Durable Transgene Expression Driven by CpG-Free and -Containing Promoters in Plasmid DNA with CpG-Free Backbone

Abstract

The duration of transgene expression from plasmid DNAs is important for gene therapy with nonviral vectors. In the present study, various cytosine-phosphate-guanine (CpG)-free and CpG-containing transcription regulatory sequences were introduced into plasmid DNAs with a CpG-free backbone. The transgenes encoding mouse secreted alkaline phosphatase and Gaussia princeps luciferase, which are both apparently non-immunogenic, were used as reporters. The plasmid DNAs were injected by the hydrodynamics-based method, and the expression was monitored for 28 d. All transcription regulatory sequences achieved long-term expression, with different expression levels depending on the sequences themselves. These results suggested that durable transgene expression at the proper level can be achieved with plasmid DNAs containing the CpG-free backbone.

The clinical implementation of gene therapy requires excellent systems that enable the adequate expression of the therapeutic transgene in target organs. Thus, sophisticated systems should be promptly established. Nonviral delivery of transgenes is attractive from the viewpoint of safety, although the transgene expression is inefficient and temporal.^1–7) These problems must be solved to achieve the clinical use of nonviral gene therapy.

Using the mouse as a model animal, hydrodynamics-based tail vein injection of naked plasmid DNA has proven to be a safe and efficient procedure for DNA delivery, and the liver is the target organ.^8,9) However, it also has the drawback that the expression from the administered DNA soon disappears. We previously reported that plasmid silencing and reduction of the exogenous DNA in the liver are responsible for the decline in expression.^10,11) In contrast to “usual” plasmid DNAs, prolonged transgene expression has been demonstrated by the hydrodynamics-based injection of some plasmid and plasmid-like DNAs. The cytosine-phosphate-guanine (CpG)-free plasmid lacks CpG dinucleotides, which are present much less frequently in mammalian cells than in bacterial cells.¹²⁾ In addition, the bacterial (plasmid) backbone is deleted in the minicircle DNA.^13,14) Although long-acting DNAs including other types of plasmids have been reported, the detailed molecular mechanisms underlying the durable expression have not been demonstrated.^15,16)

We previously injected the pCpGfree-mbSeap plasmid (plasmid 1) containing the CpG-free backbone and various CpG-lacking components: the CMV enhancer, human EF1α promoter, SV40 poly(A) signal, β-globin matrix attachment region (MAR), and interferon (IFN)-β MAR¹⁷⁾ (Table 1). The sole CpG-containing moiety is the mouse Seap transgene from the BALB/c strain with 63 CpG sites, and this gene encodes the endogenous secreted alkaline phosphatase (SEAP).^18,19) Despite the presence of CpG sequences, the SEAP expression from the plasmid was maintained in mouse liver. We subsequently found that the removal of the two MAR sequences from the plasmid affected the expression level, but not its duration. In addition, the insertion of the expression cassette of the plasmid into other plasmid backbones containing CpG dinucleotides resulted in transient SEAP expression.¹⁷⁾ Thus, we concluded that the backbone sequence of the CpG-free plasmid was mainly responsible for the sustained expression.

Table 1. Plasmid DNAs Used in This Study^a)

Plasmid	Upstream	Enhancer/promoter	Transgene	Poly(A)	Downstream	Backbone
1 pCpGfree-mbSeap	G-MAR*	CMV enh*+EF1α pro*	mouse Seap	SV40 pA*	I-MAR*	CpG-free*
2 pCpGfree-CMV pro-Seap	G-MAR*	CMV pro	mouse Seap	SV40 pA*	I-MAR*	CpG-free*
3 pCpGfree-EF1α pro-Seap	G-MAR*	EF1α pro*	mouse Seap	SV40 pA*	I-MAR*	CpG-free*
4 pCpGfree-CMV enh-Alb pro-Seap	G-MAR*	CMV enh*+Alb pro	mouse Seap	SV40 pA*	I-MAR*	CpG-free*
5 pCpGfree-Alb pro-Seap	G-MAR*	Alb pro	mouse Seap	SV40 pA*	I-MAR*	CpG-free*
6 pCpGfree-CMV enh-Alb pro-Seap MAR(−)		CMV enh*+Alb pro	mouse Seap	SV40 pA*		CpG-free*
7 pCpGfree-CMV enh-EF1α pro-Gluc	G-MAR*	CMV enh*+EF1α pro*	Gluc	SV40 pA*	I-MAR*	CpG-free*
8 pCpGfree-CMV pro-Gluc	G-MAR*	CMV pro	Gluc	SV40 pA*	I-MAR*	CpG-free*
9 pCpGfree-EF1α pro-Gluc	G-MAR*	EF1α pro*	Gluc	SV40 pA*	I-MAR*	CpG-free*
10 pCpGfree-Alb pro-Gluc	G-MAR*	Alb pro	Gluc	SV40 pA*	I-MAR*	CpG-free*

