Volume 1 Story 4 - 3/7/2008
Nanoscopy:
Shedding Light on Life
Where traditional optical microscopy fails, a new tool, the nanoscope, overcomes the last barrier: the diffraction limit. It can explore the interior of cells in 3D, non-invasively, and with nanometric resolution.

Microscope versus nanoscope. Diffraction-limited 3D focal spot in standard optical microscopy (top) compared to the 45 nm spherical one now achieved in optical nanoscopy (bottom).
Nanoscopy! Remember this word: it is here to leave its mark.
Optical microscopy at the nanoscale is no longer science fiction.
Indeed, researchers at the Max Planck Institute for Biophysical
Chemistry in Göttingen (Germany), led by Stefan W. Hell, have
significatively pushed the 3D resolution of optical microscopy
down to the nanoscale, and have been able to non-invasively
investigate the interior of cells.
"The resolution of a microscope," Hell explains, "is given by
the 3D size of the focal spot, which usually has an elliptical
shape, similar to a rugby ball — around 250 nm in the plane
parallel to the sample and around 500 nm in the perpendicular
plane. We have shown that it is possible to squeeze this ellipsoid
to a sphere, downsizing the focus from a rugby ball to a tennis
ball — around 45 nm in diameter. The squeezed spot leads to a
substantial improvement in resolution which we have used to map
out the distribution of proteins in a cellular organelle without
needing to chop the cells into slices, as other techniques, such
as electron microscopy, would have required."

Cell nano-image. Distribution of membrane (red) and matrix (green) proteins in cellular mitochondria.
Since its dawn, mankind has always been fascinated by the unknown.
Unluckily, only a tiny part of the much broader natural complexity
can be experienced by our physiological perception, the five
senses. Curiosity urged our ancestors to open the way to further
information by developing dedicated tools, such as microscopes, to
empower the sense of sight.
More than three hundred years ago, the Dutchman Antonie van
Leeuwenhoek and the Englishman Robert Hooke separately observed
the existence of tiny living organisms in drops of water using a
rudimentary microscope. Since then giant steps have been taken in
the field of optical microscopy.
Unfortunately, an optical microscope, even a perfect one, cannot
be used to see objects that are smaller than half the wavelength
of the illumination light, a few hundred nanometres. It is a
fundamental physical law: the diffraction limit. Have you ever
tried to pick up a very thin needle using your fingers alone?
Standard microscopy makes use of rather clumsy fingers (the
wavelength of light) to discern tiny objects: any object smaller
than the diffraction limit is invisible.
We need a more refined tool to break the diffraction barrier and
eventually pick up the aforementioned needle. For example,
electron microscopy has been widely employed in biology to gain
resolution at the nanoscale. Indeed, the wavelength associated
with fast electrons is much smaller than that of light. However,
these unconventional microscopes have a major drawback: they are
typically incompatible with life. As a result, the clumsy fingers
of focused light continue to provide us with the most promising
way to explore the 3D cell universe non-invasively. The challenge
is to make them as little clumsy as possible.
One way to achieve this is by taking advantage of fluorescence, an
optical property of some molecules. Fluorescent molecules absorb
light at a certain wavelength to emit it again at a longer one.
High resolutions have been achieved through several
fluorescence-based approaches, but at most in 2D. Hell's group,
for example, had previously studied the interaction between some
fluorescent molecules and a focused laser beam in the shape of a
doughnut. Even though the distribution of light is still
diffraction limited, the interaction with the fluorescent molecule
is not: instantly quenching the periphery, the excitation is
effectively confined only to the hole of the doughnut, which is
far below the diffraction limit. It is like placing ice on the rim
of a hot plate to cool the edge down whilst leaving the centre
hot.
Now Hell and coworkers have found the right recipe to place ice
not only on a 2D plate, but in a whole 3D space. They have
engineered the interference of laser beams in order to confine the
fluorescence to a sphere of 45 nm in diameter. Hell affirms that
"in theory this spot and hence the resolution can be squeezed
down to the size of a molecule, a few nanometers. However,
practical issues will probably limit it to around 10-20 nm."
Hell underlines that "the wavelength of light can be tuned so
that it is not absorbed by the sample. Besides, the light
intensity is almost three orders below that of multiphoton
microscopy, which is nowadays broadly used for live cell imaging."
Optical nanoscopy, therefore, would be a powerful tool to
visualize and quantify molecular distributions in living cells
without concern about biological damages.
At Harvard University (Massachusetts, USA), the group led by
Xiaowei Zhuang had already achieved similar results with an
alternative fluorescent-based method [1]. However, Lukas Novotny,
from Rochester University (NY State, USA), explains that "Hell's
technique has less stringent requirements and is faster in terms
of video rate, since it can image many fluorescent molecules at
once. This, I believe, will make it widely used in future." Bo
Huang, from Zhuang's group, mostly agrees with Novotny's opinion,
even though he adds that, regarding applications, "no approach
will be more successful than others in full generality. Different
problems will require different approaches."
[1] B. Huang
et al.,
Three-dimensional super-resolution imaging by stochastic optical reconstruction microscopy, Science
319, 810-813 (2008).
Giorgio Volpe
2008 © Optics & Photonics Focus
GV is currently working on his doctoral thesis at ICFO - The Institute of Photonic Sciences, Barcelona (Spain).
R. Schmidt et al., Spherical nanosized focal spot unravels the interior of cells, Nat Methods (2008) 5, 539-544 (link).