On Some Curiosities of Native Capillary Zone Electrophoresis Involving Protein-Ligand Interactions From Fragment Screening of Transthyretin

On Some Curiosities of Native Capillary Zone Electrophoresis Involving Protein-Ligand Interactions From Fragment Screening of Transthyretin | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article On Some Curiosities of Native Capillary Zone Electrophoresis Involving Protein-Ligand Interactions From Fragment Screening of Transthyretin Wenjie Chen This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8908294/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Reliable and sensitive detection of specific protein-ligand molecular interactions is vital for target-based early-stage small molecule drug discovery. Using human transthyretin as an example, the author here describes how non-denaturing Capillary Zone Electrophoresis (CZE) is used to detect weakly binding small molecule fragments. Two methodologically distinct CZE competition fragment screening assays were developed during the Fragment-based Drug Discovery (FBDD) campaign and compared. Many intriguing electropherograms, all attributed to effects of molecular binding and electrophoretic behaviour were observed during assay development and their implications are discussed. Drug Discovery, Design, & Development Capillary Zone Electrophoresis Fragment Screening Transthyretin FPPHR Affinity Capillary Electrophoresis Drug Discovery Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 1 Introduction In modern time analytical chemistry, High-performance Liquid Chromatography (HPLC) is an essential tool (analytical and preparative) used in many labs around the world. Alongside other purposely developed separation techniques such as Gas Chromatography (GC) and Capillary Electrophoresis (CE) coupled to mass spectrometry, and with NMR, they meet most of the analytical needs from chemists and biologists. For example, analytes (ions, small molecules, carbohydrates, lipids, peptides, nucleic acid, and protein) identification, quantification, purity determination, and post-translation modifications characterisation etc. were the common tasks. Through continuous instrument design and methodology development, use of CE has been extended to biomolecular interaction studies such as Affinity Capillary Electrophoresis (ACE) based on Capillary Zone Electrophoresis (CZE) [ 1 – 4 ]. In Contrast to HPLC and GC, the transcendence of CE from pure separation science to bioassay ACE is driven by a few advantageous characteristics of CZE. First and foremost, CZE separation does not require a stationary phase, which will disrupt protein-ligand interactions. Secondly, CZE requires the lowest sample injection volume (20–50 nL) and shortest separation time (3–8 minutes per run). Lastly, CZE is non-denaturing because analytes can be separated in a fully liquid phase environment at physiological pH without organic solvents and extreme heat whilst maintaining good tolerance to DMSO, metal ions, salt and detergent. By being microscale, non-denaturing and with high separation efficiency, CZE adds to the list of biophysical binding detection techniques used in FBDD with its unique set of beneficial properties. Most notably, no immobilisation or restraint on protein movement, truly label-free thanks to sensitive UV detection, and highly insusceptible to compound precipitation or fluorescence interference since CZE is a purification process per se . Their combined effect was that CZE is sensitive and remarkably specific (low false positives) in fragment binding detection, exemplified by the Free Probe Peak Height Restoration (FPPHR) method [ 5 ]. It involves injecting a small volume (≈ 20–30 nL) of a mixture containing 8-anilinonaphthalene-1-sulfonic acid (8-ANS) and the drug target transthyretin (a plasma protein) into the capillary filled with a test fragment followed by CZE separation. 8-ANS was the Probe Ligand as it is known to bind (K d1 = 1.05 µM, K d2 = 4.79 µM, two negatively cooperative binding sites) [ 6 ] transthyretin (TTR) and increment in free 8-ANS (portion not bound to TTR) peak will indicate competitive displacement by test fragment. The use of Probe Ligand and indirect binding detection renders the FPPHR CZE assay site-specific, a feature rarely offered by other biophysical fragment screening techniques. During development of the TTR CZE FPPHR fragment screening assay, the author has noticed some special cases such as appearance of vacancy peaks, focusing, tailing and delaying of UV peaks of displaced 8-ANS and loss of assay fidelity for a high affinity TTR ligand (Tafamidis). In response, another CZE assay was developed based on the CEfrag™ methodology [ 1 ], in which 8-ANS was injected and separated in a capillary filled with TTR and test fragment (Fig. 1 ). This means CEfrag™ pre-mixes TTR with test fragment instead of 8-ANS and measures how well the test fragment could block 8-ANS binding to TTR instead of displacement. Results and some interesting observations from both methods as well as their potential causes are described, compared, and discussed providing insights and guidance on future CZE fragment screening assay developments. 2 Materials and Methods 2.1 Chemicals, Reagents & Protein: Tafamidis was synthesised by Selcia Ltd. (Ongar, Essex, UK). 8-Anilino-1-naphthalenesulfonic acid ammonium salt (8-ANS) and N-Phenylanthranilic Acid (NPA) were purchased from Sigma (Poole, Dorset, UK). Diclofenac was purchased from Cayman Chemical (Michigan, USA). All other tested compounds were either from the chemical store of Selcia Ltd. or the Selcia Fragment Library (SFL) unless otherwise stated. Human transthyretin (UniProt P02766) lyophilised from 0.02 M NH 4 HCO 3 solution was purchased from SCIPAC (Sittingbourne, Kent, UK), Product Code P171-1 Lot.1802-20-1, purity > 96%. Dried TTR was weighed and dissolved with CZE Running Buffers to give a final stock concentration of 100 µM tetramers. 2.2 Capillary Electrophoresis: CZE assays development and fragment screening were performed on a Beckman P/ACE MDQ capillary electrophoresis system (Beckman Coulter Inc., Brea, CA) operated by 32 Karat™ software. All capillaries (Polymicro Technologies, Phoenix, AZ) used were internally coated with polyvinyl alcohol and of around 30 cm in length, 75 microns in internal diameter. Capillary temperature was maintained at 20°C by recirculating fluorinated fluid coolant during electrophoresis. Sample trays and buffer trays temperature were maintained at 4°C and ambient respectively. For every electrophoretic run, the capillary was pressure rinsed and prefilled with electrophoresis buffer also known as background electrolyte (termed as Running Buffer hereafter) before injection. After each electrophoretic run, the capillary was rinsed with Running Buffer and H 2 O by pressure (20 psi for 1 minute). Two Running Buffers (RB) A : 20 mM HEPES pH 7.0, 2 mM CaCl 2 and B : 20 mM HEPES pH 7.0, 2 mM CaCl 2 , 1% volume to volume (v/v) DMSO were used. All CE traces were analysed by 32Karat™ Software with a built-in automatic peak integration algorithm. 2.3 TTR FPPHR CZE Assay Development and Fragment Screening: Control 1 Inject Buffer (IB) containing 8-ANS (15 or 30 µM) was prepared by mixing 1 µL of 8-ANS DMSO stock (1.5 or 3 mM) with 99 µL of Running Buffer A in a 0.2 mL PCR tube. Control 2 Inject Buffer containing a mixture of 8-ANS (15 or 30 µM) and TTR (10 or 20 µM) was prepared by mixing 1 µL of 8-ANS DMSO stock (1.5 or 3 mM) with 10 or 20 µL of 100 µM tetrameric TTR stock in Running Buffer A and 79 or 89 µL of Running Buffer A. The final concentrations (f/c) were 8-ANS 15/30 µM, TTR 10/20 µM and 1% DMSO v/v. The maximum Free Probe Peak Height (Control 1 Peak Height) was determined by Outlet Injection of Control 1 with an injection pressure of 0.5 psi for 5 seconds (equivalent to ≈ 20–30 nL). An identical injection of Control 2 provides the minimum Free Probe Peak Height (Control 2 Peak Height) enabling calculation of % of bound and unbound i.e. free 8-ANS. To determine the 8-ANS peak height in Control 1 and Control 2, the capillary was prefilled with Buffer B and electrophoresis carried out at 10, 15, or 30 kV (Normal Polarity) with UV monitoring at 214 or 230 nm. For displacement analysis, ligand DMSO stocks (0.5–30 mM) was diluted 100x in Buffer A (f/c 5–300 µM) and used to fill the capillary as the Running Buffer. Electrophoresis commenced after injection of Control 2, thus the injected 8-ANS-TTR complex migrates inside a capillary filled with the test ligand. The observed free 8-ANS peak height is defined as Sample Peak Height and its restoration can be calculated as: $$\:\%\:of\:Free\:Probe\:Peak\:Height\:Restoration=\:\frac{Sample\:Peak\:Height-Control\:2\:Peak\:Height}{Control\:1\:Peak\:Height-Control\:2\:Peak\:Height}$$ 2.4 TTR CEfrag™ CZE Assay: Inject Buffer was prepared by mixing 1 µL of 1.5 mM 8-ANS DMSO stock with 1 µL of test compound DMSO stock (0.025–5 mM) then diluted with 198 µL of Buffer A in a 0.2 mL vial. Final concentration was 8-ANS 7.5 µM, test compound 0.25–50 µM and 1% DMSO v/v. Running Buffer was prepared by mixing 5 µL of 100 µM TTR protein stock solution in Buffer A with 2 µL of test compound DMSO stock (0.025–5 mM) in a 0.2 mL vial. The mixture was then diluted with 193 µL of Buffer A, f/c: TTR 2.5 µM, test compound 0.25–50 µM, 1% DMSO v/v. 100 µL of this solution was transferred into another 0.2 mL vial forming a Running Buffer pair. Sample injection and electrophoresis conditions were the same as the FPPHR assay. 