Everyone should bear in mind the following when considering the presence of an ionizable anion at a protein binding site:
1. The ionizable ion will exist in solution in a equilibrium distribution of ionization states, depending on the solution pH, and will exchange between them rapidly 2. The effective pH of the solution, as well as the pKa values of the ionizable ion and its hydrogen bonding partners may differ significantly from that at 25C in a cryoprotected solution which may contain a high fraction of non-aqueous components. 3. The ionization state of the bound ion may differ from the predominant solution species at the effective pH depending on the hydrogen bonding environment in the binding site. 4. The hydrogen bonding of the ionizable ion can often only be inferred from plausible interactions with potential hydrogen bonding partners on the protein. An anomalous map can help determine if there is a particular heavy atom present, but ambiguities may still remain, especially if the binding site is weak and there is not a well-defined orientation for the ionizable anion. I've seen well-defined bicarbonate ions where it is relatively straightforward to infer hydrogen bonding interactions and poorly resolved blobs that are likely loosely bound sulfate or phosphate. Cheers, Roger Rowlett Gordon & Dorothy Kline Professor, Emeritus Colgate University Department of Chemistry On Mon, Dec 25, 2023, 2:51 PM Tom Peat < 0000b7e4a7a8af49-dmarc-requ...@jiscmail.ac.uk> wrote: > Hello Dale, > > Thank you for the correction/ clarification. > I think this is still a tricky question, as in solution, this is an > average state and one doesn't have a stable hydrogen (or two) sitting > discretely on the phosphate. More specifically, the hydrogens are coming > off and popping back on constantly (just the ratios change depending on the > pH). It is likely that the phosphate is also moving in and out of the > binding site of a protein in solution. What state is captured in a crystal > structure and is that consistent across all of the proteins in that > crystal? > As you say, one needs very high resolution to determine the bond length > difference between those oxygens with and without a potential hydrogen > attached to orient a phosphate correctly in a structure, assuming that > there is only a single preferred orientation to start with. > I believe the original question was whether in fact the density supported > a phosphate ion, and I still believe that looking for some anomalous signal > may be a good way to approach that question. > Nonetheless, I stand corrected and there is likely to be some hydrogen on > phosphate ions found in crystal structures. > Happy holidays to all, tom > ------------------------------ > *From:* Dale Tronrud <de...@daletronrud.com> > *Sent:* Monday, December 25, 2023 8:52 PM > *To:* Tom Peat <t.p...@unsw.edu.au>; CCP4BB@JISCMAIL.AC.UK < > CCP4BB@JISCMAIL.AC.UK> > *Subject:* Re: [ccp4bb] Query on density fitting to phosphate > > [You don't often get email from de...@daletronrud.com. Learn why this is > important at https://aka.ms/LearnAboutSenderIdentification ] > > Hi, > > I wanted to correct a statement by Prof. Peat about the ionic state > of phosphate in solution. Phosphate has four states differing by the > number of attached hydrogen atoms. The number of hydrogen atoms depends > on the pH, or maybe it is the other way around since phosphate is used > as a buffer. I've attached a plot of the fraction of each species as a > function of pH (Citation: "By Clarolux - Own work, CC BY-SA 4.0, > https://commons.wikimedia.org/w/index.php?curid=90586171";). There you > can see that for all pH's usually seen in mother liquors the solution is > almost completely either HPO4(-2) and H2PO4(-1). A binding site may, of > course, prefer a species that is present in low concentration but such a > protein will be fighting entropy to fill its pocket. > > Unless your mother liquor has an extreme pH you should expect that > the phosphate species you are seeing in your crystal has either one or > two hydrogen atoms attached. Their presence will affect both the nature > of the hydrogen bonding of the protein to the phosphate and will change > the length of the P-O bonds (with the P-O-H bond being about 0.05 A > longer than the P=O bond). The two lengths will only be distinguishable > given very high resolution diffraction data but there are examples in > the PDB where the differences are clear. You can determine the presence > of an