A western blot detects a specific protein within a complex mixture by separating proteins by size, transferring them to a membrane, and then using antibodies to find the one you care about. The workflow has been a staple of molecular biology labs for decades, and while the core steps are straightforward, each one has decision points that can make or break your results. Getting clean, interpretable bands depends less on any single clever trick and more on attention to detail across every stage, from how you lyse your cells to how you image the final blot.
Preparing Your Protein Samples
Everything starts with extracting proteins from your cells or tissue. You lyse the cells open using a detergent-based buffer, and the choice of buffer matters depending on what you are looking for. A standard RIPA buffer works well for most targets, but if you are trying to detect phosphorylated proteins, you need to add phosphatase inhibitors to your lysis buffer to prevent those modifications from being stripped off before you even run your gel.1PubMed Central. Western blot analysis Protease inhibitors are also essential in virtually every case, since endogenous proteases will start chewing up your proteins the moment the cell membrane is disrupted.
After lysis, you spin down the debris and collect the supernatant. At this point, you need to know how much total protein you have, so you run a protein assay (a BCA or Bradford assay) on the lysate. This lets you load equal amounts of protein in each lane of your gel, which is critical for any downstream comparison. A common loading amount is somewhere between 10 and 50 micrograms of total protein per lane, though the ideal amount depends on how abundant your target is. Overloading the gel creates distorted, smeary bands; underloading means your target may be invisible.
Before loading, you mix your protein sample with a loading buffer containing a detergent and a reducing agent, then heat the mixture. This denatures the proteins, unfolding them into linear chains coated with negative charge so they will migrate through the gel based on size alone rather than their native shape or charge.
Separating Proteins by Size With SDS-PAGE
The denatured proteins are loaded into wells of a polyacrylamide gel and separated by applying an electric field. Smaller proteins move through the gel matrix faster than larger ones, so by the time you stop the run, proteins are arranged from smallest (at the bottom) to largest (near the top). The percentage of acrylamide in the gel determines its pore size: a higher percentage gel (say 12–15%) resolves small proteins well, while a lower percentage gel (6–8%) is better for large proteins. Gradient gels that transition from low to high percentage offer a wider separation range and are a convenient default when you are looking at multiple targets of different sizes on the same blot.
You always include a molecular weight marker (a pre-stained protein ladder) in at least one lane. This gives you reference bands of known sizes so you can confirm your protein of interest is migrating where it should. Running the gel too fast generates excess heat and can distort bands, so a moderate, steady voltage is preferable. You typically run until the dye front in the loading buffer reaches the bottom of the gel, or until the region around your target’s expected size is well resolved.
Transferring Proteins to a Membrane
Once proteins are separated in the gel, they need to be moved onto a solid membrane where antibodies can access them. This transfer step is done by sandwiching the gel against a membrane and applying an electric field perpendicular to the gel, which drives the negatively charged proteins out of the gel and onto the membrane surface. The two main membrane options are nitrocellulose and PVDF, and the choice is not arbitrary.
Nitrocellulose binds low molecular weight proteins more effectively than PVDF, while PVDF has an advantage for high molecular weight proteins and glycoproteins. For proteins in the middle range, the two membranes perform similarly.2PubMed Central. Comparison of the sensitivity of Western blotting between PVDF and NC membranes PVDF also has higher overall protein-binding capacity and is more physically durable, which matters if you plan to strip and reprobe the blot later. Nitrocellulose, on the other hand, tends to produce lower background, which can be a real advantage for clean images.
Transfer can be done by wet (tank) transfer, semi-dry transfer, or newer rapid-transfer systems. Wet transfer is the most traditional and forgiving approach, especially for large proteins that are hard to coax out of the gel. Semi-dry systems are faster and use less buffer, but can be finicky with very large or very small targets. Whichever method you choose, incomplete transfer is one of the most common reasons a western blot fails, so it is worth confirming transfer efficiency. Many labs stain the membrane briefly with Ponceau S, a reversible red dye that shows all transferred protein bands, before proceeding.
Blocking the Membrane
After transfer, the membrane has proteins stuck to it in the pattern the gel created, but it also has lots of empty space where the antibody could bind nonspecifically and generate background noise. Blocking fills those empty spaces with an inert protein or polymer so that only your target-specific antibody sticks where it should.
The classic blocking agent is nonfat dry milk, sometimes called BLOTTO (Bovine Lacto Transfer Technique Optimizer), a name coined in the 1984 paper that introduced it. That study showed milk was better than bovine serum albumin or gelatin at preventing nonspecific binding and did not require detergents or chaotropic agents to reduce background.3Gene Analysis Techniques. Improved technique utilizing nonfat dry milk for analysis of proteins and nucleic acids transferred to nitrocellulose Milk remains the go-to blocker in most labs because it is cheap and effective. However, milk contains casein, which is itself a phosphoprotein, so using milk to block a blot where you are probing for a phosphorylated protein can create confusing background. In that case, bovine serum albumin is the standard substitute.
