Gene expression imaging by enzymatic catalysis ... - Semantic Scholar

Gene Therapy (2007), 1–10 & 2007 Nature Publishing Group All rights reserved 0969-7128/07 $30.00 www.nature.com/gt

ORIGINAL ARTICLE

Gene expression imaging by enzymatic catalysis of a fluorescent probe via membrane-anchored b-glucuronidase Y-C Su1, K-H Chuang2, Y-M Wang3, C-M Cheng2, S-R Lin4, J-Y Wang5, J-J Hwang6, B-M Chen7, K-C Chen7, S Roffler7 and T-L Cheng1 1

Faculty of Biomedical Science and Environmental Biology, Kaohsiung Medical University, Kaohsiung, Taiwan; 2Graduate Institute of Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan; 3Faculty of Medicinal and Applied Chemistry, Kaohsiung Medical University, Kaohsiung, Taiwan; 4Graduate Institute of Medical Genetics, Kaohsiung Medical University, Kaohsiung, Taiwan; 5Faculty of Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan; 6Faculty of Biomedical Imaging and Radiological Sciences, National Yang-Ming University, Taipei, Taiwan and 7Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan

Development of nonimmunogenic and specific reporter genes to monitor gene expression in vivo is important for the optimization of gene therapy protocols. We developed a membrane-anchored form of mouse b-glucuronidase (mbG) as a reporter gene to hydrolyze a nonfluorescent glucuronide probe (fluorescein di-b-D-glucuronide, (FDGlcU) to a highly fluorescent reporter to assess the location and persistence of gene expression. A functional b-glucuronidase (bG) was stably expressed on the surface of murine CT26 colon adenocarcinoma cells where it selectively hydrolyzed the cell-impermeable FDGlcU probe. FDGlcU was also preferentially converted to fluorescent probe by (bG) on CT26 tumors. The fluorescent intensity in bG-expressing CT26 tumors was 240 times greater than the intensity in control

tumors. Selective imaging of gene expression was also observed after intratumoral injection of adenoviral bG vector into carcinoma xenografts. Importantly, mbG did not induce an antibody response after hydrodynamic plasmid immunization of Balb/c mice, indicating that the reporter gene product displayed low immunogenicity. A membraneanchored form of human bG also allowed in vivo imaging, demonstrating that human bG can be employed for imaging. This imaging system therefore, displays good selectivity with low immunogenicity and may help assess the location, magnitude and duration of gene expression in living animals and humans. Gene Therapy advance online publication, 18 January 2007; doi:10.1038/sj.gt.3302896

Keywords: gene expression imaging; membrane-anchored b-glucuronidase; reporter gene; fluorescein di-b-D-glucuronide

Introduction The availability of sensitive and specific reporter genes is critical for the continued development and practice of human gene therapy. To be clinically useful, a reporter gene should display low immunogenicity to allow repeated administration and prolonged expression. In addition, the reporter should be specific to allow unambiguous identification of the location and extent of gene expression. Several reporter genes have been developed including green fluorescent protein,1,2 luciferase,3 herpes simplex type 1 virus thymidine kinase,4 cytosine deaminase5 and b–galactosidase.6 Expression of these exogenous gene products, however, can induce immune responses that result in tissue Correspondence: Dr T-L Cheng, Faculty of Biomedical and Environmental Biology, Kaohsiung Medical University, 100 ShihChuan 1st Road, Kaohsiung, Taiwan. E-mail: [email protected] or Dr S Roffler, Institute of Biomedical Sciences, Academia Sinica, Academia Road, Section 2, No. 128, Taipei 11529, Taiwan. E-mail: [email protected] Received 20 June 2006; revised 7 November 2006; accepted 7 November 2006

damage and limit persistent gene expression and imaging.7–10 To prevent immune responses, endogenous reporter genes such as the dopamine D2 receptor10,11 and the transferrin receptor12 are under investigation. Although these gene products are not immunogenic, they offer poor specificity due to their widespread expression.13,14 To overcome these problems, here we present a novel reporter gene that was designed to retain specificity, whereas displaying low immunogenicity. This is based on anchoring the lysosomal bG on the surface of cells to allow the extracellular hydrolysis of a nonfluorescent glucuronide probe to a highly fluorescent reporter (Figure 1a). We hypothesized that specificity could be retained with this endogenous enzyme because bG is located in lysosomes15 and only very low levels of bG are found in human serum.16 Glucuronides are charged at physiological pH values which hinders their diffusion across the lipid bilayer of cells,17,18 effectively sequestering the glucuronide probe from contact with lysosomal bG. Conjugation of glucuronide moieties to xenobiotics by UDP-glucuronosyl transferases is also a major detoxification pathway in rodents and humans,19 suggesting that a glucuronide probe should be resistant to

Surface enzyme fluorescence imaging Y-C Su et al

2

by optical imaging. We then tested the hypothesis that mbG would display low immunogenicity in mice. Finally, we evaluated whether human bG (hbG) could be employed to hydrolyze FDGlcU for in vivo optical imaging. Our results show that surface expression of bG may provide a novel strategy for the assessment of vector specificity and safety.

