Passing on a link: The Virological Synapse

Virogical Synapses and HIV Transmission

A new article at MIT's Technology Review highlights a recent paper about the way HIV is transmitted between CD4 T cells. Using a GFP-expressing virus, the authors ( show direct cell-to-cell transmission of HIV via formation of what's known as the virological synapse. Check it out at http://www.technologyreview.com/blog/editors/23244/

Mini Introduction to the topic:

Viruses (as well as bacteria) can infect cells in a variety of ways. Normally, we think of the route of entry as being mediated by cell surface receptors. The virus attaches to the cell through this method and is taken into the cell, where it begins to replicate itself. Later, the cell bursts, spilling thousands of viral progeny into the host; alternatively, the virus can continually bud off from the plasma membrane. (The virus can either do this right away or form a latent infection, where viral DNA is present but little viral protein is made. In this method, the virus rides along with the cell until conditions are right to start producing viruses again.)

The key point here is that at some point in their lifetime, viruses are thought to go extracellular; viral particles have to spread out and infect cells. And when they're out in the open like that, the host immune system can see them better; more arms are available to fight the virus. For example, a virus can be neutralized by antibody or a lipid envelope can be degraded by complement (to name only a few things). If a virus wasn't exposed to this arm at all, the immune system is at a disadvantage.

Relatively recent work has proposed the existence of a virological synapse which effectively transmits virus between cells without progeny having to ever enter the extracellular space. Essentially, the virological synapse is an immune synapse (in this instance, without the priming stage). MHC II, as well as adhesion molecules, are upregulated upon T cell activation. The adhesion molecules attach to other T cells and MHC II interacts with TCR, collectively forming a tight seal between T cells. Usually, this is good for T cell priming, in addition to probably playing a pertinent role in cytotoxic T cell ability.

However, CD4 T cells are a main target for human immunodefiency virus (HIV). Instead of a protective immunological synapse forming, infected CD4 T cells interacting tightly with uninfected CD4 T cells can pass on the virus through the immunological/virological synapse. This might happen a variety of ways (i.e. trogocytosis, active infection, etc) but the thing to keep in mind is, again, that this exchange is thought to keep out antibody and other effector molecules. Incidentally, if the immunological/virological synapse is presenting cognate antigen, it will help to activate (or keep activated) the uninfected CD4 T cell, which provides an excellent environment for the virus to replicate in.

Not having read the paper (sorry to so uninformed, more pressing papers to read), I don't know whether the authors prove this beyond a doubt. (For example, notice how long it takes for the uninfected CD4 T cell to become infected. Usually an infection is quicker- maybe there was virus in the supernatant that was the cause of the late GFP explosion in that CD4 T cell.) But regardless, it certainly strengthens the argument for a virological synpase and is another avenue scientists have to explore if we are ever going to effectively combat this virus.

It’d also be interesting if they took a look at FAS/FASL on these cells. If they are going to interact that strongly, what happens when activation-induced cell death (AICD) starts to really get going? (…Or does HIV inhibit this, I just don’t know.)

The actual paper can be found at:

http://www.ncbi.nlm.nih.gov/pubmed/19325119

Prions are What?

Normally, I think of prions as a bad thing. Anyone hear of bovine spongiform encephalopathy(BSE)? BSE, or Mad Cow, was featured prominently in the news a few years ago and probably made a vegetarian out of a few people. That's because BSE is caused by prions, which are hard to detect, hard to prevent (prions are highly resistant to heat and other sterilizing devices), and ultimately result in death. So what are prions?

A prion consists of a single infective protein. Unlike other pathogens, nucleic acid isn't necessary to make future generations. Instead, a prion is capable of making more of itself by converting a particular host protein (called prion protein- PrP) into a prion. In essence, the prion protein functions to change the conformation of PrP such that it now takes on the function of a prion; these converted prions in turn recruit others. Most of the pathogenicity occurs because these prions also form aggregates in cells of the brain, ultimately resulting in death.

