Development of K562 cell clones expressing β-globin mRNA carrying the β⁰39 thalassaemia mutation for the screening of correctors of stop-codon mutations

Abstract

Nonsense mutations, giving rise to UAA, UGA and UAG stop codons within the coding region of mRNAs, promote premature translational termination and are the leading cause of approx. 30 % of inherited diseases, including cystic fibrosis, Duchenne muscular dystrophy and thalassaemia. For instance, in β⁰39-thalassaemia the CAG (glutamine) codon is mutated to the UAG stop codon, leading to premature translation termination and to mRNA destabilization through the well-described NMD (nonsense-mediated mRNA decay). In order to develop an approach facilitating translation and, therefore, protection from NMD, aminoglycoside antibiotics have been tested on mRNAs carrying premature stop codons. These drugs decrease the accuracy in the codon–anticodon base-pairing, inducing a ribosomal read-through of the premature termination codons. Interestingly, recent papers have described drugs designed and produced for suppressing premature translational termination, inducing a ribosomal read-through of premature but not normal termination codons. These findings have introduced new hopes for the development of a pharmacological approach to the therapy of β⁰39-thalassaemia. In this context, we started the development of a cellular model of the β⁰39-thalassaemia mutation that could be used for the screening of a high number of aminoglycosides and analogous molecules. To this aim, we produced a lentiviral construct containing the β⁰39-thalassaemia globin gene under a minimal LCR (locus control region) control and used this construct for the transduction of K562 cells, subsequently subcloned, with the purpose to obtain several K562 clones with different integration copies of the construct. These clones were then treated with Geneticin (also known as G418) and other aminoglycosides and the production of β-globin was analysed by FACS analysis. The results obtained suggest that this experimental system is suitable for the characterization of correction of the β⁰39-globin mutation causing β-thalassaemia.

Introduction

Nonsense mutations, giving rise to UAA, UGA and UAG stop codons within the coding region of mRNAs, promote premature translational termination [1–4] and are the leading cause of up to 30 % of inherited diseases [2,4], including cystic fibrosis [5], Duchenne muscular dystrophy [6] and thalassaemia [7–12]. For instance, in β⁰39-thalassaemia the CAG (glutamine) codon of the β-globin mRNA is mutated to the UAG stop codon [7,8], leading to premature translation termination and to mRNA destabilization through the well-described NMD (nonsense-mediated mRNA decay) [3,4]. Other examples of stop mutations of the β-globin mRNA occur at positions 15 [9], 37 [10,11] and 127 [12] of the mRNA sequence.

In order to develop an approach facilitating translation and, therefore, protection from NMD, aminoglycoside antibiotics have been tested on mRNAs carrying premature stop codons. These drugs bind the decoding centre of the ribosome and decrease the accuracy in the codon–anticodon base-pairing, inducing a ribosomal read-through of premature termination codons [13,14]. Despite their correctional potential, long-term use of aminoglycoside antibiotics is associated with nephrotoxicity and ototoxicity [13–16]. Interestingly, recent papers have described safer drugs, without structural similarity to aminoglycosides, designed and produced for suppressing premature translational termination, inducing a ribosomal read-through of premature but not normal termination codons [1,5,17]. This could be the basis for novel therapeutic approaches able to correct at the mRNA translational level the genetic defects introducing stop codons.

Accordingly, the development of experimental model systems useful for the screening of this class of molecules is of great interest for the possible identification of drugs to be employed for the experimental therapy of diseases caused by stop codons.

As far as studies on erythropoiesis, the K562 cell line [18] is well known as a useful experimental model system to study the expression of embryo-fetal globin genes, as well as their modulation [19–23]. Interestingly, especially for reaching the aims of the present study, K562 cells express the β-globin gene at very low levels both in their uninduced state as well as after erythroid differentiation stimulated by a variety of chemical inducers, such as cytosine arabinoside [24], mithramycin [25,26] and rapamycin [27,28].

In the present paper, we report the development of K562 cell lines integrating multiple copies of the human β-globin gene carrying the β⁰39-thalassaemia mutation. To this aim, we have (i) produced a lentiviral construct containing the β⁰39-thalassaemia globin gene under a minimal LCR (locus control region) control; (ii) produced several K562 cell clones with different integration copies of the normal βwt-globin (where wt is wild-type) or the β⁰39-thalassaemia constructs; (iii) analysed by quantitative real-time RT–PCR (reverse transcription–PCR) these clones for accumulation of the βwt-globin and β⁰39-globin mRNAs; (iv) treated these clones with Geneticin; (v) analysed the production of β-globin by FACS analysis; (vi) treated the best clone carrying β⁰39-globin genes and expressing the β⁰39-globin mRNA with Geneticin and other aminoglycosides in order to verify whether these experimental cellular systems are suitable for the screening of correctors of point mutations.

