Slalom Chromatography (Part 1): From Gels to Modern UHPLC Slalom Chromatography

Advertisement Home > Columns > Column: Column Watch News | Articles | April 6, 2026 (Updated: April 7, 2026) Slalom Chromatography (Part 1): From Gels to Modern UHPLC Slalom Chromatography Author(s) Fabrice Gritti , Jamuna Vaishnav , Balasubrahmanyam Addepalli +2 More https://doi.org/10.56530/lcgc.int.dk3184p9 Listen 0:00 / 45:02 Key Takeaways Early SC separations inverted SEC/HDC elution order under non-adsorptive conditions, motivating a retention concept based on DNA reorientation and hydrodynamic interactions within packed beds. Abandonment reflected missing pH-stable, mechanically robust, low-adsorption packings, bio-compatible hardware, complex mobile-phase requirements, and absent physicochemical validation of retention and mass-transfer models. SHOW MORE Part 1 of this two-part article series reviews the history of slalom chromatography (SC), the rationale for its recent revival, and the updated mechanistic framework of this technique. Rooted in the fundamental physical properties of deoxyribonucleic acid/ribonucleic acid (DNA/RNA) biopolymers and laminar flow dynamics in packed chromatographic beds, SC is highlighted for its most promising applications in cell and gene therapy. Recent two-year experimental investigations have revealed that SC operates through an out‑of‑equilibrium mechanism, coupling the entropic elasticity of double‑stranded deoxyribonucleic acid/ribonucleic acid (dsDNA/RNA) with the extension and shear forces generated within the interparticle spaces of ultrahigh-pressure liquid chromatography (UHPLC) columns. In contrast to gel permeation chromatography (GPC) or hydrodynamic chromatography (HDC), SC functions at high speed and uniquely elutes smaller dsDNA fragments first, followed by progressively larger ones. Unlike agarose gel electrophoresis (AGE), SC achieves separations of large DNAs in less than three minutes, with nearly twice the resolution power of typical AGE. These findings paved the way for the recent design of a new SC column intended for use in cell and gene therapy as an alternative to standard AGE. Further applications of this new SC column within cell and gene therapy workflows will be discussed in Part 2. In the 1980s, the most widely used technique for separating, quantifying, and preparing double-stranded deoxyribonucleic acid (dsDNA) fragments was agarose gel electrophoresis (AGE).1 However, the separation of large DNAs above 20 kbp was not achievable with conventional AGE until pulsed‑field gel electrophoresis (PFGE) was introduced.2,3 The main limitations of AGE and PFGE are their procedural complexity (gel preparation, voltage supply, staining, and DNA detection) and their long analysis times, approximately 2 h for AGE and more than 10 h for PFGE. To address these issues, in 1988, Boyes4 and Kasai5 tested the performance of gel permeation chromatography (GPC)—a technique widely used for protein separation—for large DNA separations. Remarkably, both research groups independently demonstrated that large dsDNA fragments could indeed be successfully separated using conventional gel permeation chromatography supports. Kasai termed this new separation mode slalom chromatography (SC), and the name was officially born. While this breakthrough opened the door to the development of new columns for the rapid characterization of large DNA samples, the technology failed to gain significant attention. Between 1988 and 2004, no more than 15 original research articles—fewer than one per year—were published on both the experimental6–16 and theoretical17–19 fronts. After this period, SC was abandoned and no longer considered as a potential alternative to AGE. The reasons why SC was not fully recognized or supported as a powerful separation technique over the past two decades have not been clearly addressed in the literature. This merits attention and better understanding given all the significant improvements in column technologies following the introduction of ultrahigh-pressure liquid chromatography (UHPLC). In Part 1 of this series, we first revisit the historical context of SC, explaining why it was not adopted as a new promising separation technique until 2023, when Waters Corporation invested research resources into the commercialization of the first SC column. This part specifically highlights the extensive experimental investigations undertaken to unravel the hidden retention and mass‑transfer mechanism in SC. It also demonstrates how the unique retention and mass‑transfer properties of SC are deeply rooted in the physics of single DNA biopolymers and fluid hydrodynamics in a packed chromatographic column, and why SC can now compete well with the current gold standard in DNA separation, AGE, based on the quality-by-design (QbD) of a new SC column released in 2025. The main applications in cell and gene therapy workflow of this new SC column will be discussed in Part 2. Materials and Methods The Optima-grade solvents used in this work were water and acetonitrile from Fisher Scientific. Phosphate buffered saline (PBS) tablets, sodium monophosphate salt (NaH2PO3), disodium phosphate salt (Na2HPO3), 10X Tris Acetate EDTA (TAE), sucrose, and cytosine were also acquired from Fisher Scientific. The three dsDNA sample mixtures, Lambda DNAHind III Digest (2.0, 2.3-, 4.4-, 6.6-, 9.4-, and 23.1 kbp dsDNAs), Lambda DNA-BstEII Digest (1.264-, 1.371-, 1.929-, 2.323-, 3.675-, 4.324-, 4.822-, 5.686-, 6.369-, 7.242-, and 8.454-kbp dsDNAs), and Lambda DNA-Mono Cut Mix (1.5-, 10.0-, 15.0-, 17.0-, 24.0-, 24.5-, 29.9-, 33.5-, 38.4-, and 48.5-kbp dsDNAs), were procured from New England BioLabs Inc. Three homemade research prototype columns with MaxPeak Premier Technology column hardware (4.6 mm internal diameter (i.d). x 300 mm long) were used in this work. These research prototype columns were packed with monodisperse non-porous 2.6-µm silica particles obtained on-site (Waters), porous 1.7-, 2.4-, 5.4-, and 11.3-µm BEH125 particles produced on-site (Waters), and with polydisperse non-porous 3.0-µm silica particles purchased from Glantreo. The restrictor capillaries (40 µm i.d. x 30 inch, 40 µm