Purification of supercoiled G-quadruplex pDNA for in vitro transcription

Accepted Manuscript Purification of supercoiled G-quadruplex pDNA for in vitro transcription Tiago Santos, Patrícia Pereira, Fani Sousa, João A. Queiroz, Carla Cruz PII: DOI: Reference:

S1383-5866(16)30089-2 http://dx.doi.org/10.1016/j.seppur.2016.02.036 SEPPUR 12867

To appear in:

Separation and Purification Technology

Received Date: Revised Date: Accepted Date:

21 August 2015 16 February 2016 22 February 2016

Please cite this article as: T. Santos, P. Pereira, F. Sousa, J.A. Queiroz, C. Cruz, Purification of supercoiled Gquadruplex pDNA for in vitro transcription, Separation and Purification Technology (2016), doi: http://dx.doi.org/ 10.1016/j.seppur.2016.02.036

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Purification of supercoiled G-quadruplex pDNA for in vitro transcription Tiago Santos1†, Patrícia Pereira1†, Fani Sousa1, João A. Queiroz1, Carla Cruz1*

CICS-UBI - Centro de Investigação em Ciências da Saúde, Universidade da Beira Interior, Av. Infante D. Henrique, 6200-506 Covilhã, Portugal; Tel: +351 275 329 076; Fax: +351 275 329 099; e-mail: [email protected]

†These authors contributed equally to this work.

Keywords: G-quadruplex; supercoiled isoform; amino acids-based supports; affinity chromatography; in vitro transcription; circular dichroism

1

Abstract The formation of G-quadruplex in G-rich regions of DNA can be induced by transcription and is formed within plasmid transcribed in Escherichia coli. The G-loops are more evidenced on supercoiled (sc) topology than on relaxed (oc) or linearized (ln) plasmid isoforms. Thus, the present work reports different purification strategies to efficiently purify the pPH600 sc isoform from other plasmid topologies and host contaminants present in a clarified E. coli lysate. To accomplish this purpose, two affinity supports, L-tyrosineSepharose and L-tryptophan-Sepharose were prepared by linking L-tyrosine and Ltryptophan onto epoxy-activated Sepharose CL-6B and were further characterized. The commercial support L-arginine Sepharose was also used to purify sc pPH600 since it has already been efficiently applied to separate sc isoforms of other plasmids using mild binding and elution conditions. A first screen was performed to select the best support that allows to obtain highly pure sc pPH600 from a native sample (sc+oc). By comparing the binding/elution conditions of the three supports, L-tyrosine support showed the preeminent result in separation of two isoforms, allowing the total recovery of sc pPH600, using a decreasing (NH4)2SO4 gradient in HEPES 100 mM at 10ºC. The purification of sc pPH600 directly from clarified E. coli lysate was achieved with the support L-tyrosine-Sepharose and the quality control analysis revealed that the level of E. coli impurities (other pDNA topologies, proteins, endotoxins, gDNA and RNA) present in the final sc pPH600 sample was in accordance with the guidelines of regulatory agencies. In vitro transcription was performed using the purified sc pDNA to induce G-quadruplex formation and it was confirmed by circular dichroism (CD) that the transcript adopted parallel G-quadruplex topology. Overall, this work showed that sc pPH600 can be purified using L-tyrosine support and the transcript adopted parallel G-quadruplex topology.

2

1 - Introduction DNA has the potential to form a number of secondary structures, namely, G-quadruplex, which is a four-stranded structure stabilized by G-quartets planar arrays of four hydrogenbonded guanines and coordinated with cations (especially K+) within the central cavity [1]. Thus, G4 DNA is very stable once formed. These structures have been found in telomere ends and in promoter regions of oncogenes and some of these G4 structures have been identified as repressor elements to regulate gene transcription and translation in a wide variety of genes [2]. G-rich sequences are also found in immunoglobulin switch regions and are involved in the regulation of Escherichia coli (E. coli) plasmid replication [3]. It was found the formation of G-loops within a plasmid upon transcription of the S regions in living cells, in which one strand contains a RNA/DNA hybrid and the other strand contains Gquadruplex DNA [4]. G-loop formation depends on plasmid topology and it is more efficient on supercoiled (sc) templates [4]. The pPH600 plasmid (3562 bp) is derivate from the commercial plasmid pBluescript KS+ and encodes a sequence of a G-C rich sequence of immunoglobulin switch region Sγ3 of murine that can form G-quadruplex [3]. To assess Gloop formation on sc isoform, the pPH600 containing a G-rich coding strand needs to be transcribed in vitro. After biosynthesis of pPH600 in E. coli, the isolation and purification of sc isoform from lysate is still a huge challenge. Indeed, a special downstream processing is required since most of the molecules present in the lysate share analogous chemical, physical, and structural properties with plasmid: negative charge (RNA, genomic DNA (gDNA), and endotoxins) and molecular mass (gDNA and endotoxins) [5]. In what concerns to the preparation of a plasmid DNA (pDNA) product, the requirements of the regulatory agencies establish that host proteins and RNA must be undetectable, the level of gDNA should be lower than 2 ng gDNA/μg pDNA and endotoxins should not exceed 0.1 EU/μg pDNA, in the final product [6, 7]. Moreover, the purification method should not comprise the use of organic reagents, mutagenic and toxic compounds and animal derived enzymes [8]. Also, during the manufacturing and recovery process, pDNA is subjected to several stresses that damage the sc pDNA, resulting in relaxed or open circular (oc) forms, and other variants such as linear, denatured or dimeric conformations [9]. Thus, for its final purification it is necessary to have a highly selective chromatographic process. Affinity chromatography is one of the most powerful techniques employed in the selective isolation of target molecules [10]. Chromatographic supports take advantage of the small differences between the sc pDNA and its natural impurities, on properties such as charge, size, 3

hydrophobicity, accessibility of the nucleotide bases, and topological constraints imposed by supercoiling and/or affinity, permitting the design of selective pDNA purification strategies [10]. Several amino acids have already been efficiently applied as chromatographic ligands to separate plasmid isoforms, revealing the presence of a particular recognition of sc isoform [11-13]. The aromatic amino acids show unique and important properties due to its side chain [14]. Therefore, the properties of phenol and indole groups in L-tyrosine and Ltryptophan, respectively, can be used to evaluate the separation of sc pDNA from contaminants. Here, taking advantage from these properties, we prepared two affinity matrices by immobilizing L-tyrosine and L-tryptophan onto epoxy-activated Sepharose CL-6B (see Fig. 1). The commercial support L-arginine-Sepharose was also used in the strategy to purify sc pPH600 since it has already been efficiently applied to separate this isoform using mild binding and elution conditions. Then, we investigated the retention behaviour of pPH600 isoforms on these affinity matrices and discussed the factors that influence the separation efficiency. Also, purification of sc pPH600 directly from clarified E. coli lysate was achieved with the support L-tyrosine-Sepharose, and its quality and yield was further evaluated. Finally, the G-quadruplex formation after in vitro transcription of the purified sc pPH600 was assessed by circular dichroism (CD).

