SAXS and WAXS
Proteins are very important molecules in living organisms and the 3D-structure, the so-called tertiary structure,
is what determines the basic function of the protein. There are several
methods for gaining (partial) insight about this structure. One way is
to make a crystal of the protein and then bombard the crystal with
X-rays (X-ray crystallography), capturing the scattered X-rays on film
or in a detector. The next step is to find the structure of the crystal,
given the pattern captured in the detector. There are drawbacks with
this method, it may be a major undertaking to crystallize a protein and
a crystal may limit the natural movement (changing of shape) of the
protein.
An alternative is to mix the un-crystallized protein with a medium (typically water and other substances), placing the solution in a capillary
with very thin walls. The capillary is then subjected to X-rays, as in
X-ray crystallography, but since the protein is not crystallized (and
not oriented in any special way), the detector image will look very
different from the crystal case. Depending on the scattering angle, the
technique is called SAXS (Small-angle X-ray Scattering) or WAXS
(Wide-angle X-Ray Scattering). The X-rays are usually produced by a
synchrotron, synchrotron radiation, due to the small diameter, high intensity and small
divergence of such radiation.
Since the protein is free to move in the capillary, it is possible to
excite the protein to make it change shape. The excitation can be
accomplished by a laser-pulse (in parallel with the X-ray). The
Nature-article (see below) describes this type of experiment. Three
types of proteins were examined, hemoglobin, myoglobin and cytochrome.
Hemoglobin is a large molecule. In the Wikipedia-article (see below) it says that
the molecular weight is 64000 daltons, where one dalton is the mass of
1/12 of a carbon atom. (See this link for a precise definition.)
So that would be an equivalent of more than 5300 carbon atoms.
Hemoglobin is not made up just by carbon, of course. There is a huge
number of hydrogen-, oxygen- and nitrogen-atoms as well. Finally there
is some sulfur and four iron-atoms per molecule. So, the total number
of atoms in one molecule is of order 10000 (hydrogen having approximately 1/12 the mass of carbon).
To appreciate the challenge of finding the structure I suggest you visit the Protein Data Bank and have a look at hemoglobin. Try this link.
The GUI is an interesting example of scientific visualization, as well.
Try clicking "Custom View" and set Style to "Ball and Stick" to see the atoms.
Image
format
The X-ray images were captured using a Mar 133 CCD-detector. Mar was the name
of manufacturer (Rayonix today) and CCD was the technology used, CCD = Charge Coupled Device. This explains the suffix, mccd, of the files.
Here is an image of a slightly different detector, a Mar 165, and I have worked a lot with images produced by such a detector.
Those experiments were done by Peter Berntsen, a physicist at the Lundberg Lab, in Göteborg. (Peter has recently moved to Dept. of Physics, La Trobe University, Melbourne.) Most of the experiments were done at the MAX IV Laboratory in Lund, the synchrotron closest to Göteborg.
The
image files are in machine readable form (not human readable), and they are in a so-called tiff-format, a standard format for images. The files starts with
a 2048 integer header (16-bit unsigned integers are used), this can (in part) be read by using the Matlab-command, imfinfo. In this lab you should skip the header, but in a
real application one may use the information it contains. One can,
for example, see that the detector was cooled to -68.85 Celsius, to
decrease the noise (although this information is not reported by imfinfo). The detector was also kept at a low pressure for
the same reason. Following the header, comes the image
consisting of 2048^2 16-bit unsigned integers. This part can be read by the imread-command. In order to see something, apart from a black square, you need to use the caxis-command. The hist-command (or imhist) can supply you with appropriate values. Choose a suitable colormap.
Use the hints above to visualize some the of mccd-images. You find a few (I have many more) in the local file system (not on www) in the following directory: /chalmers/groups/thomas_math/VIS/MCCD
. The laser pulse is generated at time zero and 178ns, 316ns etc. in
the file names means that those images were captured 178 ns, 316 ns
etc. after the pulse.
References
Here are some Wikipedia links if you want to know more, this is not compulsory reading.
Protein
Protein structure
Protein tertiary structure
Structural biology
Hemoglobin
Myoglobin
Cytochrome
Synchrotron radiation
WAXS, Wide-angle X-ray scattering
SAXS, Small-angle X-ray scattering
X-ray crystallography
Protein Data Bank
Here is link to an article published in Nature, the scientific
journal.Try to read the first part of the article, it gives a good
introduction. Note that it is possible to fetch the whole article as a
pdf-file.
Tracking the structural dynamics of proteins in solution using time-resolved wide-angle X-ray scattering
