark et. al. 1998) – While our knowledge of the IFN signaling system has increased greatly much is still unclear. We know a wide variety of Type I IFN subtypes (12) exist – each presumably with its own specialization.
The extent of cellular antiviral or antiproliferative activity or even a cells response to a specific virus is dependent upon kind of Type I IFN induced. We are still unclear, however, what types of activities each IFN subtype induces. It is possible, therefore, that the p100 2-5OAS activation believed central in CFS is simply due to ‘aberrant’ signaling by one of the IFN subtypes.
The ability of a manufactured dsRNA product (Ampligen) to reregulate 2-5OAS suggests, however, that the problem is not due to a dysfunctional 2-5OAS response to IFN’s (Englebienne 2003). It appears instead that certain kinds of RNA that prompt 2-5OAS to induce the p100 2-5OAS isoform instead of the p69 one, are particularly abundant in CFS patients.
(*While we talk of RNase L activation it is important to note that what we are really talking about is 2-5OAS activation. RNase L gets all attention because it does all the damage but RNase L degradation is almost inevitable (given sufficient protease activity) once the wrong type of 2-5OAS isoform is induced. The RNase L dysregulation originates in 2-5OAS dysregulation not a dysfunctional RNase L)
RNase and the STAT I degradation
Since the STAT proteins transfer the IFN signal into the nucleus, the STAT I degradation found in CFS poses a problem; how is RNase L activated if the pathway to IFN (and therefore RNase L) activation is blocked?
Once the STAT I signal is blocked then the production of 2-5OAS, RNase L, etc. mRNA presumably would cease. Since most RNase L is latent (inactive but already present) STAT I degradation would not markedly shut down RNase L production but because 2-5OAS produces RNase L’s activator (2-5A), it would appear to shut down RNase L activity. How then to account for increased RNase L activity in CFS patients?
IFN signalling pathway – on the wrong track?
IFN’s seem the logical source of the RNase L activation seen in CFS patients. Not only are they the best known inducers of 2-5OAS, RNase L and PKR but Type I IFN treatment in hepatitis patients produces a CFS-like condition. A recent study (Percario et. al. 1999) indicated, however, that a signaling pathway induced not by IFN’s but by retinoic acid (RA) can induce 2-5OAS activity as well.
Retinoic acid is a vitamin D related compound that regulates cell differentiation and proliferation; two activities RNase L is involved in. In particular, this study found that an interferon regulator factor (IRF-1) that regulates interferon stimulated genes (ISG’s) is induced by RA.
One of the genes induced by RA (via IRF-1) is 2-5OAS. 2-5OAS induction occurs when RA induces IRF-1 to bind to the promoter region of the 2-5OAS gene. RA also induces the activity of a nuclear transcription factor (NF-kB) which encodes proteins (acute phase proteins, cytokines, cell adhesion molecules, T-cell activation) involved in the acute response to pathogens or cellular stress.
NF-kB is also a key element in activating monocytes/macrophages. These are the only cells in which RNase L dysregulation has thus far been found. RA’s activation of NF-kB is intriguing because – as we shall see – NF-kB also binds to the promoter region of the p100 gene but not to any of the others. RA then appears able not only to induce 2-5OAS but is able to induce the 2-5OAS isoform believed to be at the source of RNase L deregulation in CFS.
RA could resolve the paradox of an active RNase L system and an inactive STAT I signaling pathway. It may also explain a recent study (Ikuta et. al. 2003) which indicated 2-5OAS activation in CFS patients was not correlated with IFN b activity.
*Update- A 2004 study by Suhadolnik found significantly increased RNase L fragmentation but not IFN-a, 2-5A concentration or RNase L activity in CFS patients vs. depressed patients or healthy controls.
The authors of CFS ABA suggested retinoic acid may provide a therapeutic tool in 2-5OAS and RNase L regulation in CFS and were investigating its efficacy. (See Chapter Two of “Chronic Fatigue Syndrome A Biological Approach”.)
