Development of CpG‐Depleted CFTR Plasmid‐Based Nanoparticles for Nonviral Gene Therapy in Lung Cystic Fibrosis Disease

TL;DR

Researchers developed a new nonviral gene therapy approach for lung cystic fibrosis using CpG-depleted CFTR plasmid-based nanoparticles to achieve long-lasting protein expression.

Contribution

A CpG-depleted and codon-optimized CFTR plasmid was developed to enable sustained and high-level CFTR protein expression.

Findings

CpG-depleted CFTR plasmid nanoparticles achieved a 20-fold increase in CFTR protein expression in bronchial epithelial cells.

The hEFIα promoter outperformed the CMV promoter, showing a 2.26-fold increase in CFTR expression at 7 days post-transfection.

The plasmid-based approach demonstrated high efficacy in vitro, suggesting potential for in vivo application in cystic fibrosis treatment.

Abstract

Nonviral gene therapy holds promise as a potential treatment for lung cystic fibrosis (CF). However, the transient expression of the CF transmembrane conductance regulator (CFTR) protein has limited its clinical application. To circumvent this challenge, a CpG‐depleted CFTR plasmid was developed. The CpG‐depleted CFTR plasmid could be compacted into DNA nanoparticles and modified with the addition of highly branched poly(β‐amino ester)s (HPAEs), leading to an improved and sustained CFTR protein expression. Using a CpG‐depleted and codon‐optimized CFTR sequence, around 20‐fold increase in CFTR protein production was achieved 48 h after treatment, compared with healthy human bronchial epithelial cells (16HBE14o‐). To evaluate the duration of CFTR protein expression induced by the plasmid based on human elongation factor 1α (hEFIα) and cytomegalovirus (CMV) promoters, a time course study was conducted in human CF bronchial epithelial (CFBE14o‐) cells. hEFIα promoter revealed a remarkable 2.26‐fold increase in CFTR protein expression at 7 days after transfection compared with 16HBE14o‐ cells. This level of CFTR protein expression outperformed the commonly used CMV promoter. The in vitro results demonstrated that CpG‐depleted CFTR plasmid could be used to achieve high efficacy in subsequent in vivo evaluations, which, if validated, may provide promising prospects for the development of a novel and effective treatment for lung cystic fibrosis.

Well designed CpG‐depleted CFTR nanoparticles for cystic fibrosis.

Figures

CFTR plasmids construction. (a) Workflow of CFTR plasmids construction. (b) pUC‐hCEFIα‐WT‐CFTR, pUC‐hCEFIα‐OP‐CFTR, R6K‐hCEFIα‐WT‐CFTR, and R6K‐hCEFIα‐OP‐CFTR plasmids were digested with NheI restriction enzyme to confirm the CFTR gene bands (4443 bp). These four plasmids were digested with the SpeI restriction enzyme to confirm the plasmid sizes (pUC‐hCEFIα‐WT‐CFTR [7578 bp], pUC‐hCEFIα‐OP‐CFTR [7578 bp], R6K‐hCEFIα‐WT‐CFTR [7369 bp], and R6K‐hCEFIα‐OP‐CFTR [7369 bp]). (c) The plasmid quality was evaluated by measuring the supercoiled fraction of the different CFTR plasmids. (d) The plasmid production levels of different CFTR plasmids. All data were collected from at least three independent experiments (n ≥ 3). All bar graph data are depicted as means ± SDs. *p < 0.05, p < 0.01; ns: not significant, as compared with pWT‐CFTR1.

CFTR protein expression from different CFTR plasmids. (a) CFTR gene expression was assessed in untreated 16HBE14o‐, untreated CFBE41o‐ (UNT), and CFBE41o‐ cells transfected with HPAEs/CFTR plasmids (pWT‐CFTR1, pWT‐CFTR2, pOP‐CFTR1, pOP‐CFTR2) 48 h after transfection via western blot. (b) The semiquantification of western blot was normalized with the loading control protein α‐tubulin. (c) CFTR gene expression from untreated 16HBE14o‐, untreated CFBE41o‐ (UNT), and CFBE41o‐ cells transfected with HPAEs/CFTR plasmids 48 h after transfection was assessed by immunofluorescence, using a 20× fluorescence microscopy objective; the dimension of the scale bar is 50 μm. The yellow arrows show the CFTR protein expression in treated CFBE41o‐ cells with different CFTR plasmids. Western blot and immunofluorescence images are representative of three independent experiments (n = 3). All data were collected from at least three independent experiments (n ≥ 3). All bar graph data are depicted as means ± SDs. p < 0.01, *p < 0.001; ns: not significant, as compared with pWT‐CFTR1.

CFTR plasmids construction using different promoters. (a) Schematic representation of the workflow for constructing R6K‐hCMV‐OP‐CFTR plasmids derived from the R6K‐hCEFIα‐OP‐CFTR plasmid. (b) Verification of plasmid constructs by cutting R6K‐hCEFIα‐OP‐CFTR and R6K‐hCMV‐OP‐CFTR plasmids using SbfI and HindIII restriction enzymes to confirm the hCMV enhancer and hEFIα promoter (hCEFIα) (536 bp) and hCMV enhancer and hCMV promoter (516 bp) bands. Confirmation of the plasmid sizes for R6K‐hCEFIα‐OP‐CFTR (7369 bp) and R6K‐hCMV‐OP‐CFTR (7349 bp) was achieved by using the HindIII restriction enzyme.