4

2 - Material and Methods 2.1 - Materials All solutions were freshly prepared with ultra-pure grade deionized water purified in MilliQ system from Millipore (Billerica, MA, USA). Solutions were filtered through a 0.20 µm pore size membrane (Whatman, Dassel, Germany) and degassed before use. Sodium chloride (NaCl) and HEPES Acid were purchased from Fisher Scientific (Fair Lawn, NJ, USA). Ammonium sulphate ((NH4)2SO4) and sodium carbonate were obtained from Panreac (Barcelona, Spain). Hyper Ladder I (Bioline, London, UK) was used as DNA molecular weight marker. GreenSafe Premium and NZY Maxiprep Kit were purchased from NZYTech (Lisbon, Portugal). The Maxima® SYBR Green/Fluorescein qPCR Master Mix (Thermo Fisher Scientific Inc.) was used for gDNA quantification. 2.2 - Plasmid production and isolation The plasmid pPH600 (3562 bp), kindly provided by Dr. N. Maizels, was amplified by a cell culture of E. coli DH5α. Growth was carried out overnight in shake flasks (250 rpm) at 37°C using a terrific broth medium (12 g/L Tryptone, 24 g/L Yeast extract, 4 ml/L glycerol, 0.017 M KH2PO4 and 0.072 M K2HPO4) supplemented with 100 µg/mL of ampicillin. pDNA was purified using NZYTech Plasmid Maxiprep kit according to the supplier’s protocol to obtain the native pDNA (sc and oc isoforms). The protocol is based on alkaline lysis procedure followed by the binding of pDNA to the NZYTech anion-exchange resin under appropriate low salt and pH conditions. After that, the impurities were removed by a medium salt wash. Finally, plasmid was eluted with high salt conditions and then is concentrated and desalted by isopropanol precipitation. The resulting pDNA preparations were dissolved in 10 mM Tris-HCl (pH 8.0) buffer. These samples were used for initial evaluation of chromatographic retention behaviour. Crude extract of pDNA with all of contaminants (gDNA, RNA, proteins, endotoxins) was obtained through a modified alkaline lysis method described by Diogo and collaborators [15]. The pDNA samples concentration was measured with a NANOPhotometer (Implen). All the samples were stored at -20°C. 2.3 - Immobilization of L-tryptophan and L-tyrosine on Sepharose CL-6B Sepharose CL-6B was washed with 300 mL of milli-Q water and suspended in 0.6 M NaOH with 1.32 mM NaBH4. After 15 min, 1.4-butanediol diglycidyl ether (5 mL) was added slowly with stirring at 25ºC for 6 h. The mixture was washed with distilled water and acetone to remove an oily from the surface of the gel (the remaining epoxy compound). The 5

epoxy-activated Sepharose CL-6B matrix (3 g) was used to couple L-tyrosine and Ltryptophan. Each ligand was dissolved (L-tyrosine, 8.27 mmol, 1.5 g; L-tryptophan, 14.7 mmol, 3 g) in a solution of sodium carbonate 2 M at pH approximately 9. The mixture was stirred on the orbital at 55ºC for 16 h. After that, the supports were washed extensively with a mixture of acetone-milli-Q water (1:9; 3:7; 5:5; 8:2 v/v) followed by washing with milli-Q water (3×100 mL). 2.4 - HR-MAS NMR Spectroscopy and sample preparation Approximately 10 mg of the support was placed in a 4-mm MAS zirconia rotor (50 μL). All NMR experiments were performed at room temperature using a Bruker Avance III 400 operating at 400.15 MHz for protons, equipped with a 4-mm triple resonance (HNC) HRMAS probehead. Samples were spun at the magic angle at a rate of 4.0 kHz, and all spectra were acquired under field-frequency locked conditions using that probe channel with the spectrometer’s lock hardware. Spectra were processed using Bruker Topspin 3.1. All 1H NMR spectra were referenced internally to the residual 1H signal of DMF-d7, which also serves as the swelling agent for the polymer beads (∼0.05 mL), unless stated otherwise. Carr-Purcell-Meiboom-Gill (CPMG) sequence with an echo time of 1.5 ms was used to suppress the broad signals of the matrix, experiments were acquired in 256 transients. NOESY experiments were acquired with 150 ms mixing time in 16 transients with a relaxation delay of 2.0 s and a spectral width of ca 6000 Hz in a total of 2 K data points in F2 and 256 data points in F1. 2D TOCSY experiments were acquired by MLEV-17 pulse sequence with 75 ms mixing time in 8 transients with a relaxation delay of 2.0 s and a spectral width of ca 6000 Hz, in a total of 2 K data points in F2 and 256 data points in F1. 2.5 - Preparative Chromatography The three affinity supports, L-arginine-Sepharose 4B gel (commercial), L-tryptophanSepharose CL-6B and L-tyrosine-Sepharose CL-6B were packed individually using 2.5 mL of each support in 10 mm diameter × 35 mm long/bed height columns. All preparative chromatographic experiments using the three supports were performed in Akta Pure 25 L controlled by UNICORN software, version 6.3, at a flow-rate of 1 mL/min. The absorbance was continuously measured at 260 nm. According to each experiment the desirable temperature of column was maintained (4, 10, 15 and 25ºC), connecting a water-jacket column to circulate water bath. After purification, the fractions were pooled, according to the obtained chromatograms, concentrated and desalted with Vivaspin concentrators and 6

analysed by gel electrophoresis. The columns were washed with at least 5 Column Volumes (CV) of ultra-pure grade deionized water. 2.5.1 - Screening with L-arginine-Sepharose, L-tryptophan-Sepharose and L-tyrosineSepharose The column with commercial L-arginine-Sepharose 4B gel was equilibrated with at least 6 CV of 110 mM NaCl in 10 mM Tris-HCl (pH 8.0). Then, the plasmid pPH600 sample was injected onto the column using a 100 µL loop at the same flow-rate conditions. After elution of the non-retained species with 2.6 CV of equilibration buffer, the recovery of bound biomolecules was achieved with 4.8 CV by increasing the NaCl concentration from 110 to 500 mM NaCl in 10 mM Tris-HCl (pH 8.0). The experiments performed with L-tryptophanSepharose were initiated with the equilibration of the column with 6 CV of 2.65 M (NH4)2SO4 in 100 mM HEPES acid (pH 7.4). Thereafter, plasmid pPH600 was injected onto the column and the elution of unbound species was achieved with 2.9 CV of 2.65 M (NH4)2SO4 in 100 mM HEPES acid (pH 7.4). After the elution of the non-retained species, the ionic strength of the buffer was stepwise decreased to 0 M (NH4)2SO4 in 100 mM HEPES acid (pH 7.4) during 4.4 CV, in order to recover the remaining bound biomolecules. On the other hand, in the experiments with L-tyrosine-Sepharose, the column was equilibrated with at least 6 CV of 2.25 M (NH4)2SO4 in 100 mM HEPES acid (pH 7.4). The elution of unbound species was achieved with 3.0 CV at 2.25 M (NH4)2SO4 in 100 mM HEPES acid (pH 7.4) and the elution of the retained species was accomplished with 3.9 CV by changing the ionic strength of the buffer to 0 M (NH4)2SO4 in 100 mM HEPES acid (pH 7.4). Furthermore, different concentrations of pPH600 (25, 50, 75, 100, 150 and 300 µg/mL) were injected onto the column to evaluate the effect of pDNA concentration in the selectivity towards sc pPH600.

2.5.2 - Purification of E. coli lysate using L-tyrosine-Sepharose CL-6B support The chromatographic system was prepared with 100 mM HEPES acid (pH 7.4) in pump A (mobile phase A) and 3 M (NH4)2SO4 in 100 mM HEPES acid (pH 7.4) in pump B (mobile phase B). In the sc pPH600 pDNA purification from an E. coli lysate sample, the L-tyrosineSepharose column was equilibrated with at least 6 CV at 2.25 M (NH4)2SO4 in 100 mM HEPES acid (pH 7.4). The complex lysate sample was injected onto the column using a 200 µL loop at a flow rate of 1 mL/min. The elution of unbound species was achieved with 4.9 7

CV of 2.25 M (NH4)2SO4 in 100 mM HEPES acid (pH 7.4). After, the elution of bound species was gradually carried out by decreasing the ionic strength, first to 1.95 M (NH4)2SO4 in 100 mM HEPES acid (pH 7.4) using 8.8 CV and then to 0 M (NH4)2SO4 in 100 mM HEPES acid (pH 7.4) during 6.3 CV.