Results Surface display of functional mbG We constructed a retrovirus vector, pLNCX-mbG-eB7, to direct the expression of mbG to the plasma membrane of mammalian cells (Figure 1b). CT26 murine colon adenocarcinoma cells were infected with recombinant retroviral particles and selected in G418 to obtain CT26/mbG-eB7 cells. The cells were immunofluorescence stained for the presence of the myc-epitope in mbG-eB7, incubated with FDGlcU, and then washed to assay for mbG enzymatic activity. The substrate solution become immediately fluorescent after it was added to CT26/mbG-eB7 cells (data not show). CT26/mbG-eB7 cells, but not the parental CT26 cells expressed high levels of mbG on their surface and converted nonfluorescent FDGlcU to fluorescein (Figure 1c). These results demonstrate that high levels of functional mbG can be expressed on CT26/mbG-eB7 cells.

Figure 1 Membrane-anchored bG imaging system. (a) bG anchored on the outer surface of the cellular plasma membrane can enzymatically hydrolyze the glucuronide group of a nonfluorescent probe (FDGlcU) to produce a highly fluorescent compound (fluorescein). Oligosaccharides present in the juxtamembrane spacer domain help reduce shedding of bG. TM, transmembrane domain. (b) bG transgene. The cDNA codes for an Igk chain leader sequence (LS) followed by an HA epitope (HA), the mature bG gene, a myc epitope (myc), the immunoglobulin C2-type extracellular region of B7-1 (B7 spacer) and the transmembrane and cytosolic domains of murine B7-1. Gene expression is under the control of the CMV promoter. (c) Cells surface display of functional mbG. Live CT26 and CT26/mbG-eB7 cells were immunofluorescence stained for the presence of the myc epitope in mbG-eB7 (red) and then incubated with FDGlcU probe (green) before observation under a digital fluorescence confocal microscopy system.

premature activation by endogenous bG under physiological conditions. Based on this reasoning, we fused mouse bG (mbG) to a fragment encompassing the extracellular domain, transmembrane domain and cytosolic tail of murine B7-1 to anchor the chimeric protein on the plasma membrane. We then examined if functional mbG could be expressed on cells, whether mbG could hydrolyze a nonfluorescent probe (fluorescein di-b-D-glucuronide (FDGlcU)) to a highly fluorescent imaging agent and whether mbG gene expression in vivo could be assessed Gene Therapy

In vivo imaging of mbG expression To investigate whether sites of mbG expression could be noninvasively imaged, Balb/c mice bearing established CT26 and CT26/mbG-eB7 tumors in their left and right chest regions, respectively, were intravenously (i.v.) injected with 500 mg FDGlcU. Whole-body images of the mice were acquired by performing 3 min scans. Figure 2a shows that FDGlcU was selectively converted to fluorescein at the sites of mbG expression in CT26/ mbG-eB7 tumors but not in the control CT26 tumors. Serial imaging analysis showed that the highest fluorescence was observed at 30 min after FDGlcU injection (Figure 2b). To verify the imaging results, the tumors were resected and adjacent tumor sections were examined under a fluorescence microscope or stained with X-GlcA to examine the functional expression of mbG. Figure 2c shows that the regions of active fluorescein displayed concomitant blue X-GlcA staining at CT26/ mbG-eB7 tumor sections (Figure 2c, lower panel) but not control tumor sections (Figure 2c, upper panel), consistent with selective hydrolysis of FDGlcU at sites of mbG expression in vivo. The biodistribution of activated FDGlcU The biodistribution of activated FDGlcU was examined by killing tumor-bearing mice 30 min after they received an i.v. injection of FDGlcU or fluorescein and then performing optical imaging of whole-body frozen sections. Figure 3 shows that FDGlcU was selectively converted to fluorescein in CT26/mbG-eB7 tumors but not in CT26 tumors (Figure 3a, left panel). The fluorescent intensity in CT26/mbG-eB7 tumors was about 240 times greater than in the CT26 tumors (Figure 3b). Fluorescent signals were also observed in the intestinal tract. In contrast to the biodistribution of activated FDGlcU (Figure 3a, right panel), i.v.


3

Figure 2 In vivo imaging of mbG. (a) FDGlcU was i.v. injected into mice bearing CT26 (left) and CT26/mbG-eB7 (right) tumors and wholebody images were acquired at the indicated times (30 , 300 , 600 ). (b) The pharmacokinetics of FDGlcU activation in CT26 (J) and CT26/mbGeB7 (K) tumors (n ¼ 4) was determined by measuring fluorescence intensities in 3 min scans performed over 90 min. (c) Sections of CT26 (upper panels) and CT26/mbG-eB7 (lower panels) tumors were stained with X-GlcA and nuclear fast red (NFR). Adjacent sections were viewed under phase contrast or fluorescence microscopes.

administered fluorescein largely accumulated in the intestinal tract and kidneys.

