Frequently Asked Questions

What is biomolecular aggregation?

Biomolecular aggregation is a biological phenomenon where individual macromolecules, such as protein molecules cluster and clump together, forming a new entity that might have lost its original functions or activities and have the potential to cause diseases. Specifically, protein aggregation is a result of protein mis-folding, causing individual proteins to interact with one another. In the most severe cases, protein aggregation leads to the formation of irreversible, insoluble fibril structures, e.g., β-amyloids, which are responsible for neurodegenerative diseases such as Alzheimer’s and Huntington’s.

Image of protein showing various types of three-dimensional structures

Which molecules are most susceptible to aggregation?

In non-stressed conditions, peptide formation during translation and newly expressed proteins during the folding process are susceptible to aggregation. Proteins that are under stressed conditions (I.e., heat, pH, oxidative, etc.) are also susceptible to aggregation.

What causes protein aggregation?

The direct cause of protein aggregation results from interactions between individual proteins that become thermodynamically favored. There are two main sources of such driving force behind protein aggregation:

  • The surface of proteins loses interaction with solvents (i.e., water molecules) and consequently exposes the hydrophobic groups that can interact with the surface of another protein undergoing the same process. Aggregates formed by such surface interactions are called colloidal aggregates.
  • Mis-folded proteins fail to maintain the native structure and expose the inner hydrophobic pockets. The exposed hydrophobic pockets, primarily composed of β sheets, are prone to re-form beta interactions indiscriminately with other intra- and inter-molecular β sheets. Aggregates formed by these intermolecular β-sheets are called structural aggregates.
Protein aggregation caused by solvent evaporation
Protein aggregation caused by solvent evaporation

protein aggregation caused by misfolded proteins
Protein aggregation caused by misfolded proteins

Can protein aggregation occur at all steps of the drug development process?

Yes, protein aggregation can occur at all stages of the drug development process, including early discovery, pre-clinical research and development, formulation, and clinical development. Common factors that can cause protein aggregation in early discovery stage include process temperature of proteins, freeze/thaw cycles of proteins, agitation stress of proteins, protein concentration, etc. It is also important to be able to detect early signs of protein aggregation or just the potential of protein aggregation during the early discovery phase.

What are the different types of protein aggregates and how do they form?

There are two different types of protein aggregates – colloidal protein aggregates and structural protein aggregates. Colloidal protein aggregates refer to the self-association of protein molecules that are still well folded in their native structure. Structural protein aggregates are the aggregates in which the protein molecules are no longer in their native folded states but a different structure. Conformational or structural protein aggregation often involves the formation of intermolecular β-sheets which is irreversible, while colloidal protein aggregates are often still reversible.

There are multiple pathways that lead to protein aggregation, as illustrated in the figure below by Christopher J. Roberts et al. (A) represents the process from partly unfolded protein monomers to partly unfolded clusters, while (B) represents colloidal protein clusters where proteins are folded well in their native structures. Both (A) and (B) protein aggregates are still reversible until it reaches (C) where inter-molecular β-sheets have formed, and it will function as nuclei that triggers further aggregation into oligomers, soluble filaments, and agglomerated aggregates. As a result, protein aggregation rates are governed by conformational and colloidal stabilities, demonstrated by John Carpenter, Eva Y. Chi et al, and many other researchers and scientists.

MMS provides a way to assess protein colloidal stability (via protein hydration) and structural or conformational stability (via secondary structural analysis) in one single automatic measurement. Stay tuned for an upcoming app note.

Schematic diagram below illustrates multiple non-native aggregation pathways for a multi domain protein such as a monoclonal antibody (mAb) composed of a single Fc fragment and two identical Fab fragments. Red strands denote ‘hot spot’ sequences that are prone to form strong, effectively irreversible inter-protein contacts that stabilize aggregates, but are primarily hidden or buried in fully folded monomers. Double arrows denote effectively reversible steps. Single arrows denote irreversible steps. Nuclei are defined as the smallest net-irreversible aggregates that form soluble aggregates spanning length scales on the order of 10 to 102 nm. This occurs primarily via addition of other partly unfolded monomers (upper right) or by the agglomeration of existing aggregates (lower right) [ref].