a) Alb pro, mouse minimal albumin promoter; CMV enh, CMV enhancer; CMV pro, CMV promoter; EF1α pro, human EF1α core promoter; G-MAR, human β-globin gene matrix attachment region; I-MAR, the human interferon-β gene matrix attachment region. Asterisks indicate no CpG sites within the DNA elements.

The control of the transgene expression strength is another important issue. To achieve adequate transgene expression levels in clinical use, appropriate promoters and enhancers must be selected. Previously, Ando et al. and Yin et al. reported transgene expression from plasmid DNAs containing the CpG-free backbone and various transcriptional regulatory sequences.^20,21) They showed that long-term transgene expression was driven by the human ROSA26 promoter containing CpG sequences (598 CpG sequences within 5.8 kbp). Thus, the complete deletion of the CpG nucleotides from the transgene and transcription regulatory sequence seems unnecessary, and this is desirable for steady and proper expression levels of therapeutic genes.

In this study, we examined the effects of promoters on the duration and strength of transgene expression. We constructed plasmid DNAs containing various promoters with or without the CpG-free CMV enhancer. SEAP and Gaussia princeps luciferase (Gluc) were used as reporters. We found that all of the transcription regulatory sequences examined could drive long-term transgene expression. These results suggested that the backbone sequence of the CpG-free plasmid contributes to durable expression and that many transcription regulatory sequences, including the albumin promoter, are available for the desired and appropriate expression levels.

MATERIALS AND METHODS

Materials

The pCpGfree-mbSeap plasmid was constructed in our laboratory.¹⁹⁾ Oligodeoxyribonucleotides were obtained from Fasmac (Atsugi, Japan) and Hokkaido System Science (Sapporo, Japan) in purified forms. Plasmid DNAs were purified with a Qiagen (Hilden, Germany) EndoFree Plasmid Maxi kit.

Animals

Male BALB/c mice were purchased from Japan SLC (Shizuoka, Japan). The mice were maintained in a temperature-controlled room (23°C) under a 12 : 12 h light/dark photocycle and specific pathogen-free conditions, with food and water provided ad libitum. All animal studies were approved by the Hiroshima University Institutional Animal Care and Use Committee, and conducted according to the Hiroshima University Animal Experimentation Regulations.

Construction of Plasmid DNAs

The DNA containing the CMV promoter was amplified from pCMV-mCBP,²²⁾ and the pCpGfree-CMV pro-Seap plasmid was constructed by replacing the CMV enhancer plus the EF1α promoter of pCpGfree-mbSeap with the PCR product. The CMV enhancer was removed from pCpGfree-mbSeap, to yield pCpGfree-EF1α pro-Seap. The pCpGfree-CMV enh-Alb pro-Seap plasmid was generated by the replacement of the EF1α promoter of pCpGfree-mbSeap with the PCR product containing the mouse albumin promoter of pG5-ALB-luc-G5.²³⁾ pCpGfree-CMV enh-Alb pro-Seap MAR(−) was constructed from pCpGfree-mbSeap MAR(−) by similar procedures.¹⁷⁾ Replacement of the CMV enhancer plus EF1α promoter of pCpGfree-mbSeap with the mouse albumin promoter of pG5-ALB-luc-G5 yielded pCpGfree-Alb pro-Seap.

The plasmids bearing the Gluc gene were constructed by replacement of the Seap gene fragment with the Gluc gene fragment, synthesized by GenScript (Piscataway, NJ, U.S.A.).

