Turned up to eleven: Fair and Balanced

Wednesday, May 03, 2006

Microbiology and Pop Culture

I was all set to let this Animalcules go by (again) without writing anything, when I caught the last 10 minutes of House. For those who haven't seen the show, It is about a jerk of a doctor who just happens to be a world class diagnostician (he seems to have an emphasis on infectious disease, but maybe that's just my bias!). Anyway, the episode was a two-parter, ending this evening. The agent of the mystery disease turned out to be a water borne protozoan Naegleria fowleri, which under fairly rare circumstances can cause severe CNS disease (in the show, meningitis and encephalitis. Now, I generally don't do the protozoan thing, but the reason this one caught my eye was because of a throwaway line in the show, something about how coinfection with Legionella slowed the infection down.

Why did this spur my interest. Well, in the show, this is a throwaway that every great diagnostician should know, but I'd bet my next month's salary (or something someone might want...) that the vast majority of doctors, even ID specialists, are not aware that Legionella can hide out in amoeba. That article is from I and I in 1985, so clearly the idea has been around, but it doesn't seem to may naive eye to have agained a great deal of traction. However, it is very clear now that intracellular living in environmental amoeba is a part of several pathogenic microbial lifestyles. It even made it into this lowly blog not too long ago.

Since I went over the basic ideas (especially wrt evolution of virulence) before, I won't rehash it. However, it's nice to see a fairly cutting edge facet of the convergence of general microbiology and infectious disease entering the pop culture. (I won't bore you with my pedantic complaints about their 30 second identification of a pathogenic protozoan from biopsy...)

Wednesday, April 05, 2006

Known Unknowns
Our beloved Secretary of Defense, the wise and all knowing Don Rumsfeld, once famously discussed "known knowns, known unknowns, and unknown unknowns" with respect to the know well known phantom WMD in Iraq. His comments were widely mocked at the time, but this is in fact an interesting area of thought, at least for us scientist types. Not so much with respect to phantom WMD or ill-advised foreign wars of choice, but in much more esoteric arenas.

There has been a bit more than a century of microbiological research done so far. A great deal of work occured in the pre-genetic era, and the vast majority of microbiology time is pre-molecular biology. However, the march of progress and the exponential growth of the science community (as marked in people working in the field as well as the literature) leads to a spiralling upward of information. A new entrant into the field of microbiology (such as a new grad student or young assistant professor...) might well be concerned that there are no interesting problems to solve.

So, to illustrate the depths of our ignorance, I will point you to a recent paper on lowly Escherichia coli K-12. It isn't just that E. coli is by far the most studied organism on the planet, the K-12 lab strain is essentially the progenitor of all the lab strains used for molecular biology research (if you ever cloned a gene into E. coli as part of a classroom or research project, it was almost certainly a derivative of K-12). In this particular paper, the investigators identified some genes that are crucial to biofilm formation, a recently very well studied phenotype.

Now E. coli has approximately 4 million base pairs of sequence, which has been completely identified on a base by base basis (to see a list of all the microbial genome projects, see this NCBI page). The K-12 sequence was completed in 1997, and 16 total E. coli genomes are either complete or in the assembly stage of sequencing. The total number of identified protein coding genes is 4237. If you click on the link you will go to a table of all the predicted and known proteins in this extraordinarily well studied organism. If you go down that table, you will find, very quickly, "hypothetical protein b0005" and "hypothetical protein b0006". If you go through the entire list, you will find quite literally thousands of "hypothetical" and "putative" proteins. If it is called "putative something" that means that based on sequence homology someone has predicted that sequence should encode some particular functional protein. If it is labeled "hypothetical protein" that means quite literally that no one has any idea what it does.

Get to work!!!

Wednesday, March 22, 2006

Challenges in Microbiology
Here's a little thought experiment for you. You've set yourself up a nice little system for examining the genome of an environmental isolate for genes involved in some important phenotype. Now, there are lots of ways to do such a thing, but lets keep it simple. You use a mutagenic technique (and there are many) to introduce random changes in the genome. You then use your powers of observation to identify colonies (or growth in some other form) that is unusual. When you find a wierd colony, you isolate it for further study.

Is this science?

