Buffer Compatibility

This non-empirical approach includes analysis of chemical structures of all buffer components. If bonds like C=O or C-N are present, the compound will most likely absorb in the analysis region and should be considered as potentially interfering. For example, examining the chemical structure of monocitrate, the main component of commonly used citrate buffer, easily allows its designation as an “interfering buffer” (Figure 1A). Similarly, analysis of chemical composition of Tris buffer (Figure 1B) reveals presence of C-N bond that will absorb in the Amide I region.



NOTE: Identification of potentially interfering buffer component does not guarantee that the buffer will interfere with accurate protein quantification. Analysis of chemical structure of individual buffer components should be taken as a warning that at certain concentration ratio between the protein (lower end of dynamic range) and the interfering component (varies from buffer to buffer) possibility to quantify accurately might be limited.

Differences in the pattern in which a spotted sample dries on the PTFE membrane and the buffer used to generate the standard curve may result in inaccurate quantitation. See the illustration of this “coffee ring effect.”

---

Given this possibility, spot 2 µL of the buffer onto the sample spot of an assay-free card and compare its drying pattern to the buffer used to generate the standard curve.

Next, examine the signal of this buffer spot using the Direct Detect® spectrometer (using empty spot as a blank) for a signal in the protein analysis region. All cases where buffer shows absorbance in the amide I region should be considered as potential interference.

NOTE: If a given buffer absorbs in the protein analysis region, then, at certain concentration ratios of the protein (lower end of dynamic range) and the interfering component (varies from buffer to buffer) it may not be possible to subtract buffer signal and quantify protein accurately. In such cases, perform buffer exchange using Amicon® Pro, Amicon® Ultra, or D-Tube™ devices for diafiltration/dialysis.

Detailed Methods of “Coffee Ring” Analysis

Characteristics of coffee ring formation by the analyzed buffers were assessed using five concentrations (0.05 M, 0.1 M, 0.2 M, 0.25 M and 0.5 M) of Tris-HCl, HEPES and sodium acetate. The lysis buffers were analyzed at ready to use (1x) concentration. Each concentration point was spotted, in duplicate, onto an assay-free card and cards were left at room temperature (RT) to dry.

2 µL of all investigated sodium acetate concentrations spread over entire sample spot and dried fully without any visible deposition pattern. The analysis of spots containing various concentrations of Tris-HCl buffer showed that starting from 0.25 M the buffer did not spread over the entire sample spot. Also, spots prepared using 0.5 M Tris-HCl solution did not dry fully even after 24 hours at RT. HEPES buffer showed limited spreading and drying problems starting from 0.25 M. All investigated lysis buffers spread fully and dried at RT.

Significant differences in spreading over the sample spot and drying pattern shown by analyzed buffers emphasized even further the requirement of single time system calibration for each matrix used in the sample preparation. Pre-loaded, PBS-based, curves have been generated at RT and 30% external humidity. The pre-loaded curves will deliver accurate protein quantitation if applied to samples solubilized in PBS and analyzed under the above conditions.

NOTE: While application of pre-loaded curves for quantitation from other buffers might return accurate results users are advised to calibrate the instrument with the appropriate buffer prior to analysis.

Tris (tris(hydroxymethyl)aminomethane or 2-Amino-2-hydroxymethyl-propane-1,3-diol) is extensively used as a component of buffer solutions in biochemistry and molecular biology. Tris contains C-N bond hence it absorbs in amide I region of MIR spectra and can interfere with protein quantitation by Direct Detect® spectrometer. In order to assess compatibility of Tris-HCl buffer with protein quantitation the stock (1M Tris-HCl, pH 8.0) was diluted with MilliQ® Water to produce 0.5 M, 0.25 M, 0.2 M, 0.1 M and 0.05 M solutions. Each concentration point was spotted, in duplicate, onto an assay-free card and cards were left at room temperature (RT) to dry. All spots, including the one that did not dry completely and did not spread over entire sample spot, were analyzed for MIR signal in amide I region after 24 hours from spotting. An empty membrane spot was used as a blank for all measurements. Tris buffer absorbs in amide I region of MIR spectra at all investigated concentrations, however the signal weakens significantly below 0.2 M.

In order to assess if Tris-HCl signal at amide I region can be subtracted successfully delivering accurate result of protein quantitation low concentrations of BSA, IgG and Protein A prepared by direct dilution of respective stocks to obtain 0.5 mg/mL solutions in 0.1 M Tris were quantified by Direct Detect® spectrometer delivering 0.547 mg/mL (BSA), 0.502 mg/mL (IgG) and 0.552 mg/mL (protein A).

BSA and Protein A at 0.5 mg/mL were also successfully quantified from 0.2 M Tris while attempts on quantitation of 0.5 mg/ml IgG solubilized in 0.2 M Tris-HCl failed, delivering unreliable and irreproducible results.

IgG differs from BSA and Protein A in respect to the secondary structure and the difference translates to a shift in a location of amide I band maxima. Amide I signal from IgG overlaps with Tris absorbance almost perfectly, while signals from BSA and Protein A are shifted toward higher wavenumbers. In consequence, the signal from low concentrations of IgG has been overwhelmed by presence of 0.2 M Tris-HCl buffer while absorbance from structurally different BSA and Protein A registered as a shoulder in a raw spectrum.

In case of low concentrations of BSA and Protein A in 0.2 M Tris-HCl, the buffer subtraction step resulted in spectra that could be properly integrated, while signal from the same concentration of IgG in the studied buffer could not be extracted reliably. Higher concentrations of IgG in 0.2 M Tris-HCl as well as all concentration of IgG, Protein A and BSA prepared in 0.05 M Tris-HCl were quantified accurately (data not shown).

Concentrations of Tris-HCl above 0.2 M were not investigated because the buffer spotted onto a PTFE membrane produced irreproducible coffee ring effect or/and did not dry.

In order to test compatibility of sodium acetate buffer with protein quantitation by Direct Detect® spectrometer 0.5 M, 0.1 M and 0.05 M solutions were tested. As shown, 0.5 M solution spotted on the assay-free card spread over entire sample spot and dried well. Analysis of buffer absorbance showed strong signal at around 1590 cm^-1 (just to the right of protein analysis region) with a tendency to overrun into the amide I area; however, the signal from each of investigated concentrations of acetate buffer was subtracted successfully delivering quantifiable Amide I signal.

Attempts to calibrate the system with 0.5 M and 0.1 M acetate showed that BSA and Protein A (used for all calibrations prepared for this report) aggregate in the buffer at lower concentrations. Both standards solubilized in 0.5 M acetate buffer started to aggregate if diluted below 3 mg/mL while they could be used in 0.1 M acetate down to 2 mg/mL. BSA has also aggregated in 0.05 M buffer while Protein A in such buffer was delivering reliable calibration down to 0.5 mg/mL.

Protein A based calibration in 0.05 M acetate delivered accurate quantitation results for several IgG samples (data not shown), conversely all attempts on quantitation of protein mixture delivered inaccurate results due to protein aggregation.

Interestingly, analysis of IgG concentrations prepared in 0.05 M acetate buffer against pre-loaded PBS-based calibration curves delivered accurate quantitation results suggesting strong similarities between coffee rings formed by both buffers.