[ccp4bb] pKa and electrostatic affinity
Dear colleagues: I was wondering if you could kindly share your thoughts and help me understand the relationship between pKa and affinity of a protein for a ligand. Are these two properties related? Specifically, does a lysine with a pKa of 8.5 have a greater affinity for a negatively charged ligand than a lysine with a pKa of 10.5 for the same ligand at physiological pH? Any comments would be highly appreciated. Deepak
[ccp4bb] Summary on Aspartate pKa increase and catalysis
Dear colleagues, I appreciate your help in interpreting the pKa shift. Based on all the suggestions, I have several new leads to be able to come up with a tentative mechanism. I am summarizing the discussion below – The pKa of a catalytically critical aspartic acid has increased to 6.44. It is hydrogen bonded (2.8 Angstroms) to a water molecule that is supposed to donate a proton during the catalysis. The pKa of the amino acid was estimated by the PropKa server (http://propka.ki.ku.dk/) using the co-ordinates of the crystal structure. Theoretically, pure water (assuming 55.5M) has only 1 proton in 5.55 x 10^ -6 molecules (correct me if I am wrong here). Or simply, water is not protonated at pH 7.0. Therefore, under physiological conditions, it is almost impossible for water to act as a nucleophile, acid or base by itself. An acid/base catalyst or a metal ion is usually employed by enzymes to activate water for catalysis. In this particular case, a solvent derived proton has been speculated to complete the reaction. An aspartic acid hydrogen bonded to this candidate solvent molecule seems a good choice for protonating the water molecule. There are no other residues or solvent molecules bonded to this potential proton donor water molecule. At pH 7.4, a carboxyl with a pKa of 6.4 will be 90% ionised (deprotonated) and 10% protonated. In a mechanism where the carboxylic acid donates a proton to the water which in turn donates a proton during catalysis, then elevating the pKa to 6.4 would be reasonable as the carboxyl will need to be protonated at some point in the cycle to do the donating. Raising the pKa of aspartic acid would allow a larger fraction of it to be in its protonated state at physiologically relevant pH values, although it would reduce the intrinsic effectiveness of Asp as a general acid. There should be a significant thermodynamic and kinetic advantage in having Asp participate directly in a general acid catalyzed reaction, rather than through a water molecule. Alternatively, the higher the pKa, the higher the nucleophilicity of the residue/group (higher SN2 reactivity or affinity with electrophiles, like H+, perhaps the substrate of your enzyme..?, etc). Suggested literature and tools 1. Some papers by Nick Pace's group: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2708032/?tool=pubmed http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2679426/?tool=pubmed http://www.sciencedirect.com/science/article/pii/S002228360600934X 2. Pace, C. et al. Protein Ionizable Groups: pK values and Their Contribution to Protein Stability and Solubility. J. Biol Chem. 284, 13285-13289 (May 15, 2009) 3. Harris TK Turner GJ Structural basis of pertubed pKa values of catalytic groups in enzyme active sites IUBMB life, 53 85-98 (Feb 2002) 4. Papers of John A. Gerlt, who did a lot on protonabtraction reactions. 5. http://www.jinkai.org/AAD_history.html 6. THEMATICS for pKa calculations. Avaliable at http://www.northeastern.edu/org/wp/ 7. Tools to predict protein ionization http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2578799/ 8. Fap1_BBRC.pdf Apologies if I missed out on any reference or critical point/s raised in the discussion. Thank you. Deepak Oswal
[ccp4bb] On pKa of Aspartic acid
Dear colleagues, We have solved the crystal structure of a human enzyme. The pKa of a catalytically critical aspartic acid has increased to 6.44. It is hydrogen bonded (2.8 Angstroms) to a water molecule that is supposed to donate a proton during the catalysis. Can anybody help me a) interpret the significance of this increase in pKa of the aspartic acid from 3.8 to 6.44 in context with the catalysis? Is this advantageous or detrimental? b) How is pKa related to an amino acids’ ability to force a water molecule to donate a proton? c) At pH 7.4, the aspartic acid would be de-protonated irrespective of whether the pKa is 3.8 or 6.44; isn’t that true? d) Have similar increase in pKa values observed for aspartic acids before? I would be grateful if anybody could explain or comment on the above queries. Deepak Oswal
[ccp4bb] activation of thiol group
Dear colleagues: We have a 1.4 Angstrom structure of an enzyme showing a water molecule enclosed in a triangular pocket formed by the hydroxyl oxygens of 2 serine residues and a sulfhydryl group of an essential cysteine. The water molecule is forming a 2.8 Angstrom hydrogen bond with each of the hydroxyl groups of the 2 serines and a 2.9 Angstrom hydrogen bond with the sulfhydryl group of the cysteine. Is it possible for such a water molecule to lower the pKa of the cysteine and activate the thiol group? I would appreciate any comments or suggestions or information on any literature that I need to look up : Sincerely, Deepak
[ccp4bb] Structure-function analysis
