Hi Jerry,

to summarise your problem, using (close to) physiological buffer, SPR
and ITC give you different results, you get different results in
different salt strengths and to add to your misery, the proteins
precipitate at low salt concentrations when mixed to together.

Ok.

Given the above, your interaction probably has a large electrostatic
component, and this is why the Ks are salt sensitive. This is also
possibly why you are getting precipitate. If you start with two happy
proteins, and then titrate them in together, the (presumably)
complementary electrostatic binding surfaces will interact and cancel
each other out, reducing the overall charge on the complex. If the
complex is less charged, it will be less soluble, therefore, the
complex crashes out (in 1:1 stoich).

IMHO, ITC is the more elegant experiment. Ok, so it uses *more*
material, but you aren't relying on binding surfaces to nail things
to, and you are directly(ish) measuring the heat of the interaction.

If, in an ITC cell, two proteins come together and are insoluble, they
are able to precipitate. On an SPR chip, they are already immobilised
to a surface, and so you wouldn't necessarily detect precipitation
forming, also, as you run SPR at much lower concentrations of protein
you might not induce precipitation if it is protein concentration
dependent. If I see precipitate in my sample after an ITC experiment,
I'm always weary of it - but at least I'm aware of it.

I would always choose to run an ITC experiment over an SPR (ideally
both), but sometimes, ITCs requirements for higher concentration means
that it isn't always feasible. In this case, the solubility limit of
your complex may prevent you getting good ITC data.

I would try and keep buffers consistent between your experiments
(stick to physiological salt strength - less awkward reviewer
questions), and try repeating your experiments at different pHs. I've
had complexes that precipitate at pH 7.5 & 9.5, but are nice and happy
at pH 5.5.

Hope this (rather lengthy reply) helps!

Dave


On 23/01/2008, Jerry McCully <[log in to unmask]> wrote:
>
>  Dear All:
>
>          Recently I am pursuing the crystallziation of a complex formd by
> two individual proteins and I met several interesting problems though  they
> are kind of off-topic.
>
>          Any suggestions for these problems will be highly appreciated.
>
>          BIAcore showed about submicromolar affinity(both Kinetic and
> steady-state fitting) for these two proteins in the complex. However,
> precipitates immediately appeared when these two proteins were mixed
> together even at 10uM(<0.3mg/ml) concentration in the condition of low
> salt(less than 20mM NaCl).
>  By the way, these two proteins completely precipitated when the molar ratio
> is 1:1 in this condition.
>
>   THerefore, I increased the salt concentraion step by step and finally I
> can keep both of them soluble in the solution with 25mM Tris(pH8) and 60mM
> NaCl(the minimum of salt concentration).   Wierd thing happened when ITC
> experiments were carried out to confirm the binding affinity.  20uM in the
> sample cell and 200uM in the syringe could not give enough heat for a good
> curve fitting. The optimistic estimation of the affinity is lower than 5uM,
> which is much lower than the affinity given by BIAcore in the same
> buffer(25mM Tris plus 150mM NaCl).
>
>        Now I am suspecting the capability of the interaction between these
> two proteins. However, I can not explain why these two guys precipitated
> stoichiometrically if they do not interact with each other.
>
>        Is the complex salt-sensitive therefore there was just minor binding
> in the high-salt condition revealed by ITC?
>
>         I am planning to do the ITC again in the condition of 25mMTris and
> 60mM NaCl.
>
>         What if the affinity given by ITC is still much lower than that by
> BIAcore. Which one should I choose to believe?
>
>        Are there some better ways that  I can validate the binding affinity?
>
>
>       Thanks again for your great ideas.
>
>  Jerry McCully
>
>
>
>
>
>
> ________________________________
> Need to know the score, the latest news, or you need your Hotmail(R)-get your
> "fix". Check it out.


-- 
============================
David C. Briggs PhD
Father & Crystallographer
http://www.dbriggs.talktalk.net
AIM ID: dbassophile
============================