Gene gun bombardment with dna-coated gold particles is a potential alternative to hydrodynamics-based transfection for delivering genes into superficial hepatocytes

HUMAN GENE THERAPY 19:391–395 (April 2008)
Mary Ann Liebert, Inc.
DOI: 10.1089/hum.2007.152
Technical Report
Gene Gun Bombardment with DNA-Coated Gold Particles Is a Potential Alternative to Hydrodynamics-Based Transfection for Delivering Genes into Superficial Hepatocytes MING-LING CHANG,1 JENG-CHANG CHEN,2 CHAU-TING YEH,1 MING-YU CHANG,3 CHUN-KAI LIANG,1 CHENG-TANG CHIU,1 DENG-YN LIN,1 and YUN-FAN LIAW1 ABSTRACT
Although in vivo nonviral gene delivery to the liver is critical for hepatic gene therapy, there are a number
of technical obstacles. Enhanced green fluorescent protein (EGFP)-encoding DNA was coated onto gold par-
ticles (gold–DNA), dissolved in phosphate-buffered saline (pure DNA), and prepared as a polymer adjuvant
(jetPEI)–galactosidase solution (polymer–DNA). Murine liver transfection was attempted by nonviral ap-
proaches, which included hydrodynamics-based transfection (HBT) of pure DNA, transport and transhepatic
injection of polymer–DNA, and gene gun bombardment with pure DNA, gold–DNA, and polymer–DNA. Only
HBT and gene gun bombardment yielded significant numbers of EGFPϩ hepatocytes. With the exception of
the edge of the liver, HBT had a whole-liver transfection rate of 20% under optimized conditions. HBT re-
sulted in marked hepatic infarctions, most prominently at the edge of the liver. For gene gun bombardment,
the transfection rate was pressure dependent and limited to 15% for gold–DNA. Triple or quadruple bom-
bardment at 30 psi resulted in a transfection rate comparable to that of a single bombardment at higher pres-
sure, but was associated with minimal scattered hepatic necrosis. The EGFPϩ hepatocytes were located mainly
in the superficial layers. We conclude that both HBT and gene gun bombardment yielded efficient murine
hepatocyte transfection in vivo. Severe hepatic infarction impedes foreign gene expression in the superficial
hepatocytes after HBT. Repeated bombardment with gold–DNA, using an accelerated particle gene gun at 30
psi, is a potential alternative to HBT for delivering genes to superficial hepatocytes in vivo, although gold-re-
lated hepatic necrosis is a persistent problem.
INTRODUCTION
attractive because it can be manipulated by standard recombi-nant DNA techniques and delivered by both chemical and phys- IN VIVO gene delivery to the liver is critical for both experi- ical means. However, chemical approaches such as circulating
mental and clinical applications. At present, there are two cationic vectors can attract serum proteins, leading to dynamic main modes for gene delivery: viral and nonviral (Dobson, changes in their physicochemical properties and diminished 2006). Viral vectors confer more effective expression than syn- transfection efficiency (Nishikawa and Huang, 2001). Physical thetic molecular gene vectors, albeit at the expense of infection approaches to gene transfer have improved and become as ef- and immunogenicity (Azzam and Domb, 2004). To lessen the fective as viral vectors (Wells, 2004). Hydrodynamics-based potential biohazards of viral vectors, naked DNA is considered transfection (HBT) of hepatocytes has been reported to produce 1Liver Research Center and Department of Hepatogastroenterology, Chang Gung Memorial Hospital, Taoyuan, Taiwan; and Chang Gung Uni- 2Department of Surgery, Chang Gung Children’s Hospital, Taoyuan, Taiwan 33305.
3Division of Pediatric Critical Care and Emergency Medicine, Chang Gung Children’s Hospital, Taoyuan, Taiwan 33305.