Dear colleagues: I am trying to interpret the results of the substitution of a Methionine with Alanine. Following is the kinetic data on the mutation - 1. Km increased by 0.5 fold 2. Vmax decreased by 3.5 fold 3. Kcat decreased by 4 fold 4. Kcat/Km decreased by 10 fold. 5. Activity at saturating concentration of substrate - only 15 % of the wild type. Is it possible to conclude from the data that the methionine is involved in stabilization of the transition state (the methionine is located inside the putative active site; we do not have a structure of the enzyme-substrate complex)? Is this even possible atall? Although impossible, on second thought, given the ability of the micro-environment to alter/bring down pKa s, is there any instance or possibility that a methionine could or has acted as a nucleophile? In addition to all other routine precautions to avoid experimental errors, a CD analysis of the mutant and wild type has been performed and there are no obvious differences in the spectrum. The wild type and mutant enzyme show almost identical size exclusion profiles. I would appreciate any suggestions or comments. Sincerely, Deepak
Re: [ccp4bb] Structure-function analysis
Dear Micheal and Matthew: Thank you for your inputs on the structure-function analysis data. As pointed out by Micheal - the change in Km (For Met to Ala mutation) is not much. Could this be intepreted simply that the mutation does not change the affinity of the enzyme for the substrate because a smaller side chain (ala) would not obstruct, if not facilitate, the placement of the substrate. However, a slight increase in Km would suggest a larger side chain (met) is needed to guide and position the substrate optimally? This is further supported by a decrease in Vmax. I would assume, the Vmax values calculated at lower concentrations of substrate, as used in kinetic analysis, are generally not accurate (as opposed to that Km values at lower concentrations of substrate are considered fairly accurate). However, the general trend of the kinetic analysis showed that the Vmax for the Met to Ala mutation decreased significantly. To verify this, independant assays ( as a separate experiment) were performed for the wild type and mutant under identical conditions with saturation concentration of the substrate. Here, the differences in the Vmax for the mutant and wild type were magnified as indicated by a 15 % residual activity for the Met to Ala mutation. On the other hand, Matthews seems to suggest that the differences are not significant. Would a 85 % loss in activity at saturating concentration (6.7 fold decrease) considered insignificant? This could be more to do with the sensitivity of the assay. However, since the experiments were conducted under identical conditions, wouldn't these differences be real and significant? I truly appreciate your time and valuable suggestions : Sincerely, Deepak . On Thu, Feb 18, 2010 at 12:34 AM, R. M. Garavito rmgarav...@gmail.comwrote: Deepak, The changes you see are not huge and could be in the noise. 1. Km increased by 0.5 fold -- not much of a change These seem to be a little inconsistent as Vmax = kcat*[E]: 2. Vmax decreased by 3.5 fold 3. Kcat decreased by 4 fold These are not too different. 5. Activity at saturating concentration of substrate - only 15 % of the wild type. (a 6.7 fold decrease) As activity at saturating concentration of substrate is Vmax, why is it so much lower? Did you use the same enzyme concentration for the experiments (most protein assays are not particular accurate and are variable)? Did you benchmark [E] with A280 measurements? Did you repeat the kinetic analysis of the WT enzyme along with the mutant (using the same buffers and reagents)? Even slight differences can arise between two kinetic trials a month or two apart just due to differences in buffers and reagents (age, slightly different pH, etc). Were the temperatures of the assays the same (did you use a thermostated cell)? Moreover, in choosing the amount of enzyme to use, it must be in the valid range to get good data. This means that Vmax = kcat*[E] must hold. The valid ranges for the WT and mutant enzymes may not be the same. At this point the mutant seems not to have made much of a change. Repeat the analyses back to back to get more data. Good luck. Michael ** *R. Michael Garavito, Ph.D.* *Professor of Biochemistry Molecular Biology* *513 Biochemistry Bldg. * *Michigan State University * *East Lansing, MI 48824-1319* *Office:** **(517) 355-9724 Lab: (517) 353-9125*** *FAX: (517) 353-9334Email: garav...@msu.edu* ** Matthew Merski to me show details 5:18 AM (3 hours ago) No, these data show only small changes in the activity of the enzyme (less than an order). Typically you want to see something with at least 10^4 fold effects to be sure it matters and really you should see something like loss of 99% of the activity to start arguing for an effect. The Met to Ala mutation is probably perturbing the active site a little since the shapes are different, but not doing much else. The differences you see are not large enough that you can't honestly rule out experimental error as their cause, especially if you are using literature data for the WT values or the two measurements were performed by two different people. Methionine acting like a nucleophile would be unlikely enough that I would call it impossible. Matthew Merski UCSF Shoichet Lab - Show quoted text - On Feb 17, 2010, at 10:43 AM, Deepak Oswal wrote: Dear colleagues: I am trying to interpret the results of the substitution of a Methionine with Alanine. Following is the kinetic data on the mutation - 1. Km increased by 0.5 fold 2. Vmax decreased by 3.5 fold 3. Kcat decreased by 4 fold 4. Kcat/Km decreased by 10 fold. 5. Activity at saturating concentration of substrate - only 15 % of the wild type. Is it possible to conclude from the data that the methionine is involved in stabilization