i.d. x 24 inch, and 40 µm i.d. x 16 inch) were homemade and connected to the column outlet to increase the average pressure along the column. The new column dedicated to SC applications and used in this work is the 4.6 x 300 mm, 2.5-µm Waters GTxResolve 250 Å Slalom Column, MaxPeak Premier Technology. It is packed with 2.5 µm particles in a metal column hardware modified with a hydrophilic vapor- deposited surface. The particle surface is made inert to DNA after modification with a polyethylene oxide-bonding treatment. All the chromatograms were recorded on an Acquity UPLC I-Class Plus System (Waters). It is equipped with a binary pump with solvent selection valves, a flow-through needle (FTN) autosampler, an external one-column heater module (embedding 4.6 x 300 mm columns), active solvent pre-heaters, left and right column selection valves (each with nine ports and eight positions), and a tunable wavelength UV–vis detection system (500 nL cell). The sample loop (125-µm i.d.), the needle seat capillary (75-µm i.d.), and the active pre-heater tubes (75-µm i.d.) were all surface-modified with hydrophilic hybrid organic-inorganic silica. The system is controlled by EmpowerSoftware 3.0 (Waters) and can be operated up to a maximum pressure of 15,000 psi at 1 mL/min. A Brief History of Slalom Chromatography Trials (1988–2004) In 1988, Jun Hirabayashi and Ken‑ichi Kasai of the Faculty of Pharmaceutical Sciences in Sagamiko, Japan, together with American researchers Barry Boyes, Douglas Walker, and Patrick McGeer from the Department of Psychiatry at the University of British Columbia in Vancouver, Canada, independently discovered that double‑stranded DNA (dsDNA) fragments with molecular weights exceeding 2 MDa could be separated using conventional size‑exclusion chromatography (SEC) columns. The Japanese group employed porous SEC particles of 5 µm and 9 µm with average pore diameters of 250 Å,5 while the Canadian group used 4-µm particles with the same mesopore size.4 These packing materials were originally designed for the separation of large biomolecules such as monoclonal antibodies, typically around 100 Å in size. Remarkably, under non‑adsorptive conditions, both groups observed that the elution order of dsDNA fragments, ranging from 2 kbp to 23.1 kbp in length and corresponding to hydrodynamic diameters of approximately 0.1 µm to 0.5 µm, was the exact opposite of the conventional exclusion mechanism normally associated with SEC or hydrodynamic chromatography (HDC). The very first SC chromatograms ever recorded are shown in Figure 1. At the time, neither group could provide a physicochemical explanation for this unexpected phenomenon. Figure 1. (a) First separation of large dsDNA fragments (HindIII digest of a 48.5 kbp λDNA, from 2 kbp to 23.1 kbp) reported by Boyes et al.1 in 1988, performed on a GPC DuPont Bioseries GF-250 column (9.4 mm i.d. x 250 mm) packed with 4 μm silica spheres modified with a hydrophilic diol-bonded phase. Mobile phase: 50 mM Tris buffer, 0.5 mM EDTA, pH = 7.5. Flow rate: 0.25 mL/min. Room temperature. (b) First separation of the same large dsDNA fragments shown in the left panel, reported by Kasai et al 2 in 1988, performed on a GPC Asahipak GS-510 column (7.6 mm i.d. x 500 mm) packed with 9 μm porous polymer-based particles. Mobile phase: 100 mM Tris-HCl buffer, 0.2 M NaCl, pH=7.5. Flow rate: 0.60 mL/min. Room temperature. The origins of SC are firmly historical. The term was first introduced, arbitrarily, by Hirabayashi and Kasai as a new mode of retention and separation in liquid chromatography (LC) in their 1989 publication, Size‑Dependent Chromatographic Separation of Double‑Stranded DNA Which Is Not Based on Gel Permeation Mode.6 The authors intuitively envisioned a DNA polymer behaving like a skier, “slaloming” or repeatedly turning left and right around the packed particles during its migration through the column. Based on this analogy, they proposed a separation mechanism governed by the differential reorientation times of dsDNA molecules: longer fragments would require more time to reorient than shorter ones, thereby explaining the observed order of elution. Over the following 16 years, until the technique was effectively abandoned in 2004, Hirabayashi and Kasai remained the only researchers to conduct significant experimental investigations aimed at elucidating the separation mechanism of SC. Experimental studies published in the 1990s7–9 and in 200010 demonstrated that DNA retardation was primarily influenced by DNA length, particle size, flow rate, and solvent viscosity. However, they ultimately concluded that the precise retention mechanism remained inexplicable. During the early 2000s, similar experimental investigations were performed by Peyrin and Guillaume11–16 before a handful of theoretical models could be proposed.17–19 Yet, the mathematical frameworks proposed lacked robust physicochemical validation, as they were not supported by direct observations of large dsDNA molecules migrating through packed beds, a limitation highlighted in a recent review by Gritti.20 After 2004, coinciding with the rise of ultrahigh‑performance liquid chromatography (UHPLC) and the introduction of sub-2-µm particles, experimental interest in SC declined sharply. Aside from a few retrospective reviews,21–24 no new experimental studies or reports appeared in the scientific literature. The technique was effectively abandoned due to several critical limitations: the lack of suitable packing materials with high pH stability, low non‑specific adsorption, and adequate mechanical strength; bio‑compatible column hardware; the requirement for complex mobile phases containing salts, chelating agents such as ethylenediaminetetraacetic (EDTA), and organic solvents; the absence of a clear understanding of the SC retention mechanism; and a limited application market. Additionally, several alternative separation techniques were already available, though primarily suited for smaller DNA fragments. As shown in Figure 2, analytical techniques such as ion pairing reversed‑phase (IPRP)–LC, anion exchange chromatography (AEX), hydrophobic interaction chromatography (HIC), and size exclusion