SEAP Assay

The plasmid DNA (1 or 10 pmol, see figure and table legends) was administered to BALB/c mice (male, six weeks old) under isoflurane anesthesia by the hydrodynamics-based procedure.^8,9) Blood samples (15–30 µL) were collected on days 1, 2, 7, 14, 21, and 28 after injection, and then stored for 2 h at room temperature. The supernatant obtained by centrifugation at 2000 rpm for 20 min at 4°C was diluted and analyzed for alkaline phosphatase activity with a SEAP Reporter Gene Assay, Chemiluminescent (Sigma-Aldrich, St. Louis, MO, U.S.A.). The phosphatase activity was determined by comparison to a standard curve obtained with human SEAP.

Luciferase Assay

Gaussia luciferase activity was measured with a BioLux Gaussia Luciferase Assay kit (New England Biolabs, Ipswich, MA, U.S.A.). The supernatant was obtained by centrifugation of the blood sample as described above and diluted to 20 times by phosphate-buffered saline. The diluted solution (20 µL) was added to the GLuc assay solution (50 µL) containing the substrate, coelenterazine, as recommended by the manufacturer’s protocol. The luminescence was promptly measured with a luminometer (AB-2250 Luminescencer-MCA, Atto, Tokyo, Japan).

RESULTS

Expression of SEAP Reporter Transgene from CpG-Free Backbone Plasmid DNAs

First, we constructed various plasmid DNAs from the parental plasmid, pCpGfree-mbSeap (plasmid 1).¹⁹⁾ This plasmid DNA possesses the CpG-free backbone and components (CMV enhancer, EF1α promoter, poly(A) signal, two MARs) and the CpG-containing Seap transgene (Table 1). The gene is derived from BALB/c mice and SEAP is non-immunogenic.²⁴⁾ To achieve adequate transgene expression in clinical use, the selection of the appropriate promoter and enhancer combination is highly important. Thus, we constructed plasmid DNAs with altered regulatory sequences. We selected the CMV (82 bp) and mouse albumin (319 bp) promoters, which have 8 and 3 CpG sequences, respectively. Plasmid DNAs containing either the EF1α, CMV, or albumin promoter without the CpG-free CMV enhancer were constructed. In addition, the plasmid with the CpG-free CMV enhancer plus the albumin promoter was also prepared. The plasmid DNAs used in this study are summarized in Table 1 (plasmids 1 to 6 for the SEAP experiments).

We then administered the SEAP plasmid DNAs (1 pmol) into mouse liver by hydrodynamics-based injection.^8,9) We judged expression as durable when the activity was detected on day 28. As expected, the expression from pCpGfree-mbSeap (plasmid 1) was durable, consistent with the previous reports^17,19) (Table 2 and Fig. 1). In addition, plasmid 4 containing the albumin promoter plus CMV enhancer also showed long-term SEAP expression. Thus, the CpG-free EF1α promoter was exchangeable with the native albumin promoter.

Table 2. Alkaline Phosphatase Activities after Administration of Plasmids Containing the CpG-Free Backbone ^a)

Plasmid	Mouse No.	Time post-injection (days)
		1	2	7	14	21	28
pCpGfree-mbSeap (plasmid 1)	1	8.09	4.25	1.33	1.10	0.46	0.38
	2	7.66	8.48	0.96	0.33	0.23	0.60
	3	8.60	7.56	0.90	0.26	0.27	0.45
	mean±S.D.	8.12±0.38	6.76±1.82	1.06±0.19	0.56±0.38	0.32±0.10	0.48±0.09
pCpGfree-CMV pro-Seap (plasmid 2)	1	ND b)	0.16	0.09	ND	ND	ND
	2	0.13	0.11	0.06	ND	ND	ND
	mean	(0.13)	0.14	0.08	ND	ND	ND
pCpGfree-EF1α pro-Seap (plasmid 3)	1	ND	0.09	0.06	ND	ND	ND
	2	0.10	0.04	0.09	ND	ND	ND
	mean	(0.10)	0.06	0.07	ND	ND	ND
pCpGfree-CMV enh-Alb pro-Seap (plasmid 4)	1	ND	13.60	0.75	0.84	1.18	0.66
	2	4.60	9.61	1.80	1.84	1.10	0.99
	3	6.22	4.60	7.60	4.80	5.10	4.00
	mean±S.D.	5.41	9.27±3.68	3.38±3.01	2.49±1.68	2.46±1.87	1.88±1.50
pCpGfree-Alb pro-Seap (plasmid 5)	1	ND	0.01	0.03	ND	ND	ND
	2	0.10	0.08	0.08	ND	ND	ND
	mean	(0.10)	0.05	0.05	ND	ND	ND
pCpGfree-CMV enh-Alb pro-Seap MAR(−) (plasmid 6)	1	ND	14.40	2.99	4.38	6.10	2.14
	2	2.15	5.46	1.20	1.14	1.09	1.14
	3	5.90	2.80	1.40	3.10	2.10	2.40
	mean±S.D.	4.03	7.55±4.96	1.86±0.80	2.87±1.33	3.10±2.16	1.89±0.54
Saline	1	0.05	0.06	0.10	0.10	0.05	0.04
	2	0.15	0.05	0.05	0.08	0.05	0.05
	3	0.06	0.04	0.12	0.07	0.06	0.05
	4	0.09	0.08	0.11	0.08	0.05	0.11
	mean±S.D.	0.09±0.04	0.06±0.01	0.09±0.03	0.08±0.01	0.05±0.005	0.06±0.02