2.5.3 - Dynamic binding capacity A standard 10 mm diameter × 7 mm long column was packed with L-tyrosine-Sepharose support, giving a total bed volume of 0.5 mL for the determination of the dynamic binding capacity (DBC). Capacity experiments were performed with 0.01 and 0.1 mg/mL of pPH600 solution, at 1 mL/min of flow-rate. The column was equilibrated with at least 6 CV of 3 M (NH4)2SO4 in 100 mM HEPES acid (pH 7.4) and thereafter, it was overloaded with the pDNA solution under the same equilibrium conditions. DBC was determined by recording breakthrough curves and calculating the amount of bound pDNA per mL of support at 10% and 50% of breakthrough curve. Dynamic binding capacity values were then determined by subtracting the value obtained under non-binding conditions. Afterwards, the elution of the bound pDNA was achieved by decreasing the (NH4)2SO4 concentration from 3 M to 0 M in 100 mM HEPES acid (pH 7.4).

2.6 - Analytical chromatography The content of sc pDNA present in clarified E. coli lysate and in purified fractions was determined with CIMacTM pDNA analytical column by a modified analytical method [16]. CIMacTM pDNA analytical column consists of a disk-shaped poly(glycidyl methacrylate-coethylene dimethacrylate) highly porous polymer matrix, with 0.32 mL of volume and a column of 15.0 mm length and 5.2 mm diameter. The pDNA concentration in each sample was calculated by using a calibration curve constructed with pDNA standards of 2.5 to 75 µg/mL, purified with a commercial NZYTech kit. All samples were prepared by diluting the highest concentration with 200 mM Tris-HCl (pH 8.0). The analytical column was equilibrated with 20 CV of 600 mM NaCl in 200 mM Tris-HCl buffer (pH 8.0) and after the injection of 20 μL of sample, a linear gradient of 31 CV to 700 mM NaCl in 200 mM TrisHCl buffer (pH 8.0) was established, at 1 mL/min. After the chromatographic runs, the CIMacTM pDNA analytical column was washed with at least 20 CV of ultra-pure grade deionized water. 8

2.7 - Agarose gel electrophoresis The peaks resulting from the chromatographic experiments, after desalted and concentrated, were analysed by horizontal electrophoresis using 1% agarose gel (Hoefer, San Francisco, CA, USA) stained with 0.01% GreenSafe Premium and visualized under UV light in a UVItec FireReader system (UVItec, Cambridge, UK). Electrophoresis was carried out at 120 V, for 35 min, with TAE buffer (40 mM Tris base, 20 mM acetic acid and 1 mM EDTA, pH 8.0). 2.8 - Impurities assessment The pDNA quality control consists in the quantification of impurity levels, such as, proteins, endotoxins and gDNA, present in purified sc pDNA sample. 2.8.1 - Proteins Protein quantification in pPH600 samples was performed using the micro-BCA (bicinchoninic acid) protein assay kit from Pierce, in accordance with the specifications of the manufacturer. A calibration curve was prepared with the standards of protein bovine serum albumin (BSA) (0.01-0.1 mg/mL) diluted in 10 mM Tris-HCl (pH 8.0). A fraction of each sample (10 μL) was added to 200 μL of BCA reagent in a microplate and incubated for 30 min at 37ºC and then cooled to room temperature. The absorbance was measured at 570 nm in microplate reader. 2.8.2 -Endotoxins Endotoxin contamination was assessed by using the ToxiSensorTM Chromogenic Limulus Amebocyte Lysate (LAL) Endotoxin Assay Kit (GenScript, USA, Inc.), according to the supplier’s protocol. The calibration curve (0.005 to 0.1 EU/mL) was constructed using a provided stock solution of 10 EU/mL. To avoid external endotoxin interference, all the samples were diluted or dissolved in non-pyrogenic water, which was also used as blank. All the tubes and tips used to perform this quantification were endotoxin-free and the entire procedure was performed inside of a laminar flow cabine. 2.8.3 -Genomic DNA Real-time polymerase chain reaction (PCR) was used to evaluate the existence of gDNA in the purified samples. The analyses were performed in an iQ5 Multicolor real-time PCR detection

system

(Bio-Rad),

Specific

ACACGGTCCAGAACTCCTACG-3’

and

primers antisense

(sense

-

5’5’9

CCGGTGCTTCTTCTGCGGGTAACGTCA-3’) were used to amplify a 181-bp fragment of the 16S rRNA gene. PCR amplicons were quantified by following changes in fluorescence of the DNA binding dye SYBR Green/Fluorescein. The calibration curve to achieve the gDNA concentration was constructed by serial dilutions of the E. coli DH5α gDNA sample, purified with the Wizard gDNA purification kit (Promega), in the range of 0.005 to 50 ng/µL. 2.9 - Plasmid transcription Sc pPH600 transcription was enzymatically performed using MEGAscript® Kit (Ambion), according to the manufacturer’s recommendations. Briefly, the sc pPH600 containing a 604 bp PvuII-HindIII fragment of the murine Sγ3 switch region was linearized with HindIII enzyme and concentrated to a final concentration of 1µg/µL. Then, the linearized plasmid was in vitro transcribed for 1 h at 37ºC. The reaction was terminated by adding 1/20th volume 0.5 M EDTA, 1/10th volume of 3 M sodium acetate and two volumes of ethanol, followed by incubation at –20°C for at least 15 min. After centrifugation at 4°C for 15 min at maximum speed, the supernatant was discarded. Then, the pellets were solubilized in water. 2.10 - Circular dichroism spectroscopy CD spectra were acquired in Jasco J-1850 Spectrophotometer (Jasco, Easton, MD, USA) CD spectra using a quartz rectangular cell with an optical path length of 0.1 cm at temperature of 25°C. The plasmid samples and RNA/DNA hybrid samples with about 20 µg/mL were prepared in 10 mM Tris-HCl and analysed. Thereafter, the RNA/DNA hybrid samples were titrated with 100 mM KCl and heated at 95ºC for 5 min and cooled to room temperature to favour the formation of G-quadruplex. The spectral bandwidth was kept at 1 nm. The CD spectra were recorded from 200 to 320 nm at a scan speed of 10 nm/min. All measurements were conducted under a constant nitrogen gas flow, to purge the ozone generated by the light source of the instrument. The data were collected in triplicate and the average spectra are presented for each sample after subtracting the contribution of the buffer. The CD signal was converted to ellipticity. Noise in the data was smoothed using Jasco J-1850 software, including the Fast Fourier transform algorithm, which allows enhancement the noisy spectra without distorting their peak shape.

10

3 - Results and Discussion pPH600 was the plasmid used in this work and has a 604 bp PvuII-HindIII fragment of the murine

S3

switch

region,

which

contains

the

repeat

d(CTGGGCAGCTCTGGGGGAGCTGGGGTAGGTTGGGAGTGTGGGGACCAGG), inserted just downstream of the T7 promoter [4]. The plasmid pPH600 was amplified in E. coli DH5α after transformation of competent cells by heat shock. Analysing the growth profile, it was verified that the recombinant host, transformed with this plasmid grows faster, reaching an OD600 ≈ 6 in just 4 h. This behaviour is in accordance with that described by Cooke and collaborators in which the plasmids that encode sequences able to form Gquadruplex

grow significantly faster than

others

[17].