CHANG ET AL.
a satisfactory transfection efficiency in mice (Wolff and Bud- vivo transfection via the portal vein (100 to 400 ␮l for 10 min) ker, 2005). Notably, gene guns can be used for difficult-to-trans- or direct injected into the right lobe of the liver (20 to 100 ␮l fect cells and particular in situ approaches (Johnston and Tang, for 3 min). Tail vein injection was also performed (400 ␮l for 1994). However, whether gene guns are effective for liver trans- fection is uncertain. We examined the effectiveness of murineliver transfection by gene gun bombardment with enhanced green fluorescent protein (EGFP)-encoding DNA and compared Five to 250 ␮g of EGFP DNA was injected via the tail vein the results with those obtained by other chemical or physical in a volume of saline equivalent to 8% of the body mass of the mouse (e.g., 1.6 ml for a 20-g mouse). The entire volume wasdelivered within 5 sec.
MATERIALS AND METHODS
Mice were killed 48 hr or 7 days after transfection, and their livers were harvested. The livers were either cryofixed or fixed Eight-week-old male FVB/N mice were purchased from the in 4% buffered paraformaldehyde (PFA). Unless otherwise in- Animal Center of the National Science Council (Taipei, Tai- dicated, transfection rates were evaluated 48 hr after transfec- wan). For each transfection method, 30 mice were used. The use of animals in this study was approved by the Animal Care Cryofixation was performed by immersion of tissues in ice- and Use Committee at Chang Gung Memorial Medical Center cold isopentane for 3 min, followed by freezing at -80°C. Fixed frozen samples were mounted in Tissue-Tek O.C.T. 4583 com- pound (Sakura Finetek USA, Torrance, CA). Samples were sec-tioned sequentially on a Jung Frigocut 2800N (Leica, Deerfield, EGFP plasmid (PEGFP-C1, 4.7 kDa) was purchased from IL) at a cutting interval of 6 ␮m. Samples fixed in 4% PFA Clontech (Mountain View, CA). The plasmid was cloned and were subjected to hematoxylin and eosin (H&E) staining. Sec- purified with an EndoFree plasmid kit (Qiagen, Valencia, CA).
tions were examined by either fluorescence microscopy or light Naked EGFP DNA was dissolved at 1 ␮g/␮l in phosphate- microscopy. EGFPϩ hepatocytes were observed at ϫ20 mag- buffered saline (PBS) (pure DNA). EGFP DNA-coated gold nification under the fluorescence microscope. The transfection particles (gold–DNA) were prepared by adding 5 mg of Bi- rate was defined as the number of EGFPϩ hepatocytes divided olistic 1.0-␮m gold particles (Bio-Rad, Hercules, CA) to 5 ␮l by the total number of hepatocytes within the same field on of 1-␮g/␮l plasmid solution, 20 ␮l of 0.1 M spermidine (Sigma- three randomized occasions. Mice transfected with DNA-free Aldrich, St. Louis, MO), and 20 ␮l of 0.5 M CaCl2 (Sigma- PBS (with or without gold) of the same volume were used as Aldrich). After several washes, the precipitate was dissolved in 100% alcohol for bombardment. The EGFP DNA–jetPEI–Galsolution (polymer–DNA) was prepared according to the man- ufacturer’s protocol (Polyplus Transfection, New York, NY).
Forty-eight hours after transfection, the serum alanine amino- The ratio of nitrogen residues on jetPEI to phosphates on the transferase (ALT) levels of the mice were measured with a DNA backbone (N:P ratio) ranged from 5 to 10 for 0.31 to 0.62␮ Vitros DT60 II chemistry system (Johnson & Johnson, New Gene gun transfection with pure DNA, gold–DNA, Independent sample t testing was used to compare the means After general anesthesia by intraperitoneal injection of ket- obtained for two different bombardment pressures or repeti- amine and diphenhydramine (Benadryl; Pfizer, New York, NY), tions. One-way analysis of variance (ANOVA) was used to test the mice underwent midline laparotomy, to exposure the liver the equality of the means among the three DNA groups. Dif- for gene gun bombardment. In situ liver transfections were per- ferences were regarded as significant for p Ͻ 0.05.
