Since we're all in mourning this week for the loss of Arthur C. Clarke, I thought I'd put in a good word for one of my other favorite sci-fi authors: Connie Willis. There are many reasons I love Willis' work, not least of which is her ability to mix funky, cutting-edge science (time travel, chaos theory, neuroscience) with literary and historical references (Lincoln's Dreams), witty dialogue (Bellwether), colorful characters, elements of farce and slapstick (To Say Nothing of the Dog), and somehow also manage to break your heart (Passages) -- usually all in the same book.
Her big breakthrough novel was the Hugo-Award-winning The Doomsday Book, in which a young idealistic medieval historian in futuristic Oxford, England, travels back in time to the Middle Ages and finds herself trapped in a small village just as the Black Plague reaches the area. Spoiler alert! Everybody dies. Duh. Every sci-fi fan knows about the "Grandfather paradox," and savvy writers like Willis tend to adhere to the trope that time travelers can't change the course of history. And it's a matter of historical record that most villages saw 80%-100% mortality rates. Most of the European population was wiped out during the Plague Years. The fact that people panicked and fled only helped spread the disease faster.
Panic reactions and the spread of global disease was actually the subject of a session at the recent APS March meeting in New Orleans by researchers from the Max Planck Institute for Dynamics and Self-Organization. Recognizing that human beings will "change their dispersal characteristics" in response to local infections -- i.e., panic and flee to avoid becoming infected themselves. Not surprisingly (to me, anyway), they found that "the individual rationale of avoiding an epidemic wave... actually facilitates epidemic spread" -- at least in one of their models. A more fully developed dynamical model apparently showed the same effect, but also "an increased extinction probability of the epidemic as a function of increasing dispersal response." In other words, sure, the hypothetical "epidemic" spread faster as individuals dispersed broadly to avoid infection, but this also increased the likelihood that the epidemic would die out -- hopefully before everybody fell victim to the disease.
Epidemiological models of the dynamics of epidemics are tough nuts to crack precisely because there are so many unpredictable variables: not just irrational (or even rational) human behavior, but the behavior of whatever is the cause of the disease itself, whether it be a bacteria or a virus (or a virus that targets bacteria), whether it's vector-borne, or otherwise disseminated. Viruses can mutate into many different strains, for starters, and in some of these multistrain cases -- dengue fever, the Ebola virus -- the antibodies produced by the human immune system to ward off the primary infection can actually increase one's vulnerability to acquiring a secondary infection with a different strain.
So naturally, scientists would like to have a clearer picture of human mobility patterns -- preferably one that can be universally applied. There's good news on that front! Researchers at Northeastern University and Notre Dame University have analyzed cell phone usage to demonstrate that human mobility can be described by the same universal pattern, regardless of what our individual travel habits may be. It's been tough to model this sort of thing in the past, because scientists just haven't had the tools required to monitor the movements of a large number of people in real time. The ubiquitousness of the cell phone in modern society has changed all that: now it's possible to track people's movements by following the cell phone trail.
These patterns could be useful in urban planning, traffic forecasting, and of course, the spread of diseases and viruses. The latter would include the possibility of cell phone software viruses, in which malevolent code could be transmitted either via text messaging or through Bluetooth connections between devices. Each transmission pathway would require different countermeasures, according to Pu Wang of Northeastern (one of the speakers at a March Meeting session), since text message viruses, like email, would spread through social networks rather than the physical location of the actual cell phones, whereas a Bluetooth-specific virus would spread among cell phones in close proximity.
Some primer notes: The technical definition of a virus is "a sub-microscopic infectious agent that is unable to grow or reproduce outside a host cell." (I'm sure there's a more complicated definition than that, because otherwise, most reproductive processes would fall into this category.) They cause everything from the common cold, to rabies, yellow fever, smallpox, and the flu, and they don't respond to antibiotics, although some antiviral drugs have been successfully developed to treat a few critical types of infection, and vaccines can also prevent infection in the first place (polio has almost been eradicated, although rare cases still pop up now and then).
There are various historical records dating back to the 10th century indicating that folks understood something about the infectious nature of smallpox and measles, but I was intrigued to learn that a physician named Ibn Khatima discovered in the 14th century that the bubonic plague (Black Death) and other infectious diseases were caused by micro-organisms entering the human body. And of course, by the late 18th century, Edward Jenner had figured out that a milkmaid who had caught cowpox proved immune to smallpox (a related virus), and used it to develop the first smallpox vaccine. (Smallpox has been eradicated, per the pronouncement of the World Health Organization in 1979.)
