Enhanced protein extraction and quantification protocol for microsamples: An ultra‐sensitive workflow for low‐volume, low‐concentration total protein lysates

Abstract

Accurate protein quantification at low concentrations and small volumes is critical for advancing small‐scale research, such as microvasculature studies. However, existing microscale protocols often require ≥5 μL of sample or highly concentrated lysates, limiting their applicability in contexts with scarce material. To overcome these limitations, we developed the Nano‐Extraction BCA‐Optimized Workflow (NEBOW), a novel method requiring only 2 μL of sample and capable of detecting protein concentrations as low as 0.01 mg/mL. Optimized for the NanoDrop™ One UV–Vis Spectrophotometer, this workflow demonstrated significantly enhanced sensitivity and reproducibility compared to the standard BCA assay. Paired t tests (p < 0.01) and TOST equivalence testing (15% margin) confirmed key differences, with the NEBOW method producing steeper standard curves and more consistent results at low concentrations. Bland–Altman analysis showed that standard BCA tends to overestimate protein levels, while the NEBOW method maintained accuracy across a range of low‐input samples. Western blot validation supported the improved performance of the new workflow. This approach offers a reliable, cost‐effective solution for protein quantification when sample availability is limited, without sacrificing accuracy or sample integrity.

1. INTRODUCTION

Accurate protein quantification of tissue lysate extract is essential in numerous experimental workflows, yet existing techniques often lack the precision needed when working with small sample volumes, small sample concentrations, or both. As protein research methodologies advance, the need for fast, reliable, cost‐effective, and precise quantification of small protein concentrations less than 1000 μg/mL and/or small sample volumes less than 5 μL has grown significantly.

The A280 method, which utilizes spectrophotometry, is widely regarded as one of the fastest and most reproducible techniques for protein quantification (Noble and Bailey 2009). However, it relies on the Beer–Lambert Law, which requires the use of extinction coefficients. This makes the method unsuitable for protein lysates containing mixtures of unknown proteins (Grimsley and Pace 2003; Hirte ²⁰¹⁶). In the setting of low protein concentration analysis of whole cell or tissue lysates, the 3‐(4‐carboxybenzoyl) quinoline‐2‐carboxaldehyde (CBQCA) assay is the current recommendation; however, it relies on cyanide, which is both toxic and expensive to dispose (Olson 2016). Previous studies have evaluated protein concentration determination assays using tissue extracts and reported inaccurate and non‐reproducible results with the Qubit assay, Bicinchoninic Acid (BCA), and Bicinchoninic Acid‐Reducing Agent Compatible (BCA‐RAC) methods (Broeckx et al. 2016). While mass spectrometry (MS)‐based protocols have been developed for low‐input samples, their high costs limit widespread and practical use in many research settings (Chun et al. 2011; Ludwig et al. ²⁰¹²; Merrell et al. ²⁰⁰⁴). As shown in Table 1, currently available methods for protein quantification require higher volumes of lysate (at least 5 μL and commonly 10–100 times that amount) than are commonly available in ultra‐low input samples. This limits the ability to perform appropriate technical replicates and additional experimentation. To this point, the gap in efficient, high‐precision protein quantification methods impedes progress in studies focused on small‐scale phenomena, highlighting the need for improved techniques to support advances in these areas.

TABLE 1.

Existing whole lysate micro protein quantification methods.

Method	Assay signal	Minimum sample volume	Working range	Reference
BCA	Colorimetric	150 μL	2–40 μg/mL	Micro BCA Protein Assay, ThermoFisher Scientific (Cat. No. 23235)
Bradford	Colorimetric	20 μL	100–1500 μg/mL	Bradford Protein Assay, ThermoFisher Scientific (Cat. No. 23200)
Lowry	Colorimetric	40 μL	1–1500 μg/mL	Pierce™ Modified Lowry Protein Assay Kit, ThermoFisher Scientific (Cat. No. 23240)
Biuret	Colorimetric	200 μL	150–1000 μg/ml	Biuret reagent, MilliPore Sigma (Cat. No. B3934)
Pyrogallol Red	Colorimetric	20 μL	10–2000 μg/mL	Total Protein Kit, Micro Pyrogallol Red Method, SIGMA® (Cat. No. TP0400)
NanoOrange	Fluorescence	10 μL	100 ng/mL–10 μg/mL	NanoOrange™ Protein Quantitation Kit, ThermoFisher Scientific (Cat. No. N6666)
CBQCA	Fluorescence	5 μL	50 ng–150 μg	CBQCA Plus Protein Quantitation Kit, ThermoFisher Scientific (Cat. No. A66522)

Note: A summary of existing protein quantification methods is provided, detailing the assay signal, minimum sample volume, working range, and corresponding reference information, including assay name, company, and catalog number.