formed with the low pressure-accelerated particle gene gun(Bioware Technologies, Taipei, Taiwan). A 1-cm-thick rubberring was placed on the shooting end of gene gun. Briefly,gold–DNA (5–20 ␮l) was bombarded into mouse liver at pres- sures of 20–45 psi. Alternatively, pure DNA (5–20 ␮l) wasbombarded into mouse liver at pressures of 20–45 psi. For poly- Gene gun transfection with pure DNA, gold–DNA, and mer–DNA bombardment, DNA–jetPEI–Gal solution (5 ␮l) was bombarded into mouse liver. The mouse abdomen was closed Mice transfected with the EGFP plasmid by gene gun bom- bardment did not have significant numbers of EGFPϩ hepato-cytes unless a pressure of 30 psi was used (Figs. 1A and 2).
Intravenous or direct liver injection of polymer–DNA With respect to transfection rate, gold–DNA compared favor- Mouse liver was exposed as described above. EGFP ably with pure DNA and polymer–DNA (Fig. 2). However, liver DNA–jetPEI–Gal solution (N:P ratio, 5–10) was used for in laceration increased abruptly at pressures above 30 psi, ac- GENE GUN BOMBARDMENT FOR IN SITU LIVER TRANSFECTION
(A and B) EGFPϩ hepatocytes are shown (original magnification, ϫ20) after gene gun bombardment with gold–DNA
(A) at a pressure of 30 psi, and after HBT with injection of 10 ␮g of DNA within 4 sec (B). The edge of the liver is indicated
by red arrows. (C and E) H&E staining of mouse liver after gold–DNA bombardment; a low-power field (C, ϫ100) and a high-
power field (E, ϫ400) are shown. Gold particles (red arrows) and inflammatory cells (white arrow) are scattered in an area of
necrotic hepatocytes (red arrowheads). (D and F) H&E staining of mouse liver after HBT; a low-power field (D, ϫ100) and a
high-power field (F, ϫ400) are shown. Diffuse infarctions in the hepatic parenchyma are evident (D). A representative conflu-
ent hepatic infarction (F, red arrows) is located underneath the edge of the liver. Infiltrating inflammatory cells (F, white arrows)
and calcification (F, black arrow) are associated with the infarction.
CHANG ET AL.
were scarce at the edge of the liver (Fig. 1B, arrows). H&Estaining revealed remarkable hepatic infarctions in both controland experimental animals. At the edge of the liver, confluentinfarctions were impressive and formed broad bands (Fig. 1D and F). The ALT levels of mice were 588 Ϯ 135 U/liter. Oneweek after HBT, the transfection rate decreased to 11.8%.
DISCUSSION
The chemical approach with jetPEI-Gal injection in FVB/N mice was unsatisfactory, as it gave minimal transfection rates regardless of the injection route. Successful in vivo transfec- tions by jetPEI injection have been reported in the lung (Zou et al., 2000). Although jetPEI-Gal was chosen over jetPEI for use in the current study, because of its higher affinity for he-patocytes (Robaczewska et al., 2001), our data indicate that the Relationships between the transfection rate and bom- liver represents a more robust barrier for polymer–adjuvant bardment pressure for three DNA preparations. In terms of transfection efficiency, gold–DNA compares favorably with HBT yielded the highest transfection rate of all the nonviral pure DNA and polymer–DNA at a pressure of :30 psi (p Ͻ DNA delivery methods. This is comparable to the results of 0.001, one-way ANOVA). For each DNA preparation, signifi-cant differences were observed for 25 versus 30 psi, 30 versus previous studies (Zhang et al., 1999; Yang et al., 2002). How- 35 psi, 35 versus 40 psi, and 40 versus 45 psi (p Ͻ 0.001–0.044, ever, rapid injection of a large volume via the tail vein usually t test), but not for 40 versus 45 psi in the gold–DNA group (p ϭ causes transient heart dysfunction (Zhang et al., 2004) and may lead to animal loss. Clinical application is not feasible, becausehumans lack a homolog for the tail vein. Furthermore, HBTleads to increased venous pressure (Zhang et al., 2004) and sub- counting for a mortality rate of more than 35%. The maximal sequent hepatic infarction. The infarctions had a tendency to be transfection rate achieved by a single bombardment was ap- confluent at the liver edge, where perfusion is sparser than else- proximately 15% for gold–DNA and 5–6% for pure DNA or where. Thus, it does not guarantee foreign gene expression in polymer–DNA (Fig. 2). At 30 psi, the transfection rates reached a plateau at approximately 6.2, 5.9, and 15% for pure DNA,polymer–DNA, and gold–DNA, respectively, with three or fourbombardments (Fig. 3). The mortality rate after triple bom- bardment at 30 psi was negligible and ranged from 0 to 3.3%.
