[Rutherford]

What is the formula used to calculate the isoelectric point of a protein?

[Babcock_Hall]

Do you know the formulae to calculate the isoelectric points of amino acids?

[Rutherford]

Yes, I do.
For amino acids with neutral side chain it is: pI=(pK_aNH4⁺ + pK_aCOOH)/2
For amino acids with basic side chain it is: pI=(pK_aNH4⁺ + pK_aNH4⁺)/2 the ammonium groups are different
For amino acids with acidic side chain it is: pI=(pK_aCOOH + pK_aCOOH))/2 the carboxylic groups are different.

Now how to do it for a protein e.g. Gly-Glu-Lys-Ala?

[Babcock_Hall]

Well your symbolism is a little off, but that is no big deal.  Can you find the form of the peptide with zero net charge?

[JasonHerbalExt]

I dont think there's any kind of those formulas. Because the protein is too complicated. Only an appropertate experiment can tell its isoelectric point.

[Babcock_Hall]

Jason, I don't entirely disagree.  The pK_a values of side chains are often perturbed.  However, for a short peptide, one can generate an estimate, based upon models.  It would be nice to see how the estimates compare with experimental values.

[Rutherford]

Well your symbolism is a little off, but that is no big deal.  Can you find the form of the peptide with zero net charge?

Yes, where the N terminal is positive, C terminal negative, the Glu COOH group deprotonated and Lys NH₂ group protonated. Then?

[aHerraez]

List what the ionized / ionizable groups are, and figure out what their ionization state would be at each pH, until they sum up zero. It is an extension of the charged side chain aa problem, only take into account that the peptide bond removes some groups from the count.
?

[Babcock_Hall]

Well your symbolism is a little off, but that is no big deal.  Can you find the form of the peptide with zero net charge?

Yes, where the N terminal is positive, C terminal negative, the Glu COOH group deprotonated and Lys NH₂ group protonated. Then?

Now find the two pK_a groups on either side of this form of the peptide and average them.  My understanding is that this is only an approximation of the pI.

[Rutherford]

pKa of NH₃⁺ of Gly is 9.78 and pKa of COOH of Glu is 4.07. Therefore, pI=(9.78+4.07)/2=6.925, at this pH, all COOH groups are in COO^- state and all NH₂ groups are in NH₃⁺ state, the net charge is then zero. Did I do this correct?

Is the common pattern to take the lowest pKa of NH₃⁺ and the highest pKa of COOH group that is the part of an amino acid and to average these values? That way the net charge will always be zero. True?

[aHerraez]

I don't think that makes much sense. Why are you choosing Gly and Glu?

Put the pKa's in order and make out the charges:

pH	.___.	2.35	.___.	4.07	.___.	9.78	.___.	10.53	.___.
Ala	0	-0.5	-1	-1	-1	-1	-1	-1	-1
Glu	0	0	0	-0.5	-1	-1	-1	-1	-1
Gly	+1	+1	+1	+1	+1	+0.5	0	0	0
Lys	+1	+1	+1	+1	+1	+1	+1	+0.5	0
total	+2	+1.5	+1	+0.5	0	-0.5	-1	-1.5	-2

So the peptide has approx. charge +2 below pH=2.3, etc,
The pI will be midway between 4.07 and 9.78, yes, that's your calculation.
But if the peptide had 2 aminos and 3 carboxyls, then your rule would not be correct, as the charge between "lowest amino" and "highest carboxyl" would be negative.

Certainly this is not easily ported to a protein with hundreds of ionizable groups.

(values taken from http://www.cem.msu.edu/~cem252/sp97/ch24/ch24aa.html as an example)

[Rutherford]

Understood. Thank you very much.

[Babcock_Hall]

aHerraez,

If a peptide had three carboxylic acid groups and two amino groups, one would average the pK_a values of the second and third carboxylic groups to ionize.m  There may be some cases that work better than others.
http://biochem.uvm.edu/courses/files/205_fall_2003_lecture5supplement.pdf
http://www.elcamino.edu/faculty/pdoucette/Calculating-approximate-isoelectric-points.pdf