After that,

preparative

chromatography was performed with the columns packed individually with the supports Larginine Sepharose, L-tyrosine Sepharose and L-tryptophan Sepharose. To the best of our knowledge, the last two supports were synthetized and characterized for the first time in our laboratory. The supports were synthetized by coupling L-tryptophan and L-tyrosine to epoxy (1,4-butanediol diglycidyl ether)-activated Sepharose. The supports were characterized by Resolution Magic Angle Spinning (HR-MAS) NMR spectroscopy and it was observed that the ligands are immobilized on the matrix. Indeed, in the bound sample the chemical shifts of the amino acids are clear, especially with the resonances between 7.37 and 6.95 ppm for L-tyrosine and between 8.0 to 7.0 ppm for L-tryptophan aromatic protons. The analysis of the DEPT-HSQC spectra for both the bound and unbound polymer samples allowed the identification in the unreacted sample of the terminal epoxide moiety with the methylene signals 2.98 and 2.79 ppm and the CH resonance at 3.33 ppm (Fig. 2). These resonances are absent from the spectra of the L-tyrosine and L-tryptophan bound polymer. The CH and CH2 from the linker resonate upfield at 1.30 and 1.79 ppm, respectively, which are common to both spectra, and allow the evaluation of the ligand density in the bound sample. Comparing the integration at 1.79 ppm due to two CH2 units from the linker, with the resonances at 6.67.6 ppm for L-tyrosine and 8.0-7.0 ppm for L-tryptophan (Fig.3), we obtained ligand density of 24 % and 6.2 % for L-tyrosine and L-tryptophan, respectively. In order to establish a suitable purification methodology, a first screen was performed to select the best support that allows to obtain highly pure sc pPH600.

11

3.1- Sc pPH600 separation on L-arginine Sepharose column A first screening was performed to evaluate the retention behaviour of pPH600 on the commercial L-arginine support and different mild salt conditions were tested in a range between 100-500 mM NaCl, while temperature was maintained at 4ºC. The sc pPH600 retention was observed when NaCl concentrations were kept below 110 mM in 10 mM TrisHCl (pH 8.0), being the sc pPH600 more strongly retained on the L-arginine support than oc that elutes immediately after injection of pPH600 mixture. The total elution of sc isoform was observed at 500 mM NaCl. The chromatographic profile obtained when loading a native pPH600 (oc + sc) sample onto the L-arginine support is presented in Fig. 4. The respective chromatogram shows two resolved peaks eluting at 110 mM NaCl (peak 1) and 500 mM NaCl (peak 2) (Fig. 4A). To check the identity of the plasmid isoforms eluting within each peak, the two fractions recovered were analysed by agarose gel electrophoresis (Fig. 4B). The analysis revealed that the first peak corresponds to oc pPH600 (Fig. 4B, lane 1), whereas the second peak corresponds to the sc pPH600 (Fig. 4B, lane 2). The temperature effect (10, 15 and 25ºC) in the retention of pPH600 on L-arginine support was also evaluated. The elution profiles obtained showed that when the temperature increases, the selectivity and specificity were affected, since retention increases for all the isoforms. This fact is in accordance to what was previously described in which an increase in temperature favours the involvement of hydrophobic interactions even at low salt conditions (< 300 mM NaCl) [11]. As previously reported, the predominant interactions at low temperatures are electrostatic and occur mainly with phosphate groups of pDNA [11]. Thus, besides the involvement of electrostatic interactions, the pDNA retention increased at higher temperature, suggesting the establishment of other multiple non-covalent interactions (hydrophobic, hydrogen bonds and van der walls) [18] with the L-arginine support, which enables the pPH600 isoforms separation at lower temperature and ionic strength.

3.2- Sc pPH600 separation on L-tryptophan Sepharose column L-tryptophan was chosen to be immobilized on Sepharose CL-6B since it is an aromatic and hydrophobic amino acid, and can develop multiple non-covalent interactions under different elution strategies depending on the target biomolecule. The purification parameters of pPH600 with L-tryptophan support were explored based on its hydrophobicity, similarly to what has been described for L-histidine support [13]. Hence, the binding of the native pPH600 isoforms was tested at different high salt concentrations in a range between 2.25 to 12

3 M (NH4)2SO4. The column was equilibrated with 2.65 M (NH4)2SO4 in 100 mM of HEPES acid (pH 7.4), and after oc+sc pPH600 injection, the oc isoform was immediately eluted and the sc remained bound to the support. This behaviour is consistent with the fact that at high salt concentrations pDNA is tightly interwound [13]. This is due to cations NH4+ that reduce electrostatic repulsion between phosphate groups in the DNA backbone. In fact, the supercoiling increases with salt concentration exposing the hydrophobic bases for the interaction with L-tryptophan [13]. Thus, the bases exposure occurs in the sc isoform in a higher extent than in oc isoform [13]. The elution of sc isoform was obtained at low salt concentrations, 0 M (NH4)2SO4 in 100 mM HEPES acid (pH 7.4), when supercoiling state returns to its original state, more compact than oc isoform but with less exposition of hydrophobic bases. Fig. 5A shows a chromatographic profile obtained from the injection of pPH600 containing both isoforms (sc and oc). An agarose gel electrophoresis analysis of the fractions eluting from the column (Fig. 5B) proved that the first peak of unbound material corresponds to the oc isoform (Fig. 5B, lane 1), whereas the second peak was attributed to the sc isoform (Fig. 5B, lane 2). However, the analysis revealed that the separation of sc from oc isoform was not total, eluting a small quantity of sc isoform at 2.65 M (NH4)2SO4 in 100 mM HEPES acid (pH 7.4) at 10ºC. In order to optimize the recovery of sc pDNA, the temperature was changed to 15ºC in order to increase the hydrophobic interactions. However, the separation was not possible since the retention time of sc isoform decreased. Moreover, this behaviour evidences that not only hydrophobic interactions are involved, but also hydrogen bonds, van der Waals and π-π stacking, as described by Sousa and collaborators [19]. Several experiments were also performed at 4ºC; however, the efficient separation of pPH600 isoforms was not achieved due to lack of selectivity and specificity, which reveals the presence of hydrophobic and π-π stacking interactions.

3.3- Sc pPH600 and pVAX1-LacZ separation on L-tyrosine Sepharose column The support L-tyrosine Sepharose, as well as the previous was prepared by covalent immobilization of 1,4-butanediol diglycidyl ether directly onto Sepharose CL-6B. Although less hydrophobic, this amino acid is also aromatic like L-tryptophan. A first screen was also performed to select the best salt conditions for binding/elution, using concentrations of (NH4)2SO4 between 2 and 2.5 M. Thereafter, total (oc + sc) pPH600 retention were observed when salt concentrations were kept higher than 2.5 M (NH4)2SO4. Otherwise, total elution 13