In this manuscript, we describe a new method, Nano‐Extraction BCA‐Optimized Workflow (NEBOW), that requires ≤2 μL of sample total protein lysate that can detect concentrations as low as 0.01 mg/mL (10 ng/μL). By optimizing the protocol for NanoDrop™ One UV–Vis Spectrophotometer (Thermo Scientific™, Waltham, MA) compatibility, this method not only conserves valuable samples but also improves accuracy at low concentrations. We demonstrate the method's application on using human resistance arterioles (vessels approximately 75–150 μm internal diameter, and 4–7 mm in length), where protein extraction and quantification (commonly yields <20 μL volume with concentrations around 1 mg/mL) would otherwise be prone to inaccuracies using pre‐existing methods, while preserving lysate sample volume for additional experimentation.

2. RESULTS

2.1. Significant reduction of sample volume used for NEBOW

NEBOW demonstrated the ability to accurately quantify protein lysates using as little as 2 μL of experimental sample. To validate this, protein concentrations were measured using NEBOW with 2, 5, and 10 μL sample volumes from the same set of five samples (Table 2). Across all tested volumes and dilution factors, NEBOW produced consistent concentration readings, with a coefficient of variation below 15% in all cases. Additionally, the standard deviation between experimental volumes did not exceed 0.023 mg/mL, confirming the method's precision and reliability even with minimal sample volumes.

TABLE 2.

Comparison of NEBOW protocol performance using 2, 5, and 10 μL experimental sample volumes.

Sample	DF	Concentration using 10 μL (mg/mL)	Concentration using 5 μL (mg/mL)	Concentration using 2 μL (mg/mL)	Av. (mg/mL)	SD (mg/mL)	%CV
1	10	0.288	0.294	0.331	0.304	0.023	7.65
	100	0.056	0.056	0.052	0.055	0.002	4.22
2	5	0.132	0.136	0.15	0.139	0.009	6.78
	10	0.111	0.115	0.118	0.115	0.004	3.06
	100	0.009	0.009	0.007	0.008	0.001	13.9
3	5	0.349	0.358	0.373	0.36	0.012	3.37
	10	0.313	0.329	0.339	0.327	0.013	4.01
	100	0.057	0.059	0.059	0.058	0.001	1.98
4	5	0.101	0.104	0.123	0.109	0.012	10.9
	10	0.071	0.075	0.082	0.076	0.006	7.33
5	5	0.104	0.118	0.111	0.111	0.007	6.31
	10	0.073	0.076	0.077	0.075	0.002	2.76

Note: Experimental samples were diluted and analyzed using the NEBOW protocol, with concentrations determined from 2, 5, and 10 μL sample volumes (n = 3). The average (Av.) (mg/mL), standard deviation (SD), and coefficient of variation (%CV) across all volumes are reported.

2.2. NEBOW provides larger dynamic range and increased sensitivity

To evaluate accuracy, standard curves were generated for both the gold standard BCA method and NEBOW using known concentrations of bovine serum albumin (BSA) (Figure 1). The NEBOW method exhibited a significantly higher slope compared to the gold standard BCA method, as determined by simple linear regression of the best‐fit values. The slope of the NEBOW standard curve was 5.784, whereas the gold standard BCA method yielded a slope of 0.1589. This indicates enhanced specificity in quantification across a broader concentration range, particularly at lower protein concentrations (<1 mg/mL). These findings highlight the superior suitability of NEBOW for accurately measuring low‐concentration samples.

FIGURE 1.

