Shell games

Retroviruses, the family of viruses that includes HIV, are notorious for their fast evolution. But while these viruses have been evolving for millions of years, their hosts have been changing in response. Welkin Johnson, associate professor of biology at Boston College, studies one such change in hosts: the evolution of Trim5, a protein found in humans and rhesus monkeys that can sometimes block retrovirus infection. The protein works by recognizing parts of an invading virus’s outer shell, attaching itself to the shell and preventing the virus from reproducing. A report of research conducted in Johnson’s Viruses, Genes, and Evolution Laboratory in Higgins Hall was published in the May 2013 edition of PLOS Pathogens.

Rhesus monkeys have three versions of the Trim5 protein that protect against different retroviruses, with some overlap: One version may recognize, say, viruses A, B, and C, while another recognizes B, C, and D. Johnson and colleagues at Boston College and Harvard are working to identify the specific overlapping targets on the retrovirus shells. To do this, the researchers take advantage of the fact that two of the rhesus Trim5 variants block HIV-1, the most common HIV strain, while a related strain of simian immunodeficiency virus (SIV) isn’t blocked by any of the three Trim5 variants.

Kevin McCarthy, a member of Johnson’s lab and a graduate student at Harvard (where Johnson was an associate professor until early 2012), has been working from genetic codes to systematically swap out small sections of the SIV outer shell for pieces of HIV-1 shell. By studying Trim5 responses to the various resulting chimeras—the new viruses made up of pieces from both—he has identified two segments of the HIV-1 virus that are targeted by Trim5 in the rhesus monkey.

The next step will be to figure out how the Trim5 protein recognizes these parts: What, specifically, about the segments of the HIV-1 shell—their three-dimensional shape, say, or their spatial arrangement as part of the larger shell—triggers Trim5 to engage? Determining that could bring advances in uncovering targets and mechanisms for HIV drugs.

“Trim5 has been adapting to retroviruses for millions of years and is very good at blocking many different retroviral infections,” says Johnson. “That means Trim5 must be exploiting something about the viruses that’s so critical to their nature they cannot easily change it. If we can figure out what it is, that would be an excellent target for a drug.”

Good vibrations

Temperature management is critical for electronic devices, whose components generate heat that can destroy them if it isn’t shed. Diamond is the best-known material for channeling heat away from small and sensitive electronics such as the processors in cellphones and laptops, but real diamond and synthetic diamond are expensive. Physics professor David Broido and colleagues from the Naval Research Laboratory have developed a way to search for alternatives. In a July 2013 paper in Physical Review Letters, the team described their approach, which brought unexpected results.

Broido’s group is primarily interested in non-metals, which carry heat differently than do metals. In metals, electrons transport heat, moving throughout the material and diffusing energy. In non-metal compounds such as diamond and graphite, heat transfer occurs when atoms vibrate, and the resulting “vibrational waves” carry the heat energy “from hotter to colder parts” of the material, says Broido. How freely those waves can move within a material’s atomic structure helps determine the thermal conductivity.

Broido and colleagues developed a new approach to calculating thermal conductivity, which involves complex equations that, says Broido, “require powerful computational algorithms that we developed and vast amounts of computer memory,” which, he adds, was not available a decade ago. The results for materials such as silicon and germanium, whose conductivity is well-established, confirmed that the new methodology has what Broido calls “predictive capability. You tell us the atoms, we will generate a [conductivity] number for you that should be close to what would be measured in the lab.”

Studies of certain materials have revealed unexpectedly high conductivity numbers, most notably boron arsenide (BAs). Although BAs had been considered an inadequate candidate for heat transfer applications, based in part on its high atomic weight, Broido’s calculation has shown it to have thermal conductivity 10 times its previous predicted value and as high as diamond’s. If BAs can be produced economically—it generally doesn’t “like to form good structures,” Broido notes—it could be a boon to the electronic device industry.

In the meantime, says Broido, “from our studies of BAs, we developed an extended set of criteria for selecting elements to combine into materials with potentially high thermal conductivity. We want to use this new approach to systematically search for other high thermal conductivity materials that haven’t yet been identified because nobody knew to look at these criteria.”

Michelle Sipics is a Philadelphia-based science writer.