It is a real question. Many rigorous science theorists (Kuhn, I'm looking at you) would say "no". After all, at this point, you don't have a "falsifiable hypothesis", unless you would suppose "Mutagenesis causes changes in phenotypes" is a falsifiable hypothesis, but that's a bit trivial. However, I'd say "bollocks!" to that (or at least, if I was English I'd say that).

This is the start of science. It's not a full experiment, but rather a way to build hypotheses. And, if you want my opinion, this is the most underappreciated part of the scientific method. After all, hypotheses don't fall out of the sky like apples. We don't just wake up in the middle of the night thinking "maybe Scar-2 is encodes a phospholipase C like protein" or something like that. A huge part of the process is recognizing something interesting when you see it, and then following it up with efforts to gather supporting evidence.

Why am I belaboring this? Well, I think it's worth thinking about generally, first of all, but also because it gets at something I've been thinking about a lot lately, which is "how do you know when you've found something?"

This is particularly important when studying previously uncharacterized bacteria. If you go out to a patch of uncultivated land, and dig in the dirt, you will find countless bacteria. Simple technques can allow you to culture many of them (not percentage-wise, but numerically, thousands of unique strains); alternatively, you can use molecular methods to catalogue all the different types of organisms in that sample. However, when we look in the literature, we find people don't explore all this diversity. Rather, they pick things of well understood importance, and probe them deeper. So, instead of finding a brand new never cultured previously organism from the soil, they find a new nuance to the pathogenesis of E. coli gastroenteritis. One reason for this is obviously that it is much easier to figure out what's important with E. coli (or some other well known pathogen) because the parameters have already been hashed out historically. Another reason is funding (there's much more money in studying disease).

Here's an underappreciated reason. It's tough to study previously unstudied bugs. How do you know that you've identified a mutant, and not a variable phenotype. If it has a complex lifestyle, how many person-years can you devote to figuring it out? How do you attack the basic questions of function without clear guideposts. How many things can you assume are the same as the ones known in E. coli (depends!). Will you be able to use genetic tools? How hard are you willing to try to find out? Finally, how will you know when to publish it? (that's a big one!) And how will you persuade the editors to accept it? (another big one!)

Comments? Thoughts? It's not quite the usual animalcules fare, but it explains a lot about why we know so much about so few organisms, and so little about the vast, vast majority.

Tuesday, March 07, 2006

Connecting environmental microbiology to the "important stuff"

I was perusing the wonderful blogs hosted by Seed Magazine (all of 'em are great, but Frinktank is clearly a rising star!) when I saw a new microbiology paper(subscription req'd) noted by Afarensis. The paper is about MRSA (Methicillin Resistant S. aureus) being shown to persist inside amoeba. I argued, hopefully justifiably, that you can artificially create a situation like this in many systems, and we shouldn't all run for the hills yet. Unfortunately, I haven't had a chance to read the paper (never stopped me before!), but I have to suspect that they would acknowledge my simple, naive criticism, and either deal with it experimentally, or simply admit that it needs to be studied further. Either is a very reasonable tack. I suspect someone smarter, more capable, and with better institutional access will do the thorough once over of this paper, so I'll tackle a larger question. Why would bacteria want to be able to do this, and why would it be important to us?

Because I am just not up to a long exposition, I'll tease you a bit with this, and hopefully get some comments. We know that many pathogens are highly adapted to life in the animal, plant or human host. We also know that one of the major obstacles to persistence and infection is the innate immune system, consisting of humoral responses (complement) and cellular responses (neutrophils and macrophages, cells that eat bacteria). Note, for the immunologists, that there are two analogous branches of adaptive immunity, and based on evolutionary analysis, it is thought that the innate system developed much earlier, and the adaptive system piggybacks on it (I'm simplifying here!!).

There are a lot of complex things going on in pathogenesis, some of which I've alluded to before. A major part of the process is resisting the immune response, both innate and adaptive. We'll leave the adaptive system for now, and focus on the innate system. The big danger to any invading bug is that a phagocytic cell (like I mentioned above) will eat it. So, many mechanisms for evading this response have evolved. A big one is the ability to persist in the endocytotic vesicle (depending on the cell, the endosome or the phagosome). The phagocytic cell has a vesicle called a lysosome that fuses with the phagosome to make, unsurprisingly, a phagolysosome. This fused vesicle contains acid and lytic enzymes to break the bacterium down. Defenses against this fusion, resistance to these enzymes, and acid tolerance are all great ways for bacteria to avoid this fate. In pathogenic systems, some classic examples of these mechanisms are found in Mycobacterium tuberculosis, Listeria monocytogenes, and Legionella pneumophila.

