Isolation of aptamers against recombinant PfLDH and LDH peptide
Aptamer synthesis was performed according to Rotherham et al. [
25]. A library with single-stranded DNA (ssDNA) sequences (90 bases in length, with a 49 nucleotide length of randomized sequence flanked by constant regions for primer annealing) was sourced from Integrated DNA Technologies (IDT; USA). The library had the general sequence of 5′-GCCTGTTGTGAGCCTCCTAAC(N
49)CATGCTTATTCTTGTCTCCC-3′.
Identification of oligonucleotides binding to
rPfLDH and LDHp took place via parallel SELEX processes [
17]. Selection using nitrocellulose membrane filtration at the initial phase of SELEX was adapted from previous protocols [
26,
27].
Prior to use, nitrocellulose membranes (pore size of 0.45 µm, Merck Millipore, USA) were prepared by immersion in 0.5 M KOH for 20 min, rinsing with Milli-Q H2O, further incubation in 0.1 M Tris, pH 7.4 for 45 min and finally rinsing with HMCKN buffer (2 mM HEPES, 0.2 mM MgCl2, 0.2 mM CaCl2, 0.2 mM KCl and 15 mM NaCl, pH 7.4).
The ssDNA library was prepared in HMCKN to a final concentration of 1.59 μM. This was heat denatured at 95 °C for 10 min, cooled to − 20 °C for 5 min and equilibrated at room temperature for 5 min before being passed through prepared nitrocellulose membrane to remove non-specifically binding sequences (negative selection). The eluent was incubated with solutions of 1.59 μM target protein/peptide prepared in HMCKN; candidate aptamers were allowed to bind to the target at room temperature for 1 h under mild agitation. Following incubation, target-library mixtures were then passed through fresh nitrocellulose membranes (positive selection), which were then rinsed with HMCKN buffer to remove all unbound sequences. It was assumed that all protein/target-aptamer complexes were retained on the nitrocellulose filter [
27].
Retained ssDNA-target complexes were eluted into 100 µl elution buffer (7 M urea, 100 mM citrate buffer and 3 mM EDTA, pH 8.0), by heating to 100 °C for 5 min. Eluted ssDNA was precipitated via phenol–chloroform extraction. 600 µl of phenol: chloroform:isoamyl alcohol mixture (25:24:1, saturated with 10 mM Tris, pH 8.0, 1 mM EDTA) was added to the elution buffer. This was mixed, incubated under agitation at room temperature for 30 min and centrifuged for 5 min at 7400×g. The aqueous phase was removed and set aside. To maximize ssDNA collection, an additional volume of 100 µl of sterile Milli-Q water was added to the organic phase; the suspension thoroughly remixed and centrifuged for 5 min at 7400×g. The aqueous phase was collected and combined with the previous aqueous phase. The combined aqueous phases were re-extracted using the same protocol described above.
The extracted ssDNA was precipitated in a manner similar to that described elsewhere [
28] through the addition of 30 µl of 3 M sodium acetate buffer, pH 5.2, 3.3 µl of glycogen (20 g/l) and 1 ml of absolute ethanol. This was incubated for 16 h at − 80 °C and thereafter centrifuged at 4 °C for 30 min at 7400×
g. The supernatant was carefully decanted and 1 ml of 80% (v/v) ethanol used to resuspend the white precipitate formed during centrifugation. The resultant mixture was centrifuged for a further 5 min at 7400×
g at 4 °C. The ethanol was decanted and the pellet allowed to air dry at room temperature. The ssDNA pellet was resuspended in 30 µl sterile Milli-Q water.
The concentration and purity of ssDNA obtained during SELEX was quantified using a NanoDrop 2000 Spectrophotometer (ThermoScientific, USA). The concentration of extracted ssDNA was used to calculate the total mass of ssDNA binding to the target during the selection rounds (described as “ssDNA out”). Using the mass of ssDNA initially used in the selection (described as “ssDNA in”), the yield of positively binding ssDNA was calculated using Eq.
1:
$${\text{Recovery }}\left( \% \right) = \frac{{{\text{ssDNA out}} \left( {\text{ng}} \right)}}{{{\text{ssDNA}} {\text{in}} \left( {\text{ng}} \right)}} \times 100$$
(1)
Amplification of ssDNA was performed by PCR, producing amplified double-stranded DNA (dsDNA). PCR reaction mix (GoTaq
® Flexi DNA Polymerase kit, Promega, USA) prepared as per manufacturer’s instruction, and PCR generally proceeded using: 0.2 mM dNTPs (Fermentas, Thermo Scientific, USA); 3.5 mM MgCl
2; 0.5 µM forward primer (5′-GCCTGTTGTGAGCCTCCTAAC-3′) (IDT, USA); 0.5 µM reverse primer (5′-GGGAGACAAGAATAAGCATG-3′) (IDT, USA); and, 10 µg/ml BSA (New England Biolabs, UK). The reverse primer was procured modified with an additional phosphate group at the 5′-end for lambda exonuclease digestion [
27]. The temperature profile used for PCR was as follows: 95 °C for 3 min (initiation step); 4–20 cycles at 95 °C for 1 min (denaturation), 59 °C for 1 min (annealing) and 72 °C for 1.5 min (elongation); and, 72 °C for 8 min (final elongation step) on a MJ Mini Personal Thermo Cycler (Bio-RAD, USA). The MgCl
2 concentration was decreased to 1.5 mM after the third round of SELEX to reduce the mutation rate.