There are countless strains of any one virus, and different types of viruses beyond that, which makes one wish there existed something like a periodic table of viruses just to keep everything straight. But we did learn the difference between simple viruses like the cowpea chlorotic mottle virus (CCMV), and bacterial viruses (those that attack bacteria): CCMV has a more easily compressed single-stranded RNA center, while bacterial viruses contain double-stranded DNA. That's noteworthy because a DNA strand typically measures some 17 microns in length, whereas a viral protein capsid is 60 nanometers -- 1000 times smaller than the length of the DNA. So the DNA has to fold up and squeeze into a very tiny space. That means the bacterial virus's capsid must be strong enough to contain the immense internal pressure of roughly 50 atmospheres. (Yowza!) Another difference is that unlike CCMV, bacterial viruses bind to receptor points on bacteria, and then the DNA comes shooting out, just like the cork of a just-opened bottle of champagne.
Bogdan Dragnea of Indiana University in Bloomington is one of a growing number of researchers interested in the physics of viral protein cages, or capsids. They're intriguing because they self-assemble to contain nucleic acid (RNA or DNA), and Dragnea is among those toying with constructing artificial viruses containing nanoparticles, droplets, drugs, or other elements besides nucleic acids. Among other things, he's embedded negatively charged gold nanoparticles inside viral capsids by exploiting their attraction to the positively charged proteins lining the capsids. This mimics the interaction between "anionic genetic contents" (RNA and DNA) and those same positively charged proteins in real viruses.
He's also encased fluorescent quantum dots of cadmium selenide crystals in a shell of zinc sulfide, which he then used to track how long it took for a particular virus to travel across a cell membrane (because the dots glow for extended periods). Dragnea's artificial viruses aren't "infectious" the way a natural virus might be, but he does believe it may one day be possible to replace vaccines -- made from actual viruses -- with completely artificial versions, thereby avoiding the risk of causing an outbreak of the very disease one is trying to prevent.
It's all part and parcel of an exciting new field called physical virology, which is getting its very first Gordon Conference next February (Dragnea is one of the organizers). Dragnea's early work focused on the Brome mosaic virus. The March Meeting session featured new research using CCMV which specifically infects the cowpea plant, more commonly known as the black-eyed pea. Just imagine the poor little cowpea, innocently hanging out in a field, photosynthesizing to its heart's content, only to fall victim to a tiny invader that causes yellow spots to form on the cowpea's pretty leaves. It's unsightly, and and ultimately fatal to the plant. Next time there's a shortage of black-eyed peas, blame CCMV.
CCMV is a favorite choice for physical virologists because it's so easy to replicate and use -- it's like the hydrogen atom of viruses. That's because it's so evolutionarily designed for self-assembly that, according to UCLA's Charles Knobler, you can literally break them into their constituent parts, put them into a test tube with some sort of solution, shake them up a bit, and get fresh virus out. Under the right conditions, its purified coat protein mixed together in vitro with its genetic material will spontaneously assemble into infectious particles.
Wow. That's like the T-1000 in Terminator 2: Judgment Day literally reassembling itself after being frozen solid with liquid nitrogen, blown into tiny bits, then having the bits melt back into liquid form and "find" each other. The cowpea population could be doomed. (Jen-Luc Piquant isn't that fond of black-eyed peas anyway, but the plant leaves are so pretty.)
Knobler is investigating what determines the size of a virus. The answer appears to be some combination of polymer length, molecular weight, and capsid size, in an intricate self-assembling interplay. It is possible to manipulate the length of protein building blocks in the CCMV and the size of the capsid in such a way as to use virus proteins not only to make nonbiological particles that contain foreign molecules, but also that conform to a specific intended structure.
Building on the spontaneous re-assembly that occurs when purified viral RNA and capsid proteins are mixed in a solution at just the right pH and ionic strength, Knobler tweaked the parameters a bit, changing the solution conditions to cause the formation of empty capsids, multishell structures, tubes and sheets. He also examined the self-assembly process with different molecular weights, and noted that two distinct capsid sizes seemed to be preferred: 22 nanometers for lower molecular weights, and 27 nanometers at higher molecular weights. (I'm admittedly a bit fuzzy on the significance of that, but it's nice to have things narrowed down so specifically.)
