Perhaps the relative lack of popular science writing in this area is because most of the action goes on outside a normal scale – here, we are in a world bounded on one end by quantum mechanics (at the detailed level of enzyme action, say), on the other by classical physics (at the scale of cells), and whose character is necessarily defined by evolution and ecology. This is the world of the microbial cell, the oldest form of life on the planet.
Pathogenic bacteria represent but a tiny fraction of overall microbial diversity – and are a special subclass in that the completion of their life cycle requires a host of some kind – in the case of anthrax, a mammalian host. This is in contrast to their closest ecological relations, the harmless and even beneficial microbes that inhabit the digestive tracts of all animals, assisting with food digestion and nutrient uptake (while also meeting their own nutritional needs).
The basic functional elements of any bacterial cell can be divided into the genome, the proteome, and the overall cellular structure – cell walls, cell membranes, and various internal structures of varying complexity. The genome is the information database, the proteome consists of the tools and materials produced using information stored in the genome, and the result of the proteome in action is the overall cellular structure and activity. Technically, information about the genome is easier to come by then information about the proteome, due to the new ease of DNA sequencing. This is why the study of the proteome has lagged well behind the study of the genome.
Microbes (like us) are constantly using their cellular system to sense their local environment, and when that environment changes, signals are sent back to the genome (via the proteome) which lead to new rounds of gene expression, an altered proteome, and hence an altered cellular activity and/or structure. This is clearly seen in the lifecycle of Bacillus anthracis – as it transitions from a spore to a vegetative cell in a host and back to a spore again.
Plants, animals, fungi, bacteria – all operate on this fundamental level. Obviously, complex multicellular creatures have even more complicated genome-proteome relationships than do bacteria – but for pathogenic bacteria, which have to survive within multicellular creatures, they must adapt to that complexity in order to be successful as pathogens. As we shall see, this is also true for Bacillus anthracis. Hence, some limited discussion of the human immune system is also required to understand anthracis.
In this (not-for-profit, informational-only blog) I’ve relied on the excellent work done by the web site www.theprotein.lounge.com, in particular on graphics they produced.
Pictured here is the initial stage of inhalational anthrax infection – airborne spores are first inhaled into the lungs. The infectious dose is greatly dependent on the particle size – large clumps of spores are far less likely to make it down into the alveoli of the lungs, and are more likely to be trapped in the mucus lining the upper respiratory tract. On the other hand, individual isolated spores fall into the 1-5 micron size range and can easily penetrate the lungs. The spore material used in the 10/9 Daschle-Leahy anthrax letters (and possibly the 9/18 letters as well) was chemically treated in such a manner that the spores were prevented from clumping together – which greatly increased the infectious potential. This is no simple trick, but rather a highly sophisticated approach drawing on modern knowledge of nanotechnology.
One across the lung barrier, spores are collected by macrophages (white blood cells) in the lung lymph nodes. The bacterial spores germinate within these macrophages and begin to grow.
Like any other bacteria, most of Bacillus anthracis’ genetic information is encoded on a single circular chromosome (x million base pairs?). However, anthracis keeps the major virulence factors (proteins that play critical roles in infection) on two smaller plasmids, which are (like the chromosome) circular loops of double-stranded DNA. Every time anthracis replicates itself by cellular division, all three genetic elements are copied, and thus each daughter cell gets a full complement of genetic information – and this is now taking place within a macrophage in a lung lymph node.
After using the macrophage as their first stage incubator, the newly formed cells burst out and begin to spread through the blood stream to other lymph nodes and tissues. At this point, an immunized mammal is already mounting a full-scale immune response to deal with the invader, allowing it to wipe out the bacteria before they can move on to their next, deadly stage: toxic protein synthesis. In other words, it’s a war against time – and in the non-immunized mammal, that race is typically won by the bacteria – and that’s what happened to the five victims of the anthrax letter attacks.
From the blood stream the bacteria reach many major lymph nodes, and once there, they begin ramping up protein synthesis, using the genome as the informational template. The plasmid, pX01, contains genes coding for three proteins, named lethal factor, edema factor, and protective antigen (lef, cya, pag), as well as a regulatory gene/factor, AtxA. It is 174,000 DNA base pairs in length (and contains many other gene sequences besides these three). The second plasmid, pX02, contains genes coding for enzymes that produce a poly-D-glutamic acid capsule. This crosslinked-chemical capsule helps protect the bacteria from the scavenging action of certain white blood cells. We can now picture the invader, safely enclosed in a protective membrane coat (not a spore coat, though), while actively pumping deadly toxins into the bloodstream and lymph:
Pictured here is the binding of the first factor, protective antigen, to its cellular receptor, which of course plays some other role in human biology – in essence, as with viral infections, the receptor is hijacked by the toxin. It is only called “protective” because this is apparently the main protein recognized by an anthrax-vaccinated human immune system, and that in turn allows the immune system to launch a pre-emptive assault before the bacteria has a chance to establish itself in large numbers.