So what? Well, it turns out that each one of these bacteria have close relatives that live in natural environments inside amoeba and other phagocytic protozoa. A good review (that I just found and should read!) by Jorge Galan gives info about this and other commonalities between environmental persistence and infection. So, here's the $64,000 question (and I pose it honestly as well as rhetorically); Did bacteria already well adapted to life in soil, battling for nutrients and fighting off eukaryotic predators (protozoans, fungi) evolve all of their mechanisms in that environment, and then adapt them to colonizing and infecting eukaryotes? It's a reasonable hypothesis, and the rough time frames of the rise of various types of eukaryotes and bacterial divergences are reasonably congruent with it (note, I'm no expert on the time scales of branching events in microbial phylogeny; any knowledgeable parties should please comment).

So, here I'll leave you with the big challenge. How do we gather evidence to support or invalidate this hypothesis? Can you plan out an experimental scheme to understand this system?

Obligatory anti-ID snark; no fair just saying "God did it"!
Second anti-ID snark; there is simply no rational way to understand how different pathogenic bacteria attack many different hosts in fundamentally similar ways without acknowledging the role of evolution by variation and natural selection)

Wednesday, February 22, 2006

Biofilms, Quorum Quenching, and Antibiotic Resistance
It's been a rough couple of weeks (in a good way), but here's another installment of our saga. As we learned last week bacteria often form Biofilms, aggregates on surfaces that can cause lots of trouble. I didn't get too much into the details of how these aggregations form, but I a couple of points; 1) Biofilms play major roles in lots of things we aren't to happy about, and 2) Biofilm formation is controlled by chemical signals that bacteria produce. I mentioned bacteria forming biofilms as part of pathogenesis. Another bit of bad new is that Biofilms are more resistant to antibiotics than their planktonic (free floating) counterparts.

The ways in which bacteria cause us problems are manifold. For example, when someone goes into the hospital, they often come out with an infection they didn't have before. One of the very common ways this occurs is when bacteria grow in a biofilm on one of the various invasive catheters used commonly in medicine. These include urinary catheters (UTIs are very common, although usually not fatal) and lines into the bloodstream (much more dangerous). This problem is exacerbated by the well known fact that bacteria that colonize hospital workers are much more likely to be resistant to antibiotics (an aside; when I was a grad student, we did this experiment with the med students I was teaching. The ones who were currently working in the hospital had bacteria on their skin with higher levels of resistance to common antibiotics). This is even more problematic when we recall what I mentioned one paragraph ago, that biofilm bacteria are even more resistant.

So this leaves us with the following scenario; patient comes to hospital, gets an infection from a catheter, gets treated with antibiotic, then another, then another, till they eventually get vancomycin and get better. So now all their associated bacteria have been exposed to a very strong antibiotic. At a population level, this can lead to the emergence of vancomycin resistance in bacteria (like MRSA) that are resistant to virtually everything else.

Can this get worse? Oh, yes! Bacteria in biofilms (like the ones commonly causing these infections) are not terribly picky about their neighbors. Single species biofilms are cultivated in laboratories, multiple species biofilms are found in nature (such as dental plaque). So when a multiple species biofilm occurs, DNA transfer can occur. And does it? Hoo boy, does it ever!

So, to recap; bacteria in biofilms cause diseases, especially in hospitals, where antibiotic resistance is a bigger problem. Oh yeah, and biofilm bacteria are even more resistant.