hydrogen atom at much lower resolution if the hydrogen bond is > made with an obligate hydrogen bond acceptor. > > The inappropriate identification of an ion as PO4(-3) will > significantly degrade the quality of any electrostatic potential one > calculates from the model. > > I did a quick-and-dirty search of the PDB for the various species of > phosphates in PDB entries. While I found 5979 models with PO4(-3) (ID: > PO4) I only found 42 with HPO4(-2) (ID: PI) and 27 with H2PO4(-1) (ID: > 2HP). I didn't find any H3PO4 and could not find an ID code for that > molecule. (This search was done quite a while ago.) I believe > depositors are mostly assuming the ID PO4 indicates any protonation of a > phosphate ion but that is not correct. I am unaware of any ID that is > defined as a phosphate ion with unknown protonation state. To conform > to the wwPDB standards a depositor must do their best, using the limited > data available to them, to choose one species of phosphate when they > identify the presence of one, but almost certainly that choice should > not be PO4. > > As usual, just causing trouble, > Dale E. Tronrud > > On 12/17/2023 1:05 PM, Tom Peat wrote: > > Dear Arpita, > > > > The hydrogens on phosphate, just like sodium and potassium, will come > > off the oxygens in water. > > To be more explicit, you don't have mono- or di-hydrogen phosphate in > > water (except transiently), you just have phosphate, depending somewhat > > on the pH of course. At 2.5 Angstrom resolution, there is no way to > > 'see' hydrogens with X-rays. > > Depending on the wavelength you used for your data collection, you could > > try doing an anomalous map and see if you have any anomalous signal at > > this position, which may help in identifying what the density is. > > Best of luck, tom > > > > ------------------------------------------------------------------------ > > *From:* CCP4 bulletin board <CCP4BB@JISCMAIL.AC.UK> on behalf of Arpita > > Goswami <bt.arp...@gmail.com> > > *Sent:* Sunday, December 17, 2023 9:46 PM > > *To:* CCP4BB@JISCMAIL.AC.UK <CCP4BB@JISCMAIL.AC.UK> > > *Subject:* [ccp4bb] Query on density fitting to phosphate > > > > You don't often get email from bt.arp...@gmail.com. Learn why this is > > important <https://aka.ms/LearnAboutSenderIdentification> > > > > > > Dear All, > > > > Hope you all are doing well. > > > > The density in the image (in link below) is fitted with PO4 ion, > > although the crystallization condition has both mono and dihydrogen > > phosphate which is not fitting without hydrogen. But the resolution is > > 2.5 A, so hydrogen may not be put in, or is there any way to do so? > > Otherwise placing water is the final option. > > > > https://i.postimg.cc/4N7q2K0p/Screenshot-from-2023-12-17-16-07-07.png > > <https://i.postimg.cc/4N7q2K0p/Screenshot-from-2023-12-17-16-07-07.png> > > > > Also the density is quite close to Aspartate, so PO4 may not be right. > > Can it be dihydrogen phosphate as two positively charged residues > > (Specially the lysine) are also nearby to neutralize positive charge? > > Other ions in the crystallization condition are Cl-, K+ and Na+. These > > are not put as both aspartate and lysine are at comparable distances > > from the density. The pH is 6.2 in which dihydrogen phosphate is > > reported to interact with aspartate > > (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5855859/ > > <https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5855859/>). > > > > Waiting eagerly for your reply. > > -- > > Thanks and Merry Christmas in advance. > > Best Regards, > > Arpita > > > > ------------------------------------------------------------------------ > > > > To unsubscribe from the CCP4BB list, click the following link: > > https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1 > > <https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1> > > > > > > ------------------------------------------------------------------------ > > > > To unsubscribe from the CCP4BB list, click the following link: > > https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1 > > <https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1> > > > > ------------------------------ > > To unsubscribe from the CCP4BB list, click the following link: > https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1 > ######################################################################## To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1 This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