Other inducers of 2-5OAS – IFN’s and RA are, interestingly enough, are not the only inducers of 2-5OAS. Glucocorticoids, several growth factors, protein kinase C and the anti-estrogen Tamoxifen appear to be able to either enhance or induce 2-5OAS activity (Player and Torrence 1998). Any of these (?) apparently could be the source of or exacerbate the 2-5OAS activity seen in CFS. Arachidonic acid, gangliosides and glucose are able to inhibit 2-5OAS activation.
The 2-5OAS activity (and dysregulation?) seen in CFS patients could apparently be induced by several factors other than IFN. The signal for 2-5OAS activation does not necessarily proceed via the STAT I proteins. The activation of an alternate 2-5OAS signaling pathway in CFS patients could explain why the 2-5OAS activation persists after degradation of the STAT I proteins.
II. Activation of the 100-kDa 2-5OAS gene
Once the IFN (or RA) signal reaches the nucleus it binds to and begins activating genes that will produce the mRNA needed to create the proteins that will fight the pathogen that has evoked the IFN response. One way to determine what activates a particular gene is to analyze its structure and composition.
All genes contain promoter regions that must be activated (bound) in order to start mRNA transcription. Studying the amino acid structure of this region gives us clues as to what activates it. Since our concern in CFS is figure out why p100 is upregulated in CFS, a study of what activates the promoter region of the gene encoding the p100 isoform is potentially very illuminating. A 2000 study did just that (Rebouillat et. al. 2000).
The ‘exonic’ structures of the genes expressing the p69 and p100 isoforms are very similar; the main differences between the two lie in the promoter region of the gene. The promoter regions on both p69 and p100 genes indicates they can both be induced by both IFN types but are more readily activated by type I IFN’s.
P69’s promoter region contains binding regions for an Interferon Stimulating Gene Factor (ISGF3) and Interferon Regulatory Factor (IGF3). Unlike p69, however, the promoter region of p100 contains an NF-kB motif. This suggests that the p100 enzyme is, in contrast to p69, involved not just in the IFN mediated immune response but in the NF-kB mediated immune response as well.
Since NF-kB is, in contrast to IFN, involved in inflammatory activities, this was quite surprising. It suggests that the RNA that turns on the p100 isoform maybe a product of inflammatory activities.
(Perhaps given the convoluted world of the CFS this is not that surprising. One of the main activators of NF-kB, protein kinase R (PKR), is upregulated in many CFS patients. This suggests an upregulated PKR could play a significant role in p100 activation and RNase L fragmentation.)
If I am reading the next section correctly it appears the authors suggest IFN and retinoic acid engage in synergistic activities that increases the transcriptional activity in the promoter region of the p100 gene. They suggest that P100’s role may lie in ‘priming the pump; so to speak, for 2-5OAS activity. Why exactly this is so is unclear to me although it may lie in the p100’s ability to respond to very low levels of dsRNA (see below.)
III. 2-5OAS activation: a promising study
Once the 2-5OAS gene is turned on and mRNA is produced and 2-5OAS is actually produced it still needs to be activated. Since RNase L dysregulation appears to originate in an improper activation of 2-5OAS just what is able to activate 2-5OAS is of immediate concern to CFS patients.
Once researchers determine how 2-5OAS is turned on (i.e. what binds to its binding site) they will presumably be able to identify which RNA particles dysregulate it in CFS patients. If Englebienne and DeMeirLeir et. al. are correct, doing so would bring us close to the source of CFS! Heady stuff!
A 1998 study (Hartmann et. al. 1998) explored just how RNA particles bind to the p46 isoform (alas) of 2-5OAS not the p100 one. Despite their almost tragically poor choice of isoforms this research is encouraging simply because it shows it can be done. It is possible to identify the types of RNA that p100 responds to if someone, somewhere, someday chooses to. Perhaps its being done right now.
RNase L and PKR, are unique in their ability to respond to double-stranded portions of RNA that many viruses produce as they replicate. It has become clear, however, that RNase L and PKR sometimes respond to small bits of viral RNA differently; the same RNA that activates RNase L may inhibit PKR.
This study indicated that while PKR contained dsRNA binding domains (dsRBM’s) 2-5OAS, surprisingly enough, did not. These two enzymes appeared, in fact, to have little structural similarity. This was shocking. Both enzymes appear to response to dsRNA but they were almost completely different structurally.