Each blocker has trade-offs. Milk, BSA, fish gelatin, and synthetic polymers like polyvinylpyrrolidone each differ in autofluorescence, blocking efficiency, and cost, which matters especially if you are using fluorescent rather than chemiluminescent detection.4Analytical Biochemistry. Optimization of blocking conditions for fluorescent Western blot Blocking is typically done for one hour at room temperature or overnight at 4°C, with gentle rocking to keep the solution evenly distributed.
Probing With Primary and Secondary Antibodies
The primary antibody is the one that actually recognizes your target protein. You dilute it in blocking buffer or a buffer containing a small amount of detergent, lay it over the membrane, and incubate. Most protocols call for one hour at room temperature or overnight at 4°C. But the optimal incubation time varies more than people realize. Some antibodies reach peak signal within four to eight hours, while others continue to accumulate specific signal even after 48 hours of incubation without a meaningful increase in background.5PubMed Central. Prolonged Incubation and Stacked Film Exposure Improve Sensitivity in Western Blotting If you are struggling with low signal for a particular target, extending the primary antibody incubation beyond the standard overnight step is worth trying before you start troubleshooting elsewhere.
After washing away unbound primary antibody (usually three to five washes with a buffer containing a mild detergent like Tween-20), you apply the secondary antibody. The secondary antibody recognizes the species-specific portion of the primary antibody (for example, if your primary was raised in rabbit, your secondary is an anti-rabbit antibody). The secondary antibody is conjugated to something that produces a detectable signal, most commonly horseradish peroxidase (HRP) for chemiluminescent detection or a fluorophore for fluorescent detection.
If you need extreme sensitivity, there are amplification strategies. Using a biotinylated secondary antibody followed by streptavidin conjugated to a polymer of HRP molecules has been shown to increase sensitivity by over a hundredfold compared to a standard HRP-conjugated secondary antibody, enabling detection of a target protein in as little as 50 nanograms of lysate where the conventional approach required at least 2 micrograms.6PubMed Central. Improving the sensitivity of traditional Western blotting via Streptavidin containing Poly-horseradish peroxidase (PolyHRP) That kind of boost is useful when sample is scarce or the target protein is expressed at very low levels.
Detection and Imaging
For chemiluminescent detection, you apply a substrate solution that the HRP enzyme converts into light. You then capture that light on film or, more commonly now, with a digital imager. The digital approach has largely replaced film in most labs because it provides a wider dynamic range, making quantification more reliable. Film saturates quickly, meaning a very abundant protein looks the same as a moderately abundant one because the film cannot distinguish intensities above a certain threshold.
Fluorescent detection uses secondary antibodies tagged with infrared or near-infrared fluorophores. The membrane is scanned on a fluorescent imager, and because different fluorophores emit at different wavelengths, you can probe for two proteins on the same blot simultaneously using two different channels. This is particularly useful when you want to compare your target protein and a loading control on the same membrane without stripping and reprobing. Fluorescent detection also avoids the enzyme-substrate kinetics issue that makes chemiluminescence inherently harder to quantify, since fluorescence intensity is directly proportional to the amount of bound antibody rather than subject to how fast an enzyme is turning over substrate.
Quantification and Normalization
If you just need to know whether a protein is present, you are done once you have a visible band at the right molecular weight. But most experiments require comparing how much of a protein is in one sample versus another, which means quantification. You use densitometry software to measure the intensity of each band, then normalize those values against something that accounts for differences in how much total protein was loaded per lane.
Traditionally, this normalization is done using a housekeeping protein like actin, tubulin, or GAPDH. You probe for the housekeeping protein alongside your target, and the idea is that the housekeeping protein is expressed at the same level in every sample, so any variation in its band intensity reflects loading differences rather than biology. The problem is that housekeeping proteins are not as constant as people assumed. Their levels can change with experimental treatments, disease states, and cell type, making them unreliable as normalizers in many contexts.7Analytical Biochemistry. Stain-Free total-protein normalization enhances the reproducibility of Western blot data
An increasingly preferred alternative is total-protein normalization. Instead of relying on a single housekeeping protein, you stain the membrane (or use a stain-free gel system) to visualize all the protein in each lane. The total signal per lane becomes your normalizer. This approach improves reproducibility because it is not dependent on any single protein behaving consistently across conditions.7Analytical Biochemistry. Stain-Free total-protein normalization enhances the reproducibility of Western blot data When doing densitometry, background subtraction settings also matter. Software tools let you adjust how background is calculated, and inconsistent settings between gels can introduce artifacts, so it pays to standardize your analysis workflow.8PubMed Central. A Defined Methodology for Reliable Quantification of Western Blot Data
Stripping and Reprobing
Sometimes you need to detect multiple proteins on the same membrane, especially when sample is limited. Stripping removes the bound antibodies so you can reprobe with a different primary antibody. This is routine, but the method you use affects whether it actually works and whether it damages the proteins on the membrane.