chromatography (SEC) were commonly employed for sub‑MDa DNA. For larger DNA molecules, techniques including hydrodynamic chromatography (HDC), capillary electrophoresis (CE), AGE, PFGE, analytical ultracentrifugation (AUC), and asymmetric flow field‑flow fractionation (AF4) were widely adopted despite their intrinsic limitations in terms of simplicity, speed, resolution, and accuracy. Figure 2. List of existing analytical separation techniques (bottom) commonly used for the characterization of various biopharmaceutical modalities (top) as a function of molecular size. These modalities range from small molecules to proteins, adeno associated viruses (AAVs), exosomes, virus like particles (VLPs), lipid nanoparticles (LNPs), DNA, and cells. In green, emphasis is placed on the separation of large dsDNA fragments ranging from 2 kbp to 50 kbp, or with hydrodynamic diameters from 0.1 μm to 0.7 μm, for which classical adsorption techniques such as RP IP, AEX, HIC, and HILIC are not suitable due to limited resolution. FFF lacks sufficient resolution, while HDC and AUC are both complex separation techniques. Currently, AGE and PFGE represent the gold standard methods for separating large DNA up to 10 Mbp; however, they suffer from extremely long analysis times. This limitation justifies the revival of a simple, fast, high-resolution, and accurate SC separation technique.20 The Slalom Chromatography Revival Renewed interest in this legacy separation technology has largely been driven by growing demand within the biopharmaceutical industry in the late 2010s to improve the characterization of large DNA and ribonucleic acid (RNA) biopolymers, which exceed 2 MDa in molecular weight and are often highly heterogeneous in both composition and structure. Accurate determination of size and sequence is essential for a wide range of applications, including cell and gene therapy, for example, messenger ribonucleic acid (mRNA) vaccines, DNA restriction mapping, directed evolution via protein selection, and clustered regularly interspaced short palindromic repeats-Cas9 (CRISPR‑Cas9) gene editing. These advances have underscored the need for robust analytical tools capable of resolving complex nucleic acid mixtures with high precision.25 In the early 2020s, the market was pressuring manufacturers to develop new DNA/RNA analytical techniques that were simultaneously simple, fast, high-resolution, accurate, and reproducible.26 Under these conditions, SC emerged as a highly suitable candidate for LC manufacturers compared with the separation technologies listed in Figure 2. A recent review20 highlights the main limitations of IPRP, AEX, hydrophobic interaction chromatography (HIC), hydrophilic interaction chromatography (HILIC), and size exclusion chromatography (SEC) in terms of resolution for DNA sizes greater than a few kbp. Of the others, HDC is complex and restricted to nanoflow, narrow open capillaries, and laser- induced fluorescence detection; AUC requires complex data processing; and AF4 offers only limited resolution. AGE and PFGE remain the most widely adopted techniques for achieving high-resolution separation of DNA mixtures larger than a few kbp but their analytical throughput is slow, requiring one to two hours to complete a single gel. SC emerged as a suitable technique in the early 2020s because pH-stable, non-adsorptive, and mechanically robust sub-3-µm particles were already being manufactured for SEC, designed for the characterization of various peptides, proteins, and monoclonal antibodies. At the same time, bio-inert hardware was being developed to mitigate strong metal– analyte interactions that can cause poor peak shapes and irreversible analyte losses.27–30 These attributes are essential for performing high-resolution separations and handling the many phosphate groups present in DNA/RNA modalities. Nevertheless, in the absence of a clearly established separation mechanism in SC, several critical and fundamental questions remained unanswered regarding the design of the novel SC columns: which particle size, flow rate, temperature, operating column pressure, column length, and buffer/salt concentrations are truly required to achieve simple, fast, highresolution, and reproducible separations of large DNA/RNA biopolymers by SC? Slalom Chromatography Retention and Efficiency Properties: The Road To Victory (2023–2024) To address these fundamental questions, the immediate course of action was to perform systematic experiments by considering seven different experimental parameters, varying one at a time while keeping all others constant. Supplementary Figures S1a–S1g (to be found after the references) summarize these early empirical investigations conducted prior to the commercialization of the first GTxResolve Slalom Chromatography Column. The intrinsic effects of each parameter were carefully analyzed: hydrostatic pressure P (Supplementary Figure S1a, by connecting capillary restrictors at the column outlet, with average column pressures ranging from 900 psi to 7080 psi); temperature T (Supplementary Figure S1b, ranging from 25 oC to 90 oC); mobile phase viscosity ƞ (Supplementary Figure S1c, by adding sucrose up to 40% by weight in the eluent, ranging from 0.9 cP to 5.2 cP); average interstitial linear velocity u (Supplementary Figure S1d, by increasing flow rate from 0.15 mL/min to 1.10 mL/min through a 4.6-mm-i.d. column); ionic strength I of the mobile phase (Supplementary Figure S1e, from 0.001×PBS to 1×PBS pH 7.4 buffer concentration, with both silica particles and DNA negatively charged); particle size dp (Supplementary Figure S1f, ranging from 1.8 µm to 11.3 µm, same strain force); and surface chemistry of the column hardware (Supplementary Figure S1g, bare stainless steel vs. hybrid surface technology [HST]–modified hardware). The empirical conclusions related to the dsDNA retention mechanism in SC became clear and are summarized as follows20, 31–33: 1. DNA retention is independent of hydrostatic pressure under non-adsorbed conditions. 2. DNA retention decreases with increasing temperature as a result of reduced solvent viscosity, which diminishes extensional and shear forces in between the packed particles. 