3 Results and Discussions 3.1 Vacancy Peak of Test Fragment (FPPHR) and Target Protein (CEfrag™) Zeroing of UV absorbance on the background electrolyte (known as Running Buffer hereafter) which fills the capillary occurred in all CZE separation prior to electrophoresis. This sets the basal UV level for which passing analyte separation bands are measured against. UV absorbance is the same throughout the entire capillary volume prior to sample injection, but a gap (the injection plug) is introduced after (Fig. 2 ). The gap could manifest itself as a negative absorbance peak in the electropherogram when the following two conditions are satisfied. Firstly, the Running Buffer (RB) contains species that absorb at the measured UV wavelength but absent in the injected sample (known as Inject Buffer hereafter). Secondly, such UV-absorbing species migrate in the direction towards the detection window during electrophoresis. The vacancy peak phenomenon would be ignorable if RB has no or little basal measured UV absorbance, but this is rare in the context of biomolecular screening assays because of DMSO omnipresence and extensive UV absorbance overlap of small and biological molecules. Thus, DMSO percentage in the Inject Buffer (IB) should always match to that of Running Buffer (RB), especially if an uncoated capillary is used with strong Electroosmotic Flow (EOF). Additionally, DMSO should be kept below 5% (10% max) as more will further reduce the dynamic range and lowering UV sensitivity. Notwithstanding a potential issue in CZE bioassay, the vacancy peak principle is utilised for detecting analytes of no or little UV absorbance such as metal ions. Although DMSO percentage and buffer content (pH, salt concentration, additives) could be matched in RB and IB, there however will be other unmatchable background UV-active species (the test fragment or target protein) in CZE fragment screening depending on the methodology. In the CZE FPPHR method, the test fragment is in the Running Buffer but not Inject Buffer, meaning vacancy peak of the test fragment could appear dependent on various factors. In our case, vacancy peak of any negatively charged small molecules that absorbs around UV 230 nm and migrates pass the detection window within 5 minutes will appear in the electropherogram (Fig. 3 A). As for the CEfrag™ method, target protein is mixed with test fragment in RB and used to fill the capillary but not added to IB which contains the Probe Ligand and matched concentration of test fragment. Henceforth, vacancy peak of target protein could appear under certain conditions, which it did in the TTR CEfrag™ assay described in this paper (Fig. 3 B). Having understood how differences between IB and RB contents in FPPHR and CEfrag™ method gave rise to vacancy peaks, there implies under certain circumstances the assays could fail. For instance, if the test fragment vacancy peak has the same or very close (by less than ≈ 20 s) migration time to that of the free Probe Ligand in FPPHR, accurate free probe peak height will be unobtainable. When Probe Ligand migration time is very close (by less than ≈ 1 minute) to that of target protein in CEfrag™, not only will protein vacancy peak appear, but Probe Ligand mobility shift would also be minimal hampering assay sensitivity. Test fragment vacancy peak theoretically should not occur in CEfrag™, because fragment concentration is matched in the Inject Buffer (IB) and Running Buffer (RB). However, slight concentration difference could lead to additional positive (if fragment concentration were higher in IB, S1 Fig. ) or negative peaks i.e. vacancy peak, especially if the test fragment strongly absorbs at the detected UV wavelength. Bear in mind that modern CE instruments could equip Laser Induced Fluorescence (LIF) detection, which should alleviate the UV vacancy peak problem. This however will require a fluorescent probe (not always readily available) and introduce fluorescence interference. Coincidentally, 8-ANS is a commonly used fluorescent probe to study protein conformational changes, however LIF was unavailable to the author at the time. If LIF detection were used in FPPHR, Probe Ligand and protein concentration could be further reduced making the assay more adept for high affinity ligands, but not ideal for low affinity fragments. 3.2 Activity of Tafamidis in TTR CZE FPPHR and CEfrag™ Assays Tafamidis, an approved drug for Familial Amyloidotic Polyneuropathy (a form of TTR amyloidosis) binds TTR with dissociation constants (K d ) of less than 0.2 µM [ 5 , 7 ]. This high affinity ligand was tested as a positive control during FPPHR and CEfrag™ assay development, but its apparent potency appeared to be weaker in FPPHR. Free 8-ANS peak height restoration reached 78% when 50 µM (5x [TTR]) of Tafamidis was added to the Running Buffer (Fig. 4 A Trace E ) but 8-ANS-TTR interaction was fully blocked at just 5 µM (2x [TTR]) (Fig. 4 B Trace H ) in CEfrag™. The discrepancy could result interdependently from the assay methodology, electrophoretic properties and binding kinetics of Tafamidis towards TTR. Compared to CEfrag™, TTR is never mixed with the test fragment until after sample injection, at which then molecular diffusion and 8-ANS displacement could occur for ≈ 10 s at either end of the injection plug interface before electrophoresis starts. Mixing and 8-ANS displacement would continue during early electrophoresis as the injected 8-ANS-TTR complex migrates towards the detection window, but electrophoretic property of test fragments varies. Fragments could be moving in the opposite or same direction of 8-ANS-TTR or be virtually static resulting heterogenous electrophoretic mixing. Though considered a rare event, if test fragment moves at exact velocity to that of 8-ANS-TTR, there would be no further mixing besides the initial pressure injection. This inadvertently introduces factors besides binding affinity that could determine the final observed level of 8-ANS displacement such as association rate, non-Brownian motion and relative velocity of the test fragment. The fact that Tafamidis moves at similar velocity (Fig. 4 A) to that of 8-ANS-TTR meant their electrophoretic mixing was comparatively poorer than other compounds of greater velocity difference to 8-ANS-TTR. A slow K on rate might be another contributing factor to potency underestimation of Tafamidis by FPPHR. The author pondered whether to use FPPHR for fragment screening but thought the immense excess amount of test fragment (entire capillary volume) compared to 8-ANS-TTR (small injection plug) could compensate short interaction time, especially when screen concentration is typically set to 300–1000 µM. Moreover, majority of small molecule protein bindings are of fast K on , around 10 seconds of contact time was thought to suffice, not necessarily for establishing equilibrium, but good enough for binding detection. In addition, the author expected 8-ANS-TTR complex and test fragment molecules to be in non-Brownian motion colliding each other with certain degree of order during electrophoresis, which might facilitate 8-ANS displacement. Lastly, the apparent potency of another high affinity ligand Thyroxine T 4 tested by FPPHR [ 5 ] aligned well with previously published K d values. Potency underestimation of Tafamidis by FPPHR was an unusual exception. By considering the pragmatic advantages of FPPHR (Table 1 ), the author’s intuition was that the benefits of using it as a primary screening tool (yes/no binding) outweighs the risk of potentially missing out binders of very slow K on or having inadequate electrophoretic mixing. Table 1 Practical Comparisons between TTR CZE FPPHR and CEfrag™ Assay Methodology Method FPPHR CEfrag™ Protein Consumption Low (at least 500x lower) High Protein Adsorption Unlikely Possible (could be reduced by additives, harsh rinse, or with a different capillary surface) Sample Handling Simple (samples could be prepared manually without expensive liquid handling robotics) Complex (need to prepare many pairs of protein solutions – not ideal for manual pipetting) Run Time Shorter < 5 Minutes (Probe Ligand peak height is measured instead of migration time) Longer < 10 Minutes (the method requires large mobility shift, which is time dependent) Binding Detection Measurement Quantitative (numerical peak height values) Qualitative a (8-ANS peak separation from TTR vacancy peak) a) measurement could become quantitative when there is a significant mobility shift of the Probe Ligand. In the case described here, electrophoretic velocity of 8-ANS was too close to that of TTR and their molecular binding was reflected by peak merging (Fig. 3 B ). Competitive binding from test compound took place in form of 8-ANS and TTR peak separation (Fig. 4 B ) , which the author had judged subjectively. The peak separation or shape change could be quantitatively modelled by mathematics, but it was beyond the author’s expertise. 