was observed when using 100 mM HEPES acid (pH 7.4) or 10 mM Tris-HCl (pH 8.0). A total separation of sc from oc isoform was achieved when using, in the binding step, a salt concentration of 2.25 M of (NH4)2SO4 in 100 mM HEPES (pH 7.4) at 10ºC for pPH600 and pVAX1-LacZ, respectively. The elution of sc isoform was then obtained by decreasing the ionic strength to 0 M (NH4)2SO4 in 100 mM HEPES acid (pH 7.4), when the effect of salt is negligible. Figure 6 shows a chromatographic profile obtained after injection of pPH600 containing both isoforms (sc and oc) onto the L-tyrosine support. An agarose gel electrophoresis analysis of the fractions eluting from the column (Fig. 6B) proved that the first peak of unbound material corresponds to the oc isoform (Fig. 6B, lane 1), whereas the second peak was attributed to the sc isoform (Fig. 6B, lane 2). In this case, as depicted by agarose gel electrophoresis (Fig. 6B) the total separation of plasmid isoforms was achieved. Previously chromatography study using plasmids pVAX1-LacZ and the p53-encoding onto the L-tyrosine support showed separation of oc from sc isoform using a stepwise gradient of 2.3 M of (NH4)2SO4 to elute oc and then removing the salt to recover sc [20]. Comparing the results obtained from the purification of the pVAX1-LacZ, p53-encoding and pPH600, using similar conditions, it is possible to note that the selectivity and also the retention decrease in the experiment of pVAX1-LacZ. This result suggests that it is possible to purify different plasmids with the L-tyrosine Sepharose, but the gradient conditions should be slightly adjusted. The lower hydrophobicity of L-tyrosine compared with L-tryptophan seems to interfere in the separation of isoforms, making the process more efficient. In addition, it is also suggested that multiple non-covalent interactions occur between L-tyrosine and sc pDNA being responsible for the selectivity achieved. This fact, may be related with hydroxyl group of L-tyrosine that enables the occurrence of hydrogen bonds with guanine and the π-π stacking between aromatic ring and DNA bases as well as the establishment of van der Waals interactions [19]. The temperature effect in pPH600 isoform separation on L-tyrosine was also evaluated. The elution profiles obtained showed that when the temperature increases (15 to 25ºC), the selectivity and specificity were affected, since the sc isoform retention decreased. Thus, the separation of two isoforms was only achieved at 10ºC. On the other hand, some studies were performed by loading increasing amounts of pPH600 sample (25, 50, 75, 100, 150 and 300 µg/mL) in the L-tyrosine support aiming to observe the influence of the loaded sample concentration in the separation efficiency. The experiments revealed that for pPH600 concentrations up to 150 µg/mL, the separation of pPH600 14

isoforms was efficient, revealing high selectivity and specificity, as it is possible to observe in the electrophoresis (Fig. 7A-E). On the other hand, when the pPH600 concentration injected onto the column was 300 µg/mL, it was not possible to observe the separation of pPH600 isoforms because the L-tyrosine support didn’t show enough selectivity (Fig. 7F). However, the results demonstrate that an adjustment in ionic strength may lead to an effective separation of isoforms and, thus, achieving the desired purity. Comparing the three chromatographic strategies used to purify the sc pPH600, it is evident that in the case L-arginine the purification of sc pPH600 was achieved with low ionic strength and temperature, suggesting the presence of electrostatic interactions and hydrogen bonds. For L-tryptophan and L-tyrosine it was proved that by using an (NH4)2SO4 gradient, oc pPH600 elutes, whereas sc pPH600 remains retained on the column. However, pPH600 isoforms separation is more effective with L-tyrosine support than with L-tryptophan support, suggesting that the less hydrophobic nature of L-tyrosine compared with Ltryptophan favours the separation of isoforms. On the other hand, the results obtained can also be explained by the difference in the ligand density since L-tryptophan support presents a lower ligand density than the L-tyrosine support (6.2% and 24% respectively) that may have influence in the lower selectivity observed in the separation of pDNA isoforms when compared with L-tyrosine support (24%). In a recent work, Bai and collaborators [21] analysed the selectivity of three arginine supports with different ligand densities in the separation of pDNA isoforms. In particular, for arginine beads support, the results showed that an increase in the ligand density leads to an increase in the selectivity of the support for pDNA isoforms [21]. For all these reasons, the L-tyrosine support was selected for further experiments with E. coli lysate.

3.4- Dynamic binding capacity For a complete characterization of the L-tyrosine-Sepharose CL-6B support, the DBC was evaluated, once this parameter is a critical factor for the chromatographic performance. For this purpose, breakthrough experiments were performed at 1 mL/min, with 0.01 and 0.10 mg/mL of pDNA concentrations. The results indicate that the capacity of the L-tyrosineSepharose support to bind pDNA was higher for the feed concentration of 0.10 mg/mL, where it was found a capacity of 1.288 mg/mL, at 50% (Table 1). Whilst for an initial pDNA concentration value of 0.01 mg/mL the capacity of L-tyrosine-Sepharose support was 0.532 mg/mL, at 50%. By analyzing the data in Table 1, it appears that at 10 and 50 % of the 15

breakthrough, the DBC increases with increasing pDNA concentration [22], thus, it can be argued that the binding capacity is dependent of the pDNA concentration. The increased capacity at higher concentrations can be attributed to a more compact structure of pDNA in concentrated solutions. In fact, this compact form of sc pDNA have a smaller radius of gyration, covering a smaller area when bounded to support and leading to an increase of capacity [22]. Furthermore, these DBC results found for L-tyrosine-Sepharose support are very good when compared with other values described for purified pDNA, where the commercial L-arginine and L-histidine-Sepharose supports presented a binding capacity of 0.360 and 0.53 mg/mL, at 50%, using 0.05 and 0.125 mg/mL of pDNA solution, respectively [22, 23]. 3.5- Purification of sc pPH600 from clarified E. coli lysate using L-tyrosine support Although L-arginine support is able to separate sc from oc pPH600 under mild conditions, the total recovery of sc pPH600 is not possible since a portion of sc pPH600 elutes in first peak together with oc pPH600. In this context, L-tyrosine support shows the best result in the separation of two isoforms, allowing the total recovery of sc pPH600. Based on these considerations, L-tyrosine support was chosen to purify sc pPH600 directly from clarified E. coli lysate. The influence of impurities presents in the clarified lysate (different nucleic acid species such as native oc pDNA, total E. coli RNA and gDNA, proteins and endotoxins) was studied, by loading the pPH600-containing mixture to the column at different ionic strength and temperature conditions. The column was initially equilibrated at 10ºC with 2.25 M (NH4)2SO4 in 100 mM HEPES acid (pH 7.4). After injection of the clarified lysate sample, the elution of oc pDNA occurred immediately in the first step performed with 2.25 M of (NH4)2SO4 in 100 mM HEPES acid (pH 7.4), whereas the sc pPH600 remained bound, eluting only when the salt concentration was decreased to 1.95 M of (NH4)2SO4. The RNA and gDNA were then eluted with 100 mM HEPES acid (pH 7.4), without salt. As expected, these fractions containing the total denatured gDNA and RNA were retarded in the column, since the single strands present in denatured gDNA enables a strong interaction with the support to take place, in the same way it occurs with RNA. Thus, the remaining portion of the material adsorbed in the column was only eluted after decreasing the ionic strength to 100 mM HEPES acid (pH 7.4). Fig. 8A shows the chromatographic profile obtained after the injection of pPH600 complex clarified lysate onto the L-tyrosine support. An agarose gel electrophoresis analysis of the fractions eluting from the column (Fig. 8B) proved that the first peak of unbound material 16

