Our efforts the past year have been to define experimental conditions to allow simultaneous, parallel kinetic studies on the same sample of BR using both IR and optical spectroscopies. There have been many challenges. ? 1.) Water is an essential component for normal photocycle behavior, but because of its intense IR absorbance in the same region as the amide components of proteins, only about 0.2 microliters, or less can be present in the <0.2 mg BR sample. The collection time for IR data takes several hours during which time thousands of laser flashes are used. With this small amount of H2O, evaporation could result in desiccation and perturbations in kinetic behavior. We have found a solution to this problem. A special IR cell is used that has a small ridge drilled along its outer edge and a tight-fitting cap. An aqueous solution is placed in the outer ridge to serve as a reservoir to maintain an adequate steady-state vapor pressure in a 120 micron space above the sample in the center of the cell. The extinction coefficient for the characteristic absorbance (A) of H2O is used to quantify the actual H2O content of the sample. This amount of H2O results from the equilibrium between the vapor pressure over the reservoir and the sample. By trial and error we have determined a concentration of phosphate solution in the reservoir that provides an optimal amount of H2O in the sample. Even at this concentration of H2O in the sample, the remaining (A) due to H2O decreases the transmittance of IR light to a point that results in significant background noise. We have determined that a sufficiently high signal to noise ratio to allow the required fitting of 6 exponential constants to the data requires the pooling of several hundred samples. ? 2.) Because the kinetics of the BR photocycle are markedly influenced by pH, it is essential to know the pH of the sample being assayed. But, the sample volume is only about 0.2 microliters, spread over an area of about 1 cm2 and no pH probe can be used for such a sample. Furthermore, the pH of the initially dilute sample can not be assumed to hold upon the evaporation of most of the H2O. This pH problem can not be solved by increasing buffer strength, because the dried salt residue will retain a higher H2O content upon equilibration with the reservoir. We have solved this problem by using a micro-combination probe contained in a hypodermic needle than can measure pH in a volume of near 1 microliter. We follow the change in pH as a sample of 50 microliters is taken to dryness and have learned what pH must be established in the beginning in order to achieve the desired pH at near dryness. ? ? To guarantee that the IR sample has not deteriorated during its long run because of desiccation or laser damage, we use the optical spectrometer on the identical sample both at the beginning and end of each 100-repeat run used for pooling the data. The optical kinetics provides a much higher signal to noise ratio, uses only a single laser flash for each entire photocycle and takes only about 10 minutes for 400 repeats. The next stage of the project is to obtain a super matrix of spectral data with columns representing absorbances at 600 to 900 discrete time points and rows representing absorbances at about 90 discrete optical wavelengths in the top part and at about 600 to 1000 discrete IR wave numbers at the bottom. We will then apply the mathematical, analytical procedures developed in our laboratory to try to obtain, for the first time, isolated absolute IR spectra for each intermediate. These data should provide important structural and proton-binding information to better understand conformational changed linked to electrogenic proton-pumping.? ? New directions: We have initiated a new project in collaboration with the National Institute of Biomedical Imaging and Bioengineering (NIBIB) and the National Institute of Standards and Technology (NIST) to build an instrument that can obtain entire optical and infrared spectra (IR) very rapidly (every few microsec) from a single crystal of membrane protein. We will start with the energy-transducing proton pump, BR. No instrument capable of performing such studies has been previously described. There are many reasons to have such an instrument. 1.) Quality control to verify whether a membrane protein functions the same in the crystal as in situ in the membrane. 2.) As a guide to alter the crystallization procedure to produce crystals more closely related to in situ function. 3.) To produce valid crystals to be used in time-resolved X-ray crystallography in order to obtain isolated atomic structures for each intermediate in the proton-pumping procedure so that the overall process can be better understood. The optical spectrometry approach will be developed at NIH and the IR spectroscopy at NIST. When each part is successfully working the optical equipment will be moved to NIST to integrate the joint system. Eventually, both NIH and NIST can have the same combined working system. NIST has already ordered a new IR spectrometer and a special microscope to focus both visible and IR light on the crystal sample. During the year, we have determined that the optical system instrument will be built around a charge-coupled device (CCD) camera with an attached spectrograph. The CCD has a photon detector that contains 1024 rows of 512 pixels each. Each row can record a spectrum of 512 wavelengths and each spectrum can be obtained at a different point in time. NIBIB purchased this CCD/spectrograph unit from Princeton Instruments with the understanding that it could perform the rapid spectral acquisitions required. We have since learned that no one has actually used the CCD in the way we intend, and the instrument as delivered can not. The problem is that in a kinetic cycle consisting of several sequential intermediates with different time constants, it is necessary to use a staggered collection schedule which starts with many closely spaced time points (every few microsec) and then continuously lower the collection rate ( i.e. to 500 microsec) towards the end. The device can alter the collection schedule as desired, but the line of pixels remains exposed to the monitoring light for the entire length of each time sequence. Instead of obtaining spectra at sharp points in time, there is an accumulation of photons over the entire period of exposure (i.e. 500 microsec for the slower collection times). We have found a way to remedy this problem. An image intensifier can be placed between the exit of the spectrograph and entrance to the CCD camera. A sharp square wave gate can be used to make the intensifier serve as a programmable shutter that will admit light only for an instant at the end of each timing interval. We plan to obtain this intensifier and integrate it into the system.