Researchers appeared to be at something of a loss explaining just how 2-5OAS was activated. The 2-5OAS enzyme does not give up its secrets easily!
In order to address this problem 2-5OAS was exposed to a series of ssRNA ‘aptamers’ or artificially engineered strands of RNA. Not only did six of the aptamers turn on 2-5OAS but they turned it on more completely than did previously tested dsRNA’s. An attempt to identity the amino acid motifs on the aptamers responsible for binding to 2-5OAS indicated they bore little similarity to each other!
This meant 2-5OAS was able to respond to wide variety of different RNA types. The authors suggested this was not surprising given the wide variety of pathogens the 2-5OAS system needs to be able to interact with. Two small amino acid motifs were, however, found in all the aptamers.
The study indicated that RNA strands rich in oligo (C) which contain two specific amino acid motifs are able to bind the p46 isoform. Presumably by testing these motifs against an RNA data base the researchers would be able to determine the types of RNA able on the p46 isoform and where they come from.
IV. Location of the 100-kDa isoform – a clue to RNase L dysregulation?
That the different isoforms are found in different parts of the cell suggests they interact with different pathogens and are perhaps involved in different activities. Different viruses, for instance, may not only exploit different parts of the cell but the same virus may over time occupy different parts of the cell as well. The different isoforms, then, could target different types of viral RNA or different stages of the viral life process.
Both the p69 (the ‘good isoform) and p100 (the ‘bad’ isoform) are found in the cytoplasm but p69 is concentrated around the nucleus and in specific parts of the cytoplasm. P100, on the other hand, is found diffused throughout the cytoplasm and is more strongly associated with the ribosomes (protein factories) (Marie et. al. 1990).
The diffused nature of the P100 isoform is perhaps in keeping with the suggestion it ‘primes’ the system. It may be sentry of sorts. P69 has a feature (‘myristilization’) which p100 does not, that suggests it is able to penetrate the membranes of the organelles in the cell.
Surprisingly P100 is much more sensitive (from 10-100x’s) to the presence of dsRNA than p69 is. (Remember that after IFN ‘wakes’ the 2-5OAS enzymes up they double-check their environment to determine if signs of pathogens are present. If they are then the enzymes activate and start producing 2-5A out of ATP).
One wonders if some viruses could produce just enough dsRNA to activate p100 but not activate p69. If they could do this and induce elastase and/or calpain activity they would be able to knock out RNase L and inhibit the intracellular immune response.
Such a strategy would limit the pathogen to very slow growth; if it replicated too quickly it would produce enough dsRNA to wake up p69. There are organisms in nature that are able to survive only by living under the radar. Always sparse but often widely dispersed they are quickly cropped back when they become abundant enough to support pathogens or other enemies.
P100 – a summary
I was struck throughout my readings on p100 by two things. (1) how different the p100 isoform is from the other isoforms and (2) how often researchers suggest it may have an entirely different function from the other isoforms.
Six aspects clearly differentiate the p100 isoform from the others.
(1) Location of the gene: While the genes for the other isoforms are clustered together on one chromosome the gene for the p100 isoform is separate.
(2) Gene activation: An NF-kB binding motif in the promoter region of the p100 isoform suggests it is activated in response to different situations than are the other isoforms.
(3) Structure: In contrast to the p46 isoform (a tetramer) and the p69 isoform (a dimer) the p100 isoform is a monomer. P100 lacks a structural motif (CFK) that the other isoforms have which allows them to oligomerize. This is important because oligomerization appears to be critical to the ability to form trimers or tetramers of 2-5A. Although p100 is the largest isoform it is unable to form higher oligomers of ATP than dimers because it only contains three catalytic regions.
(4) Activation threshold: the p100 isoform responds to dsRNA at levels 10-100 times lower (!) than the other isoforms do. It also exhibits optimum activation at a much more alkaline pH level (7.5) than the others do (6.5)
(5) Location: the p100 isoform is the only isoform strongly associated with ribosomes. It is also found diffused throughout the cell.
(6) Binding ability: relative to the 2-5A oligomers produced by the other isoforms the 2-5A dimers produced by p100 exhibit only a weak ability to bind with RNase L.