Time courses of CFTR protein expression of different CFTR plasmids (pOP‐CFTR2 and pOP‐CFTR2‐CMV). (a) CFTR protein expression of different CFTR plasmids with the loading control protein α‐tubulin was assessed by western blot at 24 h, 48 h, and 7 days after transfection. (b) Quantification of CFTR mRNA levels via RT‐qPCR, normalized to GAPDH and then compared with 16HBE14o‐ cells, following treatment with HPAE complexes containing different CFTR plasmids at 24 h, 48 h, and 7 days after transfection. (c) CFTR protein was semiquantified, normalized with the loading control protein α‐tubulin, and then compared with 16HBE14o‐ cells, after treatment with HPAEs complexing to different CFTR plasmids by western blot 24 h after transfection, 48 h after transfection, and 7 days after transfection. Western blot images are representative of three independent experiments (n = 3). All data were collected from at least three independent experiments (n ≥ 3). All bar graph data are depicted as means ± SDs. p < 0.001; ns: not significant, as compared with pOP‐CFTR2‐CMV.

Full text

Introduction

1

Nowadays, the life expectancy of cystic fibrosis (CF) patients has been improved worldwide with the improvements in treating lung CF [1]. However, current treatments, including bronchodilators, mucus thinners, and antibiotics, primarily serve to alleviate the symptoms of the disease, rather than providing a cure [1, 2, 3, 4, 5, 6]. Recently, modulators have been shown to improve lung function and reduce exacerbations, but their effectiveness is limited to specific mutation types (such as F508del and G551D), leaving approximately 10% of CF patients in need of new treatments [7, 8, 9]. In contrast to current treatments, gene therapy would overcome these hurdles. Gene addition, a form of gene therapy, involves the introduction of a healthy CFTR gene into diseased cells. This approach is mutation independent and can target all CF mutation types. Moreover, gene addition targets the origin of CF, producing a functional CFTR protein to replace the missing CFTR protein. Therefore, there has been a growing interest in developing gene addition strategies for treating lung CF.

The translation of gene therapies from the bench to the bedside is a complex process, requiring both effectiveness and safety when administered to CF patients. Preclinical studies have demonstrated that for gene therapy, achieving 5%–10% of wild‐type CFTR mRNA or protein levels can be effective in the clinic [10, 11, 12]. The therapeutic challenge for this strategy is to consistently attain and sustain these levels of CFTR mRNA or protein within the lungs of CF patients, and a powerful CFTR nucleic acid component is needed to circumvent this issue. In this regard, CFTR plasmids have emerged as promising candidates for CF gene addition strategies, offering distinct advantages over their RNA counterpart, particularly in terms of ease of manufacturing, cost‐effectiveness, and long‐term stability during storage [13]. Moreover, the CFTR plasmid acts as an episome following the transfection of diseased cells, making it a safer approach than integrative therapies that utilize viral vectors [14]. However, the utility of the CFTR plasmid for gene therapy is hindered by its transient expression in transfected cells [15]. Additionally, the CpG sequences in the plasmid can induce an immune response in vivo, via the activation of toll‐like receptor 9 (TL9) [16, 17]. These drawbacks limit the therapeutic potential of the CFTR plasmid and must be addressed in order to move this approach to the clinic.

To circumvent these issues, strategies involving CFTR plasmid sequence modifications, such as CpG depletion, have been proposed to enhance transgene expression, prolong the plasmid's duration in transfected cells, and ensure patient safety [18, 19, 20]. In proof‐of‐concept studies, delivering the CpG‐depleted CFTR plasmid into the lungs of CF mice demonstrated both sustained transgene expression and an absence of inflammation in vivo [21]. On the other hand, promoters also play a crucial role in determining the level and duration of CFTR protein expression. Viral promoters, such as cytomegalovirus (CMV) enhancers/promoters, have been widely utilized in pre‐clinical and clinical studies for CFTR and other target gene expression. Although the CMV promoter initially yields relatively high levels of transgene expression, this is transient [21]. Further studies have indicated that the inflammatory cytokines induced by plasmids can lead to the CMV promoter/enhancer silencing [22]. Therefore, endogenous mammalian promoters have been employed to prolong the CFTR expression [21, 23]. Combining the CMV enhancer with human elongation factor 1α (EFIα) promoter has proven effective in achieving both high protein expression levels and sustained expression in vivo [21]. In addition to CpG sequence depletion and promoter optimization within the CFTR plasmid, the utilization of S/MARs (scaffold/matrix attachment regions) sequences has been explored as another method to prolong CFTR gene expression [24, 25]. Furthermore, optimizing the target transgene sequence has the potential to enhance the translation efficiency of the CFTR protein [18, 19]. These multifaceted approaches contribute to advancing the effective and durable CFTR expression for gene therapy in CF patients.

For a treatment to be successful, the gene delivery vector plays a pivotal role in determining overall gene replacement safety and CFTR protein restoration [26]. A previous study has demonstrated that highly branched poly(β‐amino ester)s (HPAEs) designed and synthesized from Wang's group can contribute to efficient DNA delivery in other cell lines, such as recessive dystrophic epidermolysis bullosa keratinocytes (RDEBK) [27]. Thus, robust HPAEs have been chosen as the delivery vehicle for the CFTR plasmid transfection in this work (the optimization process of PAE is being summarized in another published article [28]).