129 fragments were then screened at 300 µM by FPPHR providing 16 initial hits and 4 failed (3.1%) runs due to vacancy peak (they were later confirmed negative by CEfrag™, S2 Fig. ). A hit rate of 12.4% is unlikely achieved had the assay been fundamentally flawed in hit picking due to short contact time. In hindsight, a total of 54 compounds (mostly fragments) were tested by both FPPHR and CEfrag™, and their overall agreement in terms of hit detection (yes/no) was around 90% (49/54), see Table 2 . Table 2 Correlation of FPPHR and CEfrag™ Method on Fragment Binding Detection Total Number of Compounds Tested = 54 FPPHR Method CEfrag™ Method Positive a Negative Positive b 33 c 7 d 4 c 1 d Negative 0 1 d, e 6 c 3 d a) a positive compound in the FPPHR method would have had caused a minimum of 10% free 8-ANS peak height restoration. b) a positive compound in the CEfrag™ method must have had caused an obvious separation of free 8-ANS and TTR vacancy peak, e.g. Trace E – H in Fig. 4 B. Trace C – D in Fig. 4 B would be regarded as negative. Regrettably, all peaks shape change/separation were judged subjectively based on experience. c) Tested at same concentration; d) Tested at different concentration, a maximum difference of 2-fold e) This compound (SFL000029) was tested positive by FPPHR at 300 µM and negative in CEfrag™ at 150 µM. It was found to be positive when tested again with CEfrag™ at 500 µM, and later confirmed by crystallography [ 5 ]. Furthermore, the author has later (2019–2020) developed another CZE FPPHR fragment screening assay for an undisclosed oncology target in search of first-in-class protein-protein interaction inhibitors. 1126 fragments were screened with 47 failed runs, of which 4 failed (0.35%) because of vacancy peak issue. The percentage of failed runs by vacancy peak was much lower than the TTR screen because the Probe Ligand used was a small peptide having significantly different migration time to almost all screened fragments. 44 out of the remaining 1079 showed significant Probe Ligand displacement (4% hit-rate) and multiple fragments hits were later confirmed by crystallography, biochemical assays and cell-based assays (confidential data). It is only then the author became fully confident in the effectiveness of the FPPHR method for fragment binding detection. Having a primary screening technique affording highly reliable hits would immensely streamline subsequent processes saving costs and invaluable time. 3.3 FPPHR 8-ANS Displacement by Various Compounds With Distinct Peak Profiles During the development of TTR CZE FPPHR fragment screening assay, multiple compounds known to bind TTR (typically of higher affinity than fragments, including Tafamidis) were first tested as positive controls. Many of them had displaced significant amount of TTR-bound 8-ANS leading to increased free 8-ANS peak height, of which the peak shape remained gaussian-like (Fig. 4 A). Fascinatingly, peak shape change for some compounds were non-gaussian, characterised by peak focusing, splitting, elongation and migration delay (Fig. 5 ). In the case of N-Phenylanthranilic Acid (NPA), the displaced 8-ANS was initially split from the original free 8-ANS population (Fig. 5 A, Trace B ). As [NPA] in RB increases (Fig. 5 A, Trace C – F) , the displaced 8-ANS shifts atop and became “focused” as a sharper peak. It is worth noting that NPA elutes (vacancy peak) just before the free 8-ANS peak and the peak “focusing” effect was observed in only, but not all, negatively charged compounds. Other tested compound that experienced the “focusing” phenomenon includes 4-Pyridin-3-yl-benzoic acid (SFL000046), Biphenyl-4-carboxylic acid (A00002802), and Diflunisal, they all elute close to and slightly earlier than free 8-ANS ( S3 Fig. ). Migration of displaced 8-ANS could have been affected by the reduced conductivity (dependent on test compound concentration) of the vacancy peak zone being right in front of it. Other causes related to the test compound’s electrophoretic properties and/or TTR binding kinetics are also plausible. The author went on to test a set of compounds that were being investigated by the Wolfson Drug Discovery Unit (the Royal Free Hospital) at UCL together with Diclofenac. The compounds (structure undisclosed) displayed low micromolar IC 50 values (except compound T85 > 1 mM) in their 125 I-T 4 displacement assay [ 8 ] thus test concentration was reduced accordingly in the TTR CZE FPPHR assay. From electropherograms of those 11 test compounds (Fig. 5 B), a variety of free 8-ANS peak profiles were observed. 8-ANS displacement by compound T07, T15 & T23 led to Gaussian-like increase in free 8-ANS peak size. Compound T34 and T85 were considered negative. Compound T44, T47, T58, T60, T70 and Diclofenac caused free 8-ANS peak tailing, splitting or elongation instead. Intriguingly they all move in the same direction as 8-ANS-TTR, implying their electrophoretic mixing and/or binding kinetics (during electrophoresis) might be distinctive to neutral and positively charged compound. Diclofenac elutes close to 8-ANS but soon after rather than slightly before and the peak focusing effect was absent. The inclination was that such event would only occur when competing compounds elute just before free 8-ANS. The underlying mechanism for all these displaced 8-ANS peak forms could have been a very interesting topic relevant to the field of free solution native CZE. 4 Concluding Remarks By manipulating contents of the Running and Inject Buffer, the author has developed two methodologically (FPPHR and CEfrag™) distinct CZE binding competition assays for transthyretin. FPPHR has multiple practical screening advantages over CEfrag™ and has been successfully applied to two targets of different disease area. However, it comes with a caveat that probe ligand displacement also depends on the on-rate and electrophoretic mixing of test compound. CEfrag™ on the other hand could be used in a more affinity-orientated hit confirmation step. The vacancy peak, peak merging, tailing, splitting, focusing and elongation phenomena further highlights the importance of taking the electrophoretic properties and binding kinetics of all involved analytes into proper consideration when developing CZE fragment screening assays and interpreting results thereafter. Abbreviations 8-ANS 8-anilinonaphthalene-1-sulfonic acid ACE Affinity Capillary Electrophoresis CZE Capillary Zone Electrophoresis DMSO Dimethyl Sulfoxide DSF 3,5-Dichlorobenzen Sulphonamide EOF Electroosmotic Flow FBDD Fragment-based Drug Discovery f/c Final Concentration FPPHR Free Probe Peak Height Restoration HEPES 4-(2-Hydroxyethyl)piperazine-1-ethanesulfonic acid IB Inject Buffer NMR Nuclear Magnetic Resonance NPA N-Phenylanthranilic Acid RB Running Buffer TTR Transthyretin v/v Volume to Volume Declarations Acknowledgement The research work here formed part of a Ph.D. project (2010 – 2014) co-supervised by Professor Stephen P. Wood and Dr. Carol Austin. All capillary electrophoresis experiments were performed in Selcia Ltd (now part of Eurofins). The author thanks Dr. Graham Taylor from the UCL Wolfson Drug Discovery Unit for providing test compounds. References Austin C, Pettit SN, Magnolo SK, Sanvoisin J, Chen W, Wood SP, Freeman LD, Pengelly RJ, Hughes DE (2012) Fragment screening using capillary electrophoresis (CEfrag) for hit identification of heat shock protein 90 ATPase inhibitors. J Biomol Screen 17(7):868–876 Farcas E, Bouckaert C, Servais AC, Hanson J, Pochet L, Fillet M (2017) Partial filling affinity capillary electrophoresis as a useful tool for fragment-based drug discovery: A proof of concept on thrombin. Anal Chim Acta 984:211–222 Neaga IO, Hambye S, Bodoki E, Palmieri C, Ansseau E, Belayew A, Oprean R, Blankert B (2018) Affinity capillary electrophoresis for identification of active drug candidates in myotonic dystrophy type 1. Anal Bioanal Chem 410(18):4495–4507 Rauch JN, Nie J, Buchholz TJ, Gestwicki JE, Kennedy RT (2013) Development of a capillary electrophoresis platform for identifying inhibitors of protein-protein interactions. Anal Chem 85(20):9824–9831 Chen W (2025) Fragment-based drug discovery for transthyretin kinetic stabilisers using a novel capillary zone electrophoresis method. PLoS ONE 20(5):e0323816 Ferguson RN, Edelhoch H, Saroff HA, Robbins J, Cahnmann HJ (1975) Negative cooperativity in the binding of thyroxine to human serum prealbumin. Preparation of tritium-labeled 8-anilino-1-naphthalenesulfonic acid. Biochemistry 14(2):282–289 Bulawa CE, Connelly S, Devit M, Wang L, Weigel C, Fleming JA, Packman J, Powers ET, Wiseman RL, Foss TR, Wilson IA, Kelly JW, Labaudiniere R (2012) Tafamidis, a potent and selective transthyretin kinetic stabilizer that inhibits the amyloid cascade. Proc Natl Acad Sci U S A 109(24):9629–9634 Kolstoe SE, Mangione PP, Bellotti V, Taylor GW, Tennent GA, Deroo S, Morrison AJ, Cobb AJ, Coyne A, McCammon MG, Warner TD, Mitchell J, Gill R, Smith MD, Ley SV, Robinson CV, Wood SP, Pepys MB (2010) Trapping of palindromic ligands within native transthyretin prevents amyloid formation. Proc Natl Acad Sci U S A 107(47):20483–20488 Additional Declarations The authors declare no competing interests. Supplementary Files SupportInformation.pdf Support Information Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8908294","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":593248388,"identity":"45eb4d65-e889-495d-9769-6c471cde2e80","order_by":0,"name":"Wenjie