Prions, as we know them, are dangerous, but because PrP can be so easily converted into a prion, many groups have begun asking, "What function does normal PrP have inside a cell?” Or more generally, what are these potential prions doing there?

Eleven prions have been identified in yeast so far and are a good place to start for finding out what these things do. Interestingly, a new paper published in Nature Cell Biology (http://www.nature.com/ncb/) by Dr. Susan W. Liebman's group at the University of Illinois Chicago has discovered another method of non-Mendelian inheritance via prion function (note: this is my take, the paper is more scientifically conservative and accurate in that its title is The yeast global transcriptional co-repressor protein Cyc8 can propagate as a prion).

It's a good paper and there's a insightful comment on it in the issue, so I won't go into much detail here, but I did want to list a few key points.

Cyc8 (called OCT in its prion form) is a global transcriptional co-repressor protein that, among other things, regulates yeast growth on lactate. This protein suppresses other proteins important for growth on lactate plates. However, by transiently overexpressing the putative prion domain of Cyc8 (and thus increasing the chance of misfolding for the naturally occurring Cyc8) , the authors were able to show increased colonies growing on lactate plates versus controls. The authors were then able to do further experiments using this Lac+ phenotype

One of the most important things they did was an experiment to show that these Lac+ colonies were not spontaneous mutations resulting in revertants. (i.e. mutated nonfunctional Cyc8). Overexpression of functional Cyc8 did not restore the Lac+ phenotype.

However, as an aside, there was always the possibility of a gain-of-function mutation. Perhaps the gene required for lactate growth lost its ability to be repressed by Cyc8. I feel that simple sequencing of the most important genes involved in the Lac+ pathway could have been useful, but the authors’ main concern was elsewhere and they focused their remaining experiments in proving that OCT has all the characteristics of a prion.

This was the most fundamental, and convincing, aspect of the paper. The OCT+ mutation was dominant in an OCT- cross (not characteristic of Cyc8 loss mutations), cytoplasm of OCT+ cells could transfer the Lac+ growth phenotype, and fluorescence of OCT fused to YFP showed aggregates within the cells.

This paper continues to extend the knowledge of prion function. They ain't just fer disease anymore! Instead, proteins with prion-like characteristics might serve to regulate cell function in a rapid and inherited manner, without the need to continuously monitor transcriptional and translational levels of mRNA. (In fact, this paper cites other work with the prion Swi, a protein involved in chromatin remodeling.)

While this paper involves propagation of OCT and didn't look at any possible regulation of Cyc8 to OCT conversion, the fact that there are 11 putative prions in yeast is suggestive. Instead of relying on mutation and unchangeable consequences, perhaps some prion-like proteins can be converted between the two states. (The paper might have a citation on this; I'll need to check it out further)

Again, if you're really interested, check out the paper:


Patel B.K., Gavin-Smyth J., and S.W. Liebman. The yeast global transcriptional co-repressor protein Cyc8 can propagate as a prion. Nature Cell Biology. March 11, 2009. DOI: 10.1038/ncb1843

http://www.ncbi.nlm.nih.gov/pubmed/19219034

Host Adaptation of Influenza, Part II

Other Influenza Proteins


Neuraminidasen (NA)


Neuraminidase is the other protein major glycoprotein on the surface of the virion. Unlike hemagglutinin, it's not involved in binding to the surface of the cell. Instead, this protein will actually cleave sialic acid from other carboyhydrate linkages. This feature is especially important in virus release; as new virons are made, hemagglutinin binds to sialic acid on the infected cells. The virus wouldn't be able to escape and spread without neuraminidase cutting sialic acid and freeing the virus.


The same function helps with virus entry. Think of hemagglutinin sticking too strongly; it wouldn't be able to change its conformation and fuse with the plasma membrane. It’s true, but I haven't come across too many papers on this subject.


