Chapter One: Interferon and the 2-5A Pathway by Lionel astide, Edith Demettre, Camille Martinand-Mari, and Bernard Lebleu
Chapter Two: Ribonuclease L: Overview of a Multifaceted Protein. by Patrick Englebienne, C. Vincent Herst, Simon Roelens, Ann D’Haese, Karim El Bakkouri, Karen De Smet, Marc Fremont, Lionel Bastide, Edith Demettre and Bernard Lebleu.
Overview of the interferon/2-5A PKR system
This first section provides gives some background information not available in the text and an overview of the viral inhibition process. To go straight to the synopsis click here)
It has long been evident that individuals suffering from one viral disease rarely contract another simultaneously. Cells in laboratory cultures infected with one virus can only with difficulty be infected with another. One reason for this is a group of soluble factors – the interferon’s – that cells produce in response to infection by a virus or other agents.
Two IFN induction pathways are believed to exist; one specific to viruses, the other stimulated by other agents. While IFN’s have mostly been studied with reference to viruses, they can be activated by a wide variety of other agents including some bacteria, rickettsiae, protozoa, organic compounds, antibiotics, antigens, etc.
The two components of the immune system we are mainly concerned with, RNase L and PKR, are products of the IFN response; they are activated, however, only if double stranded (ds) RNA is present.
It is important to note that the interferon response is the cells first attempt to stop a viral invasion. IFN’s are a fundamental component of what is called the ‘innate response’ of the immune system. If the innate response fails to check the virus then the ‘adaptive response’ driven mainly by T and B cells will eventually kick in. Most pathogens are, however, stopped by the innate response.
As a virus enters a cell and begins to replicate, metabolic changes in the cell cause genes coding for IFN to produce messenger RNA (mRNA) which travels to the ribosomes and codes for IFN production. The IFN’s bind to receptors inside the cell which send a signal down a complex pathway that stimulates a wide variety of what are called interferon stimulated genes (ISG’s) in the nucleus. (They also travel outside the cell and alert other cells.)
These genes induce a host of cellular activities that are designed to stop the spread of the virus. Two of the most significant by-products of IFN stimulation are the stimulation of the protein kinase R (PKR) and RNase L systems.
As mentioned earlier both RNase L and PKR are activated by IFN’s when dsRNA is present. Viral RNA in contrast to human RNA, occasionally forms loops or double strands that alert the cell to the presence of the virus. (The cell is then alerted to the presence of a virus by at least two substances; metabolites produced by the virus initiate the IFN response; dsRNA produced by the virus results in IFN induced PKR and 2-5OAS activation.)
Once activated PKR prevents the first part of mRNA from being translated at the ribosomes. When this happens the ribosomes are unable to read the mRNA and viral replication is stopped.
RNase L degrades ribonucleic acids or RNA. Once oligoadenylate synthetase (2-5OAS) is activated by type I IFN’s, it creates something called 2-5A out of bits of ATP it has chopped up. When 2-5A is present RNase L binds with it, becomes activated and begins to degrade RNA.
Because both PKR and RNase L degrade both viral and cellular RNA, protein synthesis in a cell is virtually stopped when they are upregulated. The IFN’s do not operate with surgical precision; they are blunt instruments that stop infections by disrupting major cellular functions such as protein synthesis and cell growth. If a cell cannot be cleansed of a virus it is readily sacrificed.
So here we have a two pronged attack on viruses. PKR stops the translation of their mRNA and RNase L chops up untranslated viral mRNA. Together they provide a very potent attack, an attack that is so potent, in fact, that once interferon activates these pathways, the cell is said to have acquired an ‘antiviral state’. This book is about what happens when this attack goes awry.)
The text
First discovered in the late 1950’s, interferon’s (IFN’s) protect against a wide variety of viruses in mammalian cells. They do this by activating genes (interferon stimulated genes or ISG’s) that initiate a diverse array of biological responses.
Two types of interferons are produced. Almost all cells, upon contact with a wide array of pathogens (viruses, bacteria or mycoplasma) or cytokines can produce Type I IFN’s (IFN a/b). Only immune cells (activated T-cells) on the other hand, can produce type II IFN’s (IFN-y). More than 100 ISG (interferon stimulated genes) have been identified.
IFN’s induce the production of a dsRNA dependent protein kinase (PKR) which signals for various activities to occur, one of which inhibits viral mRNA translation. Some of the substances PKR activates are also be activated in response to stress such as amino acid starvation or heme deprivation.
(This essentially means that cells act somewhat similarly to some kinds of stress and viral attack. We will see in Chapter 7 that amino acid depletion is commonly found in CFS. Increased cellular stress is believed to be a predisposing factor in CFS (Chapter eight)). Nor surprisingly, many viruses have engineered ways to counteract PKR’s antiviral activity.