So how to deal with this?
Well, in the last few years (about 5, give or take), a lot of investigators (including me, briefly) got interested in the idea of Quorum Quenching. The basic idea goes like this. A QS circuit involves a signal that is detected by a protein on the surface or in the cytoplasm (a receptor). The receptor either directly influences gene expression (a regulatory protein, or response regulator) or signals to a downstream protein that does this (Bonus points to people who recognize the similarity to hormone systems). So interfering with this would presumably prevent expression of genes that are regulated by quorum sensing (virulence factors and biofilm formation factors, to name a few). Two basic approaches have been attempted, with some success in vitro and in planta. Competitive inhibition with molecules (furanones) that bind to the receptors (perhaps you have heard of the analogous notion of estrogen-like molecules contaminating water and interfering with hormonal cycles of animals and humans), and enzymatic degradation of the signal (I worked on this type of thing). The basic notion is that if we can prevent biofilm formation, we can reduce the levels of biofilm related disease, and therefore 1) make people's lives better, and 2) reduce the use of antibiotics. So, how's it going? Well, I'm out of the business, so I can't really answer that. It's promising, and it will be very likely to work in industrial and agricultural settings, but in medicine, we'll see.

Tuesday, February 07, 2006

Quorum Sensing, Biofilms, Bacteria as multi-cellular aggregates

When you were taught microbiology (as every last one of you should have been), you were probably shown a microbial cell structure like this one from a standard textbook. The parts are all there; DNA, membrane, cell wall, flagella, ribosomes, etc. If your micro class was pretty good, you learned about what all the parts do, then you went on to learn how bacteria get energy and carbon, as well as their other nutrition. You were probably told, in perhaps not so many words, that bacteria are just really simple versions of our cells, stripped down and optimized for their niche in the world. If your class was good (again!), you might even have learned about some of those niches besides us. However, if it was typical, you learned about diseases for most of the rest of the course.

Now, learning about infectious disease is very important. As a grad student in med micro, I made a lot of effort to learn the structures and functions of all sorts of toxins, including my favorites, the Superantigens from Staphylococcus aureus. Staphylococcus aureus makes a bewildering array of virulence factors, including things on the surface that protect it from the immune system, and lots of toxins (see above review paper). The order in which these various products are made, and which ones are being made, makes a huge difference in what disease S. aureus causes, whether it is a boil or toxic shock syndrome, or more recently, necrotizing fasciitis (By the way, I've found that as interesting as I think this is, gross pictures are worth at least a thousand words, so click through, if you have a strong stomach!). When I learned about this system, I found out a peculiar thing about bacteria. A million or a billion bacterial cells is not just the same cell multiplied a million or a billion times over. Bacterial cells change and develop over time, especially in the aggregate. In Staph (including the very, very relevant Methicillin Resistant S. aureus), there is a complex system for controlling cell behavior based on the population size and density...especially the density. The system (largely figured out by a very big group of researchers centering around Richard Novick at NYU) involves a peptide ( a short string of amino acids) secreted by the bacteria as they grow. Through a fairly complex mechanism, this peptide is bound by a protein on the surface of the cell, which activates a regulatory RNA molecule (called rnaIII). This regulatory RNA molecule (a relatively rare thing in microbial genetics) is a global regulator of gene expression. What does this mean? It means that rather than a single gene encoding a single protein that does something, this RNA molecule turns on and off a whole bunch of genes, and and that massive change in what genes are being expressed is a major factor in what disease results (the other factors are what toxin genes are present, and environmental factors such as pH, oxygen level, and CO2).

The astute reader will wonder what any of this has to do with the title of this post, which is about bacteria as multi-cellular aggregates. Well, see, the Staph stuff is the tease, to get you interested. Now here comes the science. The words I used in that paragraph; population, density, global regulator, are all words used to describe a system called a quorum sensing system. This means that the bacteria act one way when they are sparse, and as their density grows, they act another way. The canonical system was identified as a symbiosis, rather than a pathogenesis, between the marine bacterium Vibrio fischeri and the squid Euprymna scolopes. It turns out, after a great deal of fine work by too many people to name, that this system revolves around the production of a chemical called an acyl-homoserine lactone (don't worry about it!). Like the peptide I mentioned before, this is secreted by the bacteria as they grow, and it binds to a receptor (this time in the cell, because these can pass through membranes (maybe)), triggering responses in genes regulated by the system. In the V. fischeri / E. scolopes interaction, this results in bioluminescence (Finding Nemo, anyone?). In the squid symbiosis, bioluminescence is thought to be a defense against predators, because the light emitted matches moonlight in the visible spectrum, thereby preventing the squid from casting a shadow as seen from below by predators.