The number of PCR cycles required optimization during every amplification stage of each round in SELEX to prevent over-amplification of the dsDNA and minimize the inclusion of amplification artefacts. This was achieved by performing a 200 µl pilot PCR run in each round, in which 20 µl reaction tubes were taken out after every second cycle following four rounds of PCR, e.g. 0, 4, 6, 8, 10, 12, 14, 16, 18, 20 cycles, with the blank control undergoing the maximum number of PCR cycles. Following identification of the optimum number of cycles, full-scale PCR with a reaction volume between 1.0 and 5.0 ml was performed.
PCR products were purified using the Nucleospin® Gel and PCR clean-up kit, following the manufacturer’s instructions (Macherey–Nagel GmbH & Co. KG, Germany). Purified dsDNA was eluted in 50 µl sterile Milli-Q water. The quality of dsDNA was determined by PAGE using 8% (w/v) polyacrylamide gels electrophoresed at ≤ 120 V for 20–30 min in TBE buffer (45 mM Tris base, 45 mM boric acid, 1.3 mM EDTA, pH 8.0). Subsequently, gels were stained with 2.5 µM ethidium bromide solution or GelRed (Biotium, USA) and visualized under UV transillumination with a ChemiDoc XRS+ Molecular Imaging System (BioRAD, USA); the concentration of DNA was separately determined spectrophotometrically. To remove amplification artefacts occurring after Round 3, amplified DNA was purified by gel excision by initially electrophoresing the dsDNA on a 2.5% agarose gel at 80 V for 1.5–2 h in TBE buffer.
Following sufficient amplification of dsDNA (≥ 4.5 µg dsDNA), dsDNA was converted to ssDNA by lambda exonuclease digestion (New England Biolabs, UK), carried out for 4 h at 37 °C with a rate of ~ 1 U exonuclease per microgram dsDNA [
28].
The ssDNA was purified using the Nucleospin® Gel and PCR clean-up kit (Macherey–Nagel GmbH & Co. KG, Germany) as per manufacturer’s instruction. The resulting ssDNA was used for selection during the following round of SELEX. Eight rounds of selection were performed during SELEX.
The dsDNA from the final three rounds of SELEX was pooled, digested with lambda exonuclease to produce ssDNA and a final selection cycle was performed. Thereafter, the amplified dsDNA pool was ligated into the pGEM-T Easy vector (Promega, USA) and transformed into competent Escherichia coli JM109 cells (Rhodes University), according to the manufacturer’s instructions. Blue/white screening was conducted in which the cells containing ligated insert DNA (white colonies) were selected.
Sixteen (16) white colonies (containing oligonucleotide fragments selected against rPfLDH) and 18 white colonies (containing oligonucleotide fragments selected against LDHp) were re-streaked onto a second set of Luria agar plates, and were hence selected for fragment length screening via PCR amplification. PCR-amplification took place using the pUC/M13 universal primers (IDT, USA) flanking the insert region of the pGEM-T Easy vector. Colonies containing the correct insert size—8 out of 16 for rPfLDH and 8 out of 18 for LDHp—were thereafter resuspended in sterile Milli-Q water, heated to 95 °C for 10 min and PCR amplified using the KAPATaq kit (Kapa Biosystems, South Africa). The PCR reaction mix was prepared according to the manufacturer’s instructions and contained 0.1 mM dNTPs (Fermentas, Thermo Scientific, USA), 1.0 mM MgCl2, 0.2 μM 5′-biotinylated forward primer (5′-GCCTGTTGTGAGCCTCCTAAC-3′) (IDT, USA), 0.2 µM 5′-phosphorylated reverse primer (5′-GGGAGACAAGAATAAGCATG-3′) (IDT, USA) and 1 µl of resuspended cell debris. 30 cycles of PCR were performed to amplify the DNA, using the same temperature profile details for aptamer amplification. Amplified dsDNA was exonuclease-digested to ssDNA as previously described.
Biotinylated ssDNA sequences obtained in the above manner were used in a preliminary binding assay using ELONA, as detailed in the proceeding subsection. Colonies containing sequences that showed positive binding at the preliminary phase—4 out of 8 for r
PfLDH and 5 out of 8 for LDHp—were sequenced (Inqaba Biotec, South Africa) using the pUC/M13 universal forward and reverse primers. These sequences were selected for commercial synthesis (IDT, USA): rLDH 1, 4, 7 and 15; LDHp 1, 3, 11, 14 and 18 and pL1 [
20,
22]; modifications for commercially-synthesized aptamers were 5′-biotin (for ELONA), and 5′-FITC (for confocal microscopy).