The primary aim of this research is to tackle the challenge associated with nonintegrative gene addition in lung CF disease, via the construction of a CFTR plasmid with efficient and sustained CFTR protein expression. Here, an optimized CpG‐depleted CFTR plasmid‐based nanoparticle was employed for nonviral gene therapy in lung CF by using an HPAE vector. To achieve this, initially, a series of CFTR plasmids was constructed. Following evaluation and comparison with the HPAEs delivery vector, a CpG‐depleted CFTR plasmid containing the profiles of optimizing CFTR cDNA, depleting the CpG sequence, S/MAR domains, and hEF1α promoter was selected. The CpG‐depleted CFTR sequence is derived from the work of Hyde et al., with several modifications made to eliminate the cryptic promoter [21]. The integration of these various profiles, including optimized CFTR cDNA, CpG sequence depletion, S/MAR domains, and hEF1α promoter, has not been previously explored in a single CpG‐depleted CFTR plasmid. The CpG‐depleted CFTR plasmid showed improved CFTR expression via HPAE delivery in CFBE41o‐ cells, indicating that over 10% CFTR protein was expressed up to 7 days after transfection.

Materials and Methods

2

Plasmid Construction

2.1

Sequence Information

2.1.1

The hCMV enhancer (Genbank acc# JQ439994) and BGH polyA (Genbank acc# JQ439997) lack the CpG sequence. The hEFIα promoter and SI126 Intron sequences are CpG‐free and were designed based on the pcpgf‐mcs plasmid (InvivoGen, CA, USA). The WT‐CFTR sequence (Genbank acc# NM_000492) was generated from the human genomic CFTR cDNA sequence and included five base pair modifications. The OP‐CFTR sequence, based on the soCFTR2 sequence, is codon optimized and CpG‐free but has been modified with six base pair changes [21]. The hCMV promoter sequence was obtained from the RG216476 plasmid and purchased from Origene (Rockville, MD, USA).

Plasmid Digestion and Ligation

2.1.2

The pUC‐hCEFIα‐WT‐CFTR plasmid consists of a human cytomegalovirus (hCMV) enhancer‐human elongation factor 1‐α (hEFIα) promoter (hCEFIα)‐SI126 Intron‐WT‐CFTR‐BGH polyA‐pUC origin sequence‐Ampicillin resistant sequence (Figure S2). These hCMV enhancer‐hEFIα promoter‐SI126 Intron‐WT‐CFTR‐BGH polyA components were commercially synthesized by Eurofins (Munich, Germany) and ligated into the pUC origin and Ampicillin resistance sequence (Eurofins backbone, pEX‐A258) (Figure S2).

The R6K‐hCEFIα‐WT‐CFTR plasmid has a CpG‐free backbone. It was constructed by cutting the hCMV enhancer‐hEFIα promoter‐SI126 Intron‐WT‐CFTR‐BGH polyA from a pUC‐hCEFIα‐WT‐CFTR plasmid with the EcoRI restriction enzyme (Brennan & Co, Dublin, Ireland). It was then inserted into the pcpgf‐mcs plasmid that contains R6K origin sequence and Zeocin resistant sequence (Figure S2).

The pUC‐hCEFIα‐OP‐CFTR plasmid consisting of hCMV enhancer‐hEFIα promoter‐SI126 Intron‐OP‐CFTR‐BGH polyA was commercially synthesized by Eurofins (Figure S2). The R6K‐hCEFIα‐OP‐CFTR plasmid has a CpG‐free backbone, which was constructed by cutting the hCMV enhancer‐hEFIα promoter‐SI126 Intron‐OP‐CFTR‐BGH polyA from a pUC‐hCEFIα‐OP‐CFTR plasmid with the EcoRI restriction enzyme. It was then inserted into the pcpgf‐mcs plasmid (Figure S2).

The R6K‐hCMV‐OP‐CFTR plasmid was constructed by replacing the hEFIα promoter of the R6K‐hCEFIα‐OP‐CFTR plasmid with hCMV promoter at SpeI and HindIII restriction enzyme (Brennan & Co) cutting sites (Figure S4).

Plasmid Production and Purification

2.2

The pUC‐hCEFIα‐WT‐CFTR plasmid and pUC‐hCEFIα‐OP‐CFTR plasmid were transformed by heat‐shock and propagated in One Shot TOP10 Chemically Competent E. coli (Thermo Fisher Scientific, Massachusetts, USA). The R6K‐hCEFIα‐WT‐CFTR plasmid and R6K‐hCEFIα‐OP‐CFTR plasmid were transformed by heat‐shock and propagated in E. coli GT115 strain (InvivoGen). For each plasmid, a starting culture was made by inoculating a single colony and antibiotics in 5 mL of Luria‐Bertani broth (LB) (Merck, NJ, USA) at 37°C and 180 rpm for 5–8 h. Ampicillin antibiotics (100 μg/mL) (Merck) were used for the TOP10 E. coli culturing, and the GT115 strain was cultured with Zeocin antibiotics (25 μg/mL) (InvivoGen). Subsequently, 1 mL of the starting culture was added to 400 mL of LB media with specific antibiotics at 37°C and 180 rpm for 15 h. Afterwards, the bacteria were centrifuged for plasmid purification, following the Qiagen Mega kit (Hilden, Germany) protocol.

Agarose Gel Electrophoresis

2.3

The CFTR sequence in CFTR plasmids was confirmed by NheI restriction enzyme (Brennan & Co) double digestion, and the promoters were confirmed by SbfI (Brennan & Co) and HindIII (Brennan & Co) restriction enzyme double digestion. The size of the CFTR plasmids was confirmed by SpeI or HindIII restriction enzyme single digestion. The plasmids without any enzyme treatment were used to confirm the plasmid quality. The DNA fragments were then run on 1% agarose gel for 1 h. The DNA bands were captured using the iBright CL750 Imaging System (Thermo Fisher Scientific). Moreover, the semiquantitative analysis of the DNA band was determined by using ImageJ Fiji software (NIH, MD, USA). The plasmid quality was evaluated by calculating the proportion of the supercoiled plasmid bands to the total bands of the associated plasmid.