Imaging of Ad/mbG-eB7 We next examined whether mbG gene expression could be detected after adenoviral-mediated infection of human hepatocellular carcinoma cells. HCC36 cells were successfully infected with Ad5/mbG-eB7 as shown by staining of membrane mbG with anti-myc antibody (Figure 4a, left panel) or FDGlcU to assay for bG

enzymatic activity (Figure 4a, right panel), demonstrating that mbG was functionally active after Ad5/mbG-eB7 infection. For in vivo imaging, 109 PFU of Ad5/mbG-eB7 was directly injected into subcutaneous HCC36 tumors. Two days later, the mice were i.v. injected with FDGlcU and whole-body optical imaging was performed. HCC36 tumors that were injected with Ad5/mbG-eB7 displayed obvious fluorescence as compared with noninfected HCC36 tumors (Figure 4b). Histological staining for mbG activity revealed strong Gene Therapy


4

eB7 (Figure 5b, striped bar), indicating that b-galactosidase was immunogenic. By contrast, no specific antibody titer was detected in mice injected with pLNCX-mbGeB7 (Figure 5b). In addition, we also examined the cellular immunity of mbG-eB7. Balb/c and Beige severe combined immuno deficient (SCID) mice were subcutaneously (s.c.) injected with CT26, CT-26/mbG-eB7 or CT26.CL25 cells, which express high levels of E. coli b-galactosidase. Figure 5c shows that CT26.CL25 tumor growth was significantly (Pp0.005 after day 4) suppressed in Balb/c mice but not Beige SCID mice as compared with CT26 tumors. By contrast, the growth rate of CT26 and CT-26/mbG-eB7 tumor were not significantly different in Balb/c and Beige SCID mice. These results suggest that cellular immunity was not induced by mbG-eB7. Thus, mbG could be stably expressed on cells in an active form and did not induce a specific immune response, prerequisites for repetitive and persistent imaging in live animals.

Figure 3 Biodistribution of active probe in vivo. (a) The distribution of fluorescein in mice was measured at 30 min after i.v. injection of FDGlcU (left panel) or fluorescein (right panel). The fluorescence intensity of whole-body sections was measured with the IVIS Imaging System 50 (Xenogen, Alameda, CA, USA). (b) The regions of interest in different organs after i.v. injection of FDGlcU or fluorescein were analyzed with Living Image software.

fluorescence in tumor sections obtained from Ad5/mbGeB7 infected HCC36 tumors (Figure 4c, lower) but not in sections obtained from noninfected tumors (Figure 4c, upper). These results show that FDGlcU probe could specifically image sites of adenoviral-mediated gene expression in vivo.

mbG-eB7 immunogenicity The immunogenicity of the membrane-anchored bG reporter was examined after hydrodynamic-based gene transfer of pLNCX-mbG-eB7 into Balb/c mice. Control groups of mice were injected with pLNCX vector or pLNCX-LacZ-eB7, which encodes a membrane form of Escherichia coli b-galactosidase. Serum samples were collected and livers were excised, embedded in Tissue-Tek OCT and cut into sections for X-GlcA or X-Gal staining to detect for functional expression of mbG or LacZ. Figure 5a shows that mbG-eB7 and LacZ-eB7 were expressed in the liver as determined by specific hydrolysis of X-GlcA or X-Gal, respectively. The humoral immune response against the enzymes was examined by detecting antibody-binding to 293 cells that were transiently transfected with pLNCX, pLNCX-mbG-eB7 or pLNCX-LacZ-eB7. The enzymes were expressed on 293 cells as shown by the binding of anti-myc antibody to the transfected cells (Figure 5b, black bars). Antibody titers were detected in mice injected with pLNCX-LacZGene Therapy

Imaging of membrane-anchored hbG To investigate whether expression of hbG on cells could allow noninvasive imaging by FDGlcU, we constructed the retroviral vector pLNCX-hbG-eB7 and generated EJ human bladder carcinoma cells (EJ/hbG-eB7) that expressed functionally active hbG on their surface (Figure 6a). FDGlcU was preferentially converted to a fluorescent reporter in EJ/hbG-eB7 tumors but not control tumors in mice as determined by optical imaging (Figure 6b), indicating that hbG can act as reporter gene for noninvasive imaging of gene expression in vivo.

Discussion We developed a novel reporter system to allow assessment of the delivery and expression of genes in living animals. mbG was anchored to the plasma membrane of cells to allow selective hydrolysis of FDGlcU to a fluorescent reporter in vitro and in vivo. Importantly, mbG did not induce detectable humoral or cellular immune responses in mice, suggesting that repeated and persistent imaging of gene expression can be achieved. Proportional expression of the gene of interest and the bG reporter gene can be attained by inserting an internal ribosomal entry site20 or furin-2A-self processing peptide21 between the genes. Glucuronides do not readily enter cells due to the presence of a charged carboxylic acid at physiological pH values,17,18 thereby preventing contact of glucuronides with lysosomal bG. Our previous results showed that glucuronide prodrugs must be enzymatically activated outside of tumor cells for maximum cytotoxicity.18 By anchoring bG on the outer surface of the plasma membrane, maximal activation of glucuronide probes can be achieved to increase imaging sensitivity. Successful use of surface bG therefore requires efficient transport of bG to the cell surface. We previously performed a series of studies to determine the optimal conditions for the expression of proteins22 and scFv23,24 on mammalian cells. Based on those studies, we anchored bG on the surface of cells by fusion of the enzyme to the Ig-hingelike domains of the B7-1 antigen (e-B7) and the B7-1 TM. The Ig-hinge-like domains of the B7-1 antigen act as a spacer to allow more flexible assembly of the bG