The astute reader who has made it this far will wonder "what does this have to do with multi-cellular aggregates, or biofilms?" After all, I wrote above about bacteria inside a squid, or in a person causing disease. But they are inside something, not on something, right? Well, your wrong, now pay attention! (kidding!) In fact, it turns out that microbiologists have pretty well been studying the wrong stuff for about the last 100 years, in some ways (we aren't going to take back the Nobel Prizes, or anything!) Most bacteria, in most environments, most of the time, live attached to something, and aggregated with other bacteria. It is a rare thing for a bacterium outside a lab to be floating in a fluid containing free nutrients. This attached lifestyle is called a biofilm. In the last 15-20 years, it has become a major field of study. I even wrote a grant about it! (yeah, I'm pretty proud of myself, why do you ask?) So what does this biofilm thing have to do with S. aureus disease? A lot, actually! It turns out that many diseases in eukaryotes, caused by bacteria, are caused by them when they form a film on a biotic surface. The biofilm is often a critical stage in the infection. Many hospital infections are entirely based on the formation of biofilms on medical devices or tissues.

It turns out (surprise, surprise) that many bacteria have complex systems for controlling the formation of these biofilms, with some significant commonalities. In many gram-negative bacteria (like Pseudomonas aeruginosa, bacteria that can cause wound infections and lung infections in cystic fibrosis) the signals that control expression of virulence genes and biofilm formation genes are one and the same, acyl-homoserine lactones. In gram-positive bacteria (like the aforementioned S. aureus) peptide signals are used. These signals in gram positives are important in transfer of DNA between bacteria, uptake of DNA from the environment, biofilm formation, and virulence factor production (at least).

I better stop here, but it is a huge, complex story. Just understanding one of these systems can take a lifetime of well-funded research (I hope!). But that last bit there is particularly interesting. The role of horizontal or lateral gene transfer in bacteria; how is this connected to quorum sensing, biofilm formation, and virulence (as well as other things)? That sounds like a topic for next time!

New Blog Post? Is he back?

New and improved blogger. Will it get me to post? Well, not exactly, but since Tara Smith (of the U. of Iowa and Aetiology) has started a new blog carnival, (tentatively called Animalcules, but maybe to be renamed) I've decided to participate. I don't promise anything more than to try to do it more than once every other week (er....), but here we go!

Sunday, April 25, 2004

Archaea and disease
Archaea (or archaebacteria) are a large and difficult to study group of microorganisms that are similar in appearance to bacteria, but biochemically very different. Based on the pioneering work of Carl Woese of the Univ. of Illinois, Champaign-Urbana, a method for studying the phylogenetic distribution of bacterial, eukaryotic, and archaeal species was determined. The method relies on a so-called "molecular clock" using the sequence of the ribosomal RNA.

So what is ribosomal RNA? The ribosome is the cell's protein factory, that converts mRNA sequences into amino acid sequences (the central dogma of molecular biology is DNA to RNA to protein) that fold into 3 dimensional proteins that do the cell's business. Clearly, a cell that doesn't have functional ribosomes isn't going to last long (or at all!). So the ribosomes, made of ribonucleic acid molecules (rRNA or ribosomal RNA) and proteins arranged in a complex 3D structure, have two very important features wrt "molecular clock" applications. 1) they are absolutely critical, so any mutation that alters their function is big trouble, and 2) they have been around a very long time, so the evolutionary tree can be rooted at the very beginning of life. Bacteria have them, eukaryotes have them, archaea have them. Some parts of the ribosome cannot be changed at all without destroying the function. These are called conserved regions. Other parts can mutate relatively freely, and are called variable regions. These variable regions form the basis for the molecular clock.

Based on this method, microorganisms (and macroorganisms!) can be identified and placed into a phylogenetic tree. Now, even in the time that Woese was first looking at this, it was clear that microbes were literally everywhere, doing literally everything. So as time passed, and the technology grew cheaper and faster, many microbes were placed in the tree. It became clear that there was much more phylogenetic (read, evolutionary) distance between microbes than there is between eukaryotes. It is clear, for example, that the last common ancestor between people and nematodes is much more recent than between Escherichia coli and Methanococcus. In fact, a whole deep evolutionary branch of microbes was identified, called the Archaea.