P100 is the ‘odd man out’ among the 2-5OAS isoforms. Its oddity has lead researchers to suggest it functions differently than do the other isoforms. One group conjectured that p100 is primarily involved not in antiviral activity but in regulating gene expression and DNA replication. In 1999 this group (Rebouillat et. al.) celebrated the manufacture of a cDNA (cloned DNA?). They indicated this breakthrough would be ‘invaluable’ in further analyses of p100. Except for a paper on the p100 gene, however, no papers since then have been published that expand our knowledge of the role the p100 isoform plays in the body.
V. RNase L selectivity
The more selectivity RNase L exhibits in its ability to degrade mRNA, the more potential for disruption a deregulated RNase L could have. This is because a more selective RNase L is more likely to participate in a wider variety of processes than a less selective RNase L.
While the full scope of RNase L’s activities are unclear RNase L’s ability to degrade mRNA and therefore regulate protein manufacture gives it the potential to be one of the master regulators of cellular processes.
This did not at first appear to be case. RNase L appeared to be such a blunt tool that it not only destroyed viral mRNA but shut down protein synthesis in the cell in its zeal to stop viral replication.. Further studies, however, have indicated that RNase wields a much finer blade than was first supposed.
RNase L is involved in cellular apoptosis and muscle cell growth and differentiation. Its anti-proliferative properties indicate it is very likely involved in cancer suppression. Our understanding of the RNase L enzymes interactions in the body is just beginning.
A recent study (Xiao-Ling et. al. 1998) not only expanded on RNase L’s mode of interaction with viruses but found that it exhibited a surprising selectivity. For the first time this study was able to pinpoint the stage at which RNase L degrades viral RNA.
RNase L targets viral RNA early in their life while they are associated with ‘active replication complexes’. After viral RNA’s are disengaged from these complexes they are immune to RNase L degradation. That is until levels of dsRNA reach a certain point at which RNase L simply begins degrading ALL the ribosomal RNA in the cell. This apparently leads to the initiation of cellular apoptosis or cell suicide.
RNase L, then, mounts a two-staged attack on viruses. In the first stage it strictly targets viral mRNA. If that is not successful then at some point it simply attacks all rRNA and shuts down protein synthesis. This means that RNase L activity is not synonymous with reduced protein synthesis as was once believed.
Another paper by some of the same authors (Xiao-Ling 2000) further broadens the range of RNase L’s selectivity and deepens the implications of RNase L deregulation in CFS patients. Over 300 Interferon Stimulated Genes (ISG’s) have been identified. Any of those that lie downstream of RNase L could be affected in CFS patients. This paper was the first to examine RNase L’s regulation of ISG genes
The results of this study suggest that RNase L is involved in the regulation of both viral and cellular mRNA’s. RNase L’s regulatory activities were identified by noting which mRNA increased in cells containing inactive as opposed to active RNase L.
Interestingly enough, RNase L did not affect the levels of mRNA produced by genes involved in cellular proliferation/apoptosis. (Not all the genes were tested) Instead RNase L effected the abundance of a heretofore unknown gene involved in ubiquitination called Interferon Stimulated Gene 43 (ISG 43).
ISG 43 produces a protease that removes ubiquitin tags from their substrate. (Ubiquitin tags (‘kill me’ signs) mark a product for destruction by proteasomes.) By removing ubiquitin tags from proteins ISG 43 is downgrading the IFN immune response.
Ubiquitin tags play a role, interestingly enough, in the destruction of the STAT I (Signal transducers and activators of transcription) protein. STAT proteins play a key role in the transcription of IFN genes. (STAT I proteins are degraded in CFS patients – see Chapter V of “Chronic Fatigue Syndrome A Biological Approach”.)
Since destruction of the STAT proteins prevents the signal for the IFN response from getting to the nucleus, STAT proteins are destroyed when it is time to downgrade the immune response. A failure to remove the ‘kill me’ signs from STAT I because of decreased RNase L activity would result in an overactive immune response. Increased RNase L activity would presumably result in an underactive immune response.