5

Figure 4 Analysis and imaging of Ad5/mbG-eB7 infection. (a) In vitro infection of HCC36 cells with Ad5/mbG-eB7. HCC36 cells (open curve) and Ad5/mbG-eB7 infected HCC36 cells (solid curve) were stained for the presence of the HA epitope in mbG-eB7 (left panel) or directly stained with FDGlcU (right panel) and then analyzed on a flow cytometer. (b) Nude mice bearing HCC36 tumors on both sides of the chest were injected in the right tumor with Ad5/mbG-eB7. FDGlcU was i.v. injected 48 h later and whole-body images were acquired at the indicated times (30 , 300 , 600 ). (c) Sections of HCC36 (upper panel) and Ad5/mbG-eB7 injected HCC36 (lower panel) tumors that were collected 30 min after FDGlcU injection were stained with X-GlcA and NFR and then viewed under phase contrast and fluorescence microscopes.

tetramer. The spacer domain also contains glycosylation sites that reduce the shedding of chimeric proteins from the cells surface.25 Reporter gene products should display low immunogenicity to prevent tissue damage by cellular immune

responses and allow repeated and persistent imaging of gene expression. The membrane-anchored bG reporter employed in our study was derived from murine lysosomal bG and did not induce a detectable antibody response in mice. In addition, the growth rates of CT26 Gene Therapy


6

Figure 5 mbG immunogenicity. (a) Mouse livers were excised 2 days after hydrodynamic-based injection of pLNCX, pLNCX-mbG-eB7 or pLNCX-LacZ-eB7 plasmids. Frozen liver sections were stained for membrane bG activity or LacZ activity and then counterstained with NFR. (b) Serum samples collected from Balb/c mice 10 days after hydrodynamic-based gene transduction were assayed by enzyme linked immunosorbent assay for the presence of antibodies against 293 cells that were transiently transfected with pLNCX, pLNCX-mbG-eB7 or pLNCX-LacZ-eB7. The binding of an anti-myc antibody to the myc epitope present in the surface enzymes was also assayed (black bars). Results show the mean absorbance values of triplicate determinations. Bars, s.e. (c) Groups of 3B4 Balb/c (left) and Beige SCID (right) mice were s.c. injected in the right flank with 2 105 CT26 (’), CT-26/mbG-eB7 (K) or CT26.CL25 (m) cells (which express E. coli b-galactosidase). Tumor volumes (length width height 0.5) were estimated at least twice a week. Significant differences between the growth rate of CT26 and CT26.CL25 tumors are indicated; *Pp0.005. Bars. s.e.

and CT-26/mG-eB7 tumors were similar in Balb/c and Beige SCID mice, indicating that cellular immunity was not induced by the enzyme. Importantly, we showed that membrane hbG also allowed imaging of gene expression in vivo. These results indicate that hbG may be appropriate for human gene therapy. Furthermore, reporter genes can be designed that exhibit low immunogenicity in different animals by using bG derived from the species of interest. A range of glucuronide probes should be applicable to gene expression imaging with bG reporter genes. Chemical conjugation of sugar groups often inactivates Gene Therapy

or shifts the fluorescence resonance or relaxivity of imaging probes. For example, Tung et al.26 described the in vivo imaging of intracellular E. coli b-galactosidase expression with a galactoside probe that displayed a red fluorescent shift. In addition, a gadolinium-based galactoside probe was described for magnetic resonance imaging in which attachment of a galactoside group converted the probe to a low-relaxivity (inactive) state.27 E. coli b-galactosidase, unfortunately, is immunogenic in animals and humans, limiting its applications. Similarly, a glucuronidase-sensitive gadolinium-based glucuronide agent could be monitored using magnetic resonance


In summary, the power of membrane-anchored bG is based on several factors including: (1) the low immunogenicity of endogenous bG to allow persistent imaging of gene expression, (2) inaccessibility of glucuronides to endogenous lysosomal bG and low serum concentrations of bG, resulting in little nonspecific probe activation, (3) low toxicity of glucuronide conjugates due to their poor transport across the lipid bilayer of cells, (4) rapid clearance from the blood allowing quicker imaging, (5) signal amplification due to the catalytic hydrolysis of probe molecules, (6) possibility of generating a range of imaging probes by attachment of glucuronide groups and (7) possibility of performing imaging and therapy with the same construct. Based on these advantages, the bG imaging system appears to possess great potential for monitoring gene expression in animals and humans.