Chen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyUlEQVRIiWNgGAWjYLACxgYGBn4gdQDM4yFWiyQQk6jF4ACxWgyO9x5g+LnDJs/4RvKDAww1dgwGZw4Q0HLmXAJj75m0YrMbaUCLjiUzGJxtwK/F7EaOATNj2+HEbTdygA5jO8BgcJ6Aw8zuvwFp+Z+4eQZIyz9itNzgAWk5kLhBAqgFyCDsMPszOQYHe9uSE2eceWZwILEvmUeSkPcl288YPvjZZpfY35788MGHb3ZyfGcSCLiMARYdIJBAZESOglEwCkbBKCAAANutSNetecu/AAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-8180-366X","institution":"","correspondingAuthor":true,"prefix":"","firstName":"Wenjie","middleName":"","lastName":"Chen","suffix":""}],"badges":[],"createdAt":"2026-02-18 10:26:38","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-8908294/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8908294/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102974699,"identity":"dc48f32f-fea7-4e0f-83a9-507003205758","added_by":"auto","created_at":"2026-02-19 07:22:05","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":165493,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCEfrag™ and FPPHR CZE Assay Methodology\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBoth CEfrag™ and FPPHR are considered CZE-based separation processes since the Running Buffer and Inject Buffer were prepared with the same diluent (buffer agent, salt, and additives like DMSO and detergents if any). The background electrolyte is homogenous throughout the capillary except for the absence or presence of the drug target or test compound. Their methodological differences arise from the allocation of the Probe Ligand, Drug Target and Test Compound in the Running and Inject Buffer as shown above.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8908294/v1/541aec660a90d1bd0bf85caa.png"},{"id":103049666,"identity":"99f37b8b-1e99-423c-8080-5b2cb97f639f","added_by":"auto","created_at":"2026-02-20 07:44:32","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":762235,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRise of Vacancy Peak in CZE\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe stared symbol represents the UV-absorbing species that was missing in the Inject Buffer but present in the Running Buffer and used to fill the capillary. When injection was from the outlet end and polarity set to Normal, all negatively charged molecules will migrate towards the detection window with sufficiently suppressed EOF. If such species from the outlet Running Buffer vial moves pass the detection window before the end of data acquisition, its vacancy peak (negative absorbance) will manifest in the electropherogram.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8908294/v1/6012dd99b7b3aeb8b1a95347.png"},{"id":103049584,"identity":"f425a59f-4e94-43c0-aba5-9b6556e11fe8","added_by":"auto","created_at":"2026-02-20 07:43:04","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":2484809,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eVacancy Peaks in TTR CZE FPPHR and CEfrag™ Method\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eA.\u003c/strong\u003eElectropherogram of multiple CZE runs of injected 8-ANS-TTR mixture (30:20 µM) in which the capillary was filled with Running Buffer containing no test compound \u003cstrong\u003e(\u003c/strong\u003eTrace A), 150 (Trace C) or 300 (Trace D) µM of 3,5-Dichlorobenezene Sulphonamide (DSF) or 300 µM of Aspirin (Trace B) are shown. 8-ANS displacement by 300 µM of DSF became visibly clear with increased free 8-ANS peak height and area. 300 µM of Aspirin caused tiny change in free 8-ANS peak height/area compared to Trace A, but there was a negative peak in the electropherogram. This was due to Aspirin absorbs UV at 230 nm albeit weakly and it was negatively charged (at pH 7.0) with a migration time of less than 2 minutes. DSF vacancy peak did not appear because it was neutral under specified assay conditions. \u003cstrong\u003eB. \u003c/strong\u003eMultiple CZE runs of injected 7.5 µM 8-ANS in Inject Buffer (IB) are shown with increasing concentration of TTR added to the Running Buffer (RB, Trace A – F). Though TTR is a 55 KDa homotetrameric macromolecule, a pKa of around 5.3 and -20 net charge have conferred on it a significant amount of electrophoretic mobility. This results it moving pass the detection window from the injection plug in around 6.5 minutes, and the appearance of its vacancy peak. Another interesting sight was the gradual merger of free 8-ANS peak at around 4 minutes with the front of TTR vacancy peak induced by their molecular binding. Mobility shift of free 8-ANS was however relatively small.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8908294/v1/d8d6050619b34e0a46eab5af.png"},{"id":102974700,"identity":"f487c3ec-b940-4290-9b73-968de1e4ff2c","added_by":"auto","created_at":"2026-02-19 07:22:05","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1102078,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCompetitive Binding between 8-ANS, TTR and Tafamidis Measured by FPPHR and CEfrag™\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eA.\u003c/strong\u003e A small volume (20 – 30 nL) of 15 µM 8-ANS diluted in Running Buffer (20 mM HEPES pH 7.0, 2 mM CaCl\u003csub\u003e2\u003c/sub\u003e, 1% v/v DMSO) was injected and separated by CZE in a capillary filled with Running Buffer (Trace A). The same CZE run (Trace B) was repeated with 15 µM 8-ANS mixed with 10 µM TTR in the Inject Buffer (IB). As well as appearance of TTR UV peak, peak height of free 8-ANS was reduced from 836 to 340 due to molecular binding to TTR. Free 8-ANS peak height increased proportionally (Trace C – E) as more Tafamidis was added to the Running Buffer (RB) indicating displacement of 8-ANS from TTR. Vacancy peak of Tafamidis was also observed after free 8-ANS and just before TTR protein peak elution, meaning Tafamidis was moving at a similar velocity to that of 8-ANS-TTR bound complex. Out of the ≈ 200 compounds tested throughout the FBDD campaign, Tafamidis was the only compound that moved right before the peak front of 8-ANS-TTR. \u003cstrong\u003eB.\u003c/strong\u003e A small volume (20 – 30 nL) of 7.5 µM 8-ANS diluted in RB (20 mM HEPES pH 7.0, 2 mM CaCl\u003csub\u003e2\u003c/sub\u003e, 1% v/v DMSO) was injected and separated by CZE in a capillary filled with RB (Trace A). The CZE run was repeated but with 2.5 µM TTR added to the Running Buffer. 8-ANS migration time was increased slightly from around 3.1 to 3.4 and its UV peak merged with TTR vacancy peak (Trace B). A series of Inject Buffer (IB) and Running Buffer (RB) pairs were then prepared with increasing concentration of Tafamidis added to IB and RB while [8-ANS] in IB and [TTR] in RB were maintained. As 8-ANS-TTR interaction becomes more intensely blocked by increasing Tafamidis concentration (Trace C – G), 8-ANS UV peak gradually separates from TTR vacancy peak and shape restored back to as if no TTR were added to RB (Trace H).\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-8908294/v1/b0b8719610d5281afedf7b38.png"},{"id":102974697,"identity":"f7877620-5d7c-425d-bd97-36f6d591004a","added_by":"auto","created_at":"2026-02-19 07:22:05","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1161947,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFPPHR 8-ANS Displacement by Various Compounds Resulting Distinct Peak Profiles\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eA.\u003c/strong\u003e A sequence of CZE separations was performed with increasing concentration of NPA added to the Running Buffer (RB). The same volume of the same Inject Buffer (containing 30 μM 8-ANS and 20 μM TTR) was injected for each run. Without any compounds added to RB, free 8-ANS elutes at around 3.5 minutes (Trace A). When 5 μM of NPA was added to the Running Buffer, there was an additional peak immediately after the free 8-ANS peak (Trace B). As the concentration of NPA increases, this extra peak gets closer to the free 8-ANS peak (Trace C – E). When NPA concentration has reached 50 μM, the extra peak appeared to be very sharp and right on top of the free 8-ANS peak (Trace F). There seemed to be a “focusing” event for displaced 8-ANS induced by NPA. \u003cstrong\u003eB.\u003c/strong\u003e A small volume (20 – 30 nL) of 30 µM 8-ANS diluted in RB (20 mM HEPES pH 7.0, 2 mM CaCl\u003csub\u003e2\u003c/sub\u003e, 1% v/v DMSO) was injected and separated by CZE in a capillary filled with RB (Trace A). The same CZE run (Trace B) was repeated with 30 µM 8-ANS mixed with 20 µM TTR in the Inject Buffer (IB). As well as appearance of TTR UV peak, peak height of free 8-ANS was reduced significantly due to molecular binding to TTR. A total of 10 undisclosed compounds plus Diclofenac were separately added to the RB and tested by multiple CZE runs of Control 2. Many of the test compounds were negatively charged evidenced by their vacancy peak and the displaced 8-ANS peaks were characterised by classic Gaussian-like increase in size, tailing, splitting, and elongation. Compound T34 and T85 appeared to be inactive.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-8908294/v1/0b9460bd3622221f37ce3187.png"},{"id":102974696,"identity":"f7089e5f-0e2b-4bc6-bcdf-9394ef7b5c9c","added_by":"auto","created_at":"2026-02-19 07:22:05","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":281703,"visible":true,"origin":"","legend":"\u003cp\u003eSupport Information\u003c/p\u003e","description":"","filename":"SupportInformation.