Ultracentrifugation: Principles, Types & Applications

Refractive index detection (Rayleigh interferometry) measures optical path length changes proportional to total mass concentration, providing universal detection insensitive to chromophore properties. For adenovirus containing both protein and DNA, refractive index response factors are nearly identical (~0.195 vs. 0.196), eliminating extinction coefficient correction requirements. Classical boundary sedimentation velocity uses two identical cells with the same buffer, analyzing samples directly in formulation buffer at low speeds (~3,000 rpm). The Lamm equation, developed in 1929 by Theodor Lamm, mathematically describes sedimentation concentration profiles for ideal solutions at specific angular velocities, relating sedimentation coefficient (s), diffusion coefficient (D), radial distance (r), and time (t). While no analytical solution exists for the full equation, modern numerical analysis software (e.g., SEDFIT) enables computational modeling on desktop computers. This allows simulation of expected data patterns for various particle species including monomers, dimers, trimers, and aggregates. Theoretical sedimentation coefficients for spherical adenovirus oligomers predict dimer values ~1.5× monomer (s ≈ 665 Svedberg units for s = 443 Svedberg units monomer).

Particle sedimentation in centrifugation involves three balancing forces: centrifugal force proportional to mass, buoyancy related to density ratios, and friction opposing motion. At terminal velocity, these forces balance, enabling derivation of sedimentation coefficients. The Stokes-Einstein relationship connects diffusion coefficients to particle diameter for spherical particles. However, real nanoparticles deviate from perfect spheres, necessitating the Lamm equation, which describes simultaneous sedimentation and diffusion. This partial differential equation is solved numerically using software like SEDFIT or ULTRASCAN. Two experimental modes exist: sedimentation velocity (dominant for nanoparticles) and sedimentation equilibrium (better for proteins). The choice depends on whether sedimentation or diffusion dominates the particle's movement during the experiment.

The Lamm equation (ΔC/Δt = (1/r)(d/dr)(ω²r²(C - D(r)))) describes concentration distribution during centrifugation. The sedimentation coefficient (s = v/(ω²r)) is expressed in Svedberg units (1 S = 10⁻¹³ seconds) and depends on partial specific volume, molecular mass, and frictional properties. Sedimentation velocity experiments use high speeds where sedimentation dominates diffusion; plotting time versus log radial distance yields ω²s slopes for coefficient determination. Sedimentation equilibrium establishes balance between centrifugal sedimentation and diffusive movement, measuring molecular mass, subunit stoichiometry, and equilibrium constants. Coefficients are standardized to 20°C in water (s₂₀,w) because values depend on solution density and viscosity.

Sedimentation velocity experiments measure boundary movement over time, analyzed using the Lamm equation incorporating diffusion coefficient, sedimentation coefficient, and concentration. Sedimentation equilibrium experiments measure particle distribution at equilibrium, with data fitting exponential curves to determine mass. The Svedberg equation relates sedimentation and diffusion coefficients to molecular mass using the buoyancy factor. Velocity experiments can also be analyzed as equilibrium experiments, yielding accurate mass measurements (within 100 Daltons of sequence mass for 12 kDa proteins).

AUC employs two complementary experimental approaches: sedimentation velocity and sedimentation equilibrium. Sedimentation velocity applies high centrifugal forces (typically 50,000 rpm) causing particles to sediment while diffusion creates characteristic boundary slopes. The sedimentation coefficient (s-value) quantifies sedimentation rate in Svedberg units (10^-13 seconds), depending on buoyant mass, partial specific volume, and frictional coefficient. The Svedberg equation s = m(1 - v̄ρ)/f incorporates these parameters, where f/f₀ (frictional ratio) typically ranges 1.2-1.5 for globular proteins. Computer programs like SEDFIT and SEDPHAT analyze sedimentation profiles by simulating expected distributions based on hydrodynamic parameters. Sedimentation equilibrium achieves balance between centrifugal sedimentation and diffusional back-diffusion at slower speeds, producing steady-state exponential concentration profiles described by c(r) = c₀exp(-sω²r²/(4D)). This method determines absolute molecular masses independent of shape effects by fitting exponential profiles at multiple speeds. Hydrodynamic parameters can be calculated from amino acid sequences or directly measured using viscometers and densitometers. These complementary approaches provide comprehensive characterization of biomolecular properties in solution.

Cryo-EM sample preparation involves removing water and flash-freezing samples at -196°C to preserve native structure. X-ray crystallography requires crystallization of protein samples, where ordered arrays form that diffract X-rays into characteristic patterns. These diffraction patterns are captured on photographic film and analyzed to reconstruct atomic positions. The crystallization process is essential but not always achievable for all proteins.

Cryo-EM sample preparation involves applying purified samples to electron microscopy grids, spreading into thin films, and rapid vitrification in liquid ethane. This process freezes water in an amorphous state, preserving native molecular structures. Electron crystallography, developed by Richard Henderson, was used to determine tubulin structure by imaging two-dimensional crystals formed by anti-parallel protofilament association. Taxol stabilizes these polymers, enabling structure determination. This approach yielded 20-angstrom resolution structures that correctly predicted lateral contacts and surface exposure.

