Truncated hemoglobin O (trHbO) is one of two trHbs in Mycobacterium tuberculosis. Remarkably, trHbO possesses two novel distal residues, in addition to the B10 tyrosine, that may be important in ligand binding. These are the CD1 tyrosine and G8 tryptophan. Here we investigate the reactions of trHbO and mutants using stopped-flow spectrometry, flash photolysis, and UV-enhanced resonance Raman spectroscopy. A biphasic kinetic behavior is observed for combination and dissociation of O(2) and CO that is controlled by the B10 and CD1 residues. The rate constants for combination (<1.0 microM(-1) s(-1)) and dissociation (<0.006 s(-1)) of O(2) are among the slowest known, precluding transport or diffusion of O(2) as a major function. Mutation of CD1 tyrosine to phenylalanine shows that this group controls ligand binding, as evidenced by 25- and 77-fold increases in the combination rate constants for O(2) and CO, respectively. In support of a functional role for G8 tryptophan, UV resonance Raman indicates that the chi((2,1)) dihedral angle for the indole ring increases progressively from approximately 93 degrees to at least 100 degrees in going sequentially from the deoxy to CO to O(2) derivative, demonstrating a significant conformational change in the G8 tryptophan with ligation. Remarkably, protein modeling predicts a network of hydrogen bonds between B10 tyrosine, CD1 tyrosine, and G8 tryptophan, with the latter residues being within hydrogen bonding distance of the heme-bound ligand. Such a rigid hydrogen bonding network may thus represent a considerable barrier to ligand entrance and escape. In accord with this model, we found that changing CD1 or B10 tyrosine for phenylalanine causes only small changes in the rate of O(2) dissociation, suggesting that more than one hydrogen bond must be broken at a time to promote ligand escape. Furthermore, trHbO-CO cannot be photodissociated under conditions where the CO derivative of myoglobin is extensively photodissociated, indicating that CO is constrained near the heme by the hydrogen bonding network.

The Yersinia protein tyrosine phosphatases (PTPase) contain a single and invariant tryptophan (W354) located at one of the hinge positions of the flexible loop (WpD loop), which is essential for catalysis. The wild-type Yersinia PTPase and an active site mutant in which the esential Cys 403 has been replaced by serine (C403S) have been examined using both time-resolved fluorescence anisotropy and steady-state UV resonance Raman (UVRR) spectroscopies. Both enzymes were examined with and without the bound inhibitor arsenate. The UVRR spectra indicate that in solution the ligand-free, wild-type PTPase exists as an equilibrium mixture of two tryptophan rotamer structures with chi2,1 dihedral angles of -4 degrees and -90 degrees. The two rotamers have been attributed to the presence of both "closed" and "open" WpD loop conformers of the ligand-free enzyme. Conversely, the UVRR spectra of the arsenate-ligated, wild-type PTPase and of ligand-free and arsenate-ligated C403S PTPase contain a single W3 band which is correlated to the -4 degrees rotamer of W354, indicating a predominance of the closed WpD loop conformer. The tryptophan fluorescence anisotropy decay measurements of the ligand-bound, wild-type Yersinia PTPase and of both ligation states of the C403S PTPase reveal a single correlation time of 30-48 ns due to the rotational motion of the protein, while the ligand-free, wild-type PTPase is found to have two correlation times of 31 and 3.8 ns. The 3.8 ns correlation time of the ligand-free enzyme is attributed to the hinged movement of the WpD loop which contains W354. These results indicate that under physiological conditions, the nonligated, wild-type Yersinia PTPase alternates between an open WpD loop and a closed loop form with a rate constant of approximately 2.6 x 10(8) s(-1). We conclude that the rate of WpD loop closure of the wild-type Yersinia PTPase is thus independent of the presence of ligand, whereas in the presence of ligand the rate of opening is dramatically reduced resulting in a closed conformation on ligand binding. In contrast, the ligand-free and ligated C403S PTPase remain in the loop closed configuration over the time course of our dynamic measurements. The lack of WpD loop motion in the C403S PTPase is believed to be due to either a loss of repulsive potential between the anionic thiolate and Asp 356 of the WpD loop and/or the formation of a hydrogen bond or water bridged hydrogen bond between Ser 403 and Asp 356.

Functionally distinct conformations of HbA (human adult hemoglobin) were probed using deoxy and diliganded derivatives of symmetric Fe-Zn hybrids of HbA. To expand the range of accessible structures, different environments were utilized including solution, sol-gel encapsulation, and crystals. Further structural and functional modulation was achieved by the addition of allosteric effectors. Functional characterization included oxygen affinity measurements, CO combination rates, and geminate and bimolecular CO recombination, after photodissociation. The conformational properties were studied using visible resonance Raman spectroscopy as a probe of local tertiary structure at the iron-containing hemes and UV resonance Raman spectroscopy as a probe of elements of the globin known to be sensitive to quaternary structure. The combined results show a pattern in which there is a progression of conformational and functional properties that are consistent with a picture in which the T quaternary structure can accommodate a range of tertiary conformations (plasticity). At one end of the distribution is the equilibrium deoxy T state conformation that has the lowest ligand reactivity. At the other end of the distribution are T state conformations with higher ligand reactivity that exhibit "loosened" T state constraints within the globin including the alpha(1)beta(2) interface and reduced proximal strain at the heme.