Dynamic range evaluation of the gold standard BCA protein quantification method compared to NEBOW. A standard curve was constructed using BSA of known concentration (mg/mL) for both the NEBOW and gold standard BCA protein quantification methods (n = 8). The NEBOW method was measured at an absorbance of 562 nm, while the gold standard BCA method was measured at 650 nm. The best‐fit line for the gold standard BCA method is represented by the equation y = 0.1589x + 0.06079, while the best‐fit line for the NEBOW method is represented by y = 5.784x + 0.09772.

2.3. Gold standard and new method are not equivalent in quantifying protein

A comparison of protein lysate concentrations reported by the gold standard BCA method and NEBOW was conducted using both a two one‐sided t test (TOST) equivalence test and a paired t test. The TOST equivalence test revealed that the two methods were not equivalent, with a non‐inferiority margin of 15%. Specifically, NEBOW did not fall within the range of 0.16 to 0.22, which would be required for equivalence to the gold standard BCA assay. The total sample (n = 56) had an average concentration of 0.20 ± 0.21 mg/mL, while the gold standard BCA assay (n = 28) reported 0.26 ± 0.26 mg/mL and NEBOW (n = 28) reported 0.13 ± 0.11 mg/mL (p = 0.82). The paired t test confirmed that the two methods produced significantly different concentration values (p < 0.01).

Results from both tests indicated that the two methods produced significantly different concentration values. Agreement between methods was further assessed using a Bland–Altman plot (Figure 2). This analysis reveals that NEBOW consistently reports lower protein concentrations than BCA for the same low‐input samples, with variability in the over‐estimation of protein concentrations reported by BCA. The average of the mean bias of these samples is 0.12 mg/mL, which represents a significant over‐estimation bias by BCA based on the average concentration of these samples that are in this range. There also appears to be greater variability in the bias as the difference between concentrations measured by the two methods increases.

FIGURE 2.

Bland–Altman plot for arteriolar tissue protein lysate concentration measurements between methods. The plot displays the difference between the new (NEBOW) and gold standard BCA methods versus the average concentration (mg/mL) of protein lysates. Data points outside the red dotted lines indicate non‐equivalency between the two methods.

2.4. Validation of efficacy of NEBOW

We provided another laboratory at our institution with the NEBOW protocol for arteriolar protein extraction for use in western blotting. Although they used the newly described extraction method, they initially quantified the lysates using the “Gold‐Standard” Bio‐Rad DC™ BCA assay with a microplate reader. For immunoblotting, 30 μg total protein lysates were loaded into the gel for electrophoresis, and GAPDH (GAPDH (D16H11) XP® Rabbit mAb, Cell Signaling #5174, 37 kD, 1:1000) was probed for. After imaging the blot, it became apparent that the band intensity was not equal across samples, and the group then opted to re‐quantify the protein lysates using NEBOW. Figure 3 shows the western blot probed for GAPDH among six samples with the amount of protein loaded as quantified by both methods. It is clear that the band intensity closely follows the protein amounts loaded, as calculated by the NEBOW method, and was not 30 μg equally loaded, as quantified by the gold standard BCA method (Figure 3).

FIGURE 3.

Arteriolar total protein lysate concentrations measured between methods using western blotting. Protein lysates, assumed to contain 30 μg protein in each well based on gold standard BCA method, were also measured by NEBOW method and probed for GAPDH in western blotting. Mean band intensity from the Western blot was quantified to assess the relative accuracy of each quantification method. ND, not determined.

3. DISCUSSION

Protein quantification in microvessels presents significant challenges. Protein extractions from these small structures typically yield volumes of 20 μL or less, with an average concentration of 1 mg/mL. Accurate quantification of protein at low concentrations and with minimal sample volumes is increasingly critical in various applications, including microvessel studies and other studies with low sample availability, commonly in studies of human tissues (such as those obtained from organ biopsies). The new method, NEBOW, addresses this need by providing a rapid and precise quantification process that requires ≤2 μL of sample (Table 2) which yields at least 2 technical replicates, a significant advantage over traditional approaches. Specifically, compared to the conventional Bio‐Rad DC BCA method, the new method demonstrates improved sensitivity at low concentrations (Figures 1 and 2), enabling better differentiation and yielding more accurate concentration values. This specificity and minimal sample requirement make it highly suitable for applications in which traditional methods may lack accuracy due to low overall protein content or consume excessive amounts of precious sample for other experiments. In particular, our new method appears to work better at protein concentrations below 1 mg/mL.