3. DNA retention increases with higher sucrose concentrations as solvent viscosity rises. 4. DNA retention volume increases with higher linear velocity as extensional and shear forces intensify and average DNA extension length (Lext) increases. 5. DNA retention decreases with lower ionic strength, owing to stronger electrostatic repulsions between the negatively charged silica surface (pH ~ 8) and the phosphate backbone of DNA polymer chains. 6. DNA retention decreases at constant extensional/shear force with increasing particle diameter. When the ratio of DNA extension length to particle diameter Lext/dp falls below about 25%, retention is governed by a pure HDC exclusion mechanism. When this ratio exceeds one, a pure SC retention mechanism applies, and the retention factor increases linearly with the average extension length best estimated from the known force–extension curve of a single DNA polymer chain.34–36,20,31 7. DNA peak shape is deteriorated and DNA mass is lost when biopolymers are exposed to stainless-steel surfaces compared with hybrid surface technology-modified hardware.27–30 It required nearly two years to plan, collect, and analyze these empirical data, and to provide a consistent interpretation throughout. Figure 3 shows the variation of the zone retention factor k1 (the reference hold-up time is the external interparticle void time) of dsDNAs of sizes ranging from 1.5 kbp to 48.5 kbp measured on the GTxResolve 250Å Slalom Column, MaxPeak Premier Technology (4.6 x 300 mm, 2.5 µm when run at 25 oC using a 10XTAE buffer pH 8.0 mobile phase and a series of flow rates from 0.15-, 0.30-, 0.50-, 0.70-, 0.90- to 1.10 mL/min.37 When the average extension length exceeds the particle diameter (dp), these results demonstrated convincingly that the retention of dsDNAs (k1 > 0.2) of varying sizes is governed by a unique extension/shear stress parameter, which is written: Equation 1 From a practical viewpoint, the experimental data indicate that the retention of DNA is unchanged as long as the stress parameter τ remains constant, irrespective of temperature, flow rate, and particle diameter used. In contrast, Lext when is shorter than approximately 25% of dp, the DNA biopolymer is too small and is simply excluded according to a conventional HDC retention mechanism (k1 < 0). For intermediate extension lengths (0.25dp < Lext < dp), a mixed HDC–SC retention mechanism occurs and the selectivity between critical pairs of linear dsDNAs reaches a maximum for DNA sizes around 6–7 kbp.31,32 As the extended length of the DNA increases, it becomes less excluded from the particle surface (hence the HDC retention factor is less negative) and the probability that the DNA chain is in proximity of the particle surface increases (hence the SC retardation increases). Supplementary Figure S2 perfectly illustrates this mixed HDC (exclusion)–SC (retardation) retention mechanism and presents the same plots as in Figure 3, obtained at a constant flow rate of 1 mL/min, but under three different ionic strengths: 4-, 40-, and 400-mM TAE. As expected, increasing the buffer concentration increases DNA retention because this more effectively screens the negative charges on the silica particle surface, allowing the extended DNA chains to be less excluded, approach closer to the particle surface, be more stretched, and experience more local velocities approaching zero. Figure 3. Retention properties of the 250 Å. BEH GTxResolve Slalom Column (4.6 mm i.d. x 300 mm, 2.5- μm) as a function of the size of linear dsDNA fragments (ranging from 1.5 kbp to 50 kbp) across seven applied flow rates: 0.15-, 0.3-, 0.5-, 0.7-, 0.9-, and 1.1mL/min. Mobile phase: 10xTAE buffer, pH 8. Room temperature. Each data point corresponds to the elution of dsDNA fragments present in three λ-DNA digest mixtures (HindIII, BstEII, and Mono Cut Mix). Three distinct retention regimes are observed depending on the ratio of the extended length of the dsDNA fragments to the particle diameter: the HDC separation mechanism applies (k1 < 0) when the ratio falls below 25%; the SC retention mechanism applies (k1) > 0) when the ratio exceeds 1. Most remarkably, selectivity is highest when a mixed HDC–SC retention mechanism occurs at intermediate values of this size ratio. For the GTxResolve Slalom Column, maximum selectivity is expected for DNA fragments around 7 kbp.37 Interestingly, as shown in Figure 4, and from a purely empirical perspective, the efficiency of the recorded DNA peaks appears to be a uniquely decreasing function of the retention factor (k1), regardless of the choice among the many possible experimental parameters. In practice, it is generally recommended to maintain k1 below 0.3 to maximize the resolution power of the slalom column, particularly around k1 ~ 0.32 Figure 4. Efficiency properties of the recently commercialized 250 Å BEH GTxResolve Slalom Column (4.6 mm i.d. x 300 mm, 2.5 μm) as a function of the retention factor (k1) of linear dsDNA fragments under varying experimental conditions of temperature (5–70 oC), flow rates (0.1–1.4 mL/min), and TAE buffer concentration (4–400 mM). Mobile phase: TAE buffer, pH 8. Each data point corresponds to the elution of dsDNA fragments present in three l-DNA digest mixtures (HindIII, BstEII, and Mono Cut Mix). A strong empirical correlation was observed between column efficiency and DNA retention, explained by the change in relaxation time (tR) required for dsDNA fragments to transition from their extended, retained, non equilibrium conformation to their random coil, free, equilibrium conformation. Maximum resolution factors, reflecting the optimal trade-off between selectivity (Figure 3) and plate counts (this figure), are consistently observed at k1 ~ 0.72.32 The translation of this empirically determined separation mechanism into a real-life application is illustrated in Figure 5 (DNA restriction mapping), which shows the highresolution separation of 13 linear dsDNA fragments (sizes from 1.3 kbp to 14.4 kbp including the cohesive ends assembly of the 5.686 and 8.454 fragments as revealed in Supplementary Figure S1d) generated by the BstII digest of λ-DNA (48.5 kbp linear dsDNA). Figure 