Negative Response Due to Analyte Injections

To give you some background, I am looking into interactions between 1 cell receptor and various bioconjugates (containing peptides that have been shown to interact with this receptor). For a few of these hydrophilic samples, I am able to dilute into my HBS/0.5 mg/mL BSA running buffer and obtain nice binding curves.

However, I am having tremendous difficulty working with hydrophobic peptide bioconjugates. In these cases, I am not able to dissolve directly into running buffer after dialysis/lyophilization (with or without BSA). However, these samples which are initially prepared in DMSO will go in if I dilute ~100x into running buffer/BSA (1 vol% DMSO). To attempt to account for this, I have been switching my running buffer to HBS/BSA/1% DMSO after ligand immobilization. The issue with this is that, I get sharp peaks and sudden dips into negative RU (probably slight buffer mismatch??). I have repeated this experiment, trying to be attentive to this problem and keeping the buffers as consistent as possible, but I am still running into issues. I made a large batch of the buffer/DMSO and took out a small volume that I can use if needed during the experiment just so I'm not opening and closing the lid of the actual running buffer bottle too much which may cause discrepancies.

TLDR: That'd be great if you could confirm if you think the two curves shown below are due to buffer mismatch. If so, what can be done to improve my experimental procedure? If possible, I'd like to stay away from dialyzing into running buffer since the concentration becomes uncertain.

Hi,
Not sure I can be of help. You did most of the things I would do to get things working.
One possibility is that the hydrophobic compounds are not properly dissolved any more when using 1% DMSO. Maybe they aggregate and induce some matrix collapse which results in a negative curve.
You could try higher DMSO concentations.

Kind regards
Arnoud

Thank you for your response, Arnoud. Unfortunately, I need to adhere to 1% DMSO because of cell culture experiments. Perhaps I can decrease the amount of peptide I have on the polymer.

So I suppose the odd response I'm seeing is not characteristic of a buffer mismatch?

In the raw data, is the reference channel also going down during analyte injection? The figure you showd is that after reference subtraction? sometimes the reference is binding more and will give a negative curve after referencing. Maybe you can inject the same DMSO dilution in buffer (without compound) to check if the reference is responding the same way to the injection as your polymer surface.
Arnoud

Actually when I look at the raw data the reference channel is not decreasing during analyte injection. It is actually increasing in a similar fashion as the experimental channel. I did do that control that you mentioned of injecting DMSO diluted in running buffer (no compound) and saw that the reference and experimental channels responded similarly, showing no significant deviation from baseline.

I think this would actually indicate that there was non-specific binding to the sensor chip, right? In my case, I did a BSA injection in the reference channel and used HBS buffer with BSA that worked well for the peptide. It could be that the bioconjugate containing peptide could have different characteristics that I'm not accounting for