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8908294/v1/ad75de1df8d4f3bb2dbd0c61.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eOn Some Curiosities of Native Capillary Zone Electrophoresis Involving Protein-Ligand Interactions From Fragment Screening of Transthyretin\u003c/p\u003e","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eIn modern time analytical chemistry, High-performance Liquid Chromatography (HPLC) is an essential tool (analytical and preparative) used in many labs around the world. Alongside other purposely developed separation techniques such as Gas Chromatography (GC) and Capillary Electrophoresis (CE) coupled to mass spectrometry, and with NMR, they meet most of the analytical needs from chemists and biologists. For example, analytes (ions, small molecules, carbohydrates, lipids, peptides, nucleic acid, and protein) identification, quantification, purity determination, and post-translation modifications characterisation etc. were the common tasks. Through continuous instrument design and methodology development, use of CE has been extended to biomolecular interaction studies such as Affinity Capillary Electrophoresis (ACE) based on Capillary Zone Electrophoresis (CZE) [\u003cspan additionalcitationids=\"CR2 CR3\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn Contrast to HPLC and GC, the transcendence of CE from pure separation science to bioassay ACE is driven by a few advantageous characteristics of CZE. First and foremost, CZE separation does not require a stationary phase, which will disrupt protein-ligand interactions. Secondly, CZE requires the lowest sample injection volume (20\u0026ndash;50 nL) and shortest separation time (3\u0026ndash;8 minutes per run). Lastly, CZE is non-denaturing because analytes can be separated in a fully liquid phase environment at physiological pH without organic solvents and extreme heat whilst maintaining good tolerance to DMSO, metal ions, salt and detergent.\u003c/p\u003e \u003cp\u003eBy being microscale, non-denaturing and with high separation efficiency, CZE adds to the list of biophysical binding detection techniques used in FBDD with its unique set of beneficial properties. Most notably, no immobilisation or restraint on protein movement, truly label-free thanks to sensitive UV detection, and highly insusceptible to compound precipitation or fluorescence interference since CZE is a purification process \u003cem\u003eper se\u003c/em\u003e. Their combined effect was that CZE is sensitive and remarkably specific (low false positives) in fragment binding detection, exemplified by the Free Probe Peak Height Restoration (FPPHR) method [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. It involves injecting a small volume (\u0026asymp;\u0026thinsp;20\u0026ndash;30 nL) of a mixture containing 8-anilinonaphthalene-1-sulfonic acid (8-ANS) and the drug target transthyretin (a plasma protein) into the capillary filled with a test fragment followed by CZE separation. 8-ANS was the Probe Ligand as it is known to bind (K\u003csub\u003ed1\u003c/sub\u003e = 1.05 \u0026micro;M, K\u003csub\u003ed2\u003c/sub\u003e = 4.79 \u0026micro;M, two negatively cooperative binding sites) [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e] transthyretin (TTR) and increment in free 8-ANS (portion not bound to TTR) peak will indicate competitive displacement by test fragment. The use of Probe Ligand and indirect binding detection renders the FPPHR CZE assay site-specific, a feature rarely offered by other biophysical fragment screening techniques.\u003c/p\u003e \u003cp\u003eDuring development of the TTR CZE FPPHR fragment screening assay, the author has noticed some special cases such as appearance of vacancy peaks, focusing, tailing and delaying of UV peaks of displaced 8-ANS and loss of assay fidelity for a high affinity TTR ligand (Tafamidis). In response, another CZE assay was developed based on the CEfrag\u0026trade; methodology [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e], in which 8-ANS was injected and separated in a capillary filled with TTR and test fragment (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). This means CEfrag\u0026trade; pre-mixes TTR with test fragment instead of 8-ANS and measures how well the test fragment could block 8-ANS binding to TTR instead of displacement. Results and some interesting observations from both methods as well as their potential causes are described, compared, and discussed providing insights and guidance on future CZE fragment screening assay developments.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"2 Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Chemicals, Reagents \u0026amp; Protein:\u003c/h2\u003e \u003cp\u003eTafamidis was synthesised by Selcia Ltd. (Ongar, Essex, UK). 8-Anilino-1-naphthalenesulfonic acid ammonium salt (8-ANS) and N-Phenylanthranilic Acid (NPA) were purchased from Sigma (Poole, Dorset, UK). Diclofenac was purchased from Cayman Chemical (Michigan, USA). All other tested compounds were either from the chemical store of Selcia Ltd. or the Selcia Fragment Library (SFL) unless otherwise stated.\u003c/p\u003e \u003cp\u003eHuman transthyretin (UniProt P02766) lyophilised from 0.02 M NH\u003csub\u003e4\u003c/sub\u003eHCO\u003csub\u003e3\u003c/sub\u003e solution was purchased from SCIPAC (Sittingbourne, Kent, UK), Product Code P171-1 Lot.1802-20-1, purity\u0026thinsp;\u0026gt;\u0026thinsp;96%. Dried TTR was weighed and dissolved with CZE Running Buffers to give a final stock concentration of 100 \u0026micro;M tetramers.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Capillary Electrophoresis:\u003c/h2\u003e \u003cp\u003eCZE assays development and fragment screening were performed on a Beckman P/ACE MDQ capillary electrophoresis system (Beckman Coulter Inc., Brea, CA) operated by 32 Karat\u0026trade; software. All capillaries (Polymicro Technologies, Phoenix, AZ) used were internally coated with polyvinyl alcohol and of around 30 cm in length, 75 microns in internal diameter. Capillary temperature was maintained at 20\u0026deg;C by recirculating fluorinated fluid coolant during electrophoresis. Sample trays and buffer trays temperature were maintained at 4\u0026deg;C and ambient respectively. For every electrophoretic run, the capillary was pressure rinsed and prefilled with electrophoresis buffer also known as background electrolyte (termed as Running Buffer hereafter) before injection. After each electrophoretic run, the capillary was rinsed with Running Buffer and H\u003csub\u003e2\u003c/sub\u003eO by pressure (20 psi for 1 minute). Two Running Buffers (RB) \u003cb\u003eA\u003c/b\u003e: 20 mM HEPES pH 7.0, 2 mM CaCl\u003csub\u003e2\u003c/sub\u003e and \u003cb\u003eB\u003c/b\u003e: 20 mM HEPES pH 7.0, 2 mM CaCl\u003csub\u003e2\u003c/sub\u003e, 1% volume to volume (v/v) DMSO were used. All CE traces were analysed by 32Karat\u0026trade; Software with a built-in automatic peak integration algorithm.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 TTR FPPHR CZE Assay Development and Fragment Screening:\u003c/h2\u003e \u003cp\u003e \u003cb\u003eControl 1\u003c/b\u003e Inject Buffer (IB) containing 8-ANS (15 or 30 \u0026micro;M) was prepared by mixing 1 \u0026micro;L of 8-ANS DMSO stock (1.5 or 3 mM) with 99 \u0026micro;L of Running Buffer A in a 0.2 mL PCR tube. \u003cb\u003eControl 2\u003c/b\u003e Inject Buffer containing a mixture of 8-ANS (15 or 30 \u0026micro;M) and TTR (10 or 20 \u0026micro;M) was prepared by mixing 1 \u0026micro;L of 8-ANS DMSO stock (1.5 or 3 mM) with 10 or 20 \u0026micro;L of 100 \u0026micro;M tetrameric TTR stock in Running Buffer A and 79 or 89 \u0026micro;L of Running Buffer A. The final concentrations (f/c) were 8-ANS 15/30 \u0026micro;M, TTR 10/20 \u0026micro;M and 1% DMSO v/v.\u003c/p\u003e \u003cp\u003eThe maximum Free Probe Peak Height (Control 1 Peak Height) was determined by Outlet Injection of Control 1 with an injection pressure of 0.5 psi for 5 seconds (equivalent to \u0026asymp;\u0026thinsp;20\u0026ndash;30 nL). An identical injection of Control 2 provides the minimum Free Probe Peak Height (Control 2 Peak Height) enabling calculation of % of bound and unbound \u003cem\u003ei.e.\u003c/em\u003e free 8-ANS. To determine the 8-ANS peak height in Control 1 and Control 2, the capillary was prefilled with Buffer B and electrophoresis carried out at 10, 15, or 30 kV (Normal Polarity) with UV monitoring at 214 or 230 nm. For displacement analysis, ligand DMSO stocks (0.5\u0026ndash;30 mM) was diluted 100x in Buffer A (f/c 5\u0026ndash;300 \u0026micro;M) and used to fill the capillary as the Running Buffer. Electrophoresis commenced after injection of Control 2, thus the injected 8-ANS-TTR complex migrates inside a capillary filled with the test ligand. The observed free 8-ANS peak height is defined as Sample Peak Height and its restoration can be calculated as:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:\\%\\:of\\:Free\\:Probe\\:Peak\\:Height\\:Restoration=\\:\\frac{Sample\\:Peak\\:Height-Control\\:2\\:Peak\\:Height}{Control\\:1\\:Peak\\:Height-Control\\:2\\:Peak\\:Height}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 TTR CEfrag\u0026trade; CZE Assay:\u003c/h2\u003e \u003cp\u003eInject Buffer was prepared by mixing 1 \u0026micro;L of 1.5 mM 8-ANS DMSO stock with 1 \u0026micro;L of test compound DMSO stock (0.025\u0026ndash;5 mM) then diluted with 198 \u0026micro;L of Buffer A in a 0.2 mL vial. Final concentration was 8-ANS 7.5 \u0026micro;M, test compound 0.25\u0026ndash;50 \u0026micro;M and 1% DMSO v/v. Running Buffer was prepared by mixing 5 \u0026micro;L of 100 \u0026micro;M TTR protein stock solution in Buffer A with 2 \u0026micro;L of test compound DMSO stock (0.025\u0026ndash;5 mM) in a 0.2 mL vial. The mixture was then diluted with 193 \u0026micro;L of Buffer A, f/c: TTR 2.5 \u0026micro;M, test compound 0.25\u0026ndash;50 \u0026micro;M, 1% DMSO v/v. 100 \u0026micro;L of this solution was transferred into another 0.2 mL vial forming a Running Buffer pair. Sample injection and electrophoresis conditions were the same as the FPPHR assay.