Schematic diagram illustrating multiple non-native aggregation pathways for a multi domain protein such as a monoclonal antibody (mAb) composed of a single Fc fragment and two identical Fab fragments.
Schematic diagram illustrating multiple non-native aggregation pathways for a multi domain protein such as a monoclonal antibody (mAb) composed of a single Fc fragment and two identical Fab fragments.

Another way to look at different types of protein aggregates, as demonstrated by Leonid Breydo et al. and Bhavana Shivu et al., protein aggregates can be divided into three structural classes: amorphous aggregates, amyloid fibrils, and oligomers. The formation of the amorphous aggregate involves the association of the solvent-exposed hydrophobic surfaces of the nativelike cores of the intermediates accompanied by the formation of intermolecular β-sheets between disordered loops, which first yields small soluble oligomers that increase in size to generate larger aggregates. Amyloid fibril formation, however, involves initial weaker hydrophobic interactions, resulting in a soluble oligomeric pre-nucleus that probably has more β-sheet formation between the disordered segments of the intermediate.

Model representing the formation of amyloid fibrils, amorphous aggregates, and oligomers.
Model representing the formation of amyloid fibrils, amorphous aggregates, and oligomers.

How is aggregation detected and measured (tools, etc)?

Although protein aggregates can be both colloidal aggregates and structural aggregates, today’s labs almost always only quantify size using HPLC, SEC, and AUC, techniques which do not differentiate the two types of protein aggregates.

Schematic diagram illustrating multiple non-native aggregation pathways for a multidomain protein such as a monoclonal antibody (mAb) composed of a single Fc fragment and two identical Fab fragments.

In the figure above, B-C-D-E will all appear to be a size change with no way to distinguish them apart. With the addition of MMS, we can distinguish A, B, C, D, E. More specifically, A has an overall structural change but little intermolecular β-sheets, B has no overall structural changes (spectra identical); C, D, E have both overall structure changes and intermolecular β-sheets.

Analytical tools used to detect protein aggregation

Several common analytical tools used to detect aggregates include: HPLC, SEC and AUC. These do not differentiate between two types of intermolecular aggregation. MMS is able to do this.

HPLC and Mass Spec Technology
High-Performance HPLC and Mass Spec
Size-exclusion Chromatography (SEC)
Size-exclusion Chromatography (SEC)
Analytical Ultra-centrifugation (AUC)
Analytical Ultra-centrifugation (AUC)
Apollo, the second-gen MMS instrument from RedShiftBio
Microfluidic Modulation Spectroscopy

Which secondary structures are associated with aggregation and the current limitations to detect such?

Proteins are comprised of chains of amino acids which, in specific order, dictate structure and function as the primary structure. Aggregation is most notably associated with β-sheet formation, a structure that is present in protein biomolecules in varying amounts. Other structures that contribute to the overall secondary structure, folding, function and activity of proteins include alpha-helices, turns, and unordered regions.

As demonstrated above, the formation of intermolecular β-sheet is key to irreversible protein aggregation. However, the attention that researchers in the biopharmaceutical space pay to protein secondary structure and intermolecular β-sheet does not match their importance. This is primarily due to the limitations of traditional tools, especially their low sensitivities. That is to say, no meaningful conclusion can be drawn unless there is a very large change in protein secondary structure that goes beyond the system error.

KBI has published a paper to compare the Limit of Quantification (LOQ) of traditional Fourier-transform infrared (FTIR) spectroscopy, Circular dichroism (CD) and Microfluidic Modulation Spectroscopy (MMS) by Brent S. Kendrick et al. LOQ is defined as the smallest amount or the lowest concentration of a substance that is possible to be determined by means of a given analytical procedure with the established accuracy, precision, and uncertainty. MMS is 30 times more sensitive than traditional FTIR and 5 times more sensitive than CD.

KBI logo

Determining Spectroscopic Quantitation Limits for Misfolded Structures

Protein secondary structures are frequently assessed using infrared and circular dichroism spectroscopies during drug development (e.g., during product comparability and biosimilarity studies, reference standard characterization, etc.) However, there is little information on the lower limits of quantitation of structural misfolds and impurities for these methods. A model system using a monoclonal antibody reference material was spiked at various levels with a protein that had a significantly different secondary structure to represent the presence of a stable and discreet structural misfold. The ability of circular dichroism, transmission Fourier transform infrared spectroscopy and microfluidic modulation spectroscopy, along with various spectral comparison algorithms, were assessed for their ability to detect the presence and quantify the amount of the misfolded structure.”