ISG 43 (and proteasomes) also play a role in antigen presentation and viral defense. (Proteasomes are responsible for preparing the peptides that bind to MHC class I proteins. Once antigen producing cells (APC’) display the peptides from pathogens on their surface they travel to the lymph nodes where T and B cells determine if the peptides are from pathogens or not.
If they are the adaptive immune response is invoked. Because it appears that ISG 43 removes the tags that tell an APC to prepare a peptide, a down regulated ISG 43 would lead to a chronically enhanced immune response and vice versa.
This appears to be another arena that is ripe for exploration. Researchers could presumably chart the effect of RNase L fragmentation on ISG 43 functioning by measuring ISG 43 mRNA in cells with and without the 37-kDa fragment.)
Since ubiquitination plays an important role in cell development, growth and cancer protection, a dysfunctional ubiquitination process could have far reaching consequences.
ISG’s ability to down regulate IFN by removing ubiquitin adducts from cells suggests that RNase L deregulation could lead to either a chronically overactive or underactive immune response).
One of the puzzles regarding RNase L’s putative involvement in cellular mRNA regulation has been its mode of activation. If RNase L (2-5OAS) is dependent upon the dsRNA structures found in viral mRNA’s for activation how could it be activated by cellular RNA’s?
The apparent requirement for dsRNA lead the researchers to model ISG 43 to determine if it had double stranded like structures that RNase L (and 2-5OAS) might be able to bind to. It turned out that ISG 43 does indeed have ‘hairpin’ structures that are reminiscent of double-stranded structures found in dsRNA.
A recent study, however, found that some types of ssRNA may, in fact, be more potent activators of 2-5OAS than some dsRNA. An investigation of the amino acid structure of ISG 43 indicates it also contains multiple copies of the two motifs found in the most potent ssRNA 2-5OAS activators. ISG 43 contains, then, multiple elements believed requisite in interacting with RNase L.
RNase L eats its own?
Since ISG 43 contains multiple copies of the motifs that appear to be able to activate 2-5OAS then RNase L deregulation, oddly enough, could potentially either prompt an inappropriate up or down regulation of 2-5OAS. (What an odd and potentially very deleterious situation.
It seems rather short-sighted of the body to put a system regulator inside the system; if the system goes bad then so does the regulator. If this scenario is correct then the RNase L fragmentation could lead to chronic RNase L activation via upregulation of ISG 43.)
A third paper that further broadens RNase L’s activities may have significant implications for CFS patients down the line (LeRoy et. al. 2001). The finding that IFN a induction leads to mitochondrial mRNA degradation indicates that RNase L is found in the mitochondria as well.
Heretofore it has been identified in the cytoplasm where it was associated with nucleic, mitochondrial and other microsomes. The authors suggest the presence of mitochondrial RNase L may explain RNase L’s role in cell apoptosis or cell suicide.
RNase L’s degradation of mitochondrial mRNA could lead to two factors – reduced protein and ATP synthesis – associated with increased mitochondrial membrane permeability. At a certain level of mitochondrial membrane permeability cytochrome c leaks out and activates an enzyme, caspase 3, that fragments most of the cells proteins.
This suggests a fragmented and/or upregulated RNase L in the mitochondria could be responsible for increased apoptosis and/or reduced ATP production in CFS patients. The P100 2-5OAS isoform that appears to be dominant in CFS patients does occur in the mitochondria.
VI. Other possible consequences of RNase L deregulation
Most interestingly, given the subsequent reports of unusual red blood cell (RBC) morphology in CFS, are studies indicating that the 2-5OAS RNase L system may be involved in regulation of RBC formation (Player and Torrence 1998).
Also intriguing, given the possibility of hypercoagulation in CFS patients, are reports that RNase L may also play a role in blood clotting.. A mitogen (growth inducing agent) called platelet derived growth factor (PGDF) that is released during the clotting process stimulates the 2-5OAS and IFN b genes.
(Does this mean that hypercoagulation could enhance RNase L fragmentation? It would be most instructive to measure RNase L fragmentation in CFS patients with and w/out hypercoagulation.)
It appears that the heat shock proteins (HSP’s) released during increased temperature and/or toxin exposure are able to induce 2-5OAS activity as well. (Ethanol is able, in fact, to induce an antiviral state in cells!) Researchers theorize that 2-5OAS and RNase L may function as regulatory agents of PGDF and HSP activity.