7

Materials and methods

Figure 6 Analysis and imaging of hbG. (a) Live EJ (open curve) and EJ/hbG-eB7 (solid curve) cells were immunofluorescence stained for the myc epitope (left panel) or incubated with FDGlcU (right panel), respectively, and analyzed on a flow cytometry. (b) FDGlcU was i.v. injected into nude mice bearing EJ (left) and EJ/hbG-eB7 (right) tumors and whole-body images were acquired at the indicated times (30 , 300 , 600 ).

in vitro.28 These studies suggest that analogous probes can be created by attaching a glucuronide group to the different imaging agents. To improve the utility of our membrane-anchored bG reporter, glucuronide-nearinfrared fluorescent probes and glucuronide-MRI contrast agents are currently under investigation to improve tissue penetration. bG may allow both imaging and therapy of cancer with a single gene product as bG has demonstrated antitumor activity in antibody-directed enzyme prodrug therapy (ADEPT) and gene-directed enzyme prodrug therapy (GDEPT). Immunoenzymes, formed by conjugating bG to antitumor antibodies, can selectively activate glucuronide prodrugs29–34 allow accumulation of high drug concentrations at the tumor site,35 produce bystander killing of antigen-negative tumor cells31 and generate long-lasting protective immunity to subsequent tumor challenge.34 Similarly, Brusselbach et al. demonstrated that cell surface display of hbG for extracellular gene-directed enzyme prodrug therapy can produce strong bystander killing and potent antitumor activity.36 We recently demonstrated that mbG expressed on tumor cells effectively activated a glucuronide prodrug and produced tumor regressions in a mouse model.37 bG is therefore an attractive enzyme for specific conversion of glucuronide prodrugs for cancer therapy. These studies suggest that the same transgene may be employed to assess the specificity and extent of gene transduction in vivo as well as for glucuronide prodrug therapy of cancer. Real-time detection of glucuronide tracers may also allow simultaneous estimation of the pharmacokinetics and therapeutic efficacy of glucuronide prodrug activation.

Cells and animals CT26 murine colon carcinoma cells, EJ human bladder carcinoma cells, HCC36 hepatocellular carcinoma cells, 293N adenovirus packaging cells and GP2-293 retrovirus packaging cells were grown in Dulbecco’s minimal essential medium (Sigma, St Louis, MO, USA) supplemented with 10% heat-inactivated bovine calf serum, 100 U/ml penicillin and 100 mg/ml streptomycin at 371C in an atmosphere of 5% CO2. Six to eight-week-old Balb/ c and nude mice were purchased from the National Laboratory Animal Center, Taipei, Taiwan. Animal experiments were performed in accordance with institute guidelines. Construction and transduction of bG reporter genes A mouse bG cDNA was fused to the B7 extracellular and transmembrane domain present in p2C11-eB725 and then inserted into the retroviral vector pLNCX (BD Biosciences, San Diego, CA, USA) to generate pLNCXmbG-eB7. pLNCX-hbG-eB7 was constructed by replacing the mouse bG cDNA with the human bG cDNA. Recombinant retroviral particles were packaged by co-transfection of pVSVG with pLNCX-mbG-eB7 or pLNCX-hbG-eB7 into GP2-293 cells (Clontech, Mountain View, CA, USA). After 48 h, the culture medium was filtered, mixed with 8 mg/ml polybrene and added to CT26 colon carcinoma cells or EJ human bladder carcinoma cells, respectively. The cells were selected in G418 and sorted on a flow cytometer to generate CT26/mbG-eB7 or EJ/hbG-eB7 cells. Immunofluorescence staining of a functional mbG on cells Functional expression of bG on cells was measured by incubating CT26 or CT26/mbG-eB7 cells with 40 mM FDGlcU (Invitrogen, Calsbad, CA, USA) in PBS containing 0.1%. BSA, pH 6.5 at 371C for 40 min and then staining the cells with anti-myc antibody (5 mg/ml, clone Myc 1-9E10.2, American Type Culture Collection, Manassas, VA, USA) at 41C for 1 h. The cells were washed with cold phosphate-buffered saline (PBS) and incubated with rhodamine-conjugated rabbit anti-mouse IgG antibody (5 mg/ml) at 41C for 1 h. The cells were washed with cold PBS, mounted with fluorescence mounting Gene Therapy


8

medium (DakoCytomation, Carpinteria, CA, USA), and viewed under a digital fluorescence confocal microscope.

the regions of interest were analyzed with Living Image software (Xenogen).

Characterization of Ad/mbG-eB7 The mbG-eB7 cassette was subcloned into the adenovirus shuttle plasmid vector pAd-CMV, which contains a cytomegalovirus promoter and the polyadenylation signal of bovine growth hormone. A recombinant adenovirus (Ad5/mbG-eB7) was generated by homologous recombination and amplified in 293 cells as described.38 Virus titers were determined in a plaque assay on a 293 cell monolayer. HCC36 hepatocellular carcinoma cells (6 105 per well) were infected with Ad5/mbG-eB7 (6 106 PFU) in six-well plates. After 48 h, the cells were stained with 10 mM FDGlcU at 371C for 30 min or stained with 5 mg/ml rat anti-HA antibody followed by 5 mg/ml FITC-conjugated goat anti-rat Ig. After removing unbound antibodies or probe by extensive washing, the fluorescence of viable cells was measured with a FACScaliber flow cytometer and fluorescence intensities were analyzed with Flowjo V3.2 (Tree Star, Inc., San Carlos, CA, USA).