corresponds to the oc isoform, a small content of gDNA and sc pPH600 (Fig. 8B, lane 1). On the other hand, the second peak only contained the sc pPH600 (Fig. 8B, lane 2) whereas in the third peak it was verified the elution of strongly bound species such as RNA and gDNA (Fig. 8B, lane 3). Finally, as depicted in the agarose gel electrophoresis (Fig. 8B), a small quantity of sc pPH600 also elutes in the third peak, despite its purification is completely achieved in the second peak. Part of the gDNA also elutes in the first peak since it is composed by fragments with different sizes (0.7-20 kbp) and hydrophobicity levels, leading to different interactions with the support [24]. The temperature effect on adsorption of pDNA isoforms and other E. coli nucleic acids on the L-tyrosine support was also evaluated. Thereby, increasing the temperature from 15 to 25ºC it was verified that oc pPH600, RNA and gDNA do not alter the retention pattern. However, the retention of sc pPH600 decreased when the temperature was increased from 4 to 10ºC allowing the separation of two isoforms with selectivity and specificity without affecting the retention time of other nucleic acids present in E. coli lysate. Comparing the results obtained in the L-tyrosine support with other supports commonly used to purify sc pDNA (supports based on hydrophobic interactions (butyl [25], octyl [25] and phenyl [26] or anion exchange interactions (DEAE [27] and Q-Sepharose [28]), it can be concluded that L-tyrosine was efficient in the separation of sc pDNA from bacterial contaminants (RNA, gDNA, endotoxins and proteins) and in the separation of sc pDNA from less active isoforms in a single step. Other supports present lack of selectivity, not allowing an

effective separation of sc pDNA in a single chromatographic step. In addition, the anion exchange chromatography supports display poor selectivity between pDNA and impurities due to the similar characteristics of the biomolecules such as, chemical composition, structure and charge [28]. Recently, it was reported that a CIM-DEAE monolith developed by BIA Separations was able to remove most of the impurities; however no separation of different pDNA isoforms and no separation from gDNA was achieved [29]. On the other hand, the hydrophobic interaction chromatography takes advantage because the several biomolecules (pDNA and impurities) present different hydrophobicity, allowing an efficient removal of impurities however, in some stationary phases the separation of the isoforms was not achieved [26]. For all these reasons, the hydrophobic interaction chromatography is usually integrated in a combined strategy with anion exchange chromatography [29]. Thus, L-tyrosine based affinity chromatography is suitable to purify sc pDNA in a single step

17

taking advantage of multiple non-covalent interactions that allow the possibility to eliminate additional steps.

3.4.1- Plasmid quantification and purity assessment The recovery yield, concentration and quality of the sc pDNA were assessed by a CIMacTM pDNA analytical column, according to the modified analytical method previously developed by Sousa and collaborators [16]. This method allows the elution of RNA species in the flowtrough at 600 mM NaCl in 200 mM Tris-HCl (pH 8.0), while oc and sc isoforms are separated during a 10 min-linear gradient from 600 to 700 mM NaCl in 200 mM Tris-HCl (pH 8.0). After analysis of each peak with CIMacTM pDNA analytical column, the results of concentration, recovery yield and purity are depicted in Table 1. The recovery of sc pPH600 in each peak was determined with relation to the initial amount present in the clarified lysate (Table 1). The purity degree of sc pPH600 was determined from the ratio between the pDNA peak area and the sum of all the peak areas that appear in analytical chromatogram. As shown in Table 1, this purification strategy allowed the recovery of sc pPH600 with 56.28% of yield and 98.23 % of purity. Thus, the sc pPH600 sample recovered can be used in therapeutic applications because it presents a homogeneity higher than 97% of sc isoform, according to the criteria recommended by regulatory agency. Comparing with other supports, we can conclude that the strategy reported in this work allows the recovery of sc pDNA with a high purity level [24, 30]. Moreover, some purification strategies using affinity supports based on amino acids show a recovery yield of only 45% for sc pDNA purification with L-histidine and L-lysine supports [12, 24] and 39% with L-arginine modified monolith [23]. Thus, the purification of sc pDNA with L-tyrosine support enables a suitable strategy to improve the recovery yield of sc pDNA.

3.4.2- Proteins, gDNA, RNA and endotoxins quantification Proteins, gDNA and endotoxins were determined by using the micro-BCA method, real-time PCR and kinetic-QCL Limulus amebocyte lysate assays, respectively. The results presented in Table 2 show a significant reduction of all impurities in the sc pPH600 fraction purified with L-tyrosine support when compared with the initial clarified lysate sample. The analysis of protein content shows that after purification with L-tyrosine support, the sample of sc pPH600 in peak 2 did not present detectable levels of proteins, corresponding 18

to the requirements of regulatory agencies and suggesting that the use of L-tyrosine support allows the elution of proteins together with other impurities. The presence of endotoxins was also evaluated and a significant decrease in endotoxins content was observed, reducing from 2.663 EU/μg sc pDNA in lysate sample to 0.033 in peak 2. Due to their hydrophobic character, endotoxins can bind more strongly to the L-tyrosine support than pDNA, probably eluting in peak 3 (Table 2). The efficiency of L-tyrosine support to eliminate gDNA was also verified because the residual amount of gDNA in the sc pPH600 sample was 0.912 ng/µg sc pDNA, significantly lower when compared with 13.541 ng/µg sc pDNA in lysate sample. Also, it is possible to observe that RNA was totally eliminated in the third peak, as presented in Fig 7B, lane 3. Overall, the content of impurities present in the sc pPH600 sample is in accordance with the requirements of regulatory agencies.

3.5- In vitro transcription and CD spectroscopy for G-quadruplex assessment After purification of sc pPH600, in vitro transcription was performed to check G-quadruplex formation in 604 bp sequence of immunoglobulin switch region. Transcription of G-rich templates such as the S regions, either in vitro or in vivo, causes characteristic large loops called G-loops [3]. G-loops contain a stable, cotranscriptional RNA-DNA hybrid on the Crich template strand and intramolecular G4 DNA interspersed within single-stranded regions on the G-rich strand [3]. In order to evaluate the formation of G-loops in transcribed sequence and consequent formation of RNA/DNA hybrid, CD spectra were recorded before and after transcription. The results for sc pPH600 reveal a positive band at 275 nm and a negative band at 245 nm, suggesting that plasmid adopts B-form (Fig. 9A) [31]. After transcription and treatment with DNase the spectrum is completely different having a positive band at 265 nm and negative bands at 240 nm and 210 nm (Fig. 9B), characteristic of RNA/DNA hybrid as described previously [32]. The addition of 100 mM KCl and heating to 95ºC, followed by ice cooling until room temperature resulted in a CD spectrum with a negative band at 210 nm and a change in the ellipticity (Fig. 9B) [33], suggesting the formation of RNA parallel G-quadruplex [34]. Finally, the transcription of pPH600 and treatment with RNase A results in a CD spectrum with a positive band at 266 nm and a negative band at 220 nm. Another band is observed at 210 nm; however, did not reach positive ellipticity (Fig. 9C). This spectrum is characteristic of parallel G-quadruplex DNA [32]. Thus, the formation of G-loop with a hybrid RNA/DNA and consequent formation of G-quadruplex after transcription of plasmid pPH600 was proved by CD spectroscopy. 19

4- Conclusions We presented the biosynthesis and purification strategies of sc pPH600, which contains a fragment of the murine immunoglobulin S3 switch region that can fold in G-quadruplex. In order to achieve the selective isolation of sc pPH600, we have prepared two affinity matrices by immobilizing L-tyrosine and L-tryptophan on derivatized Sepharose CL-6B with 1,4-butanediol diglycidyl ether. Commercial L-arginine support was also used to compare binding/elution conditions. The retention behaviour of sc on these matrices suggests that L-tyrosine support was ideal to obtain highly pure sc pPH600 from a native sample and E. coli lysate. The separation of sc pPH600 from contaminants was performed using a stepwise gradient of ammonium sulphate in HEPES. The DBC found for L-tyrosine support (1.288 mg/mL) was higher than other supports like commercial L-arginine and L-histidine-Sepharose which presented a maximum binding capacity of 0.36 mg/mL and 0.53 mg/mL, respectively. Accordingly, L-tyrosine support allowed the recovery of desirable sc pPH600 with good yield and high purity which was used for in vitro transcription studies for detecting Gquadruplex formation by CD. The CD spectrum of the transcript showed parallel Gquadruplex folding topology. Therefore, our supports demonstrated promisor results, especially L-tyrosine support in which the yield of sc pDNA was higher than previously described for other amino acids such as L-histidine and L-lysine. The chromatographic process has also advantages when compared with more conventional resins allowing the isolation of sc pDNA in a single step. Overall, these studies support the use of aromatic amino acids, especially L-tyrosine, as ligands in affinity chromatography to purify different plasmids.