Materials and methods

Site-directed mutagenesis

Mutagenesis was performed by using the QuikChange^® II Site-directed Mutagenesis kit (Stratagene, La Jolla, CA, U.S.A.) in order to introduce the β⁰39-thalassaemic point mutation into the human β-globin gene in the pSL1180DT+β vector. A couple of mutant primers were designed (β⁰39 forward primer, 5′-GGT CTA CCC TTG GAC CTA GAG GTT CTT TGA GTC-3′; β⁰39 reverse primer, 5′-GAC TCA AAG AAC CTC TAG GTC CAA GGG TAG ACC-3′), having complementary sequences and containing the desired mutated nucleotide (indicated in bold). The mutagenesis reaction was performed in a final volume of 25 μl, containing 25 ng of pSL1180DT+β plasmid template, 1× reaction buffer [20 mM Tris/HCl, pH 8.8, 2 mM MgSO₄, 10 mM KCl, 10 mM (NH₄)₂ SO₄, 0.1 mg/ml BSA and 0.1 % Triton X-100], 0.5 μl of dNTP mix and 62.5 ng of mutagenesis primers, by using 1.25 units of PfuUltra HF DNA polymerase. The thermal reaction was performed by using the GeneAmp PCR System 9600 (PerkinElmer, Waltham, MA, U.S.A.): after a first denaturation at 94 °C for 3 min, 22 cycles were performed, consisting of denaturation at 95 °C for 30 s, annealing at 55 °C for 1 min and elongation at 68 °C for 8 min. At the end of the mutagenesis reaction, the amplification products were digested with 5 units of the restriction endonuclease DpnI, at 37 °C for 1 h, so as to remove the parental not mutated DNA. A 5 μl portion of the digestion reaction was used to transform 120 μl of ultracompetent Escherichia coli JM109 bacteria: DNA and bacteria were incubated on ice for 4 h; then, after a thermic shock at 42 °C for 45 s and immediately keeping on ice for 2 min, 1 ml of LB (Luria–Bertani) medium (10 g/l bacto-tryptone, 5 g/l yeast extract and 10 g/l NaCl) was added and an incubation at 37 °C for 1 h under slow agitation was performed; finally, bacteria were plated on Petri plates containing semisolid medium (LB medium with 15 g/l bacto-agar) in the presence of 100 μg/ml ampicillin, and incubated at 37 °C overnight. The bacterial clones obtained were screened for the incorporation of the recombinant plasmid construct, whose nucleotide sequence was finally confirmed by DNA sequencing.

Heat lysis of bacterial clones

In order to rapidly obtain a plasmid template to be amplified by PCR, the bacterial clone was washed twice, then suspended in 20 μl of ultrapure water, lysed at 96 °C for 6 min and immediately chilled on ice. After centrifugation at a maximum speed (15 000 g) for 20 min at 4 °C, the incorporated plasmid was collected in the supernatant.

DNA sequencing

In each PCR reaction, 2 μl of supernatant obtained after heat lysis of bacterial clones was amplified by Taq DNA polymerase using the BGF1 (5′-AGA CCT CAC CCT GTG GAG CC-3′) and BGR1 (5′-AGT TCT CAG GAT CCA CGT GCA-3′) primers, which amplify exons 1 and 2 of the β-globin gene leading to a 601 bp PCR product (Figure 2). PCR was performed in a final volume of 100 μl, containing 50 mM KCl, 10 mM Tris/HCl (pH 8.8), 1.5 mM MgCl₂, 0.1 % Triton X-100, 33 μM dNTPs and 0.33 μM PCR primers, by using 2 units per reaction of Taq DNA polymerase. The 35 PCR cycles were as follows: denaturation, 30 s, 95 °C; annealing, 20 s, 66 °C; and elongation, 2 min, 72 °C. The β-globin PCR products were purified with Microcon^®-100 (Millipore, Billerica, MA, U.S.A.) and sequenced using the ABI PRISM^™ BigDye^™ Terminator v 1.1 Cycle Sequencing Ready Reaction kit and the ABI Prism^™ 377 DNA sequencer (Applied Biosystems, Warrington, Cheshire, U.K.).