Once the protective antigen binds to the receptor, it is cleaved in two by furin enzymes that are naturally present, creating an active binding site on the receptor-bound residue. Since the receptors are mobile in the lipid membrane, drifting around, they eventually encounter one another, creating the heptamer (seven identical units bound in a specific arrangement) seen on the right.
The toxic proteins produced from these gene sequence operate together as a function unit in order to cross the protective cellular barrier. The seven-membered structure first binds the lethal factor and the edema factor, and then translocates them across the cell membrane into the interior of the host cell via the process of endocytosis – again, a normal cellular process that Bacillus anthracis has hijacked for its own purpose. The package has now been delivered to the cellular interior.
Pictured here is the tiny lipid bubble that contains the heptamer complex. This complex now releases the toxic proteins that it carried into the intracellular mileau.
These proteins have two major effects – one binds to ATP, the major intracellular energy carrier, hydrolyzing it all the way to cyclic AMP, which acts as a signal to the white blood cell host to stop ingesting bacteria. The other destroys intracellular proteins that play key roles in maintaining the cell cycle, leading to cellular suicide – again, a normal biological process that B. anthracis has manipulated to devastating effect. In short, these toxins disrupt essential cellular processes, triggering cell death and the release of cellular contents to the extracellular space (lymph/blood), where Bacillus anthracis cells – growing as filaments – absorb the nutrients, and continue their explosive replication and toxin production.
Eventually, the host animal dies and the nutrient supply is exhausted – at which point an entire new set of B. anthracis genes are activated, leading each vegetative cell to create an internal spore structure within which the genetic material is securely protected from damage. The entire corrupt mass of fluids, dead cells and spores then leaks out of the dead animal into the soil, where the spores will lie dormant, sometimes for decades, until another host comes along – typically a grass-eating ruminant of some kind or other.
Now that we’ve considered the physiology of anthrax (and a bit of the ecology), we can consider how one might use the genetics of B. anthracis in a forensic examination.
The first thing to note about the genetics of Bacillus anthracis is the remarkable genetic homogeneity among all known isolates – but what is an isolate? That’s actually a non-trivial question, particularly in relation to the new microbial forensics techniques. It’s been known for a long time now that the artificial media cultures that microbiologists use to collect their isolates actually select, evolutionarily speaking, from among the possible isolates.
Bacillus anthracis is not as finicky as many other microorganisms, however, and if anthracis bacteria are streaked across a solid medium (sheep blood agar), isolated colonies will grow up on the surface – as round little colonies all descended clonally from a single parent cell. However, different strains of anthracis are very hard to distinguish from one another. The most common bacterial genotyping method, analysis of the ribosomal backbone coding sequences (16S and 23S rDNA) are barely able to distinguish Bacillus anthracis from its relative, Bacillus cereus. Hence, a great deal of effort has gone into developing better genetic methods for distinguishing different strains.
Here, we'll cover the efforts made in the 1990s, which should give one enough background to understand the subject of the next post, the whole genome methodology applied later.
Initial efforts began with variable number tandem repeats – short series of sequences repeated in a series within a gene (Jackson et al. 1997) – and a different method known as amplified restriction length polymorphism (Keim et al. 1997). That’s vaguely similar to the human genetic fingerprinting methods used in modern forensics. (If only Sherlock Holmes could see us now... we’ve gone from identifying bloodstains by chemical means, to extracting the human fingerprint from those bloodstains.) However, this was still a very low-resolution method for distinguishing among strains, relative to today. For example, the Sterne, Ames and Vollum strains of anthrax were indistinguishable via the VNTR methods used even though Sterne lacks the pX02 plasmid. AFLP methods were more effective, even though B. anthracis isolate similarities were very high (97%).
These methods did allow researchers to place B. anthracis in relation to its closest relatives, Bacillus cereus, Bacillus mycoides, and Bacillus thuringiensis (some GMO crops controversially express thuringiensis insect toxins in their tissues – which could plausibly result in negative ecological and health effects, as some studies indicate). Hence, as of about 1999, the AFLP method was the state-of-the-art for distinguishing B. anthracis isolates (Keim et al. 1999).