20

References 1.

M.L. Bochman, K. Paeschke, V.A. Zakian, DNA secondary structures: stability and function of G-quadruplex structures, Nature reviews. Genetics, 13 (2012) 770-780.

2.

L. Yuan, T. Tian, Y. Chen, S. Yan, X. Xing, Z. Zhang, Q. Zhai, L. Xu, S. Wang, X. Weng, B. Yuan, Y. Feng, X. Zhou, Existence of G-quadruplex structures in promoter region of oncogenes confirmed by G-quadruplex DNA cross-linking strategy, Scientific reports, 3 (2013) 1811.

3.

K.J. Neaves, J.L. Huppert, R.M. Henderson, J.M. Edwardson, Direct visualization of G-quadruplexes in DNA using atomic force microscopy, Nucleic Acids Res, 37 (2009) 6269-6275.

4.

M.L. Duquette, P. Handa, J.A. Vincent, A.F. Taylor, N. Maizels, Intracellular transcription of G-rich DNAs induces formation of G-loops, novel structures containing G4 DNA, Genes Dev, 18 (2004) 1618-1629.

5.

A. Sousa, F. Sousa, J.A. Queiroz, Differential interactions of plasmid DNA, RNA and genomic DNA with amino acid-based affinity matrices, J Sep Sci, 33 (2010) 2610-2618.

6.

J. Stadler, R. Lemmens, T. Nyhammar, Plasmid DNA purification, J. Gene. Med., 6 (2004) S54-S66.

7.

M.M. Diogo, J.A. Queiroz, D.M.F. Prazeres, Chromatography of plasmid DNA, J. Chromatogr. A, 1069 (2005) 3-22.

8.

A. Eon-Duval, G. Burke, Purification of pharmaceutical-grade plasmid DNA by anion-exchange chromatography in an RNase-free process, J Chromatogr B Analyt Technol Biomed Life Sci, 804 (2004) 327-335.

21

9.

G.N.M. Ferreira, G.A. Monteiro, D.M.F. Prazeres, J.M.S. Cabral, Downstream processing of plasmid DNA for gene therapy and DNA vaccine applications, Trends Biotechnol., 18 (2000) 380-388.

10.

F. Sousa, D.M. Prazeres, J.A. Queiroz, Affinity chromatography approaches to overcome the challenges of purifying plasmid DNA, Trends Biotechnol., 26 (2008) 518-525.

11.

F. Sousa, T. Matos, D.M. Prazeres, J.A. Queiroz, Specific recognition of supercoiled plasmid DNA in arginine affinity chromatography, Analytical biochemistry, 374 (2008) 432-434.

12.

A. Sousa, F. Sousa, J.A. Queiroz, Impact of lysine-affinity chromatography on supercoiled plasmid DNA purification, J Chromatogr B Analyt Technol Biomed Life Sci, 879 (2011) 3507-3515.

13.

F. Sousa, C.T. Tomaz, D.M. Prazeres, J.A. Queiroz, Separation of supercoiled and open circular plasmid DNA isoforms by chromatography with a histidine-agarose support, Analytical biochemistry, 343 (2005) 183-185.

14.

K.A. Wilson, J.L. Kellie, S.D. Wetmore, DNA–protein π-interactions in nature: abundance, structure, composition and strength of contacts between aromatic amino acids and DNA nucleobases or deoxyribose sugar, Nucleic Acids Res, 42 (2014) 6726-6741.

15.

M.M. Diogo, J.A. Queiroz, G.A. Monteiro, S.A.M. Martins, G.N.M. Ferreira, D.M.F. Prazeres, Purification of a cystic fibrosis plasmid vector for gene therapy using hydrophobic interaction chromatography, Biotechnology and Bioengineering, 68 (2000) 576-583.

16.

A. Sousa, A.M. Almeida, U. Cernigoj, F. Sousa, J.A. Queiroz, Histamine monolith versatility to purify supercoiled plasmid deoxyribonucleic acid from Escherichia coli lysate, J. Chromatogr. A, 1355 (2014) 125-133.

22

17.

J.R. Cooke, E.A. McKie, J.M. Ward, E. Keshavarz-Moore, Impact of intrinsic DNA structure on processing of plasmids for gene therapy and DNA vaccines, J Biotechnol, 114 (2004) 239-254.

18.

F. Sousa, D.M. Prazeres, J.A. Queiroz, Binding and elution strategy for improved performance of arginine affinity chromatography in supercoiled plasmid DNA purification, Biomed Chromatogr, 23 (2009) 160-165.

19.

F. Sousa, C. Cruz, J.A. Queiroz, Amino acids-nucleotides biomolecular recognition: from biological occurrence to affinity chromatography, Journal of molecular recognition : JMR, 23 (2010) 505-518.

20.

S. Ferreira, J. Carvalho, J.F. Valente, M.C. Corvo, E.J. Cabrita, F. Sousa, J.A. Queiroz, C. Cruz, Affinity analysis and application of dipeptides derived from ltyrosine in plasmid purification, J Chromatogr B Analyt Technol Biomed Life Sci, 1006 (2015) 47-58.

21.

J.S. Bai, S. Bai, Q.H. Shi, Y. Sun, Purification of supercoiled plasmid DNA from clarified bacterial lysate by arginine-affinity chromatography: effects of spacer arms and ligand density, J Sep Sci, 37 (2014) 1386-1395.

22.

F. Sousa, D.M. Prazeres, J.A. Queiroz, Dynamic binding capacity of plasmid DNA in histidine-agarose chromatography, Biomed Chromatogr, 21 (2007) 993-998.

23.

A. Soares, J.A. Queiroz, F. Sousa, A. Sousa, Purification of human papillomavirus 16 E6/E7 plasmid deoxyribonucleic acid-based vaccine using an arginine modified monolithic support, J. Chromatogr. A, 1320 (2013) 72-79.

24.

F. Sousa, S. Freitas, A.R. Azzoni, D.M. Prazeres, J. Queiroz, Selective purification of supercoiled plasmid DNA from clarified cell lysates with a single histidineagarose chromatography step, Biotechnology and applied biochemistry, 45 (2006) 131-140.

23

25.

T. Kitamura, S. Nakatani, Separating plasmids from contaminants using hydrophobic or hydrophobic and ion exchange chromatography, US Patent No. 6441160 B2, (27 August 2002).

26.

S.S. Freitas, J.A.L. Santos, D.M.F. Prazeres, Plasmid purification by hydrophobic interaction chromatography using sodium citrate in the mobile phase, Sep. Purif. Technol., 65 (2009) 95-104.

27.

Y. Onishi, Y. Azuma, H. Kizaki, An assay method for DNA topoisomerase activity based on separation of relaxed DNA from supercoiled DNA using high-performance liquid chromatography, Analytical biochemistry, 210 (1993) 63-68.