\u003c/p\u003e \u003c/div\u003e"},{"header":"3 Results and Discussions","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Vacancy Peak of Test Fragment (FPPHR) and Target Protein (CEfrag\u0026trade;)\u003c/h2\u003e \u003cp\u003eZeroing of UV absorbance on the background electrolyte (known as Running Buffer hereafter) which fills the capillary occurred in all CZE separation prior to electrophoresis. This sets the basal UV level for which passing analyte separation bands are measured against. UV absorbance is the same throughout the entire capillary volume prior to sample injection, but a gap (the injection plug) is introduced after (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The gap could manifest itself as a negative absorbance peak in the electropherogram when the following two conditions are satisfied. Firstly, the Running Buffer (RB) contains species that absorb at the measured UV wavelength but absent in the injected sample (known as Inject Buffer hereafter). Secondly, such UV-absorbing species migrate in the direction towards the detection window during electrophoresis. The vacancy peak phenomenon would be ignorable if RB has no or little basal measured UV absorbance, but this is rare in the context of biomolecular screening assays because of DMSO omnipresence and extensive UV absorbance overlap of small and biological molecules. Thus, DMSO percentage in the Inject Buffer (IB) should always match to that of Running Buffer (RB), especially if an uncoated capillary is used with strong Electroosmotic Flow (EOF). Additionally, DMSO should be kept below 5% (10% max) as more will further reduce the dynamic range and lowering UV sensitivity. Notwithstanding a potential issue in CZE bioassay, the vacancy peak principle is utilised for detecting analytes of no or little UV absorbance such as metal ions.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAlthough DMSO percentage and buffer content (pH, salt concentration, additives) could be matched in RB and IB, there however will be other unmatchable background UV-active species (the test fragment or target protein) in CZE fragment screening depending on the methodology. In the CZE FPPHR method, the test fragment is in the Running Buffer but not Inject Buffer, meaning vacancy peak of the test fragment could appear dependent on various factors. In our case, vacancy peak of any negatively charged small molecules that absorbs around UV 230 nm and migrates pass the detection window within 5 minutes will appear in the electropherogram (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). As for the CEfrag\u0026trade; method, target protein is mixed with test fragment in RB and used to fill the capillary but not added to IB which contains the Probe Ligand and matched concentration of test fragment. Henceforth, vacancy peak of target protein could appear under certain conditions, which it did in the TTR CEfrag\u0026trade; assay described in this paper (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eHaving understood how differences between IB and RB contents in FPPHR and CEfrag\u0026trade; method gave rise to vacancy peaks, there implies under certain circumstances the assays could fail. For instance, if the test fragment vacancy peak has the same or very close (by less than \u0026asymp;\u0026thinsp;20 s) migration time to that of the free Probe Ligand in FPPHR, accurate free probe peak height will be unobtainable. When Probe Ligand migration time is very close (by less than \u0026asymp;\u0026thinsp;1 minute) to that of target protein in CEfrag\u0026trade;, not only will protein vacancy peak appear, but Probe Ligand mobility shift would also be minimal hampering assay sensitivity. Test fragment vacancy peak theoretically should not occur in CEfrag\u0026trade;, because fragment concentration is matched in the Inject Buffer (IB) and Running Buffer (RB). However, slight concentration difference could lead to additional positive (if fragment concentration were higher in IB, \u003cb\u003eS1 Fig.\u003c/b\u003e) or negative peaks \u003cem\u003ei.e.\u003c/em\u003e vacancy peak, especially if the test fragment strongly absorbs at the detected UV wavelength.\u003c/p\u003e \u003cp\u003eBear in mind that modern CE instruments could equip Laser Induced Fluorescence (LIF) detection, which should alleviate the UV vacancy peak problem. This however will require a fluorescent probe (not always readily available) and introduce fluorescence interference. Coincidentally, 8-ANS is a commonly used fluorescent probe to study protein conformational changes, however LIF was unavailable to the author at the time. If LIF detection were used in FPPHR, Probe Ligand and protein concentration could be further reduced making the assay more adept for high affinity ligands, but not ideal for low affinity fragments.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Activity of Tafamidis in TTR CZE FPPHR and CEfrag\u0026trade; Assays\u003c/h2\u003e \u003cp\u003eTafamidis, an approved drug for Familial Amyloidotic Polyneuropathy (a form of TTR amyloidosis) binds TTR with dissociation constants (K\u003csub\u003ed\u003c/sub\u003e) of less than 0.2 \u0026micro;M [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. This high affinity ligand was tested as a positive control during FPPHR and CEfrag\u0026trade; assay development, but its apparent potency appeared to be weaker in FPPHR. Free 8-ANS peak height restoration reached 78% when 50 \u0026micro;M (5x [TTR]) of Tafamidis was added to the Running Buffer (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA \u003cb\u003eTrace E\u003c/b\u003e) but 8-ANS-TTR interaction was fully blocked at just 5 \u0026micro;M (2x [TTR]) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB \u003cb\u003eTrace H\u003c/b\u003e) in CEfrag\u0026trade;. The discrepancy could result interdependently from the assay methodology, electrophoretic properties and binding kinetics of Tafamidis towards TTR.\u003c/p\u003e \u003cp\u003eCompared to CEfrag\u0026trade;, TTR is never mixed with the test fragment until after sample injection, at which then molecular diffusion and 8-ANS displacement could occur for \u0026asymp;\u0026thinsp;10 s at either end of the injection plug interface before electrophoresis starts. Mixing and 8-ANS displacement would continue during early electrophoresis as the injected 8-ANS-TTR complex migrates towards the detection window, but electrophoretic property of test fragments varies. Fragments could be moving in the opposite or same direction of 8-ANS-TTR or be virtually static resulting heterogenous electrophoretic mixing. Though considered a rare event, if test fragment moves at exact velocity to that of 8-ANS-TTR, there would be no further mixing besides the initial pressure injection. This inadvertently introduces factors besides binding affinity that could determine the final observed level of 8-ANS displacement such as association rate, non-Brownian motion and relative velocity of the test fragment. The fact that Tafamidis moves at similar velocity (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA) to that of 8-ANS-TTR meant their electrophoretic mixing was comparatively poorer than other compounds of greater velocity difference to 8-ANS-TTR. A slow K\u003csub\u003eon\u003c/sub\u003e rate might be another contributing factor to potency underestimation of Tafamidis by FPPHR.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe author pondered whether to use FPPHR for fragment screening but thought the immense excess amount of test fragment (entire capillary volume) compared to 8-ANS-TTR (small injection plug) could compensate short interaction time, especially when screen concentration is typically set to 300\u0026ndash;1000 \u0026micro;M. Moreover, majority of small molecule protein bindings are of fast K\u003csub\u003eon\u003c/sub\u003e, around 10 seconds of contact time was thought to suffice, not necessarily for establishing equilibrium, but good enough for binding detection. In addition, the author expected 8-ANS-TTR complex and test fragment molecules to be in non-Brownian motion colliding each other with certain degree of order during electrophoresis, which might facilitate 8-ANS displacement. Lastly, the apparent potency of another high affinity ligand Thyroxine T\u003csub\u003e4\u003c/sub\u003e tested by FPPHR [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e] aligned well with previously published K\u003csub\u003ed\u003c/sub\u003e values. Potency underestimation of Tafamidis by FPPHR was an unusual exception. By considering the pragmatic advantages of FPPHR (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), the author\u0026rsquo;s intuition was that the benefits of using it as a primary screening tool (yes/no binding) outweighs the risk of potentially missing out binders of very slow K\u003csub\u003eon\u003c/sub\u003e or having inadequate electrophoretic mixing.