Figure 2. Characterization of the dTNS9-β⁰39 vector.

(A) Electrophoretic analysis of the fragments obtained by digestion of dTNS9-β⁰39 with ClaI and XbaI enzymes. M, molecular mass marker 1 kb DNA ladder (MBI Fermentas). (B) Characterization of the pCCL.β⁰39.PGW vector: electrophoretic analysis of the PCR product obtained by amplification of exons 1 and 2 of the β-globin gene and electropherogram achieved by sequencing the 601-bp PCR fragment. M, molecular mass marker pUC Mix Marker 8 (MBI Fermentas).

Virus production

Viral stocks were generated by transfection of pCCL.βwt.PGW or pCCL.β⁰39.PGW and the helper vectors pMD2.VSVG [coding for the envelope protein VSV-G (vesicular-stomatitis-virus glycoprotein)], pMDLpg.RRE [coding for the proteins GAG (group antigens polyprotein) and the reverse transcriptase POL] and pRSV.REV [coding for the protein REV (regulator of expression of viral proteins)] [29] into HEK-293T cells [HEK-293 cells (human embryonic kidney cells) expressing the large T-antigen of SV40 (simian virus 40)]. The HEK-293T (5 × 10⁶) cells were plated in 10-cm-diameter cell-culture dishes 24 h before transfection in DMEM (Dulbecco’s modified Eagle’s medium; Cambrex-Biowhittaker Europe, Milan, Italy) with 10 % (v/v) FBS (fetal bovine serum; Biowest, Nuaillé, France), 2 mM L-glutamine (Cambrex-Biowhittaker), 100 units/ml penicillin and 100 μg/ml streptomycin (Pen-Strep; Cambrex-Biowhittaker) in a humidified atmosphere of 5 % CO₂/air. The culture medium was changed to IMDM (Iscove’s modified Dulbecco’s medium; Cambrex-Biowhittaker) 2 h before transfection. A total of 25.5 μg of DNA was used for transfection of one dish: 3 μg of the envelope plasmid pMD2.VSVG, 5 and 2.5 μg of the two packaging plasmids, pMDLpg.RRE and pRSV.REV respectively and 15 μg of the lentiviral vector. A 0.1 × TE (10 mM Tris, pH 8.0, plus 1 mM EDTA) and water solution (2:1) was added to the plasmid up to a final volume of 450 μl. The precipitate was finally formed after adding 50 μl of 2.5 mM CaCl₂ solution, mixing well and adding dropwise 500 μl of 2 × Hepes-buffered saline (281 mM NaCl, 100 mM Hepes and 1.5 mM Na₂ HPO₂, pH 7.12) while vortex-mixing. The precipitate was immediately added to the cultures. The medium was replaced after 16 h; the viral supernatant was collected after another 24 h, replaced and again collected after 24 h. Each time, the supernatant was filtered through 0.2 μm pore size cellulose acetate filters, divided into aliquots and frozen. The viral titre was calculated as previously described [30].

Human K562 cell cultures and K562 cell clones carrying the βwt-globin or the β⁰39-globin gene