Cell Culture

2.4

The CFBE41o‐ human CF bronchial epithelial cell line, derived from a CF patient containing the Phe508del CFTR mutation in homozygosity, was purchased from Merck (Cat num. SCC151). The 16HBE14o‐ human bronchial epithelial cell line, purchased from Merck (Cat num. SCC150) and derived from a healthy donor, was used in this project as a positive control. Both cell lines are homogenous cell populations, which originally were immortalized with the origin‐of‐replication defective SV40 plasmid (pSVori‐) by Dr. Dieter Gruenert. Cells were grown on plates coated with a Fibronectin/Collagen/Bovine Serum Albumin (BSA) extracellular matrix cocktail (Merck) and cultured with Minimum Essential Medium (MEM) cell culture medium (Merck), supplemented with 10% fetal bovine serum (FBS) (Thermo Fisher Scientific), 2‐mM glutamine (Thermo Fisher Scientific), 100‐U/mL penicillin, and 100‐μg/mL streptomycin sulfate (Thermo Fisher Scientific). For CFBE41o‐ cell culture medium, in particular, 1× MEM nonessential amino acid (Merck) was added. Both cell lines were incubated at 5% CO_2_ and 37°C conditions.

Nanoparticles Preparation and Cell Transfection

2.5

The CFBE41o‐ cells were initially seeded onto 24‐well plates at a density of approximately 20,000 cells/cm^2^ and incubated for 24 h. Transfection was carried out once cell confluency reached 50%–70%. The cells were transfected with HPAE polymers and BrPERfect DNA Transfection Reagent (Purchased from Branca Bunús, Dublin, Ireland) [27]. For each transfection, 50 μg of HPAE polymers and 2.5 μg of DNA were separately diluted in 25 μL of 25‐mM sodium acetate (NaOAc) buffer (pH 5.2). Subsequently, the polymer and DNA solutions were mixed to form polyplexes. These polyplexes were then diluted into 450 μL of cell culture medium, resulting in a final volume of 500 μL. Finally, the CFBE41o‐ cells were transfected with the polyplexes in the culture medium.

Western Blot Analysis

2.6

Twenty‐four hours, 48 h, or 7 days after transfection, the cells were washed with DPBS. Then, radioimmunoprecipitation assay (RIPA) buffer (Thermo Fisher Scientific) with protease inhibitor (Thermo Fisher Scientific) was added to each sample to lyse the cells and solubilize the proteins. Protein quantification was performed with the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific). Samples were then loaded in 3%–8% or 7% NuPAGE Tris‐Acetate protein gels (Thermo Fisher Scientific) and later transferred to a nitrocellulose membrane (Thermo Fisher Scientific). For CFTR protein detection, the primary antibody, mouse antihuman CFTR UNC‐596 (University of North Carolina, NC, USA), was diluted at 1:500, and the secondary antibody, antimouse IgG, HRP‐linked antibody (Cell Signaling, MA, USA), was diluted at 1:2000. Human glyceraldehyde 3‐phosphate dehydrogenase (GAPDH) (Abcam, Cambridge, UK) or human α‐tubulin (Cell Signaling) were used to detect the loading control proteins at dilutions of 1:5000 or 1:1000, respectively. The secondary antimouse IgG, HRP‐linked antibody or antirabbit IgG, HRP‐linked antibodies (Cell Signaling) were diluted at 1:5000 or 1:4000, respectively. The protein bands were developed and captured using Pierce ECL Western Blotting Substrate (Thermo Fisher Scientific) and the iBright CL750 Imaging System (Thermo Fisher Scientific). Moreover, the semiquantitative analysis of CFTR (including Band B and Band C) was determined by using ImageJ Fiji software (NIH), as described in previous studies [27].

RT‐qPCR Analysis

2.7

RT‐qPCR was performed to quantify CFTR mRNA expression. Twenty‐four hours, 48 h, or 7 days after transfection, cells in a 24‐well plate were washed with DPBS. The RNeasy Mini Kit (QIAGEN, Hilden, Germany) was used for total RNA extraction from cells. cDNA was generated using a SuperScript III First‐Strand Synthesis SuperMix kit (Thermo Fisher Scientific), and qPCR was performed by using TaqMan PCR mix (Thermo Fisher Scientific) in a QuantStudio 7 Flex System (Thermofisher). CFTR TaqMan primer (Hs00357011_m1) or customized OP‐CFTR TaqMan primer (ART2DCW) (Thermo Fisher Scientific) were used for CFTR mRNA expression detection. GAPDH TaqMan primer (Hs99999905_m1) (Thermo Fisher Scientific) was used as the endogenous control. The delta–delta Ct (2^−∆∆Ct^) method was used for the CFTR mRNA quantification.

Immunofluorescence Analysis

2.8

Cells were washed three times in DPBS, fixed in 4% paraformaldehyde (Thermo Fisher Scientific), and permeabilized in 0.5% Triton X‐100 (Merck). The samples were then blocked in 3% BSA. For the detection of CFTR protein, the primary antibody, mouse antihuman CFTR UNC‐570 (University of North Carolina), was diluted at 1:200. The AlexaFluor568‐labeled secondary goat antimouse IgG (H + L) antibody (Thermo Fisher Scientific) was diluted at 1:800. Cells of the nucleus were stained with DAPI (Abcam) and imaged using an Olympus IX83 fluorescence microscope.