5. Illustration of the separation performance of a 250 Å BEH GTxResolve Slalom Column (4.6 mm i.d. x 300 mm, 2.5 μm) for DNA restriction mapping following injection of 2 mL of the BstEII λ-DNA digest containing 12 dsDNA fragments, as indicated in the chromatogram (fragment sizes shown in blue, kbp units). Notably, the 5.686 and 8.454 kbp fragments carry complementary 5’ strand overhang (12 bases) that hybridize to form a new 13th fragment of 14.14 kbp (fragment sizes shown in green), eluting at 2.55 min. Mobile phase: 10xTAE buffer, pH 8. Flow rate: 1.1 mL/min. Room temperature. In agreement with Figures 3 and 4, the peak capacity of the SC column is maximal for the separation of 4–9 kbp DNA fragments, reaching 11 in less than 2.5 min (combined HDC- SC separation mechanism, -0.15 < k1 < 0.20). For comparison, the peak capacity measured by AGE is only 7, with gel runs lasting up to 1 h (see top right corner insert). Separation is extremely limited for DNA fragments smaller than 2.5 kbp (full HDC-based separation mechanism, k1 < -0.20) and moderate for fragments larger than 11 kbp (full SC separation mechanism, k1 > 0.25). These results confirm that column efficiency decreases with increasing retention factor, as evidenced by broader peak widths, and that resolution is maximized for DNA fragments around 6–7 kbp compared with lower resolution factors observed for fragments between 1.3–3.7 kbp or above 10 kbp. Overall, an excellent peak capacity of 11 was achieved for the elution of 4.3 kbp to 8.5 kbp linear dsDNA fragments in less than 2.5 min. By contrast, the insert in Figure 6 shows the AGE separation (1% agarose gel), which required 1.5 h to complete and yielded a peak capacity of only 7. Figure 6. Changes in the conformation of a linear dsDNA fragment as a function of the force applied to one end while the other end is fixed. The force range extends from less than 10 fN (dsDNA random coil conformation) to 1 nN (ssDNA breakage). The linear dsDNA is first uncoiled through weak entropic elasticity (persistence length ~ 500 Å), reaching up to 80% of its contour length at a force of 2 pN. It is then slightly extended via strong enthalpic elasticity up to ~65 pN, at which point the double helix structure is lost and the molecule becomes overstretched. The resulting ssDNA strands are further stretched enthalpically (strong forces applied) until the phosphodiester bonds break down at ~1 nN (Bouchiat). Forces generated in the SC column never exceed 2 pN. Immersion in the Physics of Slalom Chromatography Figure 6 shows the changes in the configuration of a single linear dsDNA chain upon the application of a force at one end while the other end is fixed. When the applied force is smaller than 10 fN (1 fN=10-15Newton), the linear DNA still adopts its equilibrium random coil conformation. For example, the hydrodynamic diameter of a 48.5 kbp λ-DNA is as large as 0.7 µm , owing to the long persistence length (a measure of chain rigidity) of dsDNA, approximately 500 Å. dsDNA is a semi-flexible biopolymer compared to single-stranded (ss) DNA, which has a much lower persistence length of only about 10–20 Å. When the applied force ranges from 10 fN to about 2 pN (1 pN=10-12 Newton), the dsDNA coil is progressively undone until its extended length reaches the contour length. This regime is governed by the entropic elasticity (low spring constant) of dsDNA, which is inversely proportional to its persistence length.34–36 For example, single-stranded deoxyribonucleic acid (ssDNA) and mRNA coils cannot be undone within this force range because their persistence lengths are too short. From 2 pN to 60 pN, the extended length of dsDNA increases only slightly, owing to the enthalpic elasticity (high spring constant) of the unfolded DNA chain. Remarkably, the dsDNA length extends by approximately 90% as it overstretches and loses its double-helix structure at a constant force of about 65 pN. For forces greater than 65 pN, the ssDNA is extended according to its own enthalpic elasticity until the weakest phosphodiester bond typically breaks at a force close to 1 nN (1 pN=10-9 Newton). The question asked was then to evaluate the force involved in the interparticle volume of SC columns. Considering a 4.6-mm-i.d. column packed with 2.5-µm particles and operated at 1 mL/min, the force applied to 2.5 kbp and 50 kbp ranges typically from 100 fN to 2 pN, that is, forces are sufficient to extend significantly such dsDNAs from 50% up to 80% of their maximum contour lengths.20,31 The right panel in Figure 7 shows the relationship between the force applied and the extension of 48.5 kbp linear λ-DNA as reported by Bouchiat.36 The retention and mass transfer mechanism is then fully interpreted by the coilto-stretch transition of the linear dsDNA. Figure 7. Physics-based explanation and description of the separation mechanism in SC: (Top left panel) Evidence of local velocity gradients in the interparticle spaces of random sphere packings obtained from numerical simulations of the Navier–Stokes equations in fluid dynamics. These gradients generate a combination of extension and shear rates that stretch the dsDNA biopolymer. (Bottom left panel) Direct visualization of the coiled and extended conformations, as well as the coil–stretch transition, of a large linear 166 kbp dsDNA polymer migrating through spaces defined by 10 mm i.d. pillars arranged in a square array (Strom). (Right panel) Experimental data (open circles) showing the relationship between applied force and extension length of a linear 48.5 kbp λ-DNA biopolymer (Bouchiat). LC forces are sufficient to significantly stretch the DNA biopolymer. The left panel in Figure 7 shows the calculated velocity profile of the mobile phase across a cross‑section of a packed column and the two configurations of a large 166 kbp dsDNA polymer migrating between 10-μm-diameter pillars arranged in a cubic array.38 The left panel reveals that under laminar flow conditions, velocity gradients occur at many locations, generating continuous extension forces (along the flow direction) and shear forces (perpendicular to the flow direction) acting on the analytes. When the DNA polymer is located far from the