What are the reasons to avoid aggregation?

Protein aggregates can be immunogenic. It is generally accepted that the immunogenic risk of aggregated protein drug molecules is correlated with its reversibility. Specifically, irreversible protein aggregates are more immunogenic. For the same mass of materials, smaller (e.g. 20nm) protein aggregates have a stronger immunogenic response due to the much greater amount of them. [1, 2]

MMS can quantify intermolecular protein β-sheets with ultra-sensitivity. This will determine the amount of irreversible protein aggregates which cannot be achieved by measuring size using DLS, NTA, MFI, or SEC. Here are some app notes that demonstrate how MMS irreversible protein aggregation.

AN_850_0109 - Detection of Pressure Induced aggregation using MMS

The RedShiftBio AQS3®pro system, powered by Microfluidic Modulation Spectroscopy (MMS), enables the early detection of instability by measuring minute changes in inter-molecular β-sheet composition, an indicator of irreversible aggregation. The AQS3pro utilizes a high-power quantum cascade laser (QCL) to generate unrivaled sensitivity in secondary structure determination over a wide concentration range. The ability to see change in inter-molecular β- sheet composition through stability monitoring at selected wavenumbers by MMS creates an advantage of correlating drug development conditions to potential detrimental effects on protein function at all stages of drug development.

AN_850_0101 Thermal Denaturation Analysis of Bovine Serum Albumin over Wide Concentration Range by Microfluidic Modulation Spectroscopy

Microfluidic Modulation Spectroscopy (MMS) is a powerful new infrared spectroscopy tool for protein structural analysis developed by RedShift BioAnalytics. This technology provides significant increases in sensitivity, dynamic range, and accuracy for determination of protein secondary structure relative to conventional mid-IR and far-UV CD techniques. The analyzer utilizes a tunable quantum cascade laser to generate an optical signal > 100x brighter than the conventional sources used in FTIR spectroscopy. Brighter sources also allow the use of simpler detectors without the need for liquid nitrogen cooling. Additionally, the sample (protein) solution and a matching reference buffer stream are automatically introduced into a microfluidic flow cell, and the two fluids are rapidly modulated (1-4 Hz) across the laser beam path to produce nearly drift-free background compensated measurements.

If aggregation is so bad, how can I prevent it?

Knowing the aggregation pathway will save your time and effort in solving your protein aggregation problems, developing formulations, and narrowing down leads. To solve reversible protein colloidal association, we usually consider controlling electrostatics using salt and buffers, i.e. to increase the charges or adjust the charge distributions on the surface). To alleviate reversible protein structural unfolding, we usually consider controlling co-solute affinity which governs the desolvation step using sugars and amino acids and the like. Or if that's not working, we may go back to choose another lead that has a "better" protein structure. "Better" means the protein can form a tighter hydrophobic core.

How can solutions be stabilized to avoid aggregation?

To stabilize a protein formulation to avoid aggregation, we need to know how proteins are aggregated. For proteins to aggregate, the first step is to approach ~3 Å, the second step is desolvation – the removal of everything between the two proteins including water and other molecules in the formulation. Then proteins can go through structural changes over time and eventually form intermolecular β-sheets to be irreversibly aggregated. This serves as a nucleus for the formation of more intermolecular β-sheets and eventually large aggregates.

Excipients are the developer’s ingredients for engineering stable formulations. Different excipients provide different influence and control over different steps described above. The first step is governed by proximity energy and can be hindered via electrostatics, e.g. protein charges, therefore, controlling over pH, ionic strength and dielectric can be very effective.

The second step – desolvation – is governed by protein hydration and exclusive volume. Sugars, some amino acids, can help control protein hydration. Some polymeric chains can be linked to protein surfaces to increase the exclusive volume, therefore stabilizing the protein formulation.

Is all protein aggregation reversible/irreversible?

Protein aggregation can be both reversible and irreversible. Theoretically, if the protein molecule still maintains its natively folded structure, it can still be reversibly separated into single molecules. Therefore, colloidal aggregates are usually considered reversible. However, once intermolecular β-sheets have formed between protein molecules, there would be no way to reverse it back to individual protein’s native states, therefore, irreversible. One may test the reversibility by removing the stress and/or diluting the sample.