Statistical Analysis

2.9

The mean ± standard deviation (±SD) was expressed for continuous variables. Significance was determined using Student's t‐test. p < 0.05 was considered significant (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001). GraphPad Prism 8.0 (San Diego, CA, USA) was used for the statistical analysis.

Results

3

CFTR Plasmids Construction

3.1

During initial attempts of cloning human CFTR cDNA into the CFTR plasmid, the constructed plasmid underwent rearrangements and mutations during amplification in E. coli [29]. Subsequent findings revealed the presence of cryptic promoters within the human CFTR cDNA sequence, with a major cryptic promoter located at residue 908 of CFTR cDNA. This promoter showed a high degree of homology with the consensus E. coli promoter sequence. This cryptic promoter led to inappropriate expression and toxicity in E. coli , impeding the fermentation process and the amplification of the CFTR plasmid. To circumvent this issue, three base pairs were synonymously substituted in the human CFTR cDNA sequence, removing the major cryptic promoter while preserving the CFTR protein expression (WT‐CFTR) (Figure S1). Based on the WT‐CFTR sequence, the first generated plasmid was successfully constructed and named pUC‐hEFIα‐WT‐CFTR (pWT‐CFTR1) plasmid, which incorporated the hCMV enhancer and the human elongation factor 1‐α (hEF1α) (hCEFIα) promoters (Figure 1a,b).

*CFTR plasmids construction. (a) Workflow of CFTR plasmids construction. (b) pUC‐hCEFIα‐WT‐CFTR, pUC‐hCEFIα‐OP‐CFTR, R6K‐hCEFIα‐WT‐CFTR, and R6K‐hCEFIα‐OP‐CFTR plasmids were digested with NheI restriction enzyme to confirm the CFTR gene bands (4443 bp). These four plasmids were digested with the SpeI restriction enzyme to confirm the plasmid sizes (pUC‐hCEFIα‐WT‐CFTR [7578 bp], pUC‐hCEFIα‐OP‐CFTR [7578 bp], R6K‐hCEFIα‐WT‐CFTR [7369 bp], and R6K‐hCEFIα‐OP‐CFTR [7369 bp]). (c) The plasmid quality was evaluated by measuring the supercoiled fraction of the different CFTR plasmids. (d) The plasmid production levels of different CFTR plasmids. All data were collected from at least three independent experiments (n ≥ 3). All bar graph data are depicted as means ± SDs. *p < 0.05, *p < 0.01; ns: not significant, as compared with pWT‐CFTR1.

The following stage involved CFTR plasmid optimization. By incorporating the S/MAR sequence and replacing the original backbone of the pUC‐hEFIα‐WT‐CFTR plasmid with a CpG‐free backbone, the second plasmid was constructed and named R6K‐hCEFIα‐WT‐CFTR (pWT‐CFTR2) (Figure 2a,b). This newly generated plasmid showed a higher supercoiled fraction compared with the pWT‐CFTR1 plasmid (p = 0.0468), although there was no improvement in plasmid yield (Figure 2c,d). On the other hand, the human CFTR cDNA sequence was further synonymously substituted to improve the translation efficiency, named OP‐CFTR. Furthermore, OP‐CFTR had a CpG‐free profile, which improved plasmid safety. The third constructed plasmid contained the OP CFTR cDNA and was named pUC‐hEFIα‐OP‐CFTR (pOP‐CFTR1) plasmid (Figure 1a,b). Although the pOP‐CFTR1 plasmid did not exhibit DNA quality improvement compared with pWT‐CFTR1, it did show an increase in plasmid production (p = 0.0348) (Figure 1c,d). Combining the advantages of pWT‐CFTR2 plasmid and pOP‐CFTR1 plasmid, a CpG‐depleted CFTR plasmid named R6K‐hCEFIα‐OP‐CFTR (pOP‐CFTR2) was constructed (Figure 2a,b). This newly generated plasmid lacked CpG in its general plasmid sequence and contained the S/MAR sequence. This plasmid demonstrated a higher supercoiled fraction and improved plasmid yield compared with the pWT‐CFTR1 plasmid (Figure 1c,d).

*CFTR protein expression from different CFTR plasmids. (a) CFTR gene expression was assessed in untreated 16HBE14o‐, untreated CFBE41o‐ (UNT), and CFBE41o‐ cells transfected with HPAEs/CFTR plasmids (pWT‐CFTR1, pWT‐CFTR2, pOP‐CFTR1, pOP‐CFTR2) 48 h after transfection via western blot. (b) The semiquantification of western blot was normalized with the loading control protein α‐tubulin. (c) CFTR gene expression from untreated 16HBE14o‐, untreated CFBE41o‐ (UNT), and CFBE41o‐ cells transfected with HPAEs/CFTR plasmids 48 h after transfection was assessed by immunofluorescence, using a 20× fluorescence microscopy objective; the dimension of the scale bar is 50 μm. The yellow arrows show the CFTR protein expression in treated CFBE41o‐ cells with different CFTR plasmids. Western blot and immunofluorescence images are representative of three independent experiments (n = 3). All data were collected from at least three independent experiments (n ≥ 3). All bar graph data are depicted as means ± SDs. **p < 0.01, **p < 0.001; ns: not significant, as compared with pWT‐CFTR1.