particle surface or near the center of a flow‑through channel, the extension and shear rates are minimal, and the dsDNA relaxes and retains its coiled configuration. It relaxes into an equilibrium “free” state and migrates rapidly at the local velocity of the mobile phase. In contrast, when a fraction of the DNA polymer chain is positioned close to the particle surface or within constricted regions of the separation medium, the extension and shear forces are maximum and stretch the DNA chain. In this non-equilibrium “retained” state, the DNA migrates at a slower velocity even though no physisorption occurs. This situation is analogous to an adsorption–desorption process, which effectively describes the DNA retention and efficiency observed. Retention is governed by the size or length of dsDNA through its probability of being exposed to the highest extension and shear forces, while efficiency is determined by the effective relaxation times between the “retained” and “free” states.39 This resolves the mystery noted by Boyes and provides a complete fundamental microscopic interpretation of the retention and mass‑transfer mechanism in SC, an explanation that Kasai and others had been seeking for more than 25 years. Conclusion In part 1, the history, resurrection, and retention and mass transfer mechanism of SC has explained why this separation technique was rapidly abandoned in the early 2000s, remained dormant for two decades, and was revived in the early 2020s. The combination of strong demand from the biopharmaceutical industry to improve the characterization of large DNA/RNA molecules, the absence of a clearly established retention and mass‑transfer mechanism, and the availability of the latest high‑performance UHPLC particles (high‑class purity type A, sub‑3-µm in size, free of non‑specific adsorption sites, with HST‑modified stainless‑steel hardware) and systems (HST‑modified) propelled intensive empirical and fundamental investigations in DNA separation, ultimately leading to the design of the first commercialized slalom column in the history of chromatography. What has been learned over the last three years now provides a clear explanation for the early, previously unexplained, observations made by Boyes and Kasai regarding the separation and efficiency of SC columns. This physics mechanism is based on the slow coilto-stretch transition of dsDNA biopolymers within the interparticle volume of a randomly packed column. Retention occurs only for dsDNAs/dsRNAs, owing to their long persistence length and weak entropic elasticity, whereas ssDNA and mRNA are excluded due to their too short persistence lengths. Additionally, the contour length of retained dsDNA/dsRNA must exceed approximately 25% of the particle diameter. Retention should not be too high, as overly large dsDNA/dsRNAs exhibit slow relaxation times, causing extremely wide peaks, let alone DNA trapping in the column. Consequently, the retention of 3–25 kbp dsDNA was found to be optimal on a 4.6 × 300 mm column (bio‑inert HST‑modified hardware) packed with 2.5-µm polyethylene oxide‑surface‑modified hybrid BEH particles. The internal particle porosity (250 Å) was selected to enhance separation of the targeted dsDNA/dsRNA analytes from smaller buffer molecules, proteins, enzymes, and oligomers, which are more strongly retained according to a SEC retention mechanism. In Part 2, we will illustrate real‑world applications of the new SC column in the field of cell and gene therapies and compare its performance to that of the conventional AGE separation technique. This includes characterization of plasmids (supercoiled vs. linear) and quantification of dsRNAs in mRNA vaccines prepared by in vitro transcription (IVT). Focus will be given on the practical advantages of the SC column, that is, simplicity, speed, resolution, and analytical precision, for determining the length and mass of injected dsDNA/dsRNA. SC’s overall performance will be benchmarked against the current gold standard, AGE, to provide the biochemists and biologists with new options to enhance and expedite their daily laboratory work. Acknowledgments Waters, MaxPeak, BEH, GTxResolve, Acquity, UPLC, and Empower are trademarks of Waters Corporation or its affiliates. Optima is a trademark of Fisher Scientific Company LLC. References 1. Maniatis, T. A.; Fritsch, E. F.; Sambrook, J. Molecular Cloning: A Laboratory Manual; Cold Spring Harbor Laboratory Press, 1982. 2. Schwartz, D. C.; Cantor, C. R. Separation of Yeast Chromosome‑Sized DNAs by Pulsed Field Gradient Gel Electrophoresis. Cell 1984, 37, 67–75. DOI: 10.1016/00928674(84)90301-5 3. Carle, G. F.; Olson, M. V. Separation of Chromosomal DNA Molecules from Yeast by Orthogonal‑Field‑Alternation Gel Electrophoresis. Nucleic Acids Res 1984, 12, 5647– 5664. DOI: 10.1093/nar/12.14.5647 4. Boyes, B. E.; Walker, D. G.; McGeer, P. L. Separation of Large DNA Restriction Fragments on a Size‑Exclusion Column by a Nonideal Mechanism. Anal Biochem 1988, 170(1), 127–134. DOI: 10.1016/0003-2697(88)90099-1 5. Hirabayashi, J.; Kasai, K.‑I. Slalom Chromatography: A New Size‑Dependent Separation Method for DNA. Nucleic Acid Res Symp Ser 1988, 20, 67–68. DOI: 10.1021/bi00493a004 6. Hirabayashi, J.; Kasai, K.‑I. Size‑Dependent Chromatographic Separation of Double‑Stranded DNA Which is Not Based on Gel Permeation Mode. Anal .Biochem 1989, 178, 336–341. DOI: 10.1016/0003-2697(89)90649-0 7. Hirabayashi, J.; Ito, N.; Noguchi, K.; Kasai, K.‑I. Slalom Chromatography: Size‑Dependent Chromatographic Separation of Double‑Stranded DNA by a Hydrodynamic Phenomenon. Biochemistry 1990, 41, 9515–9521. DOI: 10.1021/bi00493a004 8. Kasai, K.‑I. Size‑Dependent Chromatographic Separation of Nucleic Acids. J Chromatogr B: Biomed Sci Appl 1993, 618, 203–221. DOI: https://doi.org/10.1016/03784347(93)80035-3 9. Hirabayashi, J.; Kasai, K.‑I. Effects of DNA Topology, Temperature and Solvent Viscosity on DNA Retardation in Slalom Chromatography. J. Chromatogr. A 2000, 893, 115–122. DOI: 10.1016/s0021-9673(00)00693-2 10. Hirabayashi, J.; Kasai, K.