pH too acidic in coarse phase diagram protocol

[JoeRadde]

The pH was measured to be in a range of 1-3 in the first run of the coarse phase diagram protocol (10.02.22, EWSRI 20 µM). The pH was expected to be pH 6.5 and 7.5 instead.

[JoeRadde]

Since the pipette aspirates and dispenses at the very bottom of the respective well, I assume that an insufficient mixing of HCl with H2O was the reason for this issue. The problem seemed resolved when we vortexed the acidic solution manually (11.02.22, FUS 48 µM).

[JoeRadde]

I am still not happy with the pH-ing of the buffer solutions using the robot.

So far, I tried 2 approaches. In both approaches, I have two buffers at pH 8.5, one with and one without KCl. To test for different salt concentrations, the robot mixes these buffers in according volumes. In order to test for pH 6.5 and 7.5, too, I used HCl to lower the pH. Here the two approaches that I followed:

1. Pipetting 1 µl of an HCl solution to each destination well

In this approach, the robot prepares two different HCl dilutions. The concentrations are calculated via the Henderson-Hasselbalch equation in a way that when the robot adds 1 µl of these dilutions to the final wells, we get pHs of 6.5 and 7.5, respectively.

I find this approach to be error-prone due to the direct addition of the pipette's minimum volume of HCl to each well. There is also no easy way to check the pH of the wells.

2. Gradually lower pH of buffer stock solutions

Here, the robot pipettes the pre-pH-ed buffers of pH 8.5. After finishing that, the robot takes a specific volume of 2.5 % HCl dilution and adjusts the pH of the remaining stock buffer solutions to pH 7.5 and distributes it to the respective wells. In a final step, it lowers it further to pH 6.5 and distributes it to the rest of the wells.

In the latter approach, we can check the pH easily before distributing it to the wells. In fact, I have tested this, but the pH varies from 3- to 7-ish, and 6- to 8-ish across the buffers of different KCl concentration.

The (main) problem:

TRIS has a good buffering capacity from 7 to 9. Since pH 6.5 is outside of that range, even a small excess of HCl will lower the pH drastically. When within the buffering range, we have more room of error.

Other approaches to think about

• We could start off using buffers that are adjusted to an pH of 6.5 and then leave the robot to increase the pH using one of the approaches mentioned above using KOH.
• We could prepare all buffer solutions manually. This, however, kind of defeats the purpose of using the robot because we would need to prepare 12 buffers if we wanted to test 3 different pHs and 4 different salt concentrations.
• We could introduce a different buffer, but that would potentially have great consequences regarding comparison to the past experiments and future purifications.

Do you guys have any thoughts on what approach is best or how we could check for pH efficiently for the first approach?

[majahn]

Hi @JoeRadde,

yes, this is going to be more complex, as you are pointing out. I remember that Titus used PIPES buffer when looking at the influence of acidic solution on the phase behaviour of SUP35. So, you are right, we won't be able to do this with Tris...

buffer_selection_guide.pdf

Here is a guide to the different buffer systems.

[majahn]

Have a look at the paper: Phase separation of a yeast prion protein promotes cellular fitness, Franzmann, 2018. There the experiments at pH 6.5 are described...

One thing I'm wondering is whether one could prepare two balanced buffer solutions at the extremes, like PIPES pH 6.5 and TRIS pH 8.5 and then mix them in different ratios. Would you be able to get a stable solution at pH 7.5 for example?

I haven't seen buffer mixing experiments. Do they exist? Or is there a good biochemical reason not to mix buffers, like TRIS and HEPES and PIPES?

[majahn]

For the screening purpose, we could prepare a subset of buffers manually and then use them first to see in which direction the effect would have the greatest influence. I think this first step does not have to be super accurate. But you are right, outside the buffer range, one could easily go from pH 7 to pH 3 with little change in composition... Tricky...

[JoeRadde]

Hi Marcus, thanks for the input!

I looked at the paper you mentioned. They used a phosphate buffer that has a buffering range from pH 5.8 to 7.4 (also around pH 2 and 12).

I don't know of any buffer mixing experiments, either. Mixing buffers is an intriguing thought, but I feel like it's complicated and can get out of hand easily. Also, when mixing different buffers we would introduce another variable that might affect our experiment. The calculation of a multi-buffered solution is described by a mathematical model in this paper, where they for example mixed HEPES and MES and got a pretty nice, almost linear pH curve from pH 4.5 to 8.5.

We could, indeed, prepare the buffers manually. Another idea that came to my mind: We could prepare three TRIS buffers of pH 6.5, 7.5 and 8.5 with 0 KCl each. That way, we'd have done the hard part manually. Instead of then mixing a high-salt buffer with a low-salt buffer, we could then use the robot to add a small amount of highly concentrated KCl solution (e.g. 1 µl) to achieve distinct KCl concentrations.

[majahn]

Hi,

interesting paper, but this seems to easily get quite complicated. This could be good in some sense because that could be an edge, but I would not go there right now.

Let's first try the simpler version with the 3 starting buffers without KCl. This is relatively straightforward and would bring us a step further with the TUFU project...

The detailed pH mixing we could then develop once we know which direction away from neutral is the interesting one...

Sounds like a good plan!

Happy Pipetting

[majahn]

One more thought: at some point, we could try to use molecular pH sensors to measure the pH by fluorescence. Don't know how this would affect the phase profile and or imaging, but something like LysoSensor could be useful:

https://www.thermofisher.com/de/de/home/references/molecular-probes-the-handbook/ph-indicators/probes-useful-at-acidic-ph.html

[majahn]

Hier z.B. in Gelb/Blau: https://www.thermofisher.com/order/catalog/product/L7545

Pretty expensive, so we should be sure that it's useful... ;-)