The classic approaches use either a detergent like SDS or a low-pH buffer to disrupt the antibody-antigen interaction. Both have downsides. SDS-based stripping can remove transferred proteins from the membrane along with the antibodies, and low-pH buffers often strip antibodies inefficiently, leaving residual signal that contaminates the next round of probing. A guanidine hydrochloride-based solution with a nondenaturing detergent and a reducing agent has been shown to strip tightly bound antibodies from PVDF membranes rapidly at room temperature without removing significant amounts of transferred protein.9Analytical Biochemistry. A solution for stripping antibodies from polyvinylidene fluoride immunoblots for multiple reprobing
For fluorescent western blots, a different strategy can work. Combining a 10% acetic acid strip with photobleaching of the fluorophore effectively eliminates the signal from the first round of probing, achieving over 90% stripping efficiency for most antibodies tested, and the acetic acid does not cause protein loss from nitrocellulose membranes.10PubMed. Combining Acid Stripping With Photobleaching for Sequential Single-Channel Fluorescent Western Blot Another recent approach uses a heat-induced antigen retrieval method with a modified Tris-EDTA buffer. This protocol has been demonstrated to strip nitrocellulose blots at least five times without causing protein loss or changes in band intensity, which is a meaningful advantage if you are running a panel of targets on precious samples.11Analytical Biochemistry. Heat-induced antigen retrieval utilizing modified Tris-EDTA buffer for reprobing of Western blots on nitrocellulose paper
Common Problems and How to Fix Them
High background is probably the most frequent complaint. It usually comes from one of three places: insufficient blocking, too much antibody, or not enough washing. Before overhauling your protocol, try increasing the number or duration of washes after the primary antibody step. If background persists, titrate down your primary antibody concentration, since many antibodies are used at unnecessarily high concentrations based on a datasheet suggestion rather than empirical optimization.
No signal at all is demoralizing but usually traceable. Walk backward through the protocol: Was the transfer successful? (Check with Ponceau staining.) Is the primary antibody validated for western blot specifically and for the species you are working with? Is the secondary antibody matched to the primary’s host species? Did you use the right substrate? Each of these is a point of failure, and the fastest way to debug is to check transfer first, since no downstream optimization can rescue a blot where the protein never made it onto the membrane.
Unexpected bands, whether extra bands or bands at the wrong molecular weight, are a different kind of puzzle. Some are real biology, such as post-translational modifications, splice variants, or protein degradation products. Others are artifacts from cross-reactivity of the primary antibody. Running a knockout or knockdown control alongside your samples is the gold-standard way to verify that the band you care about is genuinely your target and not a cross-reactive imposter.
Uneven or splotchy bands often result from air bubbles trapped between the gel and membrane during transfer, or from uneven antibody incubation. Ensuring good contact during the transfer assembly and keeping the membrane fully submerged and rocking during all incubation steps prevents most of these artifacts.
Automated and Capillary-Based Alternatives
The traditional western blot is labor-intensive and hands-on. From lysis to final image, you can easily spend a full day or two on the process. Automated systems have emerged to address this. Semi-automated platforms handle the immunoblotting steps (blocking, antibody incubations, washes) while you still pour gels and do the transfer yourself. Fully automated capillary-based systems go further, performing separation, immunoprobing, and detection in a single instrument on tiny sample volumes.
A direct comparison of traditional western blotting with a semi-automated system and a fully automated capillary-based system found that the fully automated approach saved time and offered valuable sensitivity, which is particularly beneficial when sample amounts are limited. The trade-off is cost: both the instruments and the proprietary reagent cartridges are substantially more expensive than a conventional setup.12PubMed Central. Comparison of Automated and Traditional Western Blotting Methods Capillary-based platforms have also found niche applications in areas where traditional methods are impractical, such as quantitative assays for enzyme activity where throughput and reproducibility are essential.13PubMed. Development of an ADP-ribosylation assay for residual toxicity in C. difficile binary toxin CDTa using automated capillary western blot
For most academic labs, the traditional method remains standard because the equipment is already on hand and the per-experiment cost is low. Automated systems tend to make more sense in core facilities or industry settings where many blots are run per week and the higher reagent cost is offset by savings in personnel time.
Reporting Standards and Why They Matter
Even a perfectly executed western blot loses much of its value if the results are not reported in a way that lets others evaluate them. This is a real problem in the published literature. A large-scale assessment of western blot figures in neuroscience and cell biology papers found that over 90% of published blots were cropped, and more than 80% of papers did not provide the unprocessed blot images in the supplement. Visible molecular weight markers were rare, and many blots lacked even molecular weight labels.14PLOS Biology. Blind spots on western blots: Assessment of common problems in western blot figures and methods reporting with recommendations to improve them
Why does this matter? Without visible markers, a reader cannot independently confirm that a band is at the expected molecular weight. Without uncropped images, there is no way to evaluate whether there were nonspecific bands, uneven loading, or other quality issues that might change the interpretation. Cropping is sometimes necessary for figure clarity, but the original images should always be available as supplementary data. Journals are increasingly requiring this, but the practice still lags far behind the policy.
If you are generating western blot data for publication, the easiest way to strengthen your paper is to include the uncropped blots, show the molecular weight markers, describe your antibody catalog numbers and dilutions, and state your normalization method. These details take minimal extra effort during the experiment but make the difference between data that others can trust and data that raises questions.