18 Scientific American, April 2022
Google EarthADVANCES
G L A C I O L O G YRivers of Trouble
River reversals destabilize ice shelves
Columbia University glaciologist Alexandra Boghosian spent two
years studying a meltwater river on Greenland’s Petermann Ice
Shelf. She suspected the river ended in a waterfall like the one that
cascades off the Nansen Ice Shelf in Antarctica, potentially keeping
water from accumulating in melt ponds that can damage the ice.
Instead Boghosian and her team discovered a new phenomenon:
a deep-cut river channel that could contribute to future ice-calving
events and accelerate sea-level rise.
The team first spotted the phenomenon in 2018 using Google
Earth; an overhead view showed sea ice floating in a river carving
into the ice shelf. Further work confirmed this river had cut so deep
its water actually flowed backward, from the sea up to half a mile
into the channel. “We called it an ‘estuary’ because of the evidence
of this flow reversal that mixes fresh and salt water,” Boghosian says.
She and her colleagues detailed their find in Nature Geoscience.
The researchers also noticed ice fractures running parallel to the
river along its backward-flowing section. “The type of fractures they
found can really set up ice for failure because ocean water can get in
there and erode them,” says Catherine Walker, a glaciologist at the
Woods Hole Oceanographic Institution, who was not involved with
the study. Such fractures can contribute to calving events, in which
large blocks crack off a shelf. With reduced mass, shelves cannot as
readily hold back advancing glaciers behind them, and more ice
flows into the ocean, raising sea levels faster. Increasing tempera-
tures from climate change could cause more of these rivers to run
for longer periods with more meltwater and cut too deep.
Relatively warm water flows in “upside-down rivers” underneath
some ice shelves, melting the bottom and letting surface ice settle
to form depressions on top. Water flowing through such channels
could also make shelves more prone to forming estuaries. “These
linear features mean the river is going to form in the same place
every year, allowing the water to incise deeper,” Boghosian says.
Most of the world’s ice shelves are in Antarctica and experience
less meltwater than Greenland’s do, but climate change could lessen
the gap. “I don’t think the volume of meltwater has been able to
establish one of these estuaries in Antarctica,” Walker says, “but this
study is certainly forward-looking as to what could happen to
weaken those big ice shelves.” — Theo NicitopoulosBIOTECHNOLOGY
DNA Antennas
Nanoscale indicator may speed up drug design
Developing drugs can be hit or miss, but now a tiny, DNA-based
sensor may help streamline the task. Acting as a “fluorescent
nanoantenna,” the sensor could flag in real time if a prospective
drug is binding to its target or reveal other cellular activity.
Cells use protein molecules to communicate with one another
and trigger functions throughout the body. When such a mes-
sage comes into contact with a cell’s surface protein, one of the
molecules involved changes shape like a lock opened by a key,
prompting a reaction. At just five nanometers across—one 200th
the length of a typical bacterium—fluorescent nanoantennas can
bind to and interact with proteins on a molecular level. Each
nanoantenna can target a particular protein; when that protein
changes shape, the bound nanoantenna shifts as well and emits
specific light when viewed under a fluorescence microscope.
For a study in Nature Methods, re search ers put these new
nanoantennas to work flagging when a particular digestive pro-
tein executed five different activities in a solu-
tion, such as re act ing to antibodies and
changing intestinal acidity. “It’s a nice
tool in our toolbox,” says the study’s
senior author Alexis Vallée- Bélisle,
a nanotechnology re search er
at the Université de Montréal.
Other researchers have built
nano an ten nas from metals that
attach to any protein encoun-
tered. But the new antennas’
DNA-based structure can be pro-
grammed to adhere to a specific
protein—or region on a protein—based
on a sequence of building blocks called
nucleotides. “They’re like Legos,” says Mina Yes ̧ilyurt, a physicist
at the Leibniz Institute of Photonic Technology who was not
involved in the study. “You can create endless combinations.”
Sensing structural changes in specific molecules has big im -
pli ca tions for drug development, the study authors say. Vallée-
Bélisle uses the example of a protein involved in turning cells
cancerous. Researchers could introduce fluorescent nanoanten-
nas to monitor whether a drug successfully blocks the cancer-
causing protein from binding to a healthy cell analogue in the lab.
Fluorescent nanoantennas are still subject to many of the
same limitations as older techniques, such as false positives that
arise when proteins unfold because of interference from the
antennas themselves. “There is no silver bullet that solves all of
the problems in these things,” says Ahmet Ali Yanik, a nanoplas-
monics engineer at the University of California, Santa Cruz, who
was not involved with the research.
But Yanik does think the approach will be useful—especially
given its relative affordability compared with other ways of mon-
itoring proteins, such as x-ray crystallography. “Every biology lab
has a fluorescence microscope,” he says. “So it’s definitely a
technique that can catch on.” — Joanna Thompson
Illustration by Thomas FuchsGoogle Earth image that revealed
a meltwater river’s backward flow