How do I setup an experiment to find the formulation condition where protein aggregation is starting?

Accelerated stability studies are usually performed on protein drug products as part of the formulation screen to see how resilient each formulation is to stress conditions like heat, agitation, freeze/thaw cycles, and light exposure. This type of study will also help inform the researcher on how shelf-stable the drug product is. The formulations to be tested should include a range of pH values and additives such as sugars, salts, preservatives, and stabilizers like amino acids. The goal of this study is to determine which protein drug formulation can withstand the most stress without the drug product aggregating.

What can be done if aggregation is detected?

If protein aggregation is detected, this is an alarming sign for potential product failure, e.g. due to loss of protein function and/or unwanted interaction. The course of action resembles that of how to prevent protein aggregation (See FAQ: Reasons to Avoid Aggregation) and depends critically on where along the biotherapeutic pipeline it is detected.

In the early development phase, i.e. formulation, there are still more degrees of freedom in both the buffer recipe and the protein itself, in order to stabilize the product. The strategy depends on the type of aggregation: Reversible colloidal protein association is often approached by controlling the electrostatics through salt and buffer components, structural protein aggregation might require changes to the protein sequence. In general, protein structures with a more densely packed (hydrophobic) core tend to be more stable. In manufacturing, the product recipe is already more sophisticated and potential changes are limited to fine tuning the buffer components. Detecting structural protein aggregation at this point is a major setback for the release of this product. For released products, aggregation is a serious quality issue which requires retraction of that product and adjustments in the buffer and/or the protein sequence.

Can I use micro-filtration to remove aggregates?

Protein aggregates and other large particulates can be removed from samples that are being prepared for MMS typically by using a 0.2 um filter, or centrifuging to pellet the aggregates and use the supernatant. Removing large protein aggregates and particulates will help extend the lifetime of the flow cell as large aggregates can accumulate in the flow cell and slow the flow rate over time. If the flow rate slows, the options are to perform cleaning procedures or replace the flow cell with a new one.

What does the term biomolecular stability refer to?

Biomolecular stability refers to the ability of a biomolecule to withstand external stress, i.e., to keep its designated structure and activity. Most often increasing temperature is used to determine stability, but other stresses could be used too, such as pH or agitation. The relevant stability is determined by the stress-induced loss in a structure that already gives rise to a loss of function, too. This could be due to a structural change around the active site of that protein. In addition to that, proteins often have intermediate states that still conserve some part of their native structures and with some residual stability. The figure below shows these different states as an example of a thermally stressed protein (apoflavodoxin).

Figure adapted from: Lamazares, E., Clemente, I., Bueno, M. et al. Rational stabilization of complex proteins: a divide and combine approach. Sci Rep 5, 9129 (2015))

Why is stability important?

Stability, or the ability of a protein or biologic to withstand stresses, is crucial for its activity in a final formulation. A higher tolerance towards external stresses gives rise to better activity, safety, and a longer shelf-life of the final product.  

What causes proteins to lose stability?

Unavoidable changes in temperature, pH or just moving the sample can trigger, for example, oligomerization or unfolding processes which can decrease the stability of the protein towards further stresses. The mechanisms that lead to protein instability can be diverse. The different mechanisms that can lead to a loss of stability in monoclonal antibodies are summarized in a review by Le Basle et al. (Le Basle, Yoann, et al." Physicochemical stability of monoclonal antibodies: a review." Journal of Pharmaceutical Sciences 109.1 (2020): 169-190.)

What are the consequences of loss of stability?

A loss of stability may give rise to a loss in protein activity or may even cause unwanted side effects. A less stable protein is prone to aggregation and irreversible structural changes.

Which molecules are most susceptible to changes in stability?

In general, protein molecules that are neutral in surface charge or have hydrophobic surfaces are prone to interact with each other and hence cause changes in stability. This propensity is highly dependent on the primary amino acid sequence as well as which types of amino acids are dominant in the protein. If there are more hydrophilic and charged amino acids on the surface of the proteins, they will be less susceptible to aggregation caused by surface interactions. On the other hand, if the proteins are folded in such a way that a lot of the hydrophobic amino acids are exposed on the surface, they will be more susceptible to changes in stability. As a result, proteins that have mutations in their primary sequence are likely to suffer from stability issues. In specific applications, however, site-specific mutations are designed on purpose to enhance the stability or activity of a protein.