The human leukaemia K562 cells [14] were cultured in a humidified atmosphere of 5 % CO₂/air in RPMI 1640 medium (Sigma, St Louis, MO, U.S.A.) supplemented with 10 % (v/v) FBS (Biowest), 2 mM L-glutamine (Cambrex-Biowhittaker), 100 units/ml penicillin and 100 μg/ml streptomycin (Cambrex-Biowhittaker). Cell growth was studied by determining the cell number per ml with a ZF Coulter counter (Coulter Electronics, Hialeah, FL, U.S.A.). Two lentiviral constructs were used to generate stable K562 clones integrating human normal and β⁰39-globin genes. The pCCL.β.PGW lentiviral construct carries two LTR (long terminal repeat) sequences, the SV40 origin of replication, a GFP (green fluorescent protein) gene under the control of the PGK (phosphoglycerate kinase) promoter, and the β-globin gene, under the control of the β-globin gene promoter and a minimal LCR of the human β-like globin gene cluster. The second construct was developed replacing the wild-type β-globin gene with a β⁰39-globin gene produced by site-directed mutagenesis. Transduction was carried out by plating 10⁶ K562 cells in 9.5 cm² dishes with 45 % RPMI 1640 (Sigma) and 45 % IMDM (Cambrex-Biowhittaker), 10 % FBS (Biowest), 2 mM L-glutamine (Cambrex-Biowhittaker), 100 units/ml penicillin and 100 μg/ml streptomycin (Pen-Strep; Cambrex-Biowhittaker) at 37 °C in a humidified atmosphere of 5 % CO₂/air. Cells were infected with the thawed viral supernatant at various MOI (multiplicity of infection) values (MOI = 2–4). In order to facilitate the cell infection, 10 μl of the 800 μg/μl transduction agent polybrene (Chemicon International, Millipore) was added to the plated K562 cells, which were subsequently cultured in a 5 % CO₂ incubator. After 7 days, K562 cells were cloned by limiting dilutions and GFP-producing clones were identified under a fluorescence microscope and further characterized. Treatments with aminoglycoside antibiotics were carried out by adding the appropriate drug concentrations to the cells, seeding at 30 000 or 8000 cells/ml and incubating at 37 °C in a humidified atmosphere of 5 % CO₂/air for 3–5 days. The medium was not changed during the treatment period. Geneticin was from Gibco (Invitrogen-Life Technologies, Carlsbad, CA, U.S.A.), gentamicin, streptomycin, tobramycin and amikacin from Sigma.

Real-time quantitative PCR

To quantify the amount of β-globin gene sequences in the K562 genomic DNA from βwt-globin and β⁰39-globin-K562 cell clones, we used 100–200 ng of template DNA and a standard curve in the range 50–300 ng of K562 genomic DNA as the control. The β-globin gene sequences in βwt-globin and β⁰39-globin-K562 cell clones were quantified with respect to the γ -globin target gene, using the same amount of genomic DNA. The γ -globin gene represents a good reference because of its location in the same chromosome, within the same gene cluster, of the β-globin gene. The fold of transduced β-globin genes was expressed as 2^−ΔΔCt, where C_t represents the threshold cycle value and ΔΔC_t represents the subtraction of the ΔC_t obtained using genomic DNA from untransduced K562 cells (the original K562 cell line) from the ΔC_t derived from βwt-globin and β⁰39-globin-K562 cell clones. ΔC_t is the difference between the amplification cycle equivalent to a significant fluorescence level (C_t ) obtained by the amplification of the β-globin gene and the corresponding amplification cycle of the γ -globin gene. This equation can be applied to have a relative quantification of our templates because the slopes of β-globin and γ -globin gene amplifications are comparable. PCR primers and probes were: β-globin forward primer, 5′-CAA GAA AGT GCT CGG TGC CT-3′; β-globin reverse primer, 5′-GCA AAG GTG CCC TTG AGG T-3′; β-globin probe, 5′-FAM-TAG TGA TGG CCT GGC TCA CCT GGA C-TAMRA-3′. The fluorescent reporter and the quencher were: FAM (6-carboxyfluorescein) and TAMRA (6-carboxytetramethylrhodamine) respectively. For real-time PCR analysis we used as a reference gene the γ -globin gene. PCR primers and probe were: γ -globin forward primer, 5′-TGG CAA GAA GGT GCT GAC TTC-3′;γ-globin reverse primer, 5′-TCA CTC AGC TGG GCA AAG G-3′; γ -globin probe, 5′-FAM-TGG GAG ATG CCA TAA AGC ACC TGG-TAMRA-3′. The probe was fluorescently labelled with FAM (Applied Biosystems). Quantitative real-time PCR was carried out using a 7700 Sequence Detection System version 1.7 (Applied Biosystems).