a) One pmol of plasmid was injected. Activities are shown as units/mL serum. b) Not determined.

In contrast, no serum phosphatase activities over the background were observed at day 7 when enhancer-lacking plasmid DNAs (2, 3, 5) with the CMV, albumin, and EF1α promoters were administered (Table 2).

Moreover, the SEAP expression from plasmid 6, containing the albumin promoter plus CMV enhancer but lacking the MARs, was similar to that from plasmid 4 (Fig. 1). These results indicated again that transgene expression was durable without the MARs. However, in contrast to the previous results,¹⁷⁾ the depletion of the MAR sequences seemed to have little, if any, impact on the SEAP expression level in the case of the albumin promoter plus the CMV enhancer.

Fig. 1. Alkaline Phosphatase Activities after Administration of Plasmid DNAs Containing the CpG-Free Backbone

Plasmid 1 (closed circles), plasmid 4 (open circles), and plasmid 6 (open squares) (1 pmol each) were injected into BALB/c mice by the hydrodynamics-based method. The alkaline phosphatase activities in the serum were measured on days 1, 2, 7, 14, 21, and 28 after injection. Two or three mice were analyzed per experimental group. Data are extracted from Table 2 and expressed as means+S.D. (standard deviation). The background phosphatase activity upon the injection of saline is shown by the dashed line.

These results suggested that the albumin promoter plus the CMV enhancer, with or without the MARs, achieved durable transgene expression. The fact that only weak expression was observed for plasmid 2 containing the CMV promoter at day 2, but not at day 7 (Table 2), suggested that long-term expression from the CpG-free plasmid might be dependent on the promoter, in contrast to our hypothesis. However, the CMV promoter is transiently activated by the hydrodynamic injection method itself.^11,25) We used the Seap gene as a transgene encoding an endogenous, non-immunogenic protein. However, the background phosphatase activity was high due to the presence of other phosphatases in the serum and could mask low SEAP expression. Thus, further analysis was required using a more sensitive reporter with low background.

Expression of Gluc Reporter Transgene from CpG-Free Backbone Plasmid DNAs

We next constructed plasmid DNAs containing the gene encoding Gluc (luciferase from Gaussia princeps), since the gene product is reportedly non-immunogenic.²¹⁾ To monitor the time courses of the expression driven from the CMV, albumin, and EF1α promoters, plasmid DNAs 8–10 were designed. These plasmid DNAs were injected into mice by the hydrodynamics-based method. Plasmid 7 was used as the control.

Interestingly, Gluc expression was durable for all plasmid DNAs examined, except for the plasmid containing the EF1α promoter without the enhancer (plasmid 9), upon the injection of 1 pmol of plasmid (Fig. 2 and data not shown for plasmid 9). However, durable expression was also observed when a 10 pmol dose of plasmid 9 was administered (Fig. 2). Thus, the Gluc assay indicated that the three promoters without the CMV enhancer are also useful for long-term expression. Taken together, all of the transcription regulatory sequences examined could drive sustained transgene (Seap or Gluc) expression. Thus, the CpG-free backbone would be the determinant factor for the duration.