mbG immunogenicity Groups of four to six Balb/c mice, anesthetized by pentobarbital (65 mg/kg), were injected with 10 mg pLNCX (negative control), pLNCX-mbG-eB7 or pLNCXLacZ-eB7 (positive control), a vector containing the LacZ gene fused to the eB7 domain to anchor E. coli b-galactosidase on the cell surface.40 The plasmids were i.v. injected in 2 ml PBS within 8 s for hydrodynamicbased gene transfer on days 1 and 8.41 Serum samples were collected 10 days after the second injection. Preimmune and immune serum samples (diluted 1:250 in PBS) were added to the 96-wells plates coated with 293 cells that were transiently transfected with pLNCX, pLNCX-mbG-eB7 or pLNCX-LacZ-eB7 plasmids. Binding of the serum and anti-myc antibodies to the cells was detected by serial addition of horse-radish peroxidase conjugated goat anti-mouse antibody (2 mg/ml) and 100 ml/well ABTS substrate (0.4 mg/ml 2,20 -azino-di(3ethylbenzthiazoline-6-sulfonic acid), 0.003% H2O2, 100 mM phosphate-citrate, pH 4.0) for 30 min at room temperature. The absorbance (405 nm) of the wells was measured in a microplate reader (Molecular Device, Menlo Park, CA, USA). To assess gene expression, livers were excised 48 h after hydrodynamic-based injection of plasmids, embedded in Tissue-Tek OCT in liquid nitrogen and cut into 10 mm sections. Liver sections were stained for bG activity by using the b-glucuronidase Reporter Gene Staining Kit or b-galactosidase activity with the b-gal Staining Kit (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. All sections were examined on an upright microscope (Invitrogen). To examine if cellular immunity was induced by membraneanchored mbG, groups of three to four Balb/c and Beige SCID mice were s.c. injected in the right flank with 2 105 CT26, CT-26/mbG-eB7 or CT26.CL25 (ATCC CRL-2639) cells. Tumor volumes (length width height 0.5) were estimated at least twice a week.

In vivo imaging of membrane-anchored bG Balb/c mice (n ¼ 3) bearing established CT26 and CT26/ mbG tumors (200–300 mm3) in their left and right chest regions, respectively, were i.v. injected with 500 ug FDGlcU. Whole-body images of pentobarbital-anesthetized mice were obtained by performing 3 min scans over 90 min on a Kodak IS2000MM optical imaging system. The fluorescence intensities were analyzed with KODAK 1D Image Analysis Software. To image Ad5/ mbG-eB7, nude mice (n ¼ 3) bearing established HCC36 tumors (200–300 mm3) in their left and right chest’s regions were injected in the right HCC36 tumor with 109 PFU of Ad5/mbG-eB7 in 50 ml PBS. After 2 days, the mice were i.v. injected with 500 mg of FDGlcU. Whole-body images were obtained in 3 min scans over 2 h. A human reporter gene (hbG-eB7) was imaged in nude mice (n ¼ 3) bearing established EJ and EJ/hbG-eB7 tumors (200–300 mm3) in their left and right chest regions, respectively, by i.v. injecting 500 mg FDGlcU and performing imaging and histological analysis of fluorescent intensity and bG activity as described above. Histological analysis Tumors were excised at 30 min after FDGlcU injection, embedded in Tissue-Tek OCT in liquid nitrogen, and sectioned into 10 mm slices. Adjacent tumor sections were stained for bG activity with the b-glucuronidase Reporter Gene Staining Kit (Sigma Diagnostics, St Louis, MO, USA) and counterstained with nuclear fast red. The sections were examined on an upright BX4 microscope (Olympus, Melville, NY, USA) or viewed in phase contrast and fluorescence modes on an inverted Axiovert 200 microscope (Carl Zeiss Microimaging, Thornwood, NY, USA). Biodistribution of activated FDGlcU Mice (n ¼ 3) were i.v. injected with 500 mg FDGlcU or fluorescein 30 min before the mice were dipped into isopentane at liquid nitrogen temperatures and embedded on a cryostat holder (7 5 cm) in 4% carboxylmethylcellulose.39 The fluorescence signals of 30 mm whole-body sections were measured on an IVIS Imaging System 50 (Xenogen, Alameda, CA, USA) and Gene Therapy

Statistical significance Statistical significance of differences between mean values was estimated with Excel (Microsoft, Redmond, WA, USA) using the independent t-test for unequal variances. Po0.05 were considered statistically significant.

Acknowledgements This work was supported by the National Research Program for Genomic Medicine (NRPGM), National Science Council, Taipei, Taiwan (NSC95-3112-B-037-001 and NSC94-2745-B-037-010-URD) and the Genomic and Proteomic Program, Academia Sinica, Taipei, Taiwan (94M007-2). The Molecular-Genetic Imaging Core-Cell and tissue-imaging core of National YangMing University is also gratefully acknowledged.

References 1 Misteli T, Spector DL. Applications of the green fluorescent protein in cell biology and biotechnology. Nat Biotechnol 1997; 15: 961–964.