‑I. Applied Slalom Chromatography Improved DNA Separation by the Use of Columns Developed for Reversed‑Phase Chromatography. J Chromatogr A 1996, 722, 135–142. DOI: 10.1016/0021-9673(95)00505-6 11. Peyrin, E.; Guillaume, Y.; Grosset, C.; et al. Mobile‑Phase Viscosity Dependence on DNA Separation in Slalom Chromatography. J Chromatogr A 2000, 886, 1–7. DOI: 10.1016/S0021-9673(00)00510-0 12. Peyrin, E.; Guillaume, Y. C.; Garrel, C.; et al. Effect of Temperature on DNA Fractionation in Slalom Chromatography. Talanta 2000, 52, 1105–1110. DOI: 10.1016/s0039- 9140(00)00482-3 13. Peyrin, E.; Caron, C.; Garrel, C.; et al. DNA Migration Regimes in Hydrodynamic Chromatography and Slalom Chromatography: Evidence for a Transition. Talanta 2001, 55, 291–296. DOI: 10.1016/S0039-9140(01)00419-2 14. Perrin, F. X.; Masuyer, C.; Truong, T. T.; et al. Supercoiled Circular DNA and Protein Retention in Non‑equilibrium Chromatography. Temperature and Velocity Dependence: Testimony of a Transition. J Chromatogr A 2002, 950, 281–285. DOI: 10.1016/s00219673(02)00023-7 15. Guillaume, Y.; Andre, C.; Matoga, M.; et al. Supercoiled Circular DNA Retention in Non‑equilibrium Chromatography: Viscosity and Velocity Dependence — Behavior with Proteins. J Chromatogr Sci. 2003, 41, 133–136. DOI: 10.1093/CHROMSCI/41.3.133 16. Andre, C.; Guillaume, Y. Novel Strategy in Slalom Chromatography for Studying Both the Protein Reptation Mechanism and the Compacting Agent Effect to Improve Oligonucleotide Separation. Chromatographia 2004, 59, 487–492. DOI: 10.1365/s10337-004-0242-8 17. Peyrin, E.; Guillaume, Y. C.; Villet, A.; Favier, A. Mechanism of DNA Hydrodynamic Separation in Chromatography. Anal Chem 2000, 72, 853–857. DOI: 10.1021/ac990841s 18. Guillaume, Y. C.; Peyrin, E.; Thomassin, M.; et al. Column Efficiency and Separation of DNA Fragments Using Slalom Chromatography: Hydrodynamic Study and Fractal Considerations. Anal Chem 2000, 72, 4846–4852. DOI: 10.1021/ac000281t 19. Guillaume, Y. C.; Perrin, F. X.; Guinchard, C.; et al. Separation in Slalom Chromatography: Stretching and Velocity Dependence. Anal Chem 2002, 74, 1217–1222. DOI: 10.1021/ac010852x 20. Gritti, F. Retention Mechanism in Slalom Chromatography: Perspectives on the Characterization of Large DNA and RNA Biopolymers in Cell and Gene Therapy. J Chromatogr A 2025, 1743, 465691. DOI: 10.1016/j.chroma.2025.465691 21. Dias, R. P. Size Fractionation by Slalom Chromatography and Hydrodynamic Chromatography. Recent Patents Eng. 2008, 2, 95–103. DOI: 10.2174/187221208784705251 22. Li, J.; Peng, L.; Geng, X. Two Dynamic Chromatographies and their Application. Chin J Anal Chem 2009, 37 (7), 1082–1087. DOI: 10.1016/S1872-2040(08)60118-8 23. Khanvilkar, V.; Chitnis, A.; Shirode, A.; Kadam, V. Slalom Chromatography: An Overview. Int J Pharm Res & Chem 2012, 2, 66–76. DOI: https://doi.org/10.1016/j.chroma.2025.465691 24. Diogo, M.; Queiroz, J.; Prazere, D. Chromatography of Plasmid DNA. J Chromatogr A 2005, 1069, 3–22. DOI: 10.1016/j.chroma.2004.09.050 25. Ghanem, A.; Healey, R.; Adly, G. Current Trends in Separation of Plasmid DNA Vaccines: A Review. Anal Chim Acta 2013, 760, 1–15. DOI:10.1016/j.aca.2012.11.006 26. Camperi, J.; Lippold, S.; Ayalew, L.; et al. Comprehensive Impurity Profiling of mRNA: Evaluating Current Technologies and Advanced Analytical Techniques. Anal Chem 2024, 96, 3886–3897. DOI: 10.1021/acs.analchem.3c05539 27. Delano, M.; Walter, T.; Lauber, M.; et al. Using Hybrid Organic‑Inorganic Surface Technology to Mitigate Analyte Interactions with Metal Surfaces in UHPLC. Anal Chem 2021, 93, 5773–5781. DOI: 10.1021/acs.analchem.0c05203 28. Gilar, M.; Delano, M.; Gritti, F. Mitigation of Analyte Loss on Metal Surfaces in Liquid Chromatography. J Chromatogr A 2021, 1650, 462247. DOI: 10.1016/j.chroma.2021.462247 29. Fekete, S.; Kizekai, L.; Sarisozen, Y. T.; et al. Investigating the Secondary Interactions of Packing Materials for Size‑Exclusion Chromatography of Therapeutic Proteins. J Chromatogr A 2022, 1676, 463262. DOI: 10.1016/j.chroma.2022.463262 30. Fekete, S.; Delano, M.; Harrison, A.; et al. Size Exclusion and Ion Exchange Chromatographic Hardware Modified with a Hydrophilic Hybrid Surface. Anal Chem 2022, 94, 3360–3367. DOI: 10.1021/acs.analchem.1c05466 31. Gritti, F.; Wyndham, K. Retention Mechanism in Combined Hydrodynamic and Slalom Chromatography for Analyzing Large Nucleic Acid Biopolymers Relevant to Cell and Gene Therapies. J Chromatogr A 2024, 1730, 465075. DOI: 10.1016/j.chroma.2024.465075 32. Gritti, F. Theoretical Predictions to Facilitate the Method Development in Slalom Chromatography for the Separation of Large DNA Molecules. J Chromatogr A 2024, 1736, 465379. DOI: 10.1016/j.chroma.2024.465379 33. Gritti, F. Ultra-High Pressure Slalom Chromatography: Application to the Characterization of Large DNA and RNA Samples Relevant in Cell and Gene Therapy. J. Chromatogr. A 2024, 1738, 465487. DOI: https://doi.org/10.1016/j.chroma.2024.465487 34. Bustamente, C.; Marko, J.; Siggia, E.; Smith, S. Entropic elasticity of 𝜆-phage DNA. Science 1994, 265, 1599–1600. DOI: 10.1126/science.8079175 35. Marko, J.; Siggia, E. Stretching DNA. Macromolecules 1995, 28, 8759–8770. DOI: 10.1021/ma00130a008 36. Bouchiat, C.; Wang, M.; Allemand, J.; Strick, T.; Block, S.; Croquette, V. Estimating the persistence length of a worm-like chain molecule from force-extension measurements. Biophys. J. 1999, 76, 409–413. DOI: https://doi.org/10.1016/S0006-3495(99)77207-3 37. Gritti, F.; Sawyer, K.; Vaishnav, J.; Adapelli, B.; Lauber, M.; Wyndham, K. Retention and efficiency of a novel slalom chromatography column: an alternative to agarose gel electrophoresis for DNA separation. J. Chromatogr. A 2025, 1761, 466293. DOI: 10.1016/j.chroma.2025.466293 38. Strom, O.; Beech, J.; Tegenfeldt, J. High-throughput separation of long DNA in deterministic lateral displacement arrays. Micromachines 2022, 13, 1754. DOI: https://doi.org/10.3390/mi13101754 39. Hays, J.B.; Magar, M.E.; Zimm, B.H. Persistence length of DNA. Biopolymers 1969, 8, 531–536. DOI: https://doi.org/10.1002/bip.1969.360080410 Supplementary Figures Figure S1a. Effect of average hydrostatic pressure (P) along a chromatographic column on the retention of dsDNA fragments (HindIII λ-DNA digest). A very first research prototype column (4.6 mm i.d. x 300 mm) was