RNA isolation and RT–PCR analysis

K562 cell clones were collected by centrifugation at 400 g for 5 min at 4 °C, washed with PBS and lysed in 1 ml of TRIzol^® reagent (Gibco–Invitrogen–Life Technologies), according to the manufacturer’s instructions. The isolated RNA was washed once with cold 75 % ethanol, dried and dissolved in diethylpyrocarbonate-treated water before use. For gene expression analysis 1 μg of total RNA was reverse transcribed by using random hexamers. To quantify the amount of β-globin mRNA we used 20 ng of cDNA and a standard curve of normal control DNA in the range 5–60 ng. Quantitative real-time PCR was carried out using a Custom TaqMan^® SNP Genotyping Assays kit (Applied Biosystems) in a 7700 Sequence Detection System version 1.7 (Applied Biosystems). PCR was performed in a final volume of 25 μl, containing 12.5 μl of TaqMan^® Universal PCR Master Mix and 0.625 μl of 40 × assay mix. After two initial steps, 2 min at 50 °C and 10 min at 95 °C, the 40 cycles were as follows: 15 s at 92 °C and 60 s at 60 °C. PCR primers and probes were: primer forward, 5′-CAG GCT GCT GGT GGT CTA C-3′; primer reverse, 5′-AGT GGA CAG ATC CCC AAA GGA-3′; probe βwt, 5′-VIC-AAA GAA CCT CTG GGT CCA-TAMRA; probe β⁰39, 5′-FAM-CAA AGA ACC TCT AGG TCC A-TAMRA-3′. The probes βwt and β⁰39 were fluorescently labelled with VIC and FAM (Applied Biosystems) respectively to quantify the βwt and β⁰39-globin mRNA in a single reaction. For real-time PCR analysis we always used as a reference gene the endogenous control human GAPDH kit (Applied Biosystems), in the gene expression assay. The fluorescent reporter and the quencher of the GAPDH probe were: VIC and TAMRA respectively.

FACS analysis

K562 cells treated with Geneticin were permeabilized and marked with the antibody against β-globin using the Cytofix/Cytoperm^™ kit (BD Biosciences Pharmingen, Franklin Lakes, NJ, U.S.A.). A total of 10⁶ cells were first washed with 500 μl of PBS and then incubated with 500 μl of BD Cytofix-Cytoperm solution for 20 min at 4 °C, in order to permit the cellular permeabilization. After the incubation the cells were washed twice and incubated with 300 μl of 1× PBS +1 % BSA (Sigma) solution for 1 h at room temperature (25 °C) in darkness. The BSA has the capacity to block the aspecific binding sites. The cells were then collected by centrifugation and incubated with 30 μl of β-globin-PE (phycoerythrin) monoclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA, U.S.A.), diluted 1:10 in 1 × PBS + 1 % BSA over night at 4 °C in darkness. After the incubation, the cells were washed with 500 μl of 1 × PBS and resuspended with 30 μl of 1 × PBS. The cellular suspension was transferred to a FACS tube and 500 μl of staining buffer (1 × PBS +1 % FBS) were added. The analysis was performed with FACScan (flow-activated cell sorting; Becton Dickinson), using the Cell Quest Pro software (Becton Dickinson).

Results

Production of the lentiviral construct carrying the mutated human β⁰39-globin gene

Figure 1 shows the strategy followed to produce the lentiviral constructs used for generation of the K562 cell clones. We first introduced the β⁰39-thalassaemic point mutation into the human β-globin gene in the pSL1180DT+β vector (depicted in the upper panel of Figure 1), by using a site-directed in vitro mutagenesis system (from Stratagene). Then the dTNS9-β⁰39 was produced as described in Figure 1 and characterized by restriction enzyme digestion as reported in Figure 2(A). Finally, the pCCL.β⁰39.PGW vector was generated by replacing the ClaI–XbaI fragment of the pCCL.β.PGW with the ClaI–XbaI fragment of the dTNS9-β⁰39 construct. The obtained pCCL.β⁰39.PGW construct, characterized by PCR amplification and direct sequencing as reported in Figure 2(B), displays two LTR sequences, the SV40 origin of replication, a GFP gene under the control of the PGK promoter, the β-globin gene, under the control of the β-globin gene promoter, and a minimal LCR of the human β-like globin gene cluster (Figure 1). The presence of the PGK–GFP gene allows the high-throughput screening of cells integrating this construct, giving at the same time some preliminary information on the number of integration events.

Figure 1.

Strategy followed to produce the lentiviral constructs used for generation of the K562 cell clones

Production of K562 cell clones expressing human β-globin genes carrying the β⁰39-thalassaemia mutation

The two constructs shown in Figure 1 have been employed in transduction experiments using K562 cells as the model system, and 13 clones have been identified and further characterized. Figure 3 shows the fluorescence intensities of eight of these clones (wt3, wt5, wt6 and wt7 carrying the normal β-globin gene, whereas m1, m4, m5 and m6 carry the mutated β⁰39-globin gene). As it is clearly evident, clones wt3 and m6 display the highest levels of fluorescence, suggesting the integration of high copy numbers of the respective constructs.

Figure 3.