Most of the commonly used quantification methods demand at least 5 μL per measurement, often more (Table 1), which depletes the lysate and leaves little material for replicates or further experimentation. This issue is potentiated by protocols that necessitate higher protein concentrations to reduce sample volumes, creating a trade‐off that impairs feasibility. The exhaustive use of lysate during quantification also restricts subsequent analyses. Protein quantification methods are abundant, with numerous protocols and modifications available, including techniques such as quantitative 2‐D gel electrophoresis combined with MS or MS alone. However, these approaches are often limited by high costs, reduced precision at low protein concentrations, or both issues (Björhall et al. 2005; Hamdan and Righetti ²⁰⁰²; Lill ²⁰⁰³; Maaß and Becher ²⁰¹⁶). More commonly used methods for total cell lysate quantification include the Bradford, Lowry, Biuret, and BCA assays. While effective in many scenarios, these techniques can struggle with small sample volumes and low protein concentrations, often exhibiting limitations in dynamic range, reproducibility, and accuracy (Kruger 2009; Litovchick ²⁰¹⁸; Noble et al. ²⁰⁰⁷; Okutucu et al. ²⁰⁰⁷; Sapan et al. ¹⁹⁹⁹). The newly developed NEBOW method addresses these challenges by providing a cost‐effective, rapid, accurate, and reproducible solution specifically designed for small‐volume, low‐concentration protein lysates.

The NEBOW method has limitations, including sensitivity to interfering substances, necessitating the use of the solvents specified in the protocol for compatibility with the NanoDrop One spectrophotometer. Additionally, at very high protein concentrations, the method loses accuracy and precision. As a result, NEBOW is best suited for samples with known low protein concentrations, while preexisting methods such as the BCA assay are recommended for higher protein yields (Osnes et al. 1993; Wanandy et al. ²⁰²⁴).

Despite these limitations, the NEBOW method effectively addresses a critical gap in protein quantification techniques by offering a reliable solution for small‐concentration protein lysates. With its minimal sample input requirement, rapid processing time, and high accuracy, NEBOW represents a significant advancement for researchers working with limited sample volumes.

4. METHODS

4.1. Tissue acquisition and dissection

Microvessels were obtained from consenting human samples, primarily derived from subcutaneous adipose tissue from the Medical College of Wisconsin Tissue Bank under approved Institutional Review Board guidelines. Microvessels of approximately 75–150 μm in diameter and 4–7 mm in length were isolated, flash frozen in liquid nitrogen, and stored at −80°C. Some samples consisted of vessel‐derived endothelial cells (VECs) that were scraped from the arteriolar lumen with a scalpel, flash frozen in liquid nitrogen, and stored at −80°C.

4.2. Protein whole lysate extraction

Vessel or VEC samples were kept on dry ice until removed from the storage tube for protein extraction. Throughout the extractions, reagents and tubes were kept on ice for the duration of the protocol. Previously, the gold standard method used Radioimmunoprecipitation assay (RIPA) buffer (Lysis Kit‐RIPA Buffer for Charge Assays, Aqueous Protease Inhibitor, Dimethyl Sulfoxide Inhibitor Mix, ProteinSimple, San Jose, CA) for lysis and extraction; however, there was significant interference with the wavelengths at which NanoDrop One reads absorbance. Therefore, all samples were extracted using the same lysis buffer, Tissue‐Protein Extraction Reagent (T‐PER™), composed of 100 μL T‐PER, 1 μL Halt Protease Inhibitor Cocktail, and 1 μL Ethylenediaminetetraacetic Acid (EDTA) (Thermo Scientific™, Waltham, MA). Dependent on vessel size and quantity, 10–20 μL of lysis solution was transferred to an Omni International ceramic bead tube (1.4 mm, 0.5 μL, Revvity, Waltham, MA).