28.

D.M. Prazeres, T. Schluep, C. Cooney, Preparative purification of supercoiled plasmid DNA using anion-exchange chromatography, J. Chromatogr. A, 806 (1998) 31-45.

29.

F. Smrekar, A. Podgornik, M. Ciringer, S. Kontrec, P. Raspor, A. Strancar, M. Peterka, Preparation of pharmaceutical-grade plasmid DNA using methacrylate monolithic columns, Vaccine, 28 (2010) 2039-2045.

30.

F. Sousa, D.M. Prazeres, J.A. Queiroz, Improvement of transfection efficiency by using supercoiled plasmid DNA purified with arginine affinity chromatography, J. Gene. Med., 11 (2009) 79-88.

31.

F. Sousa, D.M.F. Prazeres, J.A. Queiroz, Circular dichroism investigation of the effect of plasmid DNA structure on retention in histidine chromatography, Arch. Biochem. Biophys., 467 (2007) 154-162.

32.

M. Vorlickova, I. Kejnovska, K. Bednarova, D. Renciuk, J. Kypr, Circular dichroism spectroscopy of DNA: from duplexes to quadruplexes, Chirality, 24 (2012) 691-698.

24

33.

G. Mirihana Arachchilage, M.J. Morris, S. Basu, A library screening approach identifies naturally occurring RNA sequences for a G-quadruplex binding ligand, Chem. Commun., 50 (2014) 1250-1252.

34.

J. Vondruskova, J. Kypr, I. Kejnovska, M. Fialova, M. Vorlickova, Role of loops in the guanine quadruplex formation by DNA/RNA hybrid analogs of G4T4G4, International journal of biological macromolecules, 43 (2008) 463-467.

25

Acknowledgements This work was performed under a financial support of FCT – “Fundação para a Ciência e a Tecnologia” (project FCOMP-01-0124-FEDER-041068 - EXPL/QEQ-MED/1068/2013). P. Pereira

and

C.

Cruz

acknowledge

the

fellowships

SFRH/BD/81914/2011

and

SFRH/BPD/100015/2014, respectively from FCT. The authors thank Dr. N. Maizels for providing the pPH600 plasmid.

26

Tables Table 1 – The dynamic binding capacity of L-tyrosine-Sepharose column. The breakthrough experiments were performed with 0.010 and 0.10 mg/mL of pPH600 plasmid solution prepurified with Qiagen kit, at flow-rate of 1 mL/min. DBC (mg pDNA/mL gel) pDNA concentration (mg/mL)

10%

50%

0.010

0.113

0.532

0.10

0.462

1.288

Table 2 - Quantitative analysis of the sc pDNA recovery yield and purity in each peak recovered from clarified lysate using L-tyrosine support. Sample

sc pDNA (μg/mL)

Volume (μL)

sc pDNA (μg)

sc pDNA recovery (%)

Purity (%)

Lysate

552.23

50

27.61

-

-

Peak 1

0.67

450

0.24

0.87

-

Peak 2

25.90

600

15.54

56.28

98.23

Peak 3

6.65

500

2.99

10.84

-

Table 3 - Proteins, endotoxins and gDNA assessment from clarified lysate sample and sc pDNA samples recovered from the L-tyrosine support.

Proteins

Endotoxins

gDNA

Lysate

(μg/mL) 143.467

(EU/μg sc pDNA) 2.663

(ng/μg sc pDNA) 13.541

Peak 2 (sc pPH600)

undetectable

0.033

0.912

undetectable

< 0.1

<2

Sample

Regulatory agencies specifications [6, 7]

27

Figures Figure 1 a)

b)

c)

28

Figure 2 a)

b)

29

Figure 3 i)

a)

b)

ii)

30

Figure 4

Figure 5

31

Figure 6

32

Figure 7

33

Figure 8

Figure 9

34

Figure legends Figure 1- Chemical structures of the supports a) L-arginine Sepharose b) L-tryptophan Sepharose and c) L-tyrosine Sepharose. Figure 2- 1H 1H-NOESY HR-MAS spectrum of agarose with linker bound to a) L-tyrosine and b) L-tryptophan in DMF-d7. Figure 3- CPMG 1H HR-MAS spectra and inset with respective single pulse 1H HR-MAS of agarose with activated linker i) (a) and agarose with linker bound to L-tyrosine (b) and ii) (a) and agarose with linker bound to L-tryptophan (b) in DMF-d7. Figure 4 – (A) Chromatographic profile showing the purification of sc pPH600 with Larginine support. Elution was performed at 1 mL/min by stepwise gradient increasing the NaCl concentration in the eluent, as represented by the dashed line. (B) Agarose gel electrophoresis for the analysis of collected fractions from L-arginine support. Lane M – molecular weight marker; Lane S – pDNA native sample (oc+sc) injected onto column; Lane 1 – pDNA recovered from peak 1; Lane 2 – pDNA recovered from peak 2. Figure 5 - (A) Chromatographic profile showing the purification of sc pPH600 with Ltryptophan support. Elution was performed at 1 mL/min by stepwise gradient decreasing the (NH4)2SO4 concentration in the eluent, as represented by the dashed line. (B) Agarose gel electrophoresis for the analysis of collected fractions from L-tryptophan support. Lane M – molecular weight marker; Lane S – pDNA native sample (oc+sc) injected onto column; Lane 1 – pDNA recovered from peak 1; Lane 2 – pDNA recovered from peak 2. Figure 6 - (A) Chromatographic profile showing the purification of sc pPH600 with Ltyrosine support. Elution was performed at 1 mL/min by stepwise gradient decreasing the (NH4)2SO4 concentration in the eluent, as represented by the dashed line. (B) Agarose gel electrophoresis for the analysis of collected fractions from L-tyrosine support. Lane 1 – pDNA recovered from peak 1; Lane 2 – pDNA recovered from peak 2. Figure 7 - Agarose gel electrophoresis for the analysis of selectivity in L-tyrosine support depending on pPH600 concentration injected onto column. (A) 25 µg/mL; (B) 50 µg/mL; (C) 75 µg/mL; (D) 100 µg/mL; (E) 150 µg/mL; (F) 300 µg/mL. All agarose gel electrophoresis display Lane S – pDNA native sample (oc+sc) injected onto column; Lane 1 – pDNA recovered from first step and Lane 2 – pDNA recovered from second step.

35

Figure 8 - (A) Chromatographic profile showing the purification of sc pPH600 from clarified E. coli lysate with L-tyrosine support. Elution was performed at 1 mL/min by stepwise gradient decreasing the (NH4)2SO4 concentration in the eluent, as represented by the dashed line. (B) Agarose gel electrophoresis for the analysis of collected fractions from L-tyrosine support. Lane F – Lysate sample; Lane 1 – pDNA recovered from peak 1; Lane 2 – pDNA recovered from peak 2. Figure 9- (A) CD spectrum of sc pPH600 (green line) and ln pPH600 (blue line). (B) CD spectrum of RNA/DNA hybrid treated with DNase without KCl (green line) and with 100 mM KCl and increase in temperature to 95ºC (blue line). (C) CD spectrum of RNA/DNA hybrid treated with RNase A without KCl (green line) and with 100 mM KCl and increase in temperature to 95ºC (blue line).

36

Highlights  Screening of three affinity supports to purify pPH600.  L-tryptophan and L-tyrosine as a new affinity supports to purify pDNA.  L-tyrosine support allows the purification of sc PH600 from E. coli lysate.  G-quadruplex formation upon transcription of sc pPH600.  The transcript adopted parallel G-quadruplex topology.

37