Science News - October 8, 2011

Carolina at Chapel Hill. “And whatever
measurements we made on those mil-
lion cells would reflect what’s going on
in a single cell.”
Scientists today know that averag-
ing the constituents of a million cells
often gives no clue to what’s going on
at the cellular level. Cells thought to be
alike have been found to produce dif-
ferent numbers and types of proteins.
Beyond providing interesting insights
into basic biology, understanding this
individuality may have medical conse-
quences — illuminating characteristics
that can make a cell more amenable to
a particular drug or more vulnerable to
an allergen or viral infection.

Though in many ways single-cell analysis is still in its infancy, the field is quickly expanding. Researchers are working to improve technologies and to figure out how to make sense of the data.

Lighting the way

Of the millions of different living things in the world — large and small, plants, animals and microbes—all have in common the fact that they are made of one or more cells. Scientists have been observing cells since the mid-17th century, when Dutch scientist Antony van Leeuwenhoek discovered bacteria and, later, blood cells. Through the years, microscopes and staining methods have allowed researchers to view and describe the basic structures and internal workings of various cell types — understanding how skin cells, for example, differ from cells in the heart or kidneys.

Distinguishing unique features among cells of the same type has proven to be more difficult, because cells are so small. A typical human cell is about one one-hundredth of a millimeter across. Microbes are even smaller. Conventional light microscopes are constrained by a physical law demanding that the size of an object being imaged be no smaller than about half the wavelength of the light used to produce that image. To see anything as small as a molecule within a cell (the scale at which much of the work gets done), some tool other than traditional light microscopy is needed.

Scientists have found ways to get around this blind spot. When molecules and proteins are labeled with fluorescent dyes, the components inside a cell appear as bright colored dots against the background of a cell’s gel-like filling. Such advances have allowed researchers to see cellular processes unfolding at nanometer scales. Fluorescent probes have also brought to light the wide variation that exists within a population of seemingly alike members.

When sorting through a population of cloned lung cancer cells loaded up with fluorescent probes, systems biologist Steven Altschuler noticed that each cell carried its own combination of the molecules that send signals between cells.

“It was a surprise to us,” says
Altschuler, of the University of Texas
Southwestern Medical Center at Dallas,
who jointly runs a lab with pharmacolo-
gist Lani Wu. “It even made us wonder
what the definition of a clone was.”
The team developed a way to combine
fluorescent microscopy with a type of
software designed for face recognition
to look for patterns among the cells. The
team first labeled the cells by attaching
different fluorescent tags to proteins
involved in cell signaling. Computers
then analyzed pictures of the cells, pixel
by pixel, looking for variation in the loca-
tion and intensity of the molecules. The
scientists also trained the computers
to identify combinations of markers
that commonly appear together—
discovering the cellular equivalent of a
“blue-eyed blond.”
Using this method, the scientists were
able to identify small subpopulations of
cells, or stereotypes, based on which
molecules appeared and where. Expos-
ing the stereotypes to the anticancer
drug Taxol revealed that different sub-
populations respond differently to the
drug. In May 2010 in Molecular Systems
Biology, the team reported on how indi-
vidual cell differences could help predict
the outcome of treatment.

Spotting siblings By combining ;uorescent markers and cell-imaging techniques, researchers were able to divide a population of cloned cells into several subpopulations (four shown in chart). Members of each subpopulation showed similar patterns of ;uorescence (images).