Archaea look similar to bacteria, but are fundamentally physiologically different. They use ether rather than ester linked lipids in their cell membranes, they used different compounds in their cell walls, and they use fundamentally distinct mechanisms for acquisition of nutrients from the environment. (here is a nice overview of Archaea)

Most Archaea fall into two categories, extremophiles andmethanogens. Extremophiles are microbes found in extreme enviroments, such as sulfur springs and deep sea vents, or in the Dead Sea. They not only tolerate but thrive on very high salt contents, high temperatures, or high pressures. These organisms have been isolated by their lifestyle for eons, and this explains, to some extent, the evolutionary distance. The other physiological type of Archaea, however, methanogens, are not nearly so geographically isolated. In fact, there are methanogenic Archaea in your body right now! The are in seawater, sediment, the intestinal tracts of many animals, and can be said, in biological terms, to be "ubiquitous".

This brings up a conundrum for microbiologists. Typically, when a broad category of microbes is known, there exists within that category a pathogen, that is, a species or strain that causes disease. Of course, microbes that haven't come into contact with other living things for billions of years are unlikely to be pathogenic, but those in our guts and soils may find a good evolutionary strategy in pathogenesis. The consensus, I think, has been that we will eventually find an Archaea that causes disease, but because they are difficult to work with (cultivation and genetics of Archaea is significant challenge, it might take a while.

Well, no more! David Relman at Stanford and his associates have shown that abundant Archaea in the oral cavity is well correlated with periodontitis, which can lead to atherosclerosis, stroke, and other health issues. An excerpt from the press release.

Relman and members of his lab embarked on a comprehensive, controlled study of the archaea found in the subgingival crevice - the deep gap between the gums and teeth - where periodontitis begins. They rigorously analyzed samples from 58 patients' mouths taken by their collaborator, Gary Armitage, DDS, at the UC-San Francisco School of Dentistry, and found that more than one-third of the periodontitis patients had archaea in their diseased subgingival spaces, but nowhere else in their mouths. In addition, the relative abundance of archaea correlated with disease severity. Their findings are published in this week's issue of the Proceedings of the National Academy of Sciences.

"Of course we'd ultimately like to say archaea caused disease, but it's a horse-and-cart problem right now because we haven't shown that the archaea come before the disease," Relman said. In the future, he noted, they will collect specimens repeatedly from the same spot in the subgingival pocket in hopes of being able to pinpoint the moment when the archaea start to increase in number and then determine whether that predicts the later development of the disease.

The paper's first author, Paul Lepp, PhD, research associate in microbiology and immunology, explained that while a third of the periodontitis sufferers harbored archaea, many of the others had high levels of bacteria that - like archaea - consume hydrogen. Hydrogen consumption creates a more hospitable environment for bacteria long known to play a role in gum disease.

The group speculates that archaea may not directly cause periodontal disease. Rather, the microbes may indirectly contribute to it by helping other organisms - in this case, gum-damaging bacteria - grow more productively. Lepp said they are now looking for other hydrogen consumers to test their theory.

"In my mind, it's increasingly clear that the disease may be the result of a community disturbance rather than the presence or absence of a particular organism," Relman said.

Relman also sees a potentially broader side to this research. "Maybe we should look a little harder for evidence of archaea as promoting or causing other diseases. We certainly have them in our bodies and we are exposed to them, so the archaea have the opportunity to cause disease if they are capable of doing so. We haven't been looking for them so we wouldn't know."

As the quote says, the Archaea are probably not the single causative agents of the disease, but probably contribute to a polymicrobial etiology, or as Dr. Relman puts it, community disturbance. Later this week I will post some more material on biofilms, microbial communities, ecology, and disease. Cheers.

Thanks for the nod, P.Z.!
Thanks to Paul Myers from U. of Minnesota, Morris, for the nod to my renewed sporadic blogging. As a former denizen of the great state of Minnesota (grad school, UM-TC), it is nice to hear a strong voice for science and reason from that region. If you aren't reading Pharyngula and Panda's thumb on a regular basis, you are truly missing out.