In terms of host range, I'm not aware of any drastic changes that need to be made in NA in order for an avian strain to infect humans or vice versa. But keep in mind that the neuraminidase activity of any virus will evolve to be directly proportional to HA's ability to bind sialic acid (i.e. a weaker binding hemagglutinin has a weak neuraminidase)

Matrix (MA) and Nonstructural (NS) Proteins

Ok, these proteins are important. The matrix proteins do two things: M2 provides an ion channel. When the virus invades the cell, the low pH of the endosome is essentially transported into the virus (H+) resulting in release of ribonucleotide particles (the genome is released). M1 is definitely matrix and forms the outer capsid-like structure of the virus.

At least one of the nonstructural proteins slows down the host immune response by interfering with the interferon pathway (It stops RIG-I recognition of nucleic acid) The other frameshift, which results in NS1, does something else, like assisting in viral replication, but it's pretty vague for me.

So, while these proteins play major roles in virus biology, I'm not aware of any specific changes that need to occur in order for influenza to adapt from human to avian and vice versa. However, these changes do occur, they are just not studied extensively.

Next: Polymerase proteins and nucleoprotein.

Host range adaptation of Influenza, Part I

Introduction

Influenza is a major problem worldwide. Even without a pandemic year (the 1918 "Spanish" flu killed between 20-50million people), influenza kills 36,000 people a year in the US and results in 200,000 hospitalizations. Why, when people get vaccinated every year, is influenza such a problem?

One reason is that the virus is not the same from year to year. This virus can evade host immunity in two main ways. One mechanism, called antigenic drift, is when the virus gradually changes its proteins such that it is no longer recognized as effectively by the immune system. Although this results in some viral escape, there is usually enough cross-reactivity in the population so that there are no large-scale pandemics. Think of it as panting a car. It's now harder to find in a parking lot, but it still has the basic shape and look of the original enough so that you can tell the difference.

Antigenic drift can cause many problems for the host, but antigenic shift is much scarier. Instead of a gradual change in the virus, a completely new one is created. But to explain what antigenic shift is, it's necessary to delve a little deeper into the biology of the actual virus


Influenza virus biology.

Influenza is an enveloped, single-stranded negative sense RNA virus that mainly infects epithelial cells of the respiratory tract. This virus has ~11 proteins (new ones continue to be discovered) contained in 8 separate genomic pieces. What are those proteins, and what are they doing? What role do they play in host adaptation?


Hemagglutinin (HA)


The hemagglutinin (HA) glycoprotein on the virus surface binds sialic acid, ubiquitous on many cell types besides epithelial cells. (Though for the most part, influenza does not go systemic and remains a respiratory virus)

Hemagglutinin's ability to bind sialic acid is a major factor of host range. For example, the H5N1 avian virus binds alpha 2'3 linked sialic acid. Although it has been known to infect humans, one of the major reasons it hasn't been spread by human-to-human contact (except in rare, close contact familial cases) is that our lungs mostly contain alpha 2'6 linked sialic acid in the upper respiratory tract. Thus, most human-adapted influenza viruses are 2'6 linked while those of avian origin are 2'3 linked and these viruses rarely cross species barriers.

Worry about a pandemic H5N1 outbreak is centered on the virus mutating its binding preference to alpha 2'6 so that it can readily infect humans. Although this switch is the major concern, binding preference is not the only determinant of host range. In fact, much of the genome is involved.

Detour, so what actually is antigenic drift?

What happens if a virus that is human-adapted except for its HA protein suddenly acquires this H5? This isn't just science fiction. The nature of influenza's segmented genome means that these separate pieces can be mixed and matched and in the Darwinian struggle for evolution, those that have the advantage will survive (This mixing and matching between different viruses is what's known as antigenic shift). So, even though the H5 is still avian-adapted, are the remaining proteins enough to let the virus infect humans? Will this spur the adaptation of a human (alpha 2'6 sialic acid binding) H5 subtype HA?


This is the concern that keeps scientists up at night, and why it's important to continue to study influenza as much as possible. Of course, it's unlikely that a fully human adapted virus will acquire the H5 HA, or that an avian H5 that becomes human adapted will be as pathogenic as the original. Why? Well this goes back to the role the other viral proteins play in infection. So what are those proteins?

Next topic: Getting back to the rest of influenza's proteins