Which protein structure is most stable?

Protein stability is dictated by the type of interactions that help the folding of the entire protein molecule. The primary structure of proteins is the most stable since the protein sequence is held together by covalent bonds between individual amino acids. Some of the tertiary structures that are held together by disulfide bonds are the most stable other than the primary sequence of the protein. Among the non-covalently bound structures, protein secondary structure is the most stable due to the predominant hydrogen bonding interactions.

Can loss of stability occur at all steps of the drug development process?

Yes, loss of stability can occur at all steps of the drug development process, including early discovery, pre-clinical research and development, formulation, and clinical development. Common factors that can cause the loss of protein stability in early discovery stage include process temperature of proteins, freeze/thaw cycles of proteins, agitation stress of proteins, protein concentration, etc. In the later drug development stage, such as formulation, the goal is to maximize protein stability by choosing the best excipients and buffers for the proteins.

Which secondary structures are associated with stability loss and the current limitations to detect such?

The loss of alpha-helix or beta-sheet structure and the formation of intermolecular beta-sheet is associated with stability loss. FT-IR and CD are the most prevalent techniques to detect these secondary structures. However, FT-IR suffers from low sensitivity and reproducibility to detect small changes in the secondary structure. CD on the other hand is limited by a low concentration range and less sensitive towards the prediction of beta-sheet compared to alpha-helix structures.

Is all stability loss in proteins reversible/irreversible?

Some stability lossesin proteins are reversible but some are irreversible. A method to predict whether the stability loss is reversible or irreversible is to identify what kind of change in the protein is causing instability. If the change comes from the secondary structures, there is a high probability that the stability loss is irreversible, especially in the case of beta-amyloid formation. If the change comes from the tertiary or quaternary structures, the stability loss is more likely to be reversible. For example, reversible aggregation due to the surface interaction of proteins will cause a temporary loss of stability.  Reversible stability loss can be fixed by changing the formulation to separate protein monomers.

How is stability detected and measured (tools, etc)?

Stability can be defined in many ways based on what the applied stress is. In most  cases, stability is inferred based on the protein’s melting temperature (Tm).  If a protein can withstand higher temperatures without unfolding, that protein is said to be more stable. There are multiple tools for measuring melting temperature including DSF, DSC, and CD.

Quantification of stability through the melting temperature determined by differential scanning fluorimetry (DSF)


The figure above shows the quantification of stability through the melting temperature determined by differential scanning fluorimetry (DSF). The melting temperature is the temperature that causes the protein to be halfway unfolded, and the higher the melting temperature, the more stable the protein.  This figure shows that the melting temperature increases when it’s bound to a  ligand. (Samuel, Errol LG, et al.) "Processing binding data  using an open-source workflow." Journal of Cheminformatics 13.1  (2021): 1-11.)


However,  there are many other forms of stress other than temperature such as agitation, pH, pressure, light exposure etc. It is also important to understand the structure and stability of a biologic under all likely production conditions. MMS is another tool that can measure the protein structure after being exposed to any of these stresses and that can help predict the protein stability.

What can be done if loss of protein stability is detected?

Loss in proteinstability can manifest in multiple different ways, I.e., the protein could unfold, self-associate into oligomers, or structurally aggregate into fibrils. The strategy for addressing loss in stability will depend on how the protein is manifesting that loss. For example, colloidal aggregates can be addressed using different formulation conditions to help dissociate the oligomers. However, if the samples are unfolding or structurally aggregating, the solution may be tore-engineer or make mutations to the original construct to make it less hydrophobic.

What can help us prevent loss of stability?

Minimizing the amount of stress that a protein or biologic undergoes (temperature, agitation, pH swings) will help mitigate loss of stability. Additionally, certain excipients can be added to formulation buffers to help protect against the stresses that are unavoidable. Some examples of excipients used to increase stability are amino acids, sugars, salts, or small amounts of surfactants.