Cryo-EM sample preparation involves plunge freezing using specialized robots: grids are loaded with 2-3 μL sample, blotted to achieve 70-100 nm ice thickness, then rapidly plunged into cryogenic ethane/propane liquid. The resulting cryo-grids feature support films with ordered holes (holo-grids) where vitrified ice acts as a matrix holding particles in near-native hydrated states. Applications span viral structures (coronavirus spikes, bacteriophages, tobacco mosaic virus), engineered nanoparticles (DNA origami with lipid bilayers), and synthetic molecules. Signal averaging improves signal-to-noise ratios for high-resolution reconstructions. Cryo-EM complements crystallography by enabling study of solution-phase structures, revealing inter-particle distances, unit cells, and lattice constants through diffraction analysis.

Cryo-electron microscopy requires specialized sample preparation to preserve protein structure at near-native states. Samples are applied to copper grids coated with carbon films containing small holes, allowing the electron beam to pass through. Excess liquid is removed through blotting, leaving only a thin layer of sample solution. The critical step is vitrification: grids are rapidly plunged into liquid ethane, freezing the sample so quickly that water forms vitreous (amorphous) ice rather than crystalline ice. Crystalline ice scatters electrons and creates imaging artifacts, while vitreous ice preserves the protein structure. The grids are then stored under liquid nitrogen until imaging. This preparation method enables visualization of proteins at resolutions approaching atomic detail, revealing the intricate architecture of molecular machines like the bacterial flagellar motor.

Cryo-electron microscopy (cryo-EM) is a powerful structural biology technique that uses electrons accelerated to 100-300 keV to image biological specimens at near-atomic resolution, employing three main workflows: microelectron diffraction for small molecule and protein crystals, single particle imaging for purified protein complexes, and cryo-electron tomography for visualizing native cellular ultrastructure. The technique addresses radiation damage through low-dose methods and requires sophisticated sample preparation including cryo-FIB (focused ion beam) milling for thicker samples. EMSL provides access to state-of-the-art cryo-EM resources through competitive proposal mechanisms, supporting both individual and collaborative research projects in structural biology.

Large-scale purification of viral vectors involves sequential unit operations: upstream cell culture in HEK cells produces virus; cell disruption releases particles; depth filtration removes debris; tangential flow filtration concentrates and partially purifies; ion exchange chromatography captures and purifies viral particles. Traditional agarose bead chromatography has limitations for large viral particles (~100 nm). Nano-fiber materials with layered, non-woven cellulose structures provide larger surface areas for efficient binding and elution of adenoviruses. Successful purification achieves ~90% recovery, minimal host cell protein contamination, and maintains viral structural integrity confirmed by electron microscopy. This technology has been commercialized for industrial applications.

Large-scale purification using LentiHero1 FPLC device achieved diafiltration recovery of nearly all antiviral particles. Weak anion exchange fiber achieved sharp resolution at low conductivity, yielding 70% viral genome recovery and 53% functional recovery (measured by transduction assay). Processing 200 mL suspension and 242 mL adherent feed achieved 53% functional LV recovery (2-5×10^9 TU). Impurity removal reached 98% for HCP and 82-85% for DNA. Scalability modeling demonstrated that for 50 L bioreactor volume (clinical scale), only 250 mL nanofiber absorbent would be required, meaning one liter unit could process 200 L bioreactor in single pass, satisfying demands for large-scale lentiviral vector production.

Producing viral vectors for clinical use requires scaling from laboratory volumes to industrial production. The process involves upstream operations (cell expansion, infection/transfection, harvest) and downstream purification. Chromatography separates viral particles from contaminants based on charge properties. Tangential flow filtration concentrates and buffers viral preparations. Successful technology transfer requires integrating virology principles with engineering processes across six orders of magnitude in scale.

This comprehensive section covers the complete spectrum of viral vector manufacturing for gene therapy applications. It begins with fundamental characteristics of adeno-associated viruses (AAVs) at 18-28 nm, replication-incompetent requiring helper viruses, and lentiviruses as retroviruses integrating RNA into host genomes. The manufacturing landscape uses HEK293 producer cells with triple plasmid transfection for AAVs and four-plasmid systems with self-inactivating elements for lentiviruses. Four key challenges are addressed: high costs from expensive plasmids and transfection reagents, competitive market pressures requiring accelerated timelines, increasing regulatory requirements for quality attributes, and the need for flexible platforms accommodating engineered serotypes. Platform-based approaches provide predefined manufacturing procedures, established analytical panels, and scalable technologies from Phase 1 to commercial scale. Downstream purification employs generic approaches with two-step clarification, affinity chromatography for capture, and ion exchange chromatography for full/empty particle separation. Full/empty characterization combines size exclusion chromatography with UV detection and multi-angle light scattering. Platform implementation requires only transgene sequence and serotype input, with first visibility runs commencing immediately using established protocols. Technology transfer follows structured methodologies including gap analysis and small-scale model confirmation. High-throughput technologies combined with statistical analysis accelerate development packages, with some assays achievable in one week using next-generation sequencing. Typical AAV production yields range from 1×10^11 to 1×10^12 particles per milliliter depending on serotype and transgene.

Two main types of modified viruses are used as vectors in gene therapy: lentiviruses and adeno-associated viruses (AAV). Lentiviruses are derived from HIV but are inactivated and modified so they cannot replicate while still retaining their ability to infect cells. They are particularly effective for bone marrow and hematological conditions. AAVs are non-infectious but can efficiently infect cells and are used for cardiac diseases as well as neurological and liver conditions. The choice of vector depends on the target organ and its tropism characteristics.