9 2 Yang M, Baranov E, Jiang P, Sun FX, Li XM, Li L et al. Whole-body optical imaging of green fluorescent proteinexpressing tumors and metastases. Proc Natl Acad Sci USA 2000; 97: 1206–1211. 3 Colin M, Moritz S, Schneider H, Capeau J, Coutelle C, BrahimiHorn MC. Haemoglobin interferes with the ex vivo luciferase luminescence assay: consequence for detection of luciferase reporter gene expression in vivo. Gene Therapy 2000; 7: 1333–1336. 4 Liang Q, Nguyen K, Satyamurthy N, Barrio JR, Phelps ME, Gambhir SS et al. Monitoring adenoviral DNA delivery, using a mutant herpes simplex virus type 1 thymidine kinase gene as a PET reporter gene. Gene Therapy 2002; 9: 1659–1666. 5 Haberkorn U, Oberdorfer F, Gebert J, Morr I, Haack K, Weber K et al. Monitoring gene therapy with cytosine deaminase: in vitro studies using tritiated-5-fluorocytosine. J Nucl Med 1996; 37: 87–94. 6 Yu J, Liu L, Kodibagkar VD, Cui W, Mason RP. Synthesis and evaluation of novel enhanced gene reporter molecules: Detection of beta-galactosidase activity using (19)F NMR of trifluoromethylated aryl beta-d-galactopyranosides. Bioorg Med Chem 2006; 14: 326–333. 7 Bonini C, Ferrari G, Verzeletti S, Servida P, Zappone E, Ruggieri L et al. HSV-TK gene transfer into donor lymphocytes for control of allogeneic graft-versus-leukemia. Science 1997; 276: 1719–1724. 8 Haack K, Linnebacher M, Eisold S, Zoller M, von Knebel Doeberitz M, Gebert J. Induction of protective immunity against syngeneic rat cancer cells by expression of the cytosine deaminase suicide gene. Cancer Gene Ther 2000; 7: 1357–1364. 9 Russo V, Tanzarella S, Dalerba P, Rigatti D, Rovere P, Villa A et al. Dendritic cells acquire the MAGE-3 human tumor antigen from apoptotic cells and induce a class I-restricted T cell response. Proc Natl Acad Sci USA 2000; 97: 2185–2190. 10 Reichel MB, Bainbridge J, Baker D, Thrasher AJ, Bhattacharya SS, Ali RR. An immune response after intraocular administration of an adenoviral vector containing a beta galactosidase reporter gene slows retinal degeneration in the rd mouse. Br J Ophthalmol 2001; 85: 341–344. 11 MacLaren DC, Gambhir SS, Satyamurthy N, Barrio JR, Sharfstein S, Toyokuni T et al. Repetitive, non-invasive imaging of the dopamine D2 receptor as a reporter gene in living animals. Gene Therapy 1999; 6: 785–791. 12 Weissleder R, Moore A, Mahmood U, Bhorade R, Benveniste H, Chiocca EA et al. In vivo magnetic resonance imaging of transgene expression. Nat Med 2000; 6: 351–355. 13 Wolf G. Vitamin A functions in the regulation of the dopaminergic system in the brain and pituitary gland. Nutr Rev 1998; 56: 354–355. 14 Fleming RE, Migas MC, Holden CC, Waheed A, Britton RS, Tomatsu S et al. Transferrin receptor 2: continued expression in mouse liver in the face of iron overload and in hereditary hemochromatosis. Proc Natl Acad Sci USA 2000; 97: 2214–2219. 15 Aleksandrowicz J, Starek A, Moszczynski P, Czarnobilski Z. Changes in the activity of lysosome beta-glucuronidase in rat leukocytes treated with various doses of selenium. Patol Pol 1978; 29: 143–151. 16 Stahl PD, Fishman WH. b-D-glucuronidase. In: Bergmeyer J, Grabl M (eds). Methods in Enzymatic Analysis. Verlag Chemie: Weinheim, 1984; vol 5. pp 246–256. 17 Haisma HJ, Boven E, van Muijen M, de Jong J, van der Vijgh WJ, Pinedo HM. A monoclonal antibody-beta-glucuronidase conjugate as activator of the prodrug epirubicin-glucuronide for specific treatment of cancer. Br J Cancer 1992; 66: 474–478. 18 Cheng TL, Chou WC, Chen BM, Chern JW, Roffler SR. Characterization of an antineoplastic glucuronide prodrug. Biochem Pharmacol 1999; 58: 325–328. 19 Kiang TK, Ensom MH, Chang TK. UDP-glucuronosyltransferases and clinical drug–drug interactions. Pharmacol Ther 2005; 106: 97–132.