The application note titled “Detecting Protein Conformational Change Due to Ligand Binding and Stabilization Using MMS” details the stabilizing effects ligands can have. In this study, a protein was exposed to 50°C overnight. There were 3 different conditions: 1) protein alone(Apo), 2) protein + ligand 1, and 3) protein + ligand 2. The results show that when the protein is exposed to heat stress, there is a large increase in the the beta-sheet features and a decrease in alpha-helix. However, when in the presence of ligand 2 the effect of heat is much less dramatic, and the heat has the least impact on the sample bound to ligand 1. This indicates that ligands 1 and 2 both exhibit stabilizing effects on the protein, and ligand 1 has the most stabilizing effects.

AU Data from App Note "Detecting Protein Conformational Change Due to Ligand Binding and Stabilization Using MMS"
HOS Data from App Note "Detecting Protein Confirmational Change Due to Ligand Binding and Stabilization Using MMS"

The figures above are from the application note and show MMS data collected of a protein unbound (Apo) and bound to Ligand 1 and how 50°C heat stress overnight affects the structure of the protein. The heat stress drastically changes the structure of the Apo protein but has very little effect on the protein that is bound to ligand 1. This data shows that the ligand stabilizes the protein against heat stress.

Can different formulation buffers be used to increase stability?

Yes! There are many different additives/excipients/preservatives that can be added to formulation buffers to increase a biologic’s stability. It is important that the additives do not interfere with the function or structure of the biologic, so all functional assays and characterization must be completed in the final formulation buffer that includes any additives. Some examples of additives used to increase stability are amino acids, sugars, salts, or small amounts of surfactants.


Batabyal et al. emphasize the importance of proper buffer formulation and characterization in the article “Shaping IR Spectroscopy into a Powerful Tool for Biopharma Characterization.” In this study, polysorbate 80 (PS 80) was introduced to the formulation buffer for a monoclonal antibody. PS 80 is commonly used in aqueous drug products to increase stability; however, it is important that the introduction of PS 80 does not change the structure or function of the drug product. MMS had no issues with buffer or excipient interference and generated clean protein structural information that showed PS 80does not affect the structure of the monoclonal antibody.


A different example can be seen in the application note titled “Using MMS to Measure Buffer-Induced Structural Changes in an Alpha-helix Rich Enzyme,” where the buffering conditions did impact the structure of lysozyme. Four different buffering conditions were tested, and it was determined that lysozyme has the highest fractional contribution of alpha-helical structure when it’s in tris buffer compared to water, phosphate buffer, and PBS. Lysozyme is the most active when in tris buffer and this may be due to the optimized structure with higher alpha-helical content. Since the active site of lysozyme is sandwiched between regions of alpha-helix, it makes sense that loss in alpha-helical content would result in less activity.

Lysozyme primary and tertiary structure highlighting the active site near the Asp-52 and Glu-35 amino acids. (Gálvez-Iriqui, et al. "Lysozymes: characteristics, mechanism of action and technological applications on the control of pathogenic microorganisms." Revista mexicana de fitopatología 38.3 (2020):360-383.)

How do I setup an experiment to find the formulation condition where stability of proteins starts to decline?

Experiments that directly probe the loss in stability would be accelerated stability studies where a protein or biologic is exposed to higher-than-normal stress conditions to help predict when that drug product will fail and gauge the ttheshelf-life and storage conditions. Accelerated stability studies involve testing a range of stresses like temperature, agitation, pH, and light exposure.


An example of an accelerated stability study using MMS can be seen in the application note titled “Biosimilar Comparison and Accelerated Stability Predictions Based on <2% Secondary Structure Differences Using MMS.” In this study, insulin originator (Humalog) and a biosimilar were compared and the biosimilar was exposed to 4 or 30°C for 8 weeks to determine its structural stability. The data show small structural changes due to the heat stress and it is correlated with a loss in alpha-helix, unordered, and turn features, and increase in beta-sheet content. Structural information can be useful for stability studies to understand what happens when a protein becomes unstable.

Is it possible to increase protein solubility using formulations to prevent precipitation and protein aggregation?

Formulations can absolutely help increase solubility! One of the most important features to get right in a formulation buffer is the pH. In addition to finding the right buffer, there are also many different additives that can help stabilize or solubilize your protein. Some of these additives include salts to help balance ionic strength, sugars to help with the protein’s hydration, detergents to act as an emulsifier or to protect the protein from aggregating, and amino acids can help balance charges.

Want to Join Our Team?