Fluorescence intensities of eight representatives of the 13 clones obtained, four containing the βwt-globin gene (upper four panels) and four containing the β⁰39-globin gene (lower four panels)

Analysis of the integration units of the βwt-globin and the β⁰39-globin genes

The analysis of the integration units of the βwt-globin-K562 and β⁰39-globin-K562 cells was performed by PCR using genomic DNA isolated from the different clones and performing real-time PCR (a representative PCR experiment is shown in Figures 4A and 4B). The results obtained are shown in Figures 4(C) and 4(D), and clearly indicate that higher numbers of integrated constructs are present in wt1, wt2, wt4, wt5, wt6, wt7 (βwt-globin-K562 cells) and m1, m2, m3 (β⁰39-globin-K562 cells) clones. The highest numbers of integrated constructs (transgene fold ranging from 5 to 10) are present in wt3 (βwt-globin-K562 cells) and m4, m5 and m6 (β⁰39-globin-K562 cells) clones. This analysis has been carried out three times with similar results. Interestingly, when FACS analysis was performed on all of the clones, we noted that there is a very good correlation between the GFP median fluorescence intensity (Figures 4C and 4D, black histograms) and the number of the integrated constructs (Figures 4C and 4D, open histograms). This is not surprising, since the GFP gene is under the control of the constitutive PGK promoter. On the contrary, we cannot exclude that the site(s) of integration might influence (either positively or negatively) the transcription of the βwt and β⁰39-globin genes, which are under the control of the β-globin promoter and a minimal LCR sequence (see Figure 1). Furthermore, we are expecting the well-known phenomenon of NMD [10,11] of the β⁰39-globin mRNA in the β⁰39-globin-K562 cell clones. Therefore a careful analysis of the mRNA levels was performed by quantitative real-time RT–PCR, as described in detail elsewhere [22–28].

Figure 4. Analysis of the integration and GFP fluorescence of the βwt-globin and β⁰39-globin K562 cell clones.

(A, B) Representative real-time PCR analysis of β-globin DNA from (A) wt3 (open circles), wt1 (open squares) clones and (B) m5 (black circles) and m3 (black squares) clones; the PCR probe used amplifies human β-globin gene sequences. (C, D) White bars, integrated βwt-globin and β⁰39-globin genes in the genome of the βwt-globin-K562 (C) and β⁰39-globin-K562 (D) cells. Black bars, GFP fluorescence. The results represent the means ± S.D. for three independent determinations. In order to determine the integration efficiency of the βwt-globin and β⁰39-globin genes by real-time PCR, the obtained C_t values were compared with those obtained using primers amplifying γ-globin gene sequences.

Accumulation of the β-globin and β⁰39-globin mRNAs in βwt-globin-K562 and β⁰39-globin-K562 cell clones

The extent of accumulation of β-globin and β⁰39-globin mRNAs in the βwt-globin-K562 and β⁰39-globin-K562 cell clones was analysed by RT–PCR using as the template RNA isolated from exponentially growing cells. We were interested in identifying βwt-globin-K562 and β⁰39-globin-K562 cell clones accumulating similar amounts of β-globin mRNA sequences. The results obtained gave clear evidence that within the βwt-globin-K562 clones, high expression (relative to wt1) was detected in wt6 and low expression in wt3, wt5 and wt7 clones (Figure 5A); on the other hand, within the β⁰39-globin-K562 clones, the β-globin gene was highly expressed (in comparison with m3) in K562 cell clones m6, m5, m4 and m1 and poorly expressed in K562 cell clone m2 (Figure 5B). Despite the fact that no silencing effect was noted, no clear relationship is evident between levels of gene expression and number of integration units. This is however as expected, since transcription might depend also on the site of integration. Moreover, in β⁰39-globin-K562 cell clones the accumulation of β-globin mRNA sequences is affected by the known effect of NMD [3,4]. After comparative analysis of the RT–PCR data, clones wt3 (βwt-globin-K562 cells) and m5 (β⁰39-globin-K562 cells) were selected for the experiments employing possible correction with Geneticin, due to the fact that, among the βwt-globin and β⁰39-globin-K562 clone sets, they express similar, although not identical, levels of β-globin mRNA molecules, as depicted in Figures 5(C) and 5(D). Interestingly, clone m5 displays higher β-globin gene integration units and GFP production with respect to clone wt3, suggesting that, as expected, NMD affects β⁰39-globin mRNA. The differential expression of β-globin mRNA between m5 and wt3 is maintained when the culture conditions were changed from expansion to differentiation, by treating cells with cytosine arabinoside (F. Salvatori, G. Breveglieri, I. Lampronti, C. Zuccato, N. Bianchi, M. Borgatti and R. Gambari, unpublished work).