20 Yu Y, Annala AJ, Barrio JR, Toyokuni T, Satyamurthy N, Namavari M et al. Quantification of target gene expression by imaging reporter gene expression in living animals. Nat Med 2000; 6: 933–937. 21 Szymczak AL, Workman CJ, Wang Y, Vignali KM, Dilioglou S, Vanin EF et al. Correction of multi-gene deficiency in vivo using a single ’self-cleaving’ 2A peptide-based retroviral vector. Nat Biotechnol 2004; 22: 589–594. 22 Chou WC, Liao KW, Lo YC, Jiang SY, Yeh MY, Roffler SR. Expression of chimeric monomer and dimer proteins on the plasma membrane of mammalian cells. Biotechnol Bioeng 1999; 65: 160–169. 23 Cheng TL, Liao KW, Tzou SC, Cheng CM, Chen BM, Roffler SR. Hapten-directed targeting to single-chain antibody receptors. Cancer Gene Ther 2004; 11: 380–388. 24 Liao KW, Lo YC, Roffler SR. Activation of lymphocytes by antiCD3 single-chain antibody dimers expressed on the plasma membrane of tumor cells. Gene Therapy 2000; 7: 339–347. 25 Liao KW, Chen BM, Liu TB, Tzou SC, Lin YM, Lin KF et al. Stable expression of chimeric anti-CD3 receptors on mammalian cells for stimulation of antitumor immunity. Cancer Gene Ther 2003; 10: 779–790. 26 Tung CH, Zeng Q, Shah K, Kim DE, Schellingerhout D, Weissleder R. In vivo imaging of beta-galactosidase activity using far red fluorescent switch. Cancer Res 2004; 64: 1579–1583. 27 Louie AY, Huber MM, Ahrens ET, Rothbacher U, Moats R, Jacobs RE et al. In vivo visualization of gene expression using magnetic resonance imaging. Nat Biotechnol 2000; 18: 321–325. 28 Duimstra JA, Femia FJ, Meade TJ. A gadolinium chelate for detection of beta-glucuronidase: a self-immolative approach. J Am Chem Soc 2005; 127: 12847–12855. 29 Cheng TL, Chen BM, Chan LY, Wu PY, Chern JW, Roffler SR. Poly(ethylene glycol) modification of beta-glucuronidase-antibody conjugates for solid-tumor therapy by targeted activation of glucuronide prodrugs. Cancer Immunol Immunother 1997; 44: 305–315. 30 Chen BM, Chan LY, Wang SM, Wu MF, Chern JW, Roffler SR. Cure of malignant ascites and generation of protective immunity by monoclonal antibody-targeted activation of a glucuronide prodrug in rats. Int J Cancer 1997; 73: 392–402. 31 Cheng TL, Wei SL, Chen BM, Chern JW, Wu MF, Liu PW et al. Bystander killing of tumour cells by antibody-targeted enzymatic activation of a glucuronide prodrug. Br J Cancer 1999; 79: 1378–1385. 32 Leu YL, Roffler SR, Chern JW. Design and synthesis of watersoluble glucuronide derivatives of camptothecin for cancer prodrug monotherapy and antibody-directed enzyme prodrug therapy (ADEPT). J Med Chem 1999; 42: 3623–3628. 33 Cheng TL, Chen BM, Chern JW, Wu MF, Roffler SR. Efficient clearance of poly(ethylene glycol)-modified immunoenzyme with anti-PEG monoclonal antibody for prodrug cancer therapy. Bioconjug Chem 2000; 11: 258–266. 34 Chen BM, Cheng TL, Tzou SC, Roffler SR. Potentiation of antitumor immunity by antibody-directed enzyme prodrug therapy. Int J Cancer 2001; 94: 850–858. 35 de Graaf M, Pinedo HM, Oosterhoff D, van der MeulenMuileman IH, Gerritsen WR, Haisma HJ et al. Pronounced antitumor efficacy by extracellular activation of a doxorubicinglucuronide prodrug after adenoviral vector-mediated expression of a human antibody-enzyme fusion protein. Hum Gene Ther 2004; 15: 229–238. 36 Heine D, Muller R, Brusselbach S. Cell surface display of a lysosomal enzyme for extracellular gene-directed enzyme prodrug therapy. Gene Therapy 2001; 8: 1005–1010. 37 Chen KC, Cheng TL, Leu YL, Prijovich ZM, Chuang CH, Chen BM et al. Membrane-localized activation of glucuronide prodrugs by b-glucuronidase enzymes. Cancer Gene Ther 2006, September 15; [E-pub ahead of print]. Gene Therapy


10 38 Shyue SK, Tsai MJ, Liou JY, Willerson JT, Wu KK. Selective augmentation of prostacyclin production by combined prostacyclin synthase and cyclooxygenase-1 gene transfer. Circulation 2001; 103: 2090–2095. 39 Lin KJ, Ye XX, Yen TC, Wey SP, Tzen KY, Ting G et al. Biodistribution study of [(123)I] ADAM in mice: correlation with whole body autoradiography. Nucl Med Biol 2002; 29: 643–650.

Gene Therapy

40 Roffler SR, Wang HE, Yu HM, Chang WD, Cheng CM, Lu YL et al. A membrane antibody receptor for noninvasive imaging of gene expression. Gene Therapy 2005; 13: 412–420. 41 Andrianaivo F, Lecocq M, Wattiaux-De Coninck S, Wattiaux R, Jadot M. Hydrodynamics-based transfection of the liver: entrance into hepatocytes of DNA that causes expression takes place very early after injection. J Gene Med 2004; 6: 877–883.