Further repetitions of bombardment led to mortality due to gross liver laceration. Regardless of the composition of the DNA so- lution, EGFPϩ hepatocytes after bombardment were locatedmainly in the superficial layers (depth of 10–60 ␮m, one to three cell layers) of the liver. Despite the better transfection rate obtained for gold–DNA, H&E staining of bombarded liver tis- sues revealed several necrotic spots with deposition of gold par- ticles (Fig. 1C and E), indicating liver injury at the bombard- ment site, probably caused by the gold particles. The ALT levels of the mice were 249 Ϯ 75 U/liter (normal range, 15–84 U/liter). One week after bombardment with gold–DNA, the transfection rate decreased to 9.7%.
Intravenous and direct liver injection of polymer–DNA Transfection rates in relation to number of bombard- None of the transfections with polymer–DNA generated ment repetitions at a pressure of 30 psi. For the same number of bombardment repetitions, gold–DNA gives superior trans-fection rates compared with pure DNA and polymer–DNA (p Ͻ 0.001, one-way ANOVA). A significant increase in transfec-tion rate is observed for bombardment performed up to three The immediate mortality rate was 6.6%, despite cardiopul- times for each DNA group (p Ͻ 0.001–0.032, t test), with the monary resuscitation for more than 10 min. The highest trans- exception of polymer–DNA bombardment carried out once and fection rate for HBT was about 20% under optimized condi- twice (p ϭ 0.075, t test). The transfection rates for three and tions of :10 ␮g of DNA injected within 4 sec. EGFPϩ four bombardment repetitions are not significantly different for hepatocytes were evenly distributed over the whole liver but each group (p ϭ 0.55–1.0, t test).
GENE GUN BOMBARDMENT FOR IN SITU LIVER TRANSFECTION
The original application of the gene gun was for skin vacci- REFERENCES
nation, which induces DNA expression in the most superficiallayers of the skin (Johnston and Tang, 1994; Peachman et al., AZZAM, T., and DOMB, A.J. (2004). Current developments in gene 2003). Thus, cell sampling for gene gun bombardment should transfection agents. Curr. Drug Deliv. 1, 165–193.
focus on the superficial cells. GFPϩ hepatocytes were most DOBSON, J. (2006). Gene therapy progress and prospects: Magnetic prominent in the first three layers. In comparison with the skin, nanoparticle-based gene delivery. Gene Ther. 13, 283–287.
the liver is too fragile to bear the bombardment pressure re- JOHNSTON, S.A., and TANG, D.C. (1994). Gene gun transfection of animal cells and genetic immunization. Methods Cell Biol. 43,
quired for in situ transfection. Therefore, the pressure must be adjusted by weighing transfection efficiency against possible NISHIKAWA, M., and HUANG, L. (2001). Nonviral vectors in the liver tearing. Triple bombardment at a tolerable pressure of 30 new millennium: Delivery barriers in gene transfer. Hum. Gene Ther.