Figure 5. Expression of the β-globin gene in βwt-globin and β⁰39-globin K562 cell clones.

(A, B) RT–PCR analysis showing the β-globin mRNA content in βwt-globin (A) and β⁰39-globin-K562 (B) clones (means ± S.D. for three independent determinations). Results represent the fold change in β-globin mRNA content with respect to wt1 (A) and m3 (B). The relative content of β-globin mRNA of the original K562 cells is also shown (left side of each panel). (C, D) Profiles of quantitative RT-PCR analysis performed on RNA isolated from βwt-globin-K562 clone wt3 (open symbols) and from β⁰39-globin-K562 clone m5 (closed symbols) using primers amplifying GAPDH (C) and β-globin (D) mRNA sequences. Two different probes were used for the β-globin mRNA: a βwt probe recognizing the βwt-globin mRNA (open symbols) and a β⁰39 probe recognizing the β⁰39-globin mRNA (closed symbols).

Effects of Geneticin on β-globin production in the m5 clone

The effects of Geneticin on the β-globin production were analysed after FACS analysis of untreated compared with Geneticin-treated cells. As expected, very low levels of β-globin are detected in control K562 cells (Table 1, left column). It is well known indeed that K562 cells are committed to γ -globin gene expression and produce only very low levels of β-globin mRNA. In order to verify the effects of Geneticin on β-globin production in the wt3 βwt-globin-K562 cell clone, cells were, in a first experiment, cultured in the absence or presence of 400 ng/μl Geneticin. At the end of the treatment, cells were recovered and labelled with the PE-labelled β-globin monoclonal antibody. This labelling allows discrimination, under FACS analysis, of the green fluorescence of GFP from the red fluorescence of β-globin-PE monoclonal antibody. When FACS analysis was performed (Figures 6A and 6B) it was clearly evident that Geneticin did not induce major changes in β-globin production in wt3 βwt-globin-K562 cells. By sharp contrast, when the same experiment was carried out using the m5 β⁰39-globin K562 cell clone, a clear increase in cells producing β-globin was detected (Figures 6C and 6D). No major changes in GFP production were observed after treatment with Geneticin of both wt3 βwt-globin-K562 and m5 β⁰39-globin-K562 cell clones. The results of three independent experiments employing increasing concentrations of Geneticin are shown in Figure 7(A), confirming a concentration-dependent increase in β-globin-containing cells when the m5 β⁰39-globin-K562 cell clone is treated with Geneticin. The effect of Geneticin was found to be higher than that displayed by gentamicin (Figure 7B).

Table 1.

Induction of β-globin accumulation by aminoglycosides

Results are presented as a percentage of β-globin-containing cells.
Treatment	Cell line …	β-Globin-containing cells (%)
		K562 (original)	wt3 βwt-globin-K562 cells	m5 β039-globin K562 cells
None		1.6 ± 0.3	73.7 ± 18.3	1.5 ± 0.4
Geneticin		1.5 ± 0.4	85.3 ± 4.5	51.5 ± 22.3
Gentamicin		1.8 ± 0.1	73.4 ± 7.5	5.65 ± 0.7
Streptomycin		1.4 ± 0.2	64.5 ± 11.4	3.74 ± 0.5
Tobramycin		1.1 ± 0.2	70.1 ± 6.8	2.2 ± 0.2
Amikacin		1.5 ± 0.4	67.3 ± 8.5	2.1 ± 0.5

Figure 6. Effect of 400 ng/μl Geneticin on the production of β-globin and GFP in βwt-globin K562 clone wt3 (A, B) and β⁰39-globin-K562 clone m5 (C, D).

FACS analysis of untreated (A, C) compared with Geneticin-treated (B, D) cells is shown.

Figure 7. Effect of increasing concentrations of Geneticin (A) and gentamicin (B) on the proportion (%) of β-globin-containing β⁰39-globin-K562 cells.

FACS analyses were conducted as shown in Figure 6. The results are means ± S.D. for three independent determinations.

