What does dynamic light scattering peptide particle size actually report?
Dynamic light scattering peptide particle size is calculated, not imaged. Intensity fluctuations from Brownian motion are converted to a correlation function. For one diffusing species the field correlation decays as exp(−q²Dτ), where D is the translational diffusion coefficient and q depends on wavelength, solvent refractive index and angle. The Stokes–Einstein relation, Rh = kBT/(6πηD), converts D to a hydrodynamic radius. The reported diameter is twice that radius, so it inherits temperature, solvent viscosity and the assumption that D describes the species of interest rather than a mixture.
Polydisperse samples produce a sum of exponentials. Cumulant analysis fits the early decay and returns a z-average diffusion coefficient and a polydispersity index. That z-average is intensity-weighted, not a number-mean and not a mass-mean. Inversion algorithms turn the same curve into an intensity distribution, but the inversion is ill-posed: baseline noise and the regularisation setting can move a shoulder between bins.
Large objects dominate because, for particles much smaller than the wavelength, scattered intensity scales approximately with the sixth power of diameter. A few 50 nm assemblies can out-scatter a large excess of 2 nm peptide. A dilute monomer can instead sit near the count-rate floor, so the intercept collapses and the fit follows noise. Reviews of light-scattering methods keep dynamic, static and electrophoretic measurements distinct: static data can inform mass and form factor, while electrophoretic light scattering reports mobility related to zeta potential, not sequence and not diameter (PMID:29988578). A defensible research-lot statement names the intensity-weighted z-average, the polydispersity index and the correlation-function quality.
Why do NMR diffusion and dynamic light scattering disagree on formulated peptides?
A comparison of NMR and dynamic light scattering for diffusion coefficients of formulated insulin frames why the two hydrodynamic sizes are not replicates (PMID:28791599). Both experiments can be turned into a diameter, but they weight the sample differently. Pulsed-field-gradient NMR attenuation tracks nuclei. The monomer, or the small oligomer that holds most of the mass, dominates that curve. Dynamic light scattering weights scattered intensity, so trace larger particles, including aggregates and some excipient scatterers, pull the z-average up even when their mass fraction is small. Agreement within viscosity and temperature uncertainty supports one dominant small species and few strong scatterers. A substantially larger light-scattering diameter is a result to explain, not an automatic instrument fault. Causes include aggregates, incomplete dissolution, dust, micellar excipients, or a mismatch in concentration and buffer between the NMR tube and the cuvette. Those variables have to be aligned before the gap is called aggregation.
Neither method alone resolves weakly bound oligomers that exchange during the measurement. Polarised multidimensional fluorescence has been used to quantify weakly bound insulin oligomers in solution, which is a different observable from the intensity-weighted diameter (PMID:33161979). If a research file needs both bulk hydrodynamic size and evidence of low-order association, the orthogonal method belongs in the protocol before the z-average is locked. The light-scattering bias toward large scatterers is useful when it is labelled as an early warning. It is a documentation error when it is labelled as the size of the monomer. Report both numbers with their weighting, and do not average them.
How do laboratories follow peptide aggregation and lipopeptide assembly by light scattering?
Aggregation changes the scatterer population, so a time series of hydrodynamic diameters can show larger species even when chromatography still detects the parent peak. Dynamic light scattering has been used to measure aggregation of the peptide fragment ProIAPP(1-48) (PMID:30466917). The analytical question in that design is physical: under a stated buffer, concentration and temperature, does the intensity-weighted size grow and does the polydispersity index broaden? The trace does not identify which chemical event, if any, started the assembly. Oxidation, deamidation or a nucleation-prone sequence can feed the same increase. Sequence identity remains a mass-spectrometry result.
A usable time course states concentration as net peptide content where that figure is known, because counter-ion and moisture change molarity. It states whether the solution was centrifuged or filtered, and at what pore size, because a 0.22 µm filter removes part of the range under study. Intercept, count rate and attenuator belong with each point, so a reviewer can separate a real intensity jump from a collapsed correlation.
Lipopeptide assembly needs the same caution. Daptomycin aggregation has been examined by fluorescence, light scattering and NMR together (PMID:24464772). Scattering reports that an assembly large enough to scatter is present; it does not replace the spectroscopic evidence. Incorporating lipids or liposaccharides influences the particle size of peptide materials (PMID:20564037), so a lipidated analogue should not inherit a diameter expectation from the naked sequence. A narrow peak above an association threshold can be a micelle rather than dust. Reversibility is tested by dilution and by the orthogonal spectra, not by the z-average alone.