psi has been shown to yield a transfection efficiency compara- 12, 861–870.
ble to that obtained from a single bombardment at higher pres- PEACHMAN, K.K., RAO, M., and ALVING, C.R. (2003). Immu- sure, which usually causes gross liver laceration. However, un- nization with DNA through the skin. Methods 31, 232–242.
predictable location of gene transfer usually ensues from a ROBACZEWSKA, M., GUERRET, S., REMY, J.S., CHEMIN, I., direct strike (our unpublished data). Therefore, a rubber ring OFFENSPERGER, W.B., CHEVALLIER, M., BEHR, J.P., POD- was placed at the opening of the gene gun. It ensures good guid- HAJSKA, A.J., BLUM, H.E., TREPO, C., and COVA, L. (2001).
ance, allowing constant focusing and an attenuated blast effect.
Inhibition of hepadnaviral replication by polyethylenimine-based intravenous delivery of antisense phosphodiester oligodeoxynu- Among the various DNA preparations, gold–DNA bombard- cleotides to the liver. Gene Ther. 8, 874–881.
ment had the highest transfection rate, although it was associ- WELLS, D.J. (2004). Gene therapy progress and prospects: electropo- ated with gold particle-related necrosis. Nevertheless, the level ration and other physical methods. Gene Ther. 11, 1363–1366.
of injury, determined by histological examination and the serum WOLFF, J.A., and BUDKER, V. (2005). The mechanism of naked ALT level, was less severe than that caused by HBT. Gene gun DNA uptake and expression. Adv. Genet. 54, 3–20.
bombardment is comparable with HBT in terms of stability, YANG, P.L., ALTHAGE, A., CHUNG, J., and CHISARI, F.V. (2002).
with a 20% decrease in the transfection rate after 1 week.
Hydrodynamic injection of viral DNA: A mouse model of acute he- In conclusion, gene gun bombardment of the liver with patitis B virus infection. Proc. Natl. Acad. Sci. U.S.A. 99, 13825–13830.
gold–DNA is a potentially useful alternative to HBT for the ZHANG, G., BUDKER, V., and WOLFF, J.A. (1999). High levels of transfection of superficial hepatocytes, particularly because foreign gene expression in hepatocytes after tail vein injections of
naked plasmid DNA. Hum. Gene Ther. 10, 1735–1737.
it does not induce severe hepatic injury. Its application could ZHANG, G., GAO, X., SONG, Y.K., VOLLMER, R., STOLZ, D.B., potentially be extended to other animals, regardless of the GASIOROWSKI, J.Z., DEAN, D.A., and LIU, D. (2004). Hy- presence or absence of a tail vein; however, its application is droporation as the mechanism of hydrodynamic delivery. Gene Ther.
limited to superficial cells and animals that can tolerate lap- 11, 675–682.
ZOU, S.M., ERBACHER, P., REMY, J.S., and BEHR, J.P. (2000). Sys- temic linear polyethylenimine (L-PEI)-mediated gene delivery in the
mouse. J. Gene Med. 2, 128–134.
ACKNOWLEDGMENTS
Financial support was provided by the National Science Liver Research Unit and Department of Council, Taiwan (93-2314-B-182A-148, 94-2314-B-182A-185, and 95-3112-B-182A-002) and by Chang Gung Memorial Hos- pital, Taoyuan, Taiwan (CMRPG 33014, CMRPG 340341, and SMRPG 350081). Professor Pei-Jer Chen (Hepatitis Research Center, National Taiwan University Hospital, Taipei, Taiwan)provided helpful discussion.
Received for publication November 14, 2007; accepted after re- AUTHOR DISCLOSURE STATEMENT
No competing financial interests exist.

Source: http://www.bioware.com.tw/bioware/pdf/Human%20Gene%20Ther%202008%20Apr.pdf