Finally, Table 1 shows a summary of the comparative FACS analysis of the m5 β⁰39-globin K562 cell clone treated with the aminoglycosides Geneticin, gentamicin, streptomycin, amikacin and tobramycin. As it is evident, gentamicin and streptomycin were found to be less efficient than Geneticin. Among the tested aminoglycosides, tobramycin and amikacyn were found to be almost ineffective. These results support the concept that our cellular screening system could be useful in ranking the activity of correctors of stop mutations. As expected, and shown in Table 1 (left and middle columns), no significant increase in β-globin-positive cells was observed when the treatment was performed on original K562 cells (expressing very low amounts of β-globin), as well as on the wt3 βwt-globin-K562 cell clone (expressing a high amount of β-globin). In addition, Table 1 indicates that the FACS data concerning untreated m5 cells are very similar to those found with the original K562 cells, suggesting that the β⁰39-globin mRNA accumulated by m5 cells is not translated, unless the m5 cells are treated with aminoglycosides. Taken together, the results obtained suggest that the effects of aminoglycosides are related to a read-through effect, rather than an induction of endogenous or de novo integrated wild-type β-globin genes.

Discussion

In the present paper, we described the development of cellular K562 cell systems carrying β-globin genes with a stop mutation causing β⁰-thalassaemia. These K562 clones were generated by stable transduction with a lentiviral vector carrying wild-type or mutated β⁰39-globin genes under the control of a minimal LCR region. These clones express the GFP protein at a level correlated with the integrated copy number, whereas in several clones a direct relationship between copy number and accumulation of the βwt or β⁰39-globin mRNA was not evident, possibly due to the site of integration, which might affect the tissue-specific expression of the β-globin gene, independently of the presence of the stop mutation.

As a proof of principle of the possible application of these K562 cell clones, Geneticin was used. Geneticin, indeed, has been reported to retain the very interesting property to induce read-through of mRNAs carrying translation stop mutations leading to production of truncated and functionally inactive proteins [31,32]. The results obtained suggest that this experimental system is suitable for the characterization of correctors of the β⁰39-globin mutation causing β-thalassaemia. This was reproducibly obtained using Geneticin on K562 cell clones carrying β⁰39-globin genes. In addition, our experimental system allows ranking of the activity of several other aminoglycosides, such as gentamicin, streptomycin, tobramycin and amikacin, as we have demonstrated in Figure 7 and Table 1. No major effects on β-globin synthesis were found on the original K562 and on the K562-β-wt3 clones (Table 1).

Accordingly, the development of experimental model systems useful for the screening of this class of molecules is of great interest for the possible identification of drugs to be employed for the experimental therapy of diseases caused by stop codons. K562 clones integrating the human β⁰39-globin gene appear to be a low-throughput cellular system for the characterization of molecules selected by means of high-throughput strategies, such as those employing reporter genes carrying stop mutations mimicking molecular defects leading to human pathologies.

In this respect, we have to point out that several human pathologies caused by stop mutations are known, including cystic fibrosis [5], muscular dystrophy [6] and the Hurler syndrome [13]. Accordingly, several research groups have started projects aimed at developing molecules able to promote read-through, to be tested on in vitro and in vivo model systems. For instance, in a recent paper by Welch et al. [1], two high-throughput screens (comprising 800 000 low-molecular-mass compounds) were performed to identify compounds that promoted UGA nonsense suppression. Chemical scaffolds were identified and optimized through extensive medicinal chemistry efforts. These analyses identified PTC124 {3-[5-(2-fluorophenyl)-[1,2,4]oxadiazol-3-yl]benzoic acid; C₁₅ H₉FN₂ O₃} as a candidate for further development. The selectivity of PTC124 for premature termination codons, its well-characterized activity profile, oral bioavailability and pharmacological properties indicate that this drug may have broad clinical potential for the treatment of a large group of genetic disorders with limited or no therapeutic options.

The results shown in the present paper allow us to propose our developed K562 cell lines as a useful cellular biosensor to verify whether compounds suppressing the effects of premature termination codons might be of interest for the possible development of drugs of interest for thalassaemia.

Abbreviations used

C_t

threshold cycle value

FAM

6-carboxyfluorescein

FBS

fetal bovine serum

GAPDH

glyceraldehyde-3-phosphate dehydrogenase

GFP

green fluorescent protein

LB

Luria–Bertani

LCR

locus control region

LTR

long terminal repeat

MOI

multiplicity of infection

NMD

nonsense-mediated mRNA decay

PE

phycoerythrin

PGK

phosphoglycerate kinase

RT–PCR

reverse transcription–PCR

SV40

simian virus 40

TAMRA

6-carboxytetramethylrhodamine

wt

wild-type