Fig. 2. Gluc Activities after Administration of Plasmid DNAs Containing the CpG-Free Backbone

Plasmid 7 (closed circles), plasmid 8 (open squares), and plasmid 10 (open circles) (1 pmol each) or plasmid 9 (open triangles, 10 pmol) were injected into BALB/c mice by the hydrodynamics-based method. The Gluc activities in the serum were measured on days 1, 2, 7, 14, 21, and 28 after injection. Two (days 2 and 28 of plasmid 8 and day 2 of plasmid 9) or three (others) mice were analyzed per experimental group. Data are expressed as means+S.D. (standard deviation). The background Gluc activity upon the injection of saline is shown by the dashed line.

DISCUSSION

The objective of this study was to examine the duration of transgene expression when various transcription regulatory sequences were placed in plasmids containing the CpG-free backbone. As shown in Figs. 1 and 2, all promoters with or without the CMV enhancer drove durable transgene expression. The albumin and CMV promoters contain CpG sequences, and therefore these results indicated that CpG deletion is unessential, in line with the reports by Ando et al. and Yin et al.^20,21) Thus, the anti-silencing element of the pCpG-free vectors was the backbone lacking CpG dinucleotides. In addition, we clarified that a variety of transcription regulatory sequences are available for the proper expression levels required in clinical use.

As discussed previously, the pCpG-free and minicircle vectors seemed to contain a common structure. The head and tail moieties of the expression cassette are joined directly, without the plasmid backbone, in the minicircle DNA. Meanwhile, the expression durability was lost when a 1-kb DNA linker was present between the head and tail.²⁶⁾ Thus, the minicircle DNA resembles the pCpG-free plasmid, in that both have no/few CpG sites except for the expression cassette.

The CpG sequences in the plasmid DNAs are potential sites for the methylation of C bases. However, the involvement of C methylation in transgene silencing is still open to debate. Experimental results against and supporting this hypothesis have been reported.^10,14,27,28) The reason for the durable transgene expression from the pCpG-free and minicircle vectors might be due to the binding and modification of proteins, including histones.^11,29,30)

We deleted the two MARs from plasmid 4 to yield plasmid 6, and compared the transgene expression. The deletion did not affect the duration of expression, as observed in the previous study.¹⁷⁾ However, the deletion seemed to have no impact on the expression level, in contrast to the previous report (Fig. 1). The CMV enhancer plus the albumin promoter and the CMV enhancer plus the EF1α promoter were used in the present and previous studies, respectively. The discrepancy between the two studies appeared to be due to the different promoters. Ho et al. compared the impacts of the chicken lysozyme and human IFN-β MARs on the transgene expression levels from integrated plasmids with different promoters in cultured cells.³¹⁾ In their constructs, the expression cassettes were placed between two identical MARs, and they found that the effects of the MAR were dependent on the promoter. Thus, our discrepant results are attributable to the promoter difference.

Although long-term expression was detected with all of the promoters examined (with or without the CMV enhancer), the expression patterns were different. In particular, the expression from plasmids 8 and 9 (the CMV and EF1α promoters, respectively) decreased by more than one order of magnitude within a week (Fig. 2). The hydrodynamics-based injection method itself activates transgene expression, and this might cause the apparent decrease in the expression for plasmid 8.^11,25) However, the expression time course of plasmid 7, containing the CMV enhancer plus the EF1α promoter, was different from that of plasmid 9 and thus a complete understanding of the expression patterns remains to be attained.

In this study, the hydrodynamics-based injection method was employed for the delivery of plasmid DNAs into liver. When the firefly luciferase gene was used a transgene, luciferase activity in liver was more than 10²-fold higher than those in other organs.⁸⁾ Thus, the additional alkaline phosphatase and Gluc activities over background would reflect transgene expression in liver. Although this method has been used as one of the standard injection techniques for plasmid DNAs in many laboratories, other delivery systems would be required for the clinical use of nonviral gene therapy. Thus, it is important to examine whether the same or similar conclusions are drawn when different delivery methods, such as lipoplexes and polyplexes, are used and the transgenes are expressed in different organs. Since the albumin promoter is liver-specific, other tissue-specific promoters should be used depending on the target organs.