Confocal microscopy
Plasmodium falciparum parasites (3D7 strain) were cultured in RPMI 1640 medium supplemented with 25 mM HEPES, 22 mM glucose, 0.65 mM hypoxanthine, 0.05 mg/ml gentamicin, 0.5% (w/v) Albumax II and 3% (v/v) human red blood cells. Cultures were maintained at 37 °C in sealed culture flasks suffused with a 5% CO2, 5% O2, 90% N2 gas mixture. When the culture contained predominantly mature stage parasites (trophozoites and schizonts) as judged by light microscopy of Giemsa-stained blood smears, the red blood cells were pelleted, washed and resuspended in PBS. Round glass coverslips (12 mm diam.) were coated for 15 min with 1 mg/ml poly-l-lysine at room temperature. The glass coverslips were rinsed with 1 ml of 1× PBS, pH 7.4. P. falciparum-infected red blood cells, suspended in PBS were allowed to settle on the poly-l-lysine coated coverslips for 1 h. Unbound red blood cells were gently washed away using PBS. Bound red blood cells were lysed with 0.05% saponin for 1 min and rinsed with PBS to remove haemoglobin and the remaining cell debris from the immobilized P. falciparum parasite bodies. Parasite bodies were fixed to the glass coverslips using 1 min incubation with ice-cold methanol. Unfixed parasite bodies were removed with a PBS wash. The coverslips were then blocked with 100 mg/ml HSA in PBS for 20 min.
As a positive control, samples were also incubated for 45 min with IgY generated to the species-specific
P. falciparum epitope (described by Hurdayal et al. [
12]), followed by three washes with PBS, pH 7.4. To elicit a fluorescent response, the antibody control included a 45 min incubation with fluorescein-tagged donkey anti-chicken IgG (Biotium, Inc., USA) as secondary antibody. The fixed parasites were incubated in the dark with 200 nM heat-activated 5′-modified FITC-tagged aptamer in HMCKN buffer (as previously described) for 45 min. Coverslips were washed three times with PBS, pH 7.4. Fixed parasite bodies were incubated in 1 µg/ml DAPI in PBS for 1 min. Coverslips were briefly dipped in Milli-Q H
2O, dried, mounted with Fluoroshield
® (ImmunoBioScience Corp., USA) and allowed to dry in the dark overnight.
Cells were imaged using a Zeiss LSM780 laser scanning confocal microscope (Carl Zeiss Microscopy GmbH, Jena, Germany) using the x63 objective. All images were acquired using the same exposure and detector settings for each spectral channel. The Zen 2011 Blue software was used to acquire images from the Zeiss LSM780 microscope and to perform image overlays. ImageJ 1.50i software was used for image analysis whereby areas of interest were highlighted and the integrated density value (IntDen) and area measured. GraphPad Prism 5 was used to plot the mean fluorescent intensity (IntDen/Area) values for 12 parasites of interest across three micrograph frames.
Statistics
All measurements were performed in, at minimum, triplicate. Presented results are the means of measurements, while all reported error bars and uncertainties represent one standard deviation from the mean. Significant difference was identified using the Kruskal–Wallis H test and datasets significantly different from their counterparts identified using Dunn’s multiple comparison test statistical significance determined using a significance level, α, set to 0.05.
The apparent dissociation constants (K
D) of aptamer-target interactions were calculated using the kinetic information obtained from ELONA analysis. ELONA assay responses were fitted via nonlinear regression (Least-Squares minimization) to a variant of a previously-described Langmuir equivalent binding isotherm equilibrium formula [
32], using Statistica
®. The formula is represented in Eq.
2:
$${\text{Assay response}}\; \left( {\Delta {\text{OD}}_{{450\;{\text{nm}}}} } \right) = \left( {\frac{{\varGamma_{\text{max} } \times [{\text{aptamer}}]}}{{{\text{K}}_{\text{D}} + \left[ {\text{aptamer}} \right]}}} \right)$$
(2)
where [aptamer] is the concentration of the aptamer used during ELONA (M), ΔOD
450 nm is the change in the ELONA absorbance at a given [aptamer], relative to the assay response when [aptamer] = 0 M. These were used to calculate the K
D i.e. the apparent dissociation constant of the aptamer-target complex (M), and Γ
max, the maximal assay response for the aptamer-target complex.
Averages and standard errors of KD and Γmax are presented. In addition to presenting these values, a Wald test of the parameters was included to calculate the significance of the KD and Γmax nonlinear regression coefficients: p values less than 0.05 indicate values that the model assessed to be integral to the dependence of ΔOD450 nm on [aptamer] for a given aptamer-target complex.