CFTR Protein Expression

3.2

Numerous studies have proved that for plasmids, the codon composition and CpG sequence can impact DNA expression efficiency [18, 19, 21]. To substantiate this, the four CFTR plasmids were delivered via HPAEs into CFBE41o‐ cells for the evaluation of CFTR expression through western blot and immunofluorescence, 48 h after transfection. The pWT‐CFTR1 plasmid exhibited a remarkable 4.96‐fold CFTR protein restoration in relation to the 16HBE14o‐ cells (Figure 2a,b). The newly generated plasmid (pWT‐CFTR2) increased 2.16‐fold CFTR protein expression compared with pWT‐CFTR1 (p < 0.01) (Figure 2b). The third plasmid, pOP‐CFTR1, that incorporated the OP‐CFTR sequence demonstrated a 1.70‐fold increase in CFTR protein expression compared with pWT‐CFTR1 (p < 0.01) (Figure 2a,b). The fourth generated CFTR plasmid pOP‐CFTR2, optimized by using a CpG‐free backbone and the OP‐CFTR sequence, achieved 4.10‐fold CFTR expression compared with pWT‐CFTR1 (p < 0.005) (Figure 2a,b). The Band C of the CFTR protein represents the mature form of the CFTR protein, which could be transported to the cell membrane and recapitulate the chloride ion channel function [28]. Therefore, assessing the level of Band C expression of constructed plasmids is crucial for their optimization. The results demonstrated that the pOP‐CFTR2 exhibited 14.54‐fold higher Band C expression compared with 16HBE14o‐ cells, surpassing the initially designed pWT‐CFTR1 plasmid (p = 0.022) (Figure S5).

Additionally, immunofluorescence supported these results, with pOP‐CFTR2 displaying higher CFTR protein expression compared with the other constructed CFTR plasmids (Figure 2c). Given its superior CFTR expression levels, particularly in the mature Band C form, pOP‐CFTR2 emerges as the most promising candidate for further investigation.

hEFIα Promoter and hCMV Promoter Comparison

3.3

The hCMV promoter has been widely employed in the development of gene replacement strategies for lung CF and other diseases due to its robust expression profile and compatibility with various cell types [30, 31, 32]. In comparison, the hEFIα promoter, while possessing higher safety and providing longer‐term expression, exhibits lower potency in early gene expression. The hEFIα promoter was further replaced with the hCMV promoter in the pOP‐CFTR2 plasmid, resulting in the construction of the R6K‐hCMV‐OP‐CFTR2‐CMV (pOP‐CFTR2‐CMV) plasmid for the direct comparison of promoters (Figure 3a,b). Subsequently, the CFTR protein expression levels were evaluated from pOP‐CFTR2 plasmid and pOP‐CFTR2‐CMV at different time points (24 h, 48 h, and 7 days after transfection), using HPAEs as a DNA delivery vector in CFBE41o‐ cells. Moreover, the produced CFTR protein ratio (hEFIα/hCMV promoters) at different time points was analyzed. The western blot results demonstrated that 24 h after transfection (Figure 4a), the hCMV promoter led to higher CFTR expression compared with the hEFIα promoter, with the CFTR protein expression of hEFIα being 0.7804‐fold that of hCMV (Figures 4a and S2). However, 48 h and 7 days after transfection (Figure 4a), the hEFIα promoter surpassed the hCMV promoter in CFTR expression, with the CFTR protein expression of hEFIα being 1.138‐fold and 1.759‐fold that of hCMV, respectively (Figure S2). As a result, OP‐CFTR2 plasmid, offering the advantages of higher and longer‐term CFTR expression capacity and a theoretical safety profile, emerges as the most suitable option for CFTR restoration in CFBE41o‐ cells.

CFTR plasmids construction using different promoters. (a) Schematic representation of the workflow for constructing R6K‐hCMV‐OP‐CFTR plasmids derived from the R6K‐hCEFIα‐OP‐CFTR plasmid. (b) Verification of plasmid constructs by cutting R6K‐hCEFIα‐OP‐CFTR and R6K‐hCMV‐OP‐CFTR plasmids using SbfI and HindIII restriction enzymes to confirm the hCMV enhancer and hEFIα promoter (hCEFIα) (536 bp) and hCMV enhancer and hCMV promoter (516 bp) bands. Confirmation of the plasmid sizes for R6K‐hCEFIα‐OP‐CFTR (7369 bp) and R6K‐hCMV‐OP‐CFTR (7349 bp) was achieved by using the HindIII restriction enzyme.

*Time courses of CFTR protein expression of different CFTR plasmids (pOP‐CFTR2 and pOP‐CFTR2‐CMV). (a) CFTR protein expression of different CFTR plasmids with the loading control protein α‐tubulin was assessed by western blot at 24 h, 48 h, and 7 days after transfection. (b) Quantification of CFTR mRNA levels via RT‐qPCR, normalized to GAPDH and then compared with 16HBE14o‐ cells, following treatment with HPAE complexes containing different CFTR plasmids at 24 h, 48 h, and 7 days after transfection. (c) CFTR protein was semiquantified, normalized with the loading control protein α‐tubulin, and then compared with 16HBE14o‐ cells, after treatment with HPAEs complexing to different CFTR plasmids by western blot 24 h after transfection, 48 h after transfection, and 7 days after transfection. Western blot images are representative of three independent experiments (n = 3). All data were collected from at least three independent experiments (n ≥ 3). All bar graph data are depicted as means ± SDs. ***p < 0.001; ns: not significant, as compared with pOP‐CFTR2‐CMV.