The control of the transgene expression level is important for safe gene therapy. We showed that various expression levels are achievable by adequate transcriptional regulatory sequences that maintain the expression. Previously, we showed that the introduction of left-handedly curved sequences into plasmid DNAs and the use of positive feedback systems with plasmid-specific transcriptional activation enhanced transgene expression.^23,30,32) Since the CpG-free backbone seems to be the key component of the duration, these factors may be incorporated into the long-term expression systems.

In conclusion, CpG-free and -containing promoters, with or without the CpG-free CMV enhancer, drove prolonged transgene expression when the plasmid DNA had the CpG-free backbone. These data indicated that durable and appropriate therapeutic transgene expression is feasible when the optimal transcription regulatory sequence is chosen.

Acknowledgments

This work was supported in part by the Japan Society for the Promotion of Science (JSPS) KAKENHI Grant Number JP 25282144 (H.K.).

Conflict of Interest

The authors declare no conflict of interest.

REFERENCES

• 1) Goodwin T, Huang L. Nonviral vectors: We have come a long way. Adv. Genet., 88, 1–12 (2014).
• 2) Hatakeyama H, Akita H, Harashima H. The polyethyleneglycol dilemma: Advantage and disadvantage of PEGylation of liposomes for systemic genes and nucleic acids delivery to tumors. Biol. Pharm. Bull., 36, 892–899 (2013).
• 3) Hill AB, Chen M, Chen CK, Pfeifer BA, Jones CH. Overcoming gene-delivery hurdles: Physiological considerations for nonviral vectors. Trends Biotechnol., 34, 91–105 (2016).
• 4) Kamiya H, Akita H, Harashima H. Pharmacokinetic and pharmacodynamic considerations in gene therapy. Drug Discov. Today, 8, 990–996 (2003).
• 5) Keles E, Song Y, Du D, Dong WJ, Lin Y. Recent progress in nanomaterials for gene delivery applications. Biomater. Sci., 4, 1291–1309 (2016).
• 6) Oliveira C, Silveira I, Veiga F, Ribeiro AJ. Recent advances in characterization of nonviral vectors for delivery of nucleic acids: impact on their biological performance. Expert Opin. Drug Deliv., 12, 27–39 (2015).
• 7) Xiang Y, Oo NNL, Lee JP, Li Z, Loh XJ. Recent development of synthetic nonviral systems for sustained gene delivery. Drug Discov. Today, 22, 1318–1335 (2017).
• 8) Liu F, Song Y, Liu D. Hydrodynamics-based transfection in animals by systemic administration of plasmid DNA. Gene Ther., 6, 1258–1266 (1999).
• 9) Zhang G, Budker V, Wolff JA. High levels of foreign gene expression in hepatocytes after tail vein injections of naked plasmid DNA. Hum. Gene Ther., 10, 1735–1737 (1999).
• 10) Ochiai H, Harashima H, Kamiya H. Intranuclear disposition of exogenous DNA in vivo: silencing, methylation and fragmentation. FEBS Lett., 580, 918–922 (2006).
• 11) Ochiai H, Fujimuro M, Yokosawa H, Harashima H, Kamiya H. Transient activation of transgene expression by hydrodynamics-based injection may cause rapid decrease in plasmid DNA expression. Gene Ther., 14, 1152–1159 (2007).
• 12) Mitsui M, Nishikawa M, Zang L, Ando M, Hattori K, Takahashi Y, Watanabe Y, Takakura Y. Effect of the content of unmethylated CpG dinucleotides in plasmid DNA on the sustainability of transgene expression. J. Gene Med., 11, 435–443 (2009).
• 13) Chen ZY, He CY, Ehrhardt A, Kay MA. Minicircle DNA vectors devoid of bacterial DNA result in persistent and high-level transgene expression in vivo. Mol. Ther., 8, 495–500 (2003).
• 14) Chen ZY, He CY, Meuse L, Kay MA. Silencing of episomal transgene expression by plasmid bacterial DNA elements in vivo. Gene Ther., 11, 856–864 (2004).