They speculate that high PGDF or HSP activity may induce RNase L to destroy PGDF or HSP mRNA in order to lower stress levels. (Could a dysregulated RNase L enzyme in CFS lead to a chronic cellular stress response because of its inability to degrade the mRNA that activates the stress response?
High levels of cellular stress are considered (in CFS ABA) to play a role in CFS pathology. This suggests another potential project – correlating levels of mRNA associated with the stress response in CFS patients to RNase L fragmentation.)
Autoimmunity and RNase L
Only the 2-5A dimer was found in the blood of autoimmune rats (!). Upon administration of poly (1) poly (C) (?) (in the absence of IFN production) only 2-5A dimers were produced in the cells of autoimmune Type I diabetes patients. Longer 2-5a products required IFN administration.
(This indicates that 2-5OAS is not dependent upon IFN for activation. Could 2-5A dimer production be a natural consequence of a blocked IFN signal? Is the IFN signal necessary for 2-5A trimer production? Perhaps RA in the absence of IFN produces only dimers?)
An intriguing but unproven interaction with energy production involves RNase L’s ability to ‘adenylate’ (and thus inactivate) NAD+, an important element in cellular metabolism. Researchers have yet to find an evidence, however, that 2-5OAS plays a role in cellular metabolism.
Bisbal, C., Silhol, M., Laubenthal, H., Kaluza, T., Carnac, G., Milligan, L., LeRoy, F. and T. Salehzada. 2000. The 2-5A Oligoadenylate/RNase L/RNase L Inhibitor Pathway Regulates Both MyoD mRNA Stability and Muscle Cell Differentiation. Molecular and Cellular Biology 20 (14) 4959-4969.
Hartmann, R., Norby, P., Martensen, P., Jorgensen, P., James, M., Jacobsen, C., Moestrup, S., Clemens, M. J. and J. Justesen. 1998. Activation of 2’-5’ Oligoadenylate Synthetase by Single-standed and Double-stranded RNA Aptamers. Journal of Biological Chemistry 273 (6), 3236-3246.
Marie, I, Svab, J., Robert, N., Galabru, J., and A. G. Hovanessian. 1990. Differential Expression and Distinct Structure of 69- and 100-kDa Forms of 2-5A Synthetase in Human Cells Treated with Interferon. The Journal of Biological Chemistry 265 (30), 18601-18607.
Percario, Z., Giandomenico, V., Florucci, G., Chaintore, M., Vannucchi, S., Hiscott, J., Affabris, E. and G. Romero. 1999. Retinoid acid is able to induce interferon regulatory factor I in squamous carcinoma cells via a STAT-1 independent signaling pathway. Cell Growth and Differentiation 10 263-270.
Rebouillat, D., Hovnanian, A., Marie, I., and A. G. Hovanessian. 1999. The 100-kDa 2’,5’-Oligoadenylate Synthetase Catalyzing Preferentially the Synthesis of Dimeric pppA2’p5’A Molecules is Composed of Three Homologous Domains. The Journal of Biological Chemistry 274 (3) 1557-1565.
Rebouillat, D., Hovanian, A., David, G., Hovanessian, A. G. and B. R. G. Williams. 2000 Characterization of the Gene Encoding the 100-kDa Form of the Human 2’,5’ Oligoadenylate Synthetase. Genomics 70, 232-240.
Stark, G. R., Keck, I. M., Williams, B. R., Silverman, R. H. and R. D. Schreiber. 1998. How cells respond to interferons, Annual Review of Biochemistry 67, 227-64.
Xiao-Ling, L, Blackford, J. A. and B. Hassel. 1998 RNase L Mediates the Antiviral Effect of Interferon through a Selective Reduction in Viral RNA during Encephalomyocarditis Virus Infection. Journal of Virology 2752-2759.
Xiao-Ling, L., Blackford, J. A., Judge, C. S., Liu, M., Xiao, W., Kalvakolanu, D. V. and B. A. Hassel. 2000 RNase L Dependent Destabilization of Interferon-induced mRNA’s. Journal of Biological Chemistry 275 (12), 8880-8888.