To further confirm the CFTR restoration capacity of the developed pOP‐CFTR2 plasmid in CFBE41o‐ cells, the CFTR protein and mRNA levels were evaluated at 24 h, 48 h, and 7 days after transfection, in comparison with the pOP‐CFTR2‐CMV plasmid (Figure S3). Both plasmids delivered via HPAEs exhibited high CFTR mRNA levels 24 h and 48 h after transfection. The HPAEs/pOP‐CFTR2 demonstrated 9.73‐fold and 13.36‐fold of CFTR mRNA levels compared with healthy 16HBE14o‐ cells after 24 h and 48 h of transfection, respectively. The HPAEs/pOP‐CFTR2‐CMV obtained 66.42‐fold and 46.38‐fold of CFTR mRNA levels of healthy cells 24 h and 48 h after transfection, respectively (Figure 4b). However, 7 days after transfection, CFTR mRNA levels for both polyplexes markedly decreased to levels below 5% of total CFTR mRNA in 16HBE14o‐ cells. Specifically, HPAEs/pOP‐CFTR2 polyplexes displayed 2.1% of CFTR mRNA levels of healthy cells, whereas the HPAEs/pOP‐CFTR2‐CMV polyplex exhibited 1.4% of CFTR mRNA levels compared with healthy cells (Figure 4b). On the other hand, both developed polyplexes achieved promising CFTR protein restoration 24 h, 48 h, and 7 days after transfection when compared with 16HBE14o‐ cells. The HPAEs/pOP‐CFTR2 obtained 13.99‐fold, 25.31‐fold, and 2.26‐fold of CFTR protein restoration relative to healthy cells' protein levels at 24 h, 48 h, and 7 days after transfection, whereas HPAEs/pOP‐CFTR2‐CMV obtained 18.57‐fold, 21.78‐fold, and 1.18‐fold of CFTR healthy cells protein levels at the different times tested (Figure 4a,c).

Discussion

4

Nucleic acid components play a crucial role in nonviral gene replacement therapies for lung CF disease. The aim of this study is to optimize CFTR plasmid DNA that can be delivered as nanoparticles when combined with HPAEs in order to develop an enhanced gene replacement therapy for lung CF. It has been demonstrated that therapeutic benefits can be achieved with ~5%–10% of normal levels of wild‐type CFTR mRNA or protein [10, 11, 12]. To achieve and sustain such therapeutic levels, several strategies have been proposed to construct the CFTR plasmid, such as decreasing the CpG sequence in CFTR plasmids [18, 20, 21], using the S/MAR sequence [24], optimizing the CFTR cDNA sequence [19], and using an efficient promoter [22, 23].

In this study, these strategies were combined to construct an efficient CFTR plasmid. By utilizing a CpG‐free plasmid backbone containing the S/MAR sequence, CFTR protein expression was increased 2.16‐fold, and the immune response triggered by toll‐like receptor 9 (TLR9) was suppressed [20, 21]. In the future, alternative strategies such as utilizing nanoplasmids or minicircles can be employed and compared with CpG‐free plasmids to further improve nucleic acid safety and gene expression efficiency [33, 34]. On the other hand, the CFTR cDNA sequence also plays an important role in the CFTR expression. Using CpG‐depleted and codon‐optimized CFTR cDNA (OP‐CFTR), a 1.70‐fold increase was achieved in CFTR protein compared with the original human CFTR cDNA (WT‐CFTR). Moreover, plasmids containing OP‐CFTR were more stable than those containing WT‐CFTR when amplified in bacteria, as they eliminated all the cryptic promoters by optimizing the CFTR sequence [29].

Promoter modifications represent another strategy to improve and prolong the CFTR protein expression. In earlier in vitro and in vivo studies, and human clinical trials, viral promoters, such as CMV or RSV, were commonly used due to their potent gene expression capabilities [30, 35, 36, 37, 38]. However, viral promoters had limitations due to the transient CFTR expression and safety concerns [21, 22]. To address the limitations and improve the duration and safety profile of CFTR protein expression, human promoters have been used in several pre‐clinical and clinical studies. These promoters include phosphoglycerate kinase (PGK) [19], EF1α [20, 21], or polyubiquitin C (UbC) promoters [18]. Here, the hEF1α promoter was chosen for CFTR expression and compared with another commonly used hCMV promoter. When evaluating the CFTR restoration over a defined time course analysis, pOP‐CFTR2 exhibited higher CFTR protein levels than pOP‐CFTR2‐CMV 48 h and 7 days after transfection, although it initially lagged behind pOP‐CFTR2‐CMV during the first 24 h. These results were anticipated, given that the CMV is an immediate promoter and incapable of sustaining long‐term protein expression in transfected cells [39].

Although the pOP‐CFTR2 exhibited therapeutic CFTR protein expression (10% of WT CFTR) [12] at 7 days after transfection, the levels significantly declined compared with the 48‐h posttransfection. Regardless of reports that the S/MAR sequence may stabilize CFTR plasmid maintenance, the reduction of CFTR expression can be attributed to the plasmid acting as an episome upon transfection into target cells, leading to loss of transfected plasmids during cell division [40]. Despite the reduction, the CFTR protein levels were around 2.26‐fold of CFTR protein restoration relative to 16HBE cells after 7 days of culturing. Therefore, to avoid the impact of cell division, future studies should evaluate the CFTR expression duration of the pOP‐CFTR2 in the air–liquid interface (ALI) culture model and in vivo. Moreover, a recent study indicated that the restored essential epithelial functions of gain‐of‐function CFTR (K978C) were more effective than codon‐optimized forms of CFTR for CF gene therapy, which could be another target in future studies [41].

On the other hand, the pOP‐CFTR2 mRNA levels were lower than the pOP‐CFTR2‐CMV 24 h and 48 h after transfection. Considering the CFTR protein expression, these results may be explained by the CMV promoter having a lower translation efficiency than EF1α, as the promoters play an essential role not only in transcription but also in influencing the translation process [42].