Soon, however, the advent of whole-genome sequencing would lead to new and better possibilities. With B. anthracis, the pX01 plasmid (181,654 bp) from the Sterne strain was the first large-scale sequencing effort in this area. Suddenly, huge amounts of data began to roll in from these methods – and organizing and making sense of this data became the central challenge. This was also critical data for determining all the variable regions of the B. anthracis genome.
Keep in mind, that back in pre-whole genome era, the goal was mainly to trace naturally occurring outbreaks of anthrax back to their source regions in order to better understand the epidemiology of the disease – how it spread from host to host, in other words. If you consider the biology of anthrax, as described above, then the following might make sense:
“The high lethality of anthrax, with a concurrent massive and efficient multiplication over a few days, and the absence of chronic anthrax infections provide little opportunity for host immunological selection for altered types. When B. anthracis is not growing in a host organism, it is thought to be quiescent as spores and, hence, perhaps evolving at a reduced rate.” – Keim et al. 1997Here, another key point emerges – when B. anthracis is grown on sheep-blood agar or similar media, it doesn’t really need its set of lethal proteins in order to survive – that work has already been done for it by the microbiologist. Over time or in large cultures, therefore, the rate of mutation may increase and those mutations may be tolerated to a greater extent then in the natural infectious cycle.
The first reported application of these genetic methods to biowarfare-related issues focused on the infamous Sverdlovsk anthrax outbreak in the Soviet Union. (Jackson et al. 1998) Preserved tissues from the Sverdlovsk victims were subjected to DNA extraction and polymerase chain reaction amplification. Their most remarkable finding was that each victim had been infected with multiple strains of anthrax. That would in turn indicate that the Sverdlovsk bioweapons facility had mixed isolates up during the production process, or was working with seed cultures that contained a wide variety of isolates.
This fact is worth keeping in mind – thirty years after Sverdlovsk, the issue of impure seed cultures has become a very critical one in the analysis of the 2001 anthrax letter attacks. The Sverdlovsk result was later verified by a detailed PCR study based on the pag gene, encoding the protective antigen protein. (Price et al. 1999)
This business of multiple strains infecting the same host is critically important in diseases such as swine flu, but it seems very rare with B. anthracis under natural conditions. The reason it matters is that multiple strains can swap their genetic material within a host (this applies to viruses like swine flu as well as bacterial pathogens) – which, incidentally, is one of the main epidemiological dangers involved with factory farming pigs, cows, chickens, etc.
What can we say in conclusion? By the late 1990s genetic methods for typing anthrax strains were gradually becoming more sophisticated, but they were seriously hampered by the low genetic diversity of the gene regions examined. The next major technological breakthrough - whole genome sequencing - would change all that.
The question then becomes, was that approach correctly applied in the anthrax investigation, or not? To answer that will require another long post, so let's leave it at that for now.
However, there is one other issue that some might be unaware of - namely, that anthrax was the first disease ever conclusively proven to arise from the action of a microbe. That was the famous discovery of Robert Koch, who was the first to cultivate the microbe in isolation, show that it infected mice and guinea pigs and sheep, and demonstrate that sporulation allowed it to survive extremes of heat and cold. Paul de Kruif described this in his book, "Microbe Hunters", published in 1926 - here is a short excerpt of some interest on sporulation:
"To himself Koch muttered guttural curses. "Other microbes have doubtless gotten into my hanging-drop," he grumbled, but when he looked very carefully he saw that wasn't true, but when he looked very carefully he saw that wasn't true, for the shiny little beads were inside the threads - the bacilli that made up the threads have turned into these beads! He dried this hanging-drop, and put it away carefully, for a month or so, and then as luck would have it, looked at it once more through his lens. The strange strings of beads were still there, shining as brightly as ever. Then an idea for an experiment got hold of him - he took a drop of pure fresh watery fluid from the eye of an ox. He placed it on the dried-up smear with its months-old bacilli that had turned into beads. His head swam with confused surprise as he looked, and watched the beads grow back into the ordinary bacilli, and then into long thin threads once more..."
This is of interest for several reasons - but I'd like to hear the FBI and Battelle microbiologists explain how the spores could end up coated with silica by any "natural" process - as the silica would have to be taken up into the microbial cell and incorporated into the spore as it was forming inside the vegetative cell - and there's zero evidence for any such process.
I also wonder what Koch would have to say about those who would use these spores as a weapon against other human beings...
