Quite often, after hydrogen atoms have been added to a protein, the formal, i.e., Lewis structure, ionization state of various sites are not what is wanted. This fault can be corrected by setting the ionization state of various sites in a protein using SITE=(text). This adds or removed individual hydrogen atoms; because of this the job cannot continue.
By default, the resulting structure will be resequenced. If this is not wanted, add NORESEQ.
Four methods of specifying sites are supported. Method A is recommended when starting with a new protein. Method B is very specific, but writing the keyword requires the most work; method C is less specific but the keyword is easy to write; method D affects all suitable residues, and its keyword is easy to write. Use option A first, then, if necessary and if possible, use C, and only when specific atoms are involved use B. Option D is sometimes useful in identifying all ionizable sites.
(A) Adding Salt Bridges. Salt bridges can be added using SITE=(SALT[=n.n]). All pairs of potentially charged sites (residues only) within 4 Ångstroms, or n.n Ångstroms if "=n.nn" is present, will be created. The default of 4.0 Angstroms was chosen to conform with Barlow, D. J. and Thornton, J. M. (1983) J. Mol. Biol., 168, 867–885. If a charged site already exists it will not be considered as a potential charged site. This option can be used with ADD-H, for example as "HTML ADD-H SITE=(SALT)" Here, ADD-H will generate a system with all sites neutral, then SITE=(SALT)will create salt bridges using (Nζ of Lys, Nε, Nη1 or Nη2 of an Arg, Nδ1 or Nε2 of a His, or a terminal -NH2 group) and a carboxylic acid group. This is equivalent to the definition used in "Hydrogen bonds and salt bridges across protein–protein interfaces", Dong Xu, Chung-Jung Tsai and Ruth Nussinov, Protein Engineering vol.10 no.9 pp.999–1012, 1997. Note that sulfate, phosphate and other ionizable sites are not considered when this option is used.
(B) Modifying specific residues. Individual residues to be ionized or de-ionized. In this format, text has the form "Annn(q1[q2])[,Bmmm(q3[q4])]..." where "A" is the chain letter, "nnn" is the residue number, and "q1" and "q2" are the charges, either +, 0, or –. If residue 123 in chain B is an Arg, then, to ionize it, use SITE=(B123(+)). If the first residue in that chain, i.e., the -NH2 end, is also to be ionized, then use SITE=(B123(+),B1(+)). If a carboxylic acid group, e.g. an Asp at residue site D234, should also be ionized, then use SITE=(B123(+),B1(+),D234(–)). A salt bridge can be created using a construction of the form SITE=(A12(+),B34(-)). To de-ionize a site use charge "(0)" thus if the sites just described were ionized, they could then be de-ionized by using SITE=(B123(0),B1(0),D234(0)). If a residue has two ionizable sites, such as a Lys at the start of a chain, or Asp at the end of a chain, use two charges, to indicate the ionization state of each site, e.g., SITE=(B1(++),B100(--)). No specific tautomer is defined - if it's not the one wanted, use method A. An alternative would be to make the change outside MOPAC, to do this, edit the resulting <file>.out, <file>.arc, or <file>.pdb to make a new data set. This should work for sulfate and phosphate groups, but is easily confused by hetero-groups, e.g., ADP or ATP.
(C) Modifying specific atoms. This option should only be used when option B cannot be used, because it requires more effort to define the atom involved. Individual oxygen and nitrogen atoms can be ionized or neutralized by adding or removing hydrogen atoms. This method is completely general in that it can be used for all ionizable oxygen and nitrogen atoms, and completely specific in that the only atom affected is the one specified. To achieve this generality, this option requires the atom to be specified in a very precise way. The format for this option is: SITE=("text"(n)) where text is the PDB name of the atom, this is given in PDB files in columns 13 to 26 for an atom. An alternative is to use the JSmol format. With JSmol, the label starts with a "[" and ends at the character before " #". As with the PDB format, spaces are not important when the JSmol format is used, i.e., SITE=("[LYS]4:A.NZ"(+),"[ASP] 43: A .OD2"(0)). As with option B, the charge, n, can be + or 0 (for nitrogen) and +, 0 or – (for oxygen). If n=1 for the nitrogen atom in pyridine, then the nitrogen would become a cation, and the pyridine would become pyridinium. If n=1 for methylamine, the result would be [CH3-NH3]+, if n=0 for a positively-charged nitrogen atom, then that atom would lose a hydrogen to become neutral. If n= – for an oxygen in acetic acid, the result would be acetate. If n=0 for a negatively-charged oxygen atom in a compound, the result would be a hydroxy group. To convert a hydroxyl group bonded to an atom into a water molecule, use "+" for the charge on the oxygen atom.