The 2-5A/RNase L pathway was discovered in the mid 1970’s by researchers attempting to understand the protein inhibition they observed in IFN treated cells. The production of dsRNA by viruses stimulates type I IFN (IFN’s A, B) production. IFN’s essentially prompt cells to respond; they interact with the receptors on cellular membranes to trigger a wide variety of effects, some of which inhibit viral replication and cell growth.
Once a type I IFN binds to its receptor on a membrane, it initiates a signaling cascade that induces the transcription of genes in the nucleus that code for, among other things, 2-5A synthetase (2-5OAS).
Type I and II IFN’s induce the transcription of several 2-5OAS synthetase isozymes. (Transcription involves constructing an RNA molecule in the nucleus that travels into the protein producing ‘factories’ (the ribosomes) in the cytoplasm. Proteins are extremely complex substances that do the work of the cells.
Isozymes are enzymes that are chemically distinct but perform the same function). The isozymes are quite different; they are synthesized from different genes, they are located in different parts of the cell, they are activated by different types of dsRNA, and they synthesize different lengths of 2-5A oligomers.
2-5OAS is produced in response to IFN a/b but is activated either by ss or ds RNA. (The requirement for dsRNA and perhaps for certain types of ssRNA provides a check on the system. Since RNase L is a blunt instrument, it is hopefully not wielded unless absolutely necessary.)
Three forms of 2-5OAS exist; small (p40), medium (p69), and large (p100). The different isoforms are triggered by different kinds of ss or ds RNA and they produce different lengths of 2-5A. Upon activation the 2-5A’s ‘oligomerize’ (form repeating units.) 2-5OAS cuts ATP up into pieces and then binds them together again to form oligomers.
Oligomers are simply a number of units of the same compound; i.e. a dimer contains two units; a tetramer four. Longer strands of dsRNA induce the p69 isoform to primarily produce 2-5A trimers (2-5A tripled). Lower sized strands of dsRNA induce the p100 isoform to mostly create 2-5A dimers (2-5A doubled).
Unless RNase L is bound by its inhibitor it will, upon binding to the appropriate 2-5A oligomer, become activated and then bind to another RNase L molecule (it ‘homodimerizes’), at which point it is able to degrade the viral (and other) mRNA in the cell and stop the viral attack (as well as shut down protein synthesis.)
As mentioned earlier RNase L needs to bind with 2-5A oligomers to become activated. In order to better understand the binding and activation sites of RNase L a three-dimensional model was constructed. The model indicated that when RNase L binds with a 2-5A trimer an internal clamp is released causing a dramatic change in the shape of the protein and uncovering RNase L’s catalytic sites.
Only after this occurs is RNase L able to dimerize and become activated. The 2-5Adimer is able to bind with RNase L, but according to the model produced by the authors, it is too short to release RNase L’s internal clamp. This will turn out to be an important bit of information.
By binding with RNase L the 2-5 dimersinhibit RNase L activation by the 2-5A trimers. In a sense they keep it frozen in its monomeric state, something that we shall see has some very negative consequences.
An inhibitor for RNase L (RLI) has recently been identified. Because RNase L inhibition is dependent upon the ratio of RNase L and RLI in the cell, any RNase L production quickly outstrips RLI’s capacity to bind and inactivate it. This system ensures that RNase L is active only when it is upregulated. Some viruses are able to induce RLI production and inhibit RNase L activation.
RNase L in chronic fatigue syndrome
An up regulation of RNase L activity in CFS was first reported in the early 1990’s. Further studies indicated that the upregulation was due to the presence of low molecular weight (LMW) variants of RNase L in peripheral blood mononuclear cells (PBMC).
PBMC’s are the focus of almost all the tests in this book. Mononuclear cells include monocytes/macrophages and T and B lymphocytes. These LMW variants (42, 37-kDa versus the native 83-kDa RNase L) were found to be capable of binding to the 2-5A oligomers synthesized by 2-5OAS and upon binding they became active just as the native RNase L does.
After much research it was determined (recently) that the 37-kDa RNase L is produced by the cleavage of the native 83-kDa RNase L. The 37-kDa protein has recently been verified as a biomarker for CFS.
Putting the 83-kDa RNase L into the PBMC’s of controls and CFS patients indicated that while the native RNase L remained whole in the controls, most of it was broken into the 37-kDa fragment within 30 minutes in CFS patients.
In order to identify the enzyme responsible for fragmenting it, RNase L was incubated with several enzymes and 2-5OAS. Three enzymes (m-calpain, human leukocyte elastase (HLE)) and cathepsin G) were capable of cleaving RNase L into fragments identical to those found in CFS patients.