Aside from focusing on the promoters or CFTR sequence, another strategy for enhancing CFTR protein expression involves optimizing the 3′ untranslated region (3′UTR), as this sequence is important in influencing the mRNA stability and regulation [43]. Future work could be directed toward employing an optimized 3′UTR sequence to improve the efficiency of CFTR plasmid expression [18, 20].

Gene delivery vectors also play a critical role in nonviral gene therapy for lung CF. By using HPAEs, when complexed to the pOP‐CFTR2 plasmid, the polyplex system can achieve around 25‐fold CFTR expression and 13.36‐fold of CFTR mRNA levels 48 h after transfection. These results, obtained in CFBE41o‐ cells, are highly promising for lung CF gene therapy, as they demonstrate a restoration of CFTR function in vitro that is 10% of the wildtype CFTR protein levels [12]. Furthermore, this system achieved 2.26‐fold of CFTR protein expression 7 days after transfection when compared with wild‐type human bronchial epithelial cells. In our previous studies, we evaluated the cytotoxicity and CFTR functional activity of the HPAEs–pOP‐CFTR2 system in CFBE41o‐ cells. These experiments demonstrated no detectable cytotoxicity compared with untransfected CFBE41o^−^ cells and achieved approximately 30% recovery of CFTR function relative to 16HBE14o^−^ cells [28]. These results indicate that the developed system could have CFTR protein levels over an extended period, which may overcome the main challenge of gene replacement in lung CF due to the transient CFTR expression [15]. However, the pOP‐CFTR2 plasmid needs to be further evaluated in vivo to assess its efficiency, duration, and safety in CFTR expression [21].

Conclusion

5

The challenges associated with transient CFTR expression were partially addressed through iterative optimization of the bacterial backbone, CFTR DNA sequence, and the use of the EF1α promoter within the CFTR plasmid. The newly developed CpG‐depleted CFTR plasmid demonstrated the potential for high CFTR protein expression in the CFBE41o‐ cell line. Moreover, using the HPAEs complexed to the CpG‐depleted CFTR, the system indicated a 10% increase in CFTR protein expression up to 7 days after transfection, overcoming the hurdle of the transient CFTR expression in gene replacement therapies. Thus, the CpG‐depleted CFTR‐based nanoparticles could be used for nonviral gene therapy treatment in lung CF disease.

Author Contributions

Conceptualization, B.Q. and I.L.‐S.; Methodology, B.Q. and X.W.; Software, B.Q., Y.L., and Z.H.; Validation, B.Q.; Formal Analysis, B.Q.; Investigation, X.W., Z.H., and Y.L.; Resources, Y.L, I.L.‐S., and W.W.; Data Curation, B.Q.; Writing – Original Draft Preparation, B.Q.; Writing – Review and Editing, B.Q., ML, X.W., Y.L., Z.H., I.L.‐S., and W.W.; Visualization, B.Q.; Supervision, I.L.‐S. and W.W.; Project Administration, I.L.‐S. and W.W.; Funding Acquisition, I.L.‐S. and W.W.

Funding

This work was supported by Enterprise Ireland–Disruptive Technologies Innovation Fund (DTIF) (164846/RR) and the Chinese Government–China Scholarship Council (202108300017).

Ethics Statement

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: (a) The major cryptic promoter located in the human CFTR cDNA sequence. (b) The strategy of removing major cryptic promoter by using synonymous substitution. Figure S2: Workflow of R6K‐hCEFIα‐WT‐CFTR (or R6K‐hCEFIα‐OP‐CFTR) plasmids construction. (I) hCMV enhancer‐hEFIα promoter‐SI126 Intron‐WT‐CFTR‐BGH polyA (or hCMV enhancer‐hEFIα promoter‐SI126 Intron‐OP‐CFTR‐BGH polyA) was obtained from pUC‐hCEFIα‐WT‐CFTR plasmid (OR hCMV enhancer‐hEFIα promoter‐SI126 Intron‐OP‐CFTR‐BGH polyA) by EcoRI restriction enzyme digestion. (II) The CpG‐free backbone, mainly containing the R6K origin Zeocin‐resistant sequence, was excised by the EcoRI restriction enzyme. (III) Ligation of hCMV enhancer‐hEFIα promoter‐SI126 Intron‐WT‐CFTR‐BGH polyA (or hCMV enhancer‐hEFIα promoter‐SI126 Intron‐OP‐CFTR‐BGH polyA) with CpG‐free backbone at EcoRI restriction enzyme cut site to construct R6K‐hCEFIα‐WT‐CFTR (or R6K‐hCEFIα‐OP‐CFTR) plasmids. Figure S3: CFTR protein expression ratio of R6K‐hCEFIα‐OP‐CFTR/R6K‐hCMV‐OP‐CFTR (hCEFIα/hCMV) of CFBE41o‐ cells by 24 h, 48 h, and 7 days after transfection using semiquantified western blot results (n = 3). Figure S4: Workflow of R6K‐hCMV‐OP‐CFTR plasmids construction. (I) hCMV promoter obtained by PCR using RG216476 as a template. (II) Removal of the hEFIα promoter in R6K‐hCEFIα‐OP‐CFTR and ligation of the hCMV promoter at SpeI and HindIII restriction enzyme cutting sites to produce R6K‐hCMV‐OP‐CFTR. Figure S5: The semiquantification of the western blot and the CFTR Band C was normalized with the loading control protein GAPDH.