• 15) Wolff LJ, Wolff JA, Sebestyén MG. Effect of tissue-specific promoters and microRNA recognition elements on stability of transgene expression after hydrodynamic naked plasmid DNA delivery. Hum. Gene Ther., 20, 374–388 (2009).
• 16) Aliño SF, Crespo A, Dasí F. Long-term therapeutic levels of human alpha-1 antitrypsin in plasma after hydrodynamic injection of nonviral DNA. Gene Ther., 10, 1672–1679 (2003).
• 17) Kanda GN, Miyamoto S, Kobayashi M, Matsuoka I, Harashima H, Kamiya H. Anatomy of plasmid DNAs with anti-silencing elements. Int. J. Pharm., 464, 27–33 (2014).
• 18) Kanda GN, Togashi R, Harashima H, Kamiya H. Effects of endogenous proteins and microRNA target sequence in a positive feedback system. Biol. Pharm. Bull., 35, 1534–1538 (2012).
• 19) Togashi R, Harashima H, Kamiya H. Correlation between transgene expression and plasmid DNA loss in mouse liver. J. Gene Med., 15, 242–248 (2013).
• 20) Ando M, Takahashi Y, Nishikawa M, Watanabe Y, Takakura Y. Constant and steady transgene expression of interferon-γ by optimization of plasmid construct for safe and effective interferon-γ gene therapy. J. Gene Med., 14, 288–295 (2012).
• 21) Yin Y, Takahashi Y, Ebisuura N, Nishikawa M, Takakura Y. Removal of transgene-expressing cells by a specific immune response induced by sustained transgene expression. J. Gene Med., 16, 97–106 (2014).
• 22) Kanda G, Ochiai H, Harashima H, Kamiya H. CREB-binding protein transcription activation domain for enhanced transgene expression by a positive feedback system. J. Biotechnol., 157, 7–11 (2012).
• 23) Ochiai H, Harashima H, Kamiya H. Positive feedback system provides efficient and persistent transgene expression. Mol. Pharm., 7, 1125–1132 (2010).
• 24) Mælandsmo GM, Ross PJ, Pavliv M, Meulenbroek RA, Evelegh C, Muruve DA, Graham FL, Parks RJ. Use of a murine secreted alkaline phosphatase as a non-immunogenic reporter gene in mice. J. Gene Med., 7, 307–315 (2005).
• 25) Nishikawa M, Nakayama A, Takahashi Y, Fukuhara Y, Takakura Y. Reactivation of silenced transgene expression in mouse liver by rapid, large-volume injection of isotonic solution. Hum. Gene Ther., 19, 1009–1020 (2008).
• 26) Lu J, Zhang F, Xu S, Fire AZ, Kay MA. The extragenic spacer length between the 5′ and 3′ ends of the transgene expression cassette affects transgene silencing from plasmid-based vectors. Mol. Ther., 20, 2111–2119 (2012).
• 27) Pringle IA, Raman S, Sharp WW, Cheng SH, Hyde SC, Gill DR. Detection of plasmid DNA vectors following gene transfer to the murine airways. Gene Ther., 12, 1206–1214 (2005).
• 28) Argyros O, Wong SP, Niceta M, Waddington SN, Howe SJ, Coutelle C, Miller AD, Harbottle RP. Persistent episomal transgene expression in liver following delivery of a scaffold/matrix attachment region containing non-viral vector. Gene Ther., 15, 1593–1605 (2008).
• 29) Riu E, Chen ZY, Xu H, He CY, Kay MA. Histone modifications are associated with the persistence or silencing of vector-mediated transgene expression in vivo. Mol. Ther., 15, 1348–1355 (2007).
• 30) Fukunaga S, Kanda G, Tanase J, Harashima H, Ohyama T, Kamiya H. A designed curved DNA sequence remarkably enhances transgene expression from plasmid DNA in mouse liver. Gene Ther., 19, 828–835 (2012).
• 31) Ho SCL, Mariati, Yeo JH, Fang SG, Yang Y. Impact of using different promoters and matrix attachment regions on recombinant protein expression level and stability in stably transfected CHO cells. Mol. Biotechnol., 57, 138–144 (2015).
• 32) Nishihara M, Kanda GN, Suzuki T, Yamakado S, Harashima H, Kamiya H. Enhanced transgene expression by plasmid-specific recruitment of histone acetyltransferase. J. Biosci. Bioeng., 123, 277–280 (2017).