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePractical Comparisons between TTR CZE FPPHR and CEfrag\u0026trade; Assay Methodology\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMethod\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFPPHR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCEfrag\u0026trade;\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eProtein Consumption\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLow\u003c/p\u003e \u003cp\u003e(at least 500x lower)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHigh\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eProtein Adsorption\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUnlikely\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePossible\u003c/p\u003e \u003cp\u003e(could be reduced by additives, harsh rinse, or with a different capillary surface)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSample Handling\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSimple\u003c/p\u003e \u003cp\u003e(samples could be prepared manually without expensive liquid handling robotics)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eComplex\u003c/p\u003e \u003cp\u003e(need to prepare many pairs of protein solutions \u0026ndash; not ideal for manual pipetting)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRun Time\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eShorter\u0026thinsp;\u0026lt;\u0026thinsp;5 Minutes\u003c/p\u003e \u003cp\u003e(Probe Ligand peak height is measured instead of migration time)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLonger\u0026thinsp;\u0026lt;\u0026thinsp;10 Minutes\u003c/p\u003e \u003cp\u003e(the method requires large mobility shift, which is time dependent)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eBinding Detection Measurement\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQuantitative\u003c/p\u003e \u003cp\u003e(numerical peak height values)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eQualitative \u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e \u003cp\u003e(8-ANS peak separation from TTR vacancy peak)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"3\"\u003e\u003csup\u003ea)\u003c/sup\u003e measurement could become quantitative when there is a significant mobility shift of the Probe Ligand. In the case described here, electrophoretic velocity of 8-ANS was too close to that of TTR and their molecular binding was reflected by peak merging (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB\u003cb\u003e).\u003c/b\u003e Competitive binding from test compound took place in form of 8-ANS and TTR peak separation (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB\u003cb\u003e)\u003c/b\u003e, which the author had judged subjectively. The peak separation or shape change could be quantitatively modelled by mathematics, but it was beyond the author\u0026rsquo;s expertise.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e129 fragments were then screened at 300 \u0026micro;M by FPPHR providing 16 initial hits and 4 failed (3.1%) runs due to vacancy peak (they were later confirmed negative by CEfrag\u0026trade;, \u003cb\u003eS2 Fig.\u003c/b\u003e). A hit rate of 12.4% is unlikely achieved had the assay been fundamentally flawed in hit picking due to short contact time. In hindsight, a total of 54 compounds (mostly fragments) were tested by both FPPHR and CEfrag\u0026trade;, and their overall agreement in terms of hit detection (yes/no) was around 90% (49/54), see Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCorrelation of FPPHR and CEfrag\u0026trade; Method on Fragment Binding Detection\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTotal Number of Compounds Tested\u0026thinsp;=\u0026thinsp;54\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"4\" nameend=\"c6\" namest=\"c3\"\u003e \u003cp\u003eFPPHR Method\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eCEfrag\u0026trade; Method\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003ePositive \u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePositive \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e33 \u003csup\u003ec\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7 \u003csup\u003ed\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4 \u003csup\u003ec\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1 \u003csup\u003ed\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1 \u003csup\u003ed, e\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e6 \u003csup\u003ec\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e3 \u003csup\u003ed\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e \u003cp\u003e\u003csup\u003ea)\u003c/sup\u003e a positive compound in the FPPHR method would have had caused a minimum of 10% free 8-ANS peak height restoration.\u003c/p\u003e \u003cp\u003e\u003csup\u003eb)\u003c/sup\u003e a positive compound in the CEfrag\u0026trade; method must have had caused an obvious separation of free 8-ANS and TTR vacancy peak, \u003cem\u003ee.g.\u003c/em\u003e Trace E \u0026ndash; H in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB. Trace C \u0026ndash; D in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB would be regarded as negative. Regrettably, all peaks shape change/separation were judged subjectively based on experience.\u003c/p\u003e \u003cp\u003e\u003csup\u003ec)\u003c/sup\u003e Tested at same concentration; \u003csup\u003ed)\u003c/sup\u003e Tested at different concentration, a maximum difference of 2-fold\u003c/p\u003e \u003cp\u003e\u003csup\u003ee)\u003c/sup\u003e This compound (SFL000029) was tested positive by FPPHR at 300 \u0026micro;M and negative in CEfrag\u0026trade; at 150 \u0026micro;M. It was found to be positive when tested again with CEfrag\u0026trade; at 500 \u0026micro;M, and later confirmed by crystallography [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e].\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eFurthermore, the author has later (2019\u0026ndash;2020) developed another CZE FPPHR fragment screening assay for an undisclosed oncology target in search of first-in-class protein-protein interaction inhibitors. 1126 fragments were screened with 47 failed runs, of which 4 failed (0.35%) because of vacancy peak issue. The percentage of failed runs by vacancy peak was much lower than the TTR screen because the Probe Ligand used was a small peptide having significantly different migration time to almost all screened fragments. 44 out of the remaining 1079 showed significant Probe Ligand displacement (4% hit-rate) and multiple fragments hits were later confirmed by crystallography, biochemical assays and cell-based assays (confidential data). It is only then the author became fully confident in the effectiveness of the FPPHR method for fragment binding detection. Having a primary screening technique affording highly reliable hits would immensely streamline subsequent processes saving costs and invaluable time.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e3.3 FPPHR 8-ANS Displacement by Various Compounds With Distinct Peak Profiles\u003c/h2\u003e \u003cp\u003eDuring the development of TTR CZE FPPHR fragment screening assay, multiple compounds known to bind TTR (typically of higher affinity than fragments, including Tafamidis) were first tested as positive controls. Many of them had displaced significant amount of TTR-bound 8-ANS leading to increased free 8-ANS peak height, of which the peak shape remained gaussian-like (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). Fascinatingly, peak shape change for some compounds were non-gaussian, characterised by peak focusing, splitting, elongation and migration delay (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn the case of N-Phenylanthranilic Acid (NPA), the displaced 8-ANS was initially split from the original free 8-ANS population (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA, \u003cb\u003eTrace B\u003c/b\u003e). As [NPA] in RB increases (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA, \u003cb\u003eTrace C \u0026ndash; F)\u003c/b\u003e, the displaced 8-ANS shifts atop and became \u0026ldquo;focused\u0026rdquo; as a sharper peak. It is worth noting that NPA elutes (vacancy peak) just before the free 8-ANS peak and the peak \u0026ldquo;focusing\u0026rdquo; effect was observed in only, but not all, negatively charged compounds. Other tested compound that experienced the \u0026ldquo;focusing\u0026rdquo; phenomenon includes 4-Pyridin-3-yl-benzoic acid (SFL000046), Biphenyl-4-carboxylic acid (A00002802), and Diflunisal, they all elute close to and slightly earlier than free 8-ANS (\u003cb\u003eS3 Fig.\u003c/b\u003e). Migration of displaced 8-ANS could have been affected by the reduced conductivity (dependent on test compound concentration) of the vacancy peak zone being right in front of it. Other causes related to the test compound\u0026rsquo;s electrophoretic properties and/or TTR binding kinetics are also plausible.