M-calpain is a cysteine protease that is particularly active during apoptosis.Cysteine proteases are able to cleave proteins at their cysteine sites. Because CFS patients exhibit increased apoptotic activity, it is not surprising that enhanced calpain activity has been found in CFS patients. None of the other apoptotic proteases tested fragmented RNase L.
Elastase and cathepsin G are serine proteases found in the azurophilic granules of polymorphonuclear leukocytes or granulocytes. Granulocytes are white blood cells with granules that contain enzymes, antibiotics, etc. that destroy and degrade foreign particles that the cells have phagocytosed or ingested.
Lurking underneath the surface tissues they man the front lines of the bodies defenses.These proteins are also, interestingly enough, involved in two processes that appear to upregulated in CFS; host defense and inflammation.
Very quickly we appear to be very near the source of the problem. The question was what causes the fragmentation of RNase L? At least three enzymes that participate in immune defense are at least capable of doing that. But why would these enzymes all of sudden start to tear apart RNase L?
Are they more abundant than before? Or is there something the matter with RNase L? The authors will leave us hanging here a bit until the end of this section while they backtrack a bit and clear up some questions about this fragment.
In order to determine if some cells have higher levels of the RNase L fragments, the PBMC cells were separated according to their CD classification into T-cells (CD3) or monocytes (CD14). The LMW RNase L fragments were primarily found in monocytes.
Because monocytes give rise to antigen presenting cells (APC’s), which are responsible for alerting T-cells that an intracellular invader is present, a disregulation in them could result in the distorted TH1/TH2 response seen in CFS.
APC’s digest the invaders, then display bits of them on their surface for T and B cells to determine if an immune response should be launched. If the APC’s are dysfunctional then the ThI arm of the immune system – which responds to intracellular invaders such as viruses could be down regulated. Down regulation of the ThI arm and the consequent upregulation of the Th2 arm of the immune system is what is seen in CFS.
The monomeric and dimeric forms of RNase L were then examined to determine if either were more susceptible to cleavage. It was believed that the folding that accompanied dimerization might hide the cleavage points and render the dimer more resistant to cleavage.
Interestingly enough, only the monomeric RNase L was fragmented in CFS patients. This is a big step! The mystery is at least partially solved; RNase does not dimerize and is left in an unprotected state and upregulated apoptotic and/or inflammatory enzymes chop it up.
Conclusions and prospects
The probable involvement of apoptotic and inflammatory proteases (m-calpain, HLE, cathepsin-G) in RNase L cleavage illustrates some of the effects that increased levels of cellular stress and inflammation may have in CFS. The RNase L abnormalities found in monocytes provides a mechanism whereby the incapacity of T-cells to activate natural killer cells via nitric oxide mediation might be explained.
In this case the T helper cells simply do not receive a signal for activation because the dysfunctional monocytes do not properly evolve into APC’s. These observations suggest that therapeutic approaches using protease inhibitors and regulators of calcium homeostasis may be fruitful in CFS.
The finding that the monomeric form of RNase L is more susceptible to cleavage than the homodimeric form, indicates that 2-5 oligoadenylate synthetase (2-5OAS) plays a major role in the syndrome. If improper activation of 2-5OAS resulted in a lack of RNase L homodimerization, then RNase L would remain in its latent monomeric form, a form that is susceptible to cleavage by inflammatory and apoptotic proteases.
These observations suggest that immmunomodulators able to (properly) activate the 2-5OAS system such as bile salts or retinoic acid(vitamin A) derivatives may be helpful.
This suggests that ‘improper’ 2-5OAS activation is very near the heart of this disorder and that RNase L is more an unwitting victim than an instigator. What would cause ‘improper’ 2-5OAS activation? Remember that 2-5A trimers activate RNase L; 2-5A dimers only bind to it.
It appears that short sections of RNA induce 2-5OAS to produce 2-5A dimers that leave the RNase L enzyme in an unprotected state just as the cell is being flooded with apoptotic and/or inflammatory enzymes. Where do these short sections of RNA come from? The authors give us little clue but do note that short sections of dsRNA that are often associated with ‘subcellular fractions’ including mitochondria, nucleus and microsomes.
Are these small bits of RNA coming from disrupted mitochondrial activity? Are they fragments from increased apoptotic activity? There is evidence of mitochondrial dysfunction in CFS. Whatever is going on, it seems just as we appear to be getting to the heart of the problem, another layer appears.
What is causing CFS? It appears to be these little strands of RNA. But where do they come from, and what is causing them to appear???
The 40 and 37-kDa fragments lack the regions (the protein kinase like and ankyrin repeats) that are needed not only for dimerization but also for regulation. The loss of the ankyrin ‘clamp’ means that RNase L’s internal regulating mechanism is gone and that the 37-kDa fragment is, unless it is inhibited by RLI, always active.