A.) Combined IR and optical spectrometry This part of the project has been completed and the first paper is scheduled for the September issue of Applied Spectroscopy. Its unique feature is using linear algebraic procedures, developed in my laboratory, to obtain for the first time, absolute visible and IR spectra for all intermediates in the bacteriorhodopsin (BR) photocycle. The ability to isolate spectra for each intermediate is based on the parallel cycle model supported in my laboratory, rather than the alternative single cycle model accepted by a large group of research laboratories. With these isolated spectra, we were able to obtain the transitional difference spectra between consecutive intermediates, rather than the difference spectra between each intermediate and the ground state, as done previously. We were also able to separate (for the first time), the spectra and kinetics of two very similar forms of the M intermediate. Using these new sequential spectra, we were able to describe new features and functions of the BR photocycle. B.) Development of instrumentation and procedures for comparing visible kinetics of the BR photocycle in membrane protein crystals to that of in situ tiny membrane fragments. Crystals of functional proteins are widely perceived and used as models for how proteins act in vivo. This is a particularly tenuous assumption for membrane proteins because they are most often separated from the membrane using detergents. We think it is essential to establish how near or far such crystals mimic the activities of the in situ protein. Towards this end, we have been trying to develop instrumentation and software to provide an answer to this question. The basis of our approach is to apply the new combined visible and IR spectroscopic approach described above, to crystals of BR and to similarly small samples of BR in its native purple membrane (PM) environment. This has proven to be a challenge that we are still trying to meet. In the studies described in Part A, the sample was in a circle of 1 cm diameter. For crystals, the microscopic sample will be in a circle of 50 or 100 microns. For any given intensity of monitoring light passing through a 50 micron orifice there will be only 1/40000th as much light and through a 100 micron opening, 1/10000th . We have been pursuing two separate approaches to obtain visible kinetic spectra;one based on the unique spectrophotometer built for me at NIH several years ago, and the other using a Princeton Instruments CCD camera/spectrometer with and without an image intensifier. The kinetic experiment must be monitored with a light intensity not high enough to initiate any turnover of the BR photocycle. A single turnover of the cycle is then initiated by an intense 5 ns laser pulse. Using both fiber optics and lenses, we were able to conduct light from the microscope to the NIH-built spectrometer. These efforts to obtain visible kinetics with the NIH spectrometer have not succeeded. The intensity of monitoring light needed to record sufficient photon counts was above the threshold for initiating turnover. The commercial CCD/spectrometer has presented a number of other problems;1.) Its response to increasing light intensity is linear only up to about 1/2 of its range. 2.) Using constantly spaced time intervals with a sample of air or water, the count level at each time point unexpectedly rises from about mid-point in the schedule, instead of remaining constant. 3.) When using kinetics that involve different time intervals for collecting data, the recorded photon counts are not linearly related to the dwell times of the different intervals. In principle, the image intensifier (ii) that we have should solve these problems by acting as a shutter or gate. In this mode, the ii placed between the spectrometer and CCD camera blocks all light except that which is programmed for a constant scheduled pulse width of 1-2 microsec. In addition, it provides a very significant amplification for the photon level it receives. The ii works as follows: Photons strike a photocathode that produces a current that is greatly amplified as it passes through a microchannel plate and the enhanced current strikes a phosphorescent screen that emits photons for the CCD to record. The problem is that if the enhanced current is too high, it could permanently damage the ii. To avoid this, a control circuit automatically cuts in (without warning) to decrease the gain. The resultant changes in photon counts totally obscure the changes due to the turnover of BR. After much discussion with the manufacturer, changes were made in the circuitry that gives us more room to use the pulse mode of operation safely. This does work, and the problems listed above for the CCD disappeared. But, the ii has introduced a new serious problem. The quantum efficiency (qe) of its phosphor is very low in the range near 412 nm where one of the most important intermediates of the photocycle occurs. The qe is drastically higher at the higher wavelengths where the other intermediates are found. With the 12-bit A/D converter, our maximum count level must be lower than 65536 counts to avoid saturation. This limits us to too few counts for following the 412 nm intermediate. There is one solution that should overcome this problem. We need a custom filter made just for the emission spectrum of the ii phosphor such that its transmission will compensate at all wavelengths to provide a near constant level of transmittance counts. With this, we can expect to increase the approximately 1500 counts we now obtain at 412 nm to about 50000 counts. I tested this idea using a variety of filters and other means to modify the photon emission spectrum and was able to raise the count level at 412 nm to 5000 counts. With the custom filter, we expect a further 10-fold enhancement in the count level. C.) At the CARB facility of NIST. Development of instrumentation for studying IR kinetics of the BR photocycle in single membrane/protein crystals and tiny membrane fragments. The newly acquired Bruker IR spectrometer and microscope are working and we can obtain IR kinetics using the 100 micron window. Our crystallographer has acquired the skill to make BR crystals. Brucker is making modifications that should allow us to work with a 50 micron mask. Once we have solved the existing problems with the visible system, we will move it to NIST for the combined IR/visible kinetic studies on crystals and membrane fragments.
|Hendler, Richard W; Meuse, Curtis W; Gallagher, Travis et al. (2015) Stray light correction in the optical spectroscopy of crystals. Appl Spectrosc 69:1106-11|
|Hendler, Richard W; Meuse, Curtis W; Smith, Paul D et al. (2013) Further studies with isolated absolute infrared spectra of bacteriorhodopsin photocycle intermediates: conformational changes and possible role of a new proton-binding center. Appl Spectrosc 67:73-85|