The microvessel was added to the bead tube directly into the lysis solution using clean and dry, fine forceps. The remainder of the lysis solution was saved for use in the quantification assays as a reference control. The sample was lysed and homogenized with the Omni International Bead Ruptor 12 (Revvity, Waltham, MA) with parameters as follows: Speed: 6.0 m/s, Mill Time: 45 s, Dwell Time: 20 s, Cycles: 3. Further protein extraction was executed with a cold incubation on ice for 30 min with brief vortexing and quick spin every 5 min. Lysates were carefully transferred to new 1.7 mL tubes and were centrifuged at 4°C at 14,000g for 20 min. Following centrifugation, the supernatant was transferred to a new 1.7 mL tube without disturbing the pellet debris. The protein extracts were either flash‐frozen in liquid nitrogen and stored at −80°C for later use or directly used in the quantification assays.

4.3. Protein quantification via bicinchoninic acid assay

During the protein quantifications using two different bicinchoninic acid (BCA) methods, the reagents and tubes were kept at room temperature per manufacturer's instructions. Samples were kept on ice and were prepared with multiple dilutions in water which may include undiluted (1:1), 1:5, 1:10, 1:100, and 1:1000, depending on volumes or predicted concentration range. BSA standards were prepared according to Tables 3 and 4 for the gold standard and new microassay (NEBOW), respectively.

TABLE 3.

BSA standards for the DC™ BCA assay.

Tube	Volume of diluent	Volume of BSA	Final [BSA] (mg/mL)	Notes
A	0 μL	40 μL	2.0
B	20 μL	20 μL of 2.0 mg/mL BSA (Tube A)	1.0
C	20 μL	20 μL of 1.0 mg/mL BSA (Tube B)	0.5
D	20 μL	20 μL of 0.5 mg/mL BSA (Tube C)	0.25
E	20 μL	20 μL of 0.25 mg/mL BSA (Tube D)	0.125
F	20 μL	20 μL of 0.125 mg/mL BSA (Tube E)	0.0625
G	20 μL	20 μL of 0.0625 mg/mL BSA (Tube F)	0.03125
H	20 μL	20 μL of 0.3125 mg/mL BSA (Tube G)	0.015625
I	40 μL	0 μL	0	Blank (Water)
J	0 μL	0 μL	0	Lysis Buffer (TPER)

Note: The volume of water and BSA in each standard tube was prepared using a serial dilution for concentration extrapolation, including controls: a blank composed of molecular‐grade water and a reference composed of TPER. After dilution, reagents are combined, and measurements are obtained with a microplate reader.

TABLE 4.

BSA standards for NEBOW method.

Tube	Volume of diluent	Volume of BSA	Final [BSA] (mg/mL)
A	20 μL	20 μL of 2.0 mg/mL BSA	1.0
B	20 μL	20 μL of 1.0 mg/mL BSA (Tube A)	0.5
C	20 μL	20 μL of 0.5 mg/mL BSA (Tube B)	0.25
D	20 μL	20 μL or 0.25 mg/mL BSA (Tube C)	0.125
E	20 μL	10 μL of 0.125 mg/mL BSA (Tube D)	0.03125
F	20 μL	10 μL of 0.03125 mg/mL BSA (Tube E)	0.0078125
G	20 μL	20 μL of 0.0078125 mg/mL BSA (Tube F)	0.00390625

Note: Standard tubes containing serial dilutions of water and BSA were prepared for concentration extrapolation. After dilution, the working reagent was added, and measurements were obtained using the NanoDrop One instrument, with TPER serving as the reference.

4.3.1. Gold‐standard method: DC™ BCA assay (Bio‐Rad Laboratories, Des Plains, IL)

A BSA standard curve was prepared each time the assay was performed according to the standard dilution series listed in Table 3. Samples were diluted with molecular‐grade water as outlined above, based on availability. DC™ BCA Reagent A was prepared in a new 1.7 mL tube consisting of 1 mL Reagent A (alkaline copper tartrate solution) and 20 μL Reagent B for approximately every 40 wells in the 96‐well assay microplate. In a clean, dry 96‐well assay microplate, 5 μL of standards and samples were pipetted into each well in triplicate, along with 25 μL of Reagent A and 200 μL of Reagent B (Folin Reagent), which were mixed by gentle agitation on the benchtop. After 15 min, the absorbances were read at 650 nm on the SpectraMax iD3 microplate reader (Molecular Devices, San Jose, CA). Instrument absorbance readings underwent quadratic calibration for relative protein concentration derivation (Hogan 2003).