Which settings decide whether a peptide hydrodynamic diameter is interpretable?
The Stokes–Einstein step is only as good as viscosity and temperature. Near room temperature the viscosity of water changes by about two per cent per degree, so a 25 °C table applied to a 20 °C run shifts every diameter. Glycerol, sucrose, dimethyl sulfoxide, high salt and acetic acid diluents cannot use a pure-water preset. If the software solvent is left on water while the sample is a mixed diluent, the diameter is a calculation error.
Concentration has two failure modes. Too low, and a small peptide yields a noisy baseline and a poor intercept. Too high, and multiple scattering, virial interactions and reversible association move D away from the infinite-dilution value the equation assumes. A short dilution series, viscosity held constant, shows whether the z-average is stable, grows on dilution (dust becoming relatively louder) or shrinks (reversible association). Nanoparticle light-scattering practice already treats concentration, contrast and data treatment as part of the method (PMID:29988578).
A fine filter is reasonable when the question is the size of a fully dissolved peptide and dust must be excluded. The same filter invalidates an aggregate question. A solvent blank in the same cell is the control for fingerprints, scratches and unfiltered rinse water. Backscatter and 90-degree optics are not interchangeable. Cumulant z-average and polydispersity index are the summary statistics to archive; number distributions are transformations and amplify error in the small-size bins. Analytical considerations for globule size distributions of cyclosporine ophthalmic emulsions underline that the reported distribution depends on the method chosen (PMID:30125649). A peptide method should name the metric, the temperature and the solvent model, not merely particle size by DLS.
When should orthogonal methods sit beside a DLS diameter?
Dynamic light scattering is most useful for suspended diffusers from a few nanometres to a few hundred nanometres when the correlation function is sound. It counts poorly in the 1–100 µm subvisible window, and it separates monomer from dimer poorly when both are small and a trace of larger material is present. Those gaps set the orthogonal panel.
Water proton NMR has been developed for in situ detection of insulin aggregates (PMID:26344698). It follows a water-relaxation response rather than peptide scattered light, so it can flag aggregate-related change without reporting a diameter. A relaxation rate should not be converted into a pretend hydrodynamic size. Holographic characterisation has been applied to protein aggregates as an optical method distinct from a batch diffusion coefficient (PMID:26886303). A clean z-average does not certify that those larger particles are absent, and a holographic result does not describe the monomer.
Weak oligomers are the opposite problem. Polarised multidimensional fluorescence has been used to quantify weakly bound insulin oligomers in solution (PMID:33161979). Those oligomers need not shift the z-average. Size-exclusion chromatography still provides the usual view of soluble high-molecular-weight species, with adsorption, on-column dilution and exclusion-limit caveats of its own. No one correlation function covers covalent aggregates, reversible oligomers, micelles and subvisible particles together.
Analytical similarity work, including a published adalimumab biosimilar comparison (PMID:27461107), proceeds attribute by attribute. A hydrodynamic diameter belongs in such a comparison only if solvent model, angle, concentration and fit match. A matched z-average does not show that related-substance profiles match. Research lots should keep size results separate rather than averaging them into one score.
How should a batch record state dynamic light scattering peptide particle size?
The record should make the run reconstructable. Minimum fields are lot identifier; peptide name and counter-ion; concentration, marked as gross powder or net peptide content; diluent; filtration or centrifugation with pore size or relative centrifugal force and time; temperature; viscosity and refractive-index model; wavelength and angle if not fixed by the instrument; number of runs and independent preparations; z-average hydrodynamic diameter; polydispersity index; and a quality line for intercept, count rate and the solvent blank. An attached distribution should name the inversion algorithm and be labelled intensity-weighted.
Pass requires a written limit. An internal research limit might require the intensity-weighted z-average and the polydispersity index to fall below stated values for a fully dissolved aliquot prepared by a fixed method. That is a method-specific control limit. It is not a medicine release specification, and it says nothing about particles in the dry cake before dissolution. Two Australian research lots can be compared on size only when both certificates used the same definition. A filtered 10 nm result is not evidence of a smaller peptide than an unfiltered 40 nm result from a different buffer.