\u003c/p\u003e \u003cp\u003eThe author went on to test a set of compounds that were being investigated by the Wolfson Drug Discovery Unit (the Royal Free Hospital) at UCL together with Diclofenac. The compounds (structure undisclosed) displayed low micromolar IC\u003csub\u003e50\u003c/sub\u003e values (except compound T85\u0026thinsp;\u0026gt;\u0026thinsp;1 mM) in their \u003csup\u003e125\u003c/sup\u003eI-T\u003csub\u003e4\u003c/sub\u003e displacement assay [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e] thus test concentration was reduced accordingly in the TTR CZE FPPHR assay.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFrom electropherograms of those 11 test compounds (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB), a variety of free 8-ANS peak profiles were observed. 8-ANS displacement by compound T07, T15 \u0026amp; T23 led to Gaussian-like increase in free 8-ANS peak size. Compound T34 and T85 were considered negative. Compound T44, T47, T58, T60, T70 and Diclofenac caused free 8-ANS peak tailing, splitting or elongation instead. Intriguingly they all move in the same direction as 8-ANS-TTR, implying their electrophoretic mixing and/or binding kinetics (during electrophoresis) might be distinctive to neutral and positively charged compound. Diclofenac elutes close to 8-ANS but soon after rather than slightly before and the peak focusing effect was absent. The inclination was that such event would only occur when competing compounds elute just before free 8-ANS. The underlying mechanism for all these displaced 8-ANS peak forms could have been a very interesting topic relevant to the field of free solution native CZE.\u003c/p\u003e \u003c/div\u003e"},{"header":"4 Concluding Remarks","content":"\u003cp\u003eBy manipulating contents of the Running and Inject Buffer, the author has developed two methodologically (FPPHR and CEfrag\u0026trade;) distinct CZE binding competition assays for transthyretin. FPPHR has multiple practical screening advantages over CEfrag\u0026trade; and has been successfully applied to two targets of different disease area. However, it comes with a caveat that probe ligand displacement also depends on the on-rate and electrophoretic mixing of test compound. CEfrag\u0026trade; on the other hand could be used in a more affinity-orientated hit confirmation step. The vacancy peak, peak merging, tailing, splitting, focusing and elongation phenomena further highlights the importance of taking the electrophoretic properties and binding kinetics of all involved analytes into proper consideration when developing CZE fragment screening assays and interpreting results thereafter.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003e8-ANS\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;8-anilinonaphthalene-1-sulfonic acid\u003c/p\u003e\n\u003cp\u003eACE\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Affinity Capillary Electrophoresis\u003c/p\u003e\n\u003cp\u003eCZE\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Capillary Zone Electrophoresis\u003c/p\u003e\n\u003cp\u003eDMSO\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Dimethyl Sulfoxide\u003c/p\u003e\n\u003cp\u003eDSF\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;3,5-Dichlorobenzen Sulphonamide\u003c/p\u003e\n\u003cp\u003eEOF\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Electroosmotic Flow\u003c/p\u003e\n\u003cp\u003eFBDD\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Fragment-based Drug Discovery\u003c/p\u003e\n\u003cp\u003ef/c\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Final Concentration\u003c/p\u003e\n\u003cp\u003eFPPHR\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Free Probe Peak Height Restoration\u003c/p\u003e\n\u003cp\u003eHEPES\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;4-(2-Hydroxyethyl)piperazine-1-ethanesulfonic acid\u003c/p\u003e\n\u003cp\u003eIB\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Inject Buffer\u003c/p\u003e\n\u003cp\u003eNMR\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Nuclear Magnetic Resonance\u003c/p\u003e\n\u003cp\u003eNPA\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;N-Phenylanthranilic Acid\u003c/p\u003e\n\u003cp\u003eRB\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Running Buffer\u003c/p\u003e\n\u003cp\u003eTTR\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Transthyretin\u003c/p\u003e\n\u003cp\u003ev/v\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Volume to Volume\u003c/p\u003e\n"},{"header":"Declarations","content":"\u003cp\u003eAcknowledgement\u003c/p\u003e\n\u003cp\u003eThe research work here formed part of a Ph.D. project (2010 \u0026ndash; 2014) co-supervised by Professor Stephen P. Wood and Dr. Carol Austin. All capillary electrophoresis experiments were performed in Selcia Ltd (now part of Eurofins). The author thanks Dr. Graham Taylor from the UCL Wolfson Drug Discovery Unit for providing test compounds.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAustin C, Pettit SN, Magnolo SK, Sanvoisin J, Chen W, Wood SP, Freeman LD, Pengelly RJ, Hughes DE (2012) Fragment screening using capillary electrophoresis (CEfrag) for hit identification of heat shock protein 90 ATPase inhibitors. J Biomol Screen 17(7):868\u0026ndash;876\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFarcas E, Bouckaert C, Servais AC, Hanson J, Pochet L, Fillet M (2017) Partial filling affinity capillary electrophoresis as a useful tool for fragment-based drug discovery: A proof of concept on thrombin. Anal Chim Acta 984:211\u0026ndash;222\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNeaga IO, Hambye S, Bodoki E, Palmieri C, Ansseau E, Belayew A, Oprean R, Blankert B (2018) Affinity capillary electrophoresis for identification of active drug candidates in myotonic dystrophy type 1. Anal Bioanal Chem 410(18):4495\u0026ndash;4507\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRauch JN, Nie J, Buchholz TJ, Gestwicki JE, Kennedy RT (2013) Development of a capillary electrophoresis platform for identifying inhibitors of protein-protein interactions. Anal Chem 85(20):9824\u0026ndash;9831\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen W (2025) Fragment-based drug discovery for transthyretin kinetic stabilisers using a novel capillary zone electrophoresis method. PLoS ONE 20(5):e0323816\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFerguson RN, Edelhoch H, Saroff HA, Robbins J, Cahnmann HJ (1975) Negative cooperativity in the binding of thyroxine to human serum prealbumin. Preparation of tritium-labeled 8-anilino-1-naphthalenesulfonic acid. Biochemistry 14(2):282\u0026ndash;289\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBulawa CE, Connelly S, Devit M, Wang L, Weigel C, Fleming JA, Packman J, Powers ET, Wiseman RL, Foss TR, Wilson IA, Kelly JW, Labaudiniere R (2012) Tafamidis, a potent and selective transthyretin kinetic stabilizer that inhibits the amyloid cascade. Proc Natl Acad Sci U S A 109(24):9629\u0026ndash;9634\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKolstoe SE, Mangione PP, Bellotti V, Taylor GW, Tennent GA, Deroo S, Morrison AJ, Cobb AJ, Coyne A, McCammon MG, Warner TD, Mitchell J, Gill R, Smith MD, Ley SV, Robinson CV, Wood SP, Pepys MB (2010) Trapping of palindromic ligands within native transthyretin prevents amyloid formation. Proc Natl Acad Sci U S A 107(47):20483\u0026ndash;20488\u003c/span\u003e\u003c/li\u003e \u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"University College London","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Capillary Zone Electrophoresis, Fragment Screening, Transthyretin, FPPHR, Affinity Capillary Electrophoresis, Drug Discovery","lastPublishedDoi":"10.21203/rs.3.rs-8908294/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8908294/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eReliable and sensitive detection of specific protein-ligand molecular interactions is vital for target-based early-stage small molecule drug discovery. Using human transthyretin as an example, the author here describes how non-denaturing Capillary Zone Electrophoresis (CZE) is used to detect weakly binding small molecule fragments. Two methodologically distinct CZE competition fragment screening assays were developed during the Fragment-based Drug Discovery (FBDD) campaign and compared. Many intriguing electropherograms, all attributed to effects of molecular binding and electrophoretic behaviour were observed during assay development and their implications are discussed.\u003c/p\u003e","manuscriptTitle":"On Some Curiosities of Native Capillary Zone Electrophoresis Involving Protein-Ligand Interactions From Fragment Screening of Transthyretin","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-19 07:21:56","doi":"10.21203/rs.3.rs-8908294/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"f2e26a9d-e01f-4c99-bf42-e86cf1a27e37","owner":[],"postedDate":"February 19th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":63128402,"name":"Drug Discovery, Design, \u0026 Development"}],"tags":[],"updatedAt":"2026-02-19T07:21:56+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-19 07:21:56","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8908294","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8908294","identity":"rs-8908294","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}