4.3.2. New low input method: NEBOW

Standards were diluted in a 7‐point series using 2 mg/mL of pure BSA (Pierce™ BCA Protein Assay Kit) and water (Table 4). The working reagent was combined using the Pierce™ BCA Protein Assay Kit reagents in a 50:1 ratio of Reagent A to Reagent B. In a fresh tube, the working reagent and samples were mixed in a 1:1 ratio.

Assessment of minimal sample volume required

The BCA microassay on NanoDrop One protocol supplementary document (T138 Rev. 28 June 2017; Thermo Scientific™, Waltham, MA) instructed users to combine 10 μL of the reference, standards, or samples with 10 μL of the working reagent. Since the protocol would rapidly deplete the sample volume, we aimed to reduce the sample volume input significantly. While utilizing the 1:1 working reagent to sample ratio, we used volumes of 10, 5, and 2 μL, if available, and measured concentrations on the NanoDrop One.

NEBOW utilizing 2 μL sample volume

For all samples, we added 2 μL of sample (undiluted and/or diluted) and 2 μL of working reagent (1:1). For diluted samples, the required sample volume can be significantly reduced (e.g., 0.5 μL of sample combined with 4.5 μL of water). For the reference control (T‐PER™ lysis solution) and BSA standards, we added 10 μL of the reference or diluted standards to 10 μL of working reagent, per manufacturer's instructions. The sample‐reagent BCA mixtures were vortexed, spun down, and followed by a 30‐min incubation in a 37°C water bath. After the incubation period was completed, the samples were vortexed and quick‐spun to obtain a homogeneous mixture. The assay controls (blank: molecular‐grade water (No BCA), reference: T‐PER™, and the standard curve: a 7‐point series of dilutions of BSA) and samples were mixed, and measurements were made with NanoDrop One on the BCA Assay tool setting within 10 min. For all measurements made on the NanoDrop One, 2 μL volumes were used with up to three replicates for the reference and standard curve samples. Absorbance values were manually recorded for the 562 nm wavelength of reference and standard samples, as they are not reported in NanoDrop One data files. This was useful for the re‐calculation of concentrations in case of outliers or technical replicate removal from the reference or standard curve series to improve quantification of sample concentrations.

4.4. Immunoblotting

Lysates were quantified using the gold standard Bio‐Rad DC™ BCA assay with a microplate reader and the newly described extraction method utilizing T‐PER™. For immunoblotting, 30 μg total protein lysates were loaded into the gel for electrophoresis, and Glyceraldehyde‐3‐phosphate dehydrogenase (GAPDH) (GAPDH (D16H11) XP® Rabbit mAb, Cell Signaling #5174, 37 kD, 1:1000) was probed for. Western blot band intensity was quantified using ImageJ2 (version 2.12.0/1.54f ).

4.5. Statistical analyses

Summary statistics of the gold standard and the new standard were reported using mean and standard deviation. These two standards were then compared using the TOST equivalence test to assess if the two standards could be considered equivalent to each other. We also ran a paired t test to assess if the two standards were significantly different. We then created a Bland–Altman plot to further assess the equivalency of the two standards by taking the average difference between the two and plotting these data points. The confidence range to be considered equivalent was created using the 1.96 critical value multiplied by the standard deviation of differences.

AUTHOR CONTRIBUTIONS

Taylor Wilcox: Writing – original draft; formal analysis; data curation; writing – review and editing; visualization; conceptualization; investigation. Michael E. Widlansky: Conceptualization; funding acquisition; supervision; resources; writing – review and editing. Justin Westhoff: Investigation; writing – review and editing. Jingli Wang: Investigation; writing – review and editing. Rong Ying: Investigation; writing – review and editing. Abigail Thorgerson: Formal analysis; writing – review and editing. Michelle L. Roberts: Conceptualization; investigation; methodology; validation; writing – review and editing; formal analysis; data curation; supervision.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

Wilcox T, Widlansky ME, Westhoff J, Wang J, Ying R, Thorgerson A, et al. Enhanced protein extraction and quantification protocol for microsamples: An ultra‐sensitive workflow for low‐volume, low‐concentration total protein lysates. Protein Science. 2025;34(6):e70161. 10.1002/pro.70161

DATA AVAILABILITY STATEMENT

Data sharing is not applicable to this article as no new data were created or analyzed in this study.