Dynamic light scattering peptide particle size does not confirm sequence, related substances, residual solvents, counter-ion content, endotoxin or bioburden. A blend gives one correlation function, not an assignment of which component scattered. Buyers checking documentation can ask for the distribution export and the blank. Cuvette temperature is an analytical input because viscosity depends on it; recording it is not a transport instruction. Research material remains research material; a diameter on the file supports no human-use claim.
Frequently asked questions
Is a dynamic light scattering diameter the monomer size?
Not by default. The reported z-average is an intensity-weighted hydrodynamic diameter of whatever scattered under the stated conditions. A trace of aggregates, dust or micelles can shift it far above the monomer. Agreement with an NMR diffusion coefficient, measured in the same buffer and at a matched concentration, is what supports a small-species interpretation. The DLS number alone does not identify the monomer.
Why do two laboratories report different peptide particle sizes for one lot?
Common causes are analytical, not necessarily two different solids. Temperature, viscosity model, concentration, filtration, scattering angle and cumulant versus distribution fitting all change the number. One laboratory may have filtered dust out; another may have measured an associating lipopeptide above its association threshold. Compare certificates only when those method fields match. An unqualified particle-size label cannot be ranked against a documented z-average.
Does a low polydispersity index show that the peptide is chemically pure?
No. Polydispersity index describes the width of the diffusion-coefficient fit, not related substances, deletion sequences or counter-ion content. A chemically impure sample of one hydrodynamic size can look narrow. A chemically uniform peptide can look broad if it associates or if dust is present. Purity remains a chromatographic and mass-spectrometric result. DLS is a physical-state measurement.
Should peptide solutions be filtered before dynamic light scattering?
Only if filtration matches the question. A fine filter is appropriate when the aim is the size of a fully dissolved peptide and dust must be excluded, and the pore size must be recorded. The same step removes species relevant to an aggregation study and makes a large-particle conclusion invalid. Centrifugation is an alternative only if force and time are documented. The solvent blank should follow the same path.
Can dynamic light scattering replace size-exclusion chromatography for aggregates?
No. The methods see different things. DLS is intensity-weighted and sensitive to a few strong scatterers, but it does not separate species into chromatographic peaks or identify them. Size-exclusion chromatography profiles soluble high-molecular-weight material and can miss particles that never enter the column, or dissociate species by dilution. A research aggregate file uses them as orthogonal results, not as substitutes. Subvisible particles need yet another method.
What should a buyer in Australia ask for beside the headline diameter?
Ask for the intensity-weighted z-average, the polydispersity index, temperature, solvent viscosity model, concentration basis, filtration history, replicate range and a solvent-blank statement. Ask whether any distribution plot is intensity-weighted and which algorithm produced it. Those fields allow a like-for-like comparison of research lots. They do not replace the HPLC and mass-spectrometry identity package, and they do not support any human-use interpretation.
References
- PMID:29988578 — Application of Light Scattering Techniques to Nanoparticle Characterization and Development — Front Chem — 2018
- PMID:28791599 — Comparison of NMR and Dynamic Light Scattering for Measuring Diffusion Coefficients of Formulated Insulin: Implications for Particle Size Distribution Measurements in Drug Products — AAPS J — 2017
- PMID:33161979 — Quantitative analysis of weakly bound insulin oligomers in solution using polarized multidimensional fluorescence spectroscopy — Anal Chim Acta — 2020
- PMID:30466917 — Aggregation of the diabetes-related peptide ProIAPP(1-48) measured by dynamic light scattering — J Trace Elem Med Biol — 2019
- PMID:24464772 — Evaluation of lipopeptide (daptomycin) aggregation using fluorescence, light scattering, and nuclear magnetic resonance spectroscopy — J Pharm Sci — 2014
- PMID:20564037 — The influence of incorporating lipids or liposaccharides on the particle size of peptide therapeutics — Biopolymers — 2011
- PMID:30125649 — Analytical considerations for measuring the globule size distribution of cyclosporine ophthalmic emulsions — Int J Pharm — 2018
- PMID:26344698 — Water Proton NMR for In Situ Detection of Insulin Aggregates — J Pharm Sci — 2015
- PMID:26886303 — Holographic Characterization of Protein Aggregates — J Pharm Sci — 2016
- PMID:27461107 — Assessing Analytical Similarity of Proposed Amgen Biosimilar ABP 501 to Adalimumab — BioDrugs — 2016
Research use only
This article is provided for laboratory research and educational purposes only. Products referenced are not for human or veterinary use. ClaraScience makes no therapeutic, medical, or efficacy claims, and nothing here constitutes medical advice.