This option works for any oxygen or nitrogen atom, including those in sulfate and phosphate, etc.
If the PDB label for the atom is not unique, the first such label will be used. To avoid this potential error, make the appropriate atom's label unique. Spaces are not important, but using them makes it easier to read the atom label.
Examples are: SITE=("OG1 THR G 166"(0)) and SITE=("OG SER G 195"(0),"OD2 ASP G 194"(-),"[PO4]1157:B.O8"(-)).
Wildcards are allowed for the residue name and for the chain letter. When a wildcard is used, a letter is replaced by an asterisk, "*", thus: SITE=("OG *** G 195"(0),"OD2 *** * 194"(-),"[***]1157:*.O8"(-)). The first occurrence that matches will be used. Using wildcards allows similar systems to be modified easily, for example if 8-Oxo-guanine monophosphate and Guanine monophosphate are present, and the N7 has to be protonated or deprotonated (steps in converting from one to the other), and the residue names are 8OG and GMP, and they have the same residue number, then rather than use the full name, wildcards can be used, thus SITE=("N7 *** * 1157"(0)) or SITE=("[***]1157:*.N7"(+)).
If an attempt is made to modify an atom that can't be modified, e.g., to make the oxygen atom in CH3-O-CH3 into an anion, an error message will be generated, but be aware that not all possible cases have been defined.
These three formats, A, B, and C can be combined, although if SALT is used, it should be first, because it automatically constructs salt bridges. A simpler way to define ionization states is to use two or more SITE keywords. For example, if all common salt bridges are wanted and one or more atoms are to be protonated or deprotonated, then the two keywords SITE=(SALT) and SITE=("OG1 THR G 166"(0),etc.) could be used. If a large number of changes need to be made, use two or more SITE commands.
(D) Modifying all specified ionizable groups. This format applies to all residues with the specified groups. Text in this case can be one or more of the entries in the table below. Entries should be separated by a comma or by a semicolon.
To neutralize all sites in a protein, use SITE=(COOH,NH2,ARG,HIS). To ionize all sites, use SITE=(COO,NH3,ARG(+),HIS(+),SO4,PO4) or, to avoid extra typing, use SITE=(IONIZE). Note: This option is equivalent to SITE=(COO,NH3,ARG(+),SO4,PO4), i.e., it will not automatically ionize His. Serine, threonine, tyrosine, and the NH2 side-chain of asparagine cannot be modified using this format. The SO4 option will delete hydrogen atoms attached to an oxygen on SO4, PO4 will delete a hydrogen atom attached to an oxygen on PO4, if one exists, to give the mono-anion, i.e., CH3-O-P(OH)2O would become ]CH3-O-P(OH)O2]-, and H3PO4 would become [H2PO4]-. No specific tautomer is defined - if it's not the one wanted, make the change using method C. The pKa of H3PO4 is 2.12, of [H2PO4]- = 7.21, and of [HPO4]= = 12.67.
|COOH||Add a hydrogen atom to -COO groups|
|COO||Remove the hydrogen atom from -COOH groups|
Add a hydrogen atom to -NH2 groups (except in -CO-NH2)
|NH2||Remove a hydrogen atom from -NH3 groups|
|Arg(+)||Add a hydrogen atom to -Arg-|
|Arg||Remove a hydrogen atom from -Arg(+)-|
|His(+)||Add a hydrogen atom to -His-|
|His||Remove a hydrogen atom from -His(+)-|
|SO4||Remove any hydrogen atoms from a sulfate group|
|PO4||Remove any hydrogen atoms from a phosphate group|
Option D cannot be used with options A, B, or C. Option D can be useful during preliminary work on proteins, to identify all ionizable sites.
see also atom numbers, ADD-H