Adam Zlotnick and his colleagues at the University of Oklahoma Health Sciences Center can manipulate the CCMV coat protein in such a way as to redirect its self-assembly to produce tubular structures. They didn't rely simply on static models based solely on the structure of the CCMV virus, which Zlotnick believes can lead to false predictions, namely, that CCMV capsids are extremely stable, and the assembly relies critically on hexamer (6 linked molecules) formation. That turns out not to be the case. "Experimentally, we have found that capsids are based on a network of extremely weak pairwise interactions and that pentamer (five linked molecules) formation is the critical step in assembly kinetics," he said. Far from being static structures, viruses are "very dynamic molecule machines based on weak energy interactions."
Furthermore, because those interactions are weak, it is possible to interrupt the assembly process to generate, say, tubular structures in addition to spheres -- or even keep the virus from forming completely in the first place. (Low pH seems to be one way to do the trick.) His kinetic models more closely match those experimental observations. And the fact that it's possible to disrupt and manipulate the process bodes well for the bottom-up production of manmade nanostructures via self-assembly for any number of applications, including the development of new antiviral drugs.
Per Zlotnick: "Knowing the structure of a virus gives us a snapshot, but add the knowledge about the kinetic process of assembly, and we have a much more complete picture" of how a virus works. And the more we know, the more we can manipulate and control. Perhaps one day, we could even defeat the common cold. But let's start with the saving the humble cowpea.
I have to say I didn't think the doomsday book was that great a novel. The characterization of some of the heroes and villains were weak, and another failure of the book (particularly as the book was written in the early 1990s), no one in the future apparently uses a cell phone or email to communicate. Now ok some of the Cambridge colleges are a bit behind the times, but they are not that far behind. So from that perspective it was a weak plot point to use the inability of the hero professor to communicate with his colleagues. Saying that I did think the description of medieval England was pretty interesting.
Posted by: Paul Guinnessy | March 20, 2008 at 09:31 AM
A minor point concerning your black death mortality numbers. "Most of the European population" was not wiped out by the Black Death. Far from it. The number was closer to 1/3 (at least according to most historians and economists nowadays). Great post otherwise.
Posted by: kleo | March 20, 2008 at 11:39 AM
I don't suppose some crazy (say Kim Jong IL or Mahmūd Ahmadinejād) might get ahold of some russian SS-25 missles, and use the info in the above study to gauge maximun effectiveness of spread of nuclear contamination?
Posted by: eingram | March 20, 2008 at 12:55 PM
Doggone it I didn't want to but I guess I need to well, answer the first comment. I was fortunate enough to have an Uncle who received "Popular Mechanics" back in the 30's. I've got old issues in my storage room (anybody wishing to confirm may delve my address from the author, and thus she and I might fill our respective fuel tanks) and none of them foresaw cell phones and "enhanced 911" or GPS.
I seem to recall a history lesson where this guy named Ghengis catapulted bodies over the walls of Constantinople, apparently he had some idea of disease transmission.
Boccacio. He described people "getting the hell outta Dodge" and by the way gave us (I think) our first case of child whatever, Rustico and -I can't recall her name.
Anyway, Arthur C. did it for me (the book) "The Foundation Trilogy" and the jokes that fostered the first inter-racial consensus jokes by Bradbury are included. Ya'll know, the "Martians Discovered" stuff. I guess ya'll can tell I'm from...
JK
Posted by: JK | March 20, 2008 at 11:45 PM
Love, love, love Connie Willis!
Posted by: Coturnix | March 21, 2008 at 10:48 AM
I, too, am a Connie Willis fan of many years. And I agree with each of your assessments of her various books. My two favourites are "The Doomsday Book', and 'To say nothing of the dog' For the stories, the plausibility (strained only a little at times), and the characters that I come to like.
My basic comment though has to do with the spread of SARS and a few other 'plagues'. My take on the total is, "Had the SARS pathogen been just a little slower to infect, say an additional incubation time of 24 hours, the spread would have been unstoppable."
There is something critical in the time of infectiousness and onset of symptoms, too rapid and deadly(ebola) and hosts are killed before they can relay or hosts are terrified enough to do something about it.
SARS nearly fulfilled all of our worst nightmares and is a near perfect case for us to study.
BUT,,,, the all time winner of human nature being woven into an outcome is the Canadian physician who after studying the epidemic in Hong Kong, did not quarantine himself an flew straight home to friends and family. Air travelers, hospital workers, friends, family, strangers,,, all paid for this bit of hubris. Had SARS been just a bit slower to exhibit, it would have spelled catastrophe.
I like shoes too.
Dean
Posted by: Dean Unick | April 03, 2008 at 10:08 PM