Mabbott Research Group:  Analytical Chemistry

Students in our research group have been developing new micro-analytical techniques for forensic and bioanalytical applications.  Recently we have explored spectral imaging techniques for obtaining chemical information through a microscope.  In our work this has meant using a camera to record images over a range of wavelengths and applying computer routines to reconstruct the intensity vs. wavelength profile afterwards for any object in the picture.

 

For example, here is an image of some 6 micron fluorescent beads.


Fluorescent objects generally emit light at longer wavelengths than the light that they absorb.  In our work we have concentrated on measuring the profile for the absorbed light, commonly called the excitation spectrum.  Here is the excitation and emission spectra for the beads taken in a conventional fluorescence spectrometer and the
excitation spectrum taken through the microscope below.

  One of the applications of our work is a rapid method for comparing dyes from fibers.  This type of comparison is an important aspect of establishing evidence that a suspect was at a scene of a crime.   Fibers that appear to be the same color often have a different spectral profile.  Conventional methods for comparing the dyes from fibers involve tedious and time-consuming steps such as extracting the dye.  Extraction destroys the sample and limits the amount of testing that the forensic scientist has time to do.  Spectral imaging, on the other hand, lets the examiner rapidly make many comparisons for a multitude of fibers anywhere within the image.

 

Similar techniques might also be applied to the challenge of measuring the concentration of interesting compounds in very small volumes such as biological cells.  How a drug distributes itself among the various compartments within a cell can be an important factor in determining its effectiveness or toxicity.  Our group is currently collaborating with Professor Edgar Arriaga and his students at the University of Minnesota.  Of special interest to this group is the widely used anti-cancer drug, doxorubicin.  Drugs have their limitations.  Some types of cancer cells are resistant to doxorubicin.  Furthermore, high levels can lead to damage to heart muscle.  Elucidating the mechanisms underlying these problems is important to improving the safety and effectiveness of the drug's clinical use.  Studying where the drug goes once inside the cell is an important aspect of that work.  Although doxorubicin is strongly fluorescent, direct measurement of its concentration distribution from fluorescence images of a treated cell is not currently possible.  The principal obstacle to this approach is the fact that the fluorescence efficiency for doxorubicin is strongly dependent on its environment.  Both the quantum yield and the spectral profile vary depending upon whether the drug is freely dissolved in the cytoplasm or associated with DNA, protein, or lipid.  Consequently, a single calibration factor at a single combination of excitation and emission wavelengths is not adequate for determining the concentration distribution, nor even the total concentration of doxorubicin in the cell.  We propose to characterize the excitation spectra for doxorubicin in various micro-environments within a cell in order to isolate the various sources that contribute to the fluorescent image and establish calibration factors for each source.  Ultimately, this strategy would permit us to extract doxorubicin distributions from spectral images of treated cells. 

We built a different type of imaging system in which the emission monochromator collects information from a slice of the image.

Microscope with monochromator in foreground.

Inside the monochromator. (Camera on the right.)

As shown in the image below we were able to use the new equipment to obtain emission spectra of fluorescent dyes and the drug, doxorubicin.

The image on the left shows some adsorbent particles stained with doxorubicin as viewed through the eyepiece of the microscope. The image on the right was taken with a very sensitive digital camera viewing only a slim slice of the same image through the entrance slit of the monochromator. In this case the monochromator was collecting the light from a single particle over a span of about 3 microns (about 0.0001 inches). A grating spreads out the light entering the monochromator along the horizontal axis as a function of wavelength. (The vertical axis corresponds to the physical position of objects along the length of the slice of the image passing through the entrance slit.) The white streak indicates that the emitted light is composed of a band of color. Superimposed on the black and white image is a set of green and red cross-hairs that indicate lines along which the computer has calculated the corresponding intensities. The red curve shows how the intensity varies with position along the red horizontal line at the position indicated. In a separate experiment this axis was calibrate in wavelength units. The red peak is centered at about 592 nanometers (orange light) and the shoulder on its right corresponds to about 634 nanometers (red light). In order to gather information about other objects in the full image (represented by the color photo on the left) we shift the position of the entrance slit and record another photo through the monochromator until we have mapped out the region that we are interested in. We hope to be able to use the position and shapes of emission peaks to distinguish the type of environment that the drug is in.

Please send me an email if you are interested in joining with my group in this work.