packed with polydisperse 3.0 μm non porous Glantreo silica particles. Mobile phase: 1xPBS, pH 7.4. Flow rate: 0.2 mL/min. Room temperature. Post column, 40 mm i.d. restrictor capillaries were connected in series to increase pressure: 1825 psi (no restrictor), 7040 psi (one 30 inch capillary added), 11,140 psi (one 30 inch and one 24 inch capillary added), and 14,200 psi (one 30 inch, one 24 inch, and one 16 inch capillary added). Fact 1 ➔ DNA retention is pressure independent. Effect of average hydrostatic pressure (P) along a chromatographic column on the retention of dsDNA fragments (HindIII λ-DNA digest). A very first research prototype column (4.6 mm i.d. x 300 mm) was packed with polydisperse 3.0 mm non porous Glantreo silica particles. Mobile phase: 1xPBS, pH 7.4. Flow rate: 0.2 mL/min. Room temperature. Post column, 40 mm i.d. restrictor capillaries were connected in series to increase pressure: 1825 psi (no restrictor), 7040 psi (one30 inch capillary added), 11,140 psi (one 30 inch and one 24 inch capillary added), and 14,200 psi (one 30 inch, one 24 inch, and one 16 inch capillary added). Fact 1 ➔ DNA retention is pressure independent. Figure S1b. Effect of the temperature (25–90 oC) on the retention of dsDNA fragments (HindIII λ-DNA digest). A research prototype column (4.6 mm i.d. x 300 mm) was packed with monosized 2.6 μm non-porous silica particles. Mobile phase: 1xPBS, pH 7.4, with 90:10 (v/v) acetonitrile. Flow rate: 1.0 mL/min. Fact 2 ➔ DNA retention decreases progressively with increasing temperature. Search... Figure S1c. Effect of mobile phase viscosity (0.9–5.2 cP, corresponding to a sucrose mass Choose Topics fragments (HindIII λ-DNA digest). A fraction of 0–40%) on the retention of dsDNA research prototype column (4.6 mm i.d. x 300 mm) was packed with monosized 2.6 μm non-porous silica particles. Mobile phase: 1xPBS, pH 7.4. Flow rate: 0.4 mL/min. Room temperature. Fact 3 ➔ DNA retention increases progressively with increasing viscosity of the mobile phase. Figure S1d. Effect of flow rate (0.15–1.1 mL/min, for 0.001xPBS-1xPBS) on the retention of dsDNA fragments (BstEII λ-DNA digest). The commercialized 250 Å. BEH GTxResolve Slalom Column (4.6 mm i.d. x 300 mm, 2.5-μm) was used. Mobile phase: 10xTAE, pH 8.0. Room temperature. Fact 4 ➔ DNA retention increases progressively with increasing flow rate. Figure S1e. Effect of ionic strength (0.156–156 mM, duplicate measurements) on the retention of negatively charged dsDNA fragments (HindIII λ-DNA digest). Two research prototype columns (4.6 mm i.d. x 300 mm) were packed with monosized, negatively charged 2.6 μm non-porous silica particles and connected in series. Mobile phase: PBS buffers, pH 7.4. Flow rate: 0.6 mL/min. Room temperature. Fact 5 ➔ DNA retention increases progressively with increasing ionic strength. Figure S1f. Effect of particle size (1.8, 2.4, 5.4, and 11.3 μm) on the retention of dsDNA fragments (λ-DNA Monocut Mix) under constant extension/shear forces (the ratio of interstitial linear velocity to particle diameter was kept constant) across a wide range of average extension lengths (Lext: 0.05–13.5 μm). Four research prototype columns (4.6 mm i.d. x 300 mm) were packed with 125 Å BEH particles of different sizes. Mobile phases: 100 mm phosphate buffer, pH 8.0. Flow rates: 0.262 mL/min (1.8 μm), 0.436 mL/min (2.4 μm), 0.921 mL/min (5.4 μm), and 2.0 mL/min (11.3 μm). Room temperature. Fact 6 ➔ DNA retention is independent of particle size when operating at constant extension/shear force and when the extension length exceeds the particle diameter. Fact 7 ➔ Conversely, DNA retention decreases with increasing particle size at constant extension/shear force when the extension length is smaller than the particle diameter, providing evidence of a mixed HDC–SC retention mechanism. Figure S1g. Effect of surface modification of stainless steel column hardware with hybrid surface technology (HST) bonding on the peak shape of dsDNA fragments (HindIII λ-DNA digest). The same 2.5 μm, 250 Å BEH particles and column dimensions (4.6 mm i.d. x 300 mm) were used. Mobile phase: 1xTAE, pH 8.0. Flow rate: 1.4 mL/min. Temperature: 40 oC. (Top) Bare stainless-steel column hardware. (Bottom) HST-modified column hardware. Fact 8 ➔ The integrity of DNA peak shape and retention is maintained with HST modified stainless steel column hardware. Figure S2. Effect of TAE buffer concentration (4–400 mM) on the retention of negatively charged dsDNA fragments (from HindIII λ-DNA digest, BstEII λ-DNA digest, and λ-DNA Monocut Mix) ranging in size from 1.5 to 48.5 kbp. The commercialized 2.5 μm, 250 Å BEH GTx- Resolve Slalom Column (4.6 mm i.d. x 300 mm) was used. Mobile phases: 0.1xTAE, 1xTAE, and 10xTAE, pH 8.0. Flow rate: 1.1 mL/min. Room temperature. Again, as in Figure S1e, DNA retention increases with increasing ionic strength. Newsletter Join the global community of analytical scientists who trust LCGC for insights on the latest techniques, trends, and expert solutions in chromatography. Subscribe Now! Column: Column Watch Related Content SPONSORED Clarus® Nova GC: One Platform. More Capability. Less Complexity. Discover how a unified GC platform can help laboratories improve uptime streamline workflows and stay prepared for… , Learn More , Pits as Protectors: LC-MS/MS Unlocks Antioxidant and Anti-Browning Potential By John Chasse April 17th 2026 About Advertise Author Guidelines Contact Us Editorial Advisory Board Ethics Policy Do Not Sell My Personal Information Privacy Policy Permissions Subscriptions Terms and Conditions Cookie settings Contact Info 259 Prospect Plains Rd, Bldg H Cranbury, NJ 08512 Best of the Week: Sustainable Analytical Methods, Chemometrics and Generative AI By John Chasse April 17th 2026 © 2026 MJH Life Sciences® All rights reserved. Home About Us News Contact Us Invasive Burmese Pythons: Sentinels for PFAS Pollution in the Everglades via HPLCMS/MS Analysis By John Chasse April 17th 2026 The SWIM Approach Extended By Gauthier Eppe, Kate Jones April 17th 2026 Chromatographic Profiling of Steroid Hormone Variability Across Menstrual Cycle and Hormonal Contraceptive Use in a Female Athlete By John Chasse April 16th 2026