Nuclear Medicine and Biology 41 (2014) 171–178

Contents lists available at ScienceDirect

Nuclear Medicine and Biology

j ourna l homepage: www.e lsev ie r .com/ locate /nucmedbio

Feasibility of baculovirus-mediated reporter gene delivery for efficient monitoringof islet transplantation in vivo☆

Shuai Liu a, Yu Pan a, Jing Lv a, Haifei Wu a, Jingyan Tian b, Yifan Zhang a,⁎a Department of Nuclear Medicine, Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai 200025, Chinab Institute of Endocrinology and Metabolism, Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai 200025, China

a b s t r a c ta r t i c l e i n f o

☆ This work was supported by grants from the NationaChina (No. 81171367) and the Innovation Program of SCommission (No. 12YZ041). Conflicts of interest and sou⁎ Corresponding author. Tel.: +86 1381 8172 489; fa

E-mail address: zhang_yifan@126.com (Y. Zhang).

0969-8051/$ – see front matter © 2014 Elsevier Inc. Alhttp://dx.doi.org/10.1016/j.nucmedbio.2013.10.009

Article history:

Received 18 August 2013Received in revised form 21 September 2013Accepted 15 October 2013

Keywords:BaculovirusSodium iodine symporterIsletTransplantationMolecular imaging

Objective: The objective of this study was to explore the feasibility of baculovirus vector-mediated sodiumiodide symporter (NIS) gene delivery to monitor islet transplantation.Methods: Baculovirus vectors expressing green fluorescent protein (GFP) or NIS (Bac-GFP and Bac-NIS) wereestablished using the Bac-to-Bac baculovirus expression system. The GFP expression of Bac-GFP-infected ratislets was observed in vitro by fluorescence microscopy. Iodine uptake and inhibition of iodine uptake byNaClO4 in Bac-NIS-infected islets were dynamically monitored in vitro. Bac-GFP- or Bac-NIS-infected isletswere implanted into the left axillary cavity of NOD-SCID mice, and fluorescence imaging and 125I NanoSPECT/CT imaging were subsequently performed in vivo.Results: Bac-GFP efficiently infected rat islets (over 95% infected at MOI = 40), and the expression of GFPlasted approximately two weeks. NaClO4 could inhibit iodine uptake by Bac-NIS-infected islets. In vivo

imaging revealed that the fluorescence intensity of the transplant sites in Bac-GFP-infected groups wassignificantly higher than in the non-infected group. Grafts could be clearly observed by 125I NanoSPECT/CTimaging for up to 8 h.Conclusion: Baculovirus vectors are powerful vehicles for studying rat islets in gene delivery. It is feasible touse a baculovirus vector to delivery an NIS gene for non-invasive monitoring transplanted islets in vivo by theexpression of the target gene.

© 2014 Elsevier Inc. All rights reserved.

1. Introduction

Diabetes, a group of metabolic disorders characterized byhyperglycemia, is a common disease that seriously threatens humanhealth. Type I diabetes is primarily caused by autoimmune diseasesthat result in the destruction of β-cells and absolute insulin deficiency.Insulin replacement therapy is commonly used to manage diabetesbut can be inconvenient and cannot completely recover a stablemetabolic state.

Islet transplantation is a simple, promising technique that caneffectively restore glucose homeostasis and prevent diabetic compli-cations. In 2000, the Canada Edmonton group reported seven diabetespatients who accepted islet transplantation therapy obtained stableglucose levels independent of exogenous insulin [1]. Since then, islettransplantation has been recognized as a great potential treatment fordiabetes; however, the effects vary depending on the number oftransplanted islets and immune rejection [2]. Traditionally, trans-planted islets are assessed by indirect methods such as measuring

l Natural Science Foundation ofhanghai Municipal Educationrces of funding: none declared.x: +86 21 6433 3548.

l rights reserved.

glucose level, C-reactive protein, and first-phase insulin secretion [3],which cannot indicate islet survival or function, or immune rejectionat an early stage. Biopsy is an alternative direct approach, but theprocedure is highly invasive [4].

Molecular imaging has developed rapidly in recent years and isincreasingly being used in vivo. Adenovirus and lentivirus vectors aregenerally used in imaging to deliver target genes in vivo. Lentivirusvectors can enable long-term expression because of integration intothe host chromosome; however, this insertion occurs randomly andmay result in the altered expression of genomic genes and lead totumorigenicity [5]. Adenovirus vectors are another type of genetransfer vector that have become widely used in molecular imaging.Although these types of vectors can transduce into a broad spectrumof cells with high infection efficiency and do not integrate intohuman DNA, the capsid proteins of adenoviral vectors are toxic tohost cells and may raise host immune response against infected cellsin vivo [6].

Baculovirus vectors are non-cytotoxic gene vectors with a largepackaging capacity (38 kb). These types of vectors cannot randomlyinsert into DNA to induce cell mutations [7]. High-quality and high-titer recombinant baculovirus can be prepared simply and rapidly inroutine laboratory settings [8–11]. Ma et al. performed an in vitrostudy on human and mouse islets transfected with recombinantbaculovirus expressing the GFP or Lac-Z genes without cytotoxicity.

172 S. Liu et al. / Nuclear Medicine and Biology 41 (2014) 171–178

The results revealed that the baculovirus vector may be a useful toolfor facilitating the transfer of foreign genes and can infect islets [12].

In our previous study, we observed that FTC-133 (follicular thyroidcarcinoma) cells infected by a recombinant baculovirus-mediated NISreporter gene in vitro can successfully express the target gene at a highlevel [13]. No studies have explored whether baculovirus-mediatedreporter gene expression can be used to monitor baculovirus-infectedtransplanted islets in vivo. The objective of this study is to evaluate theefficiency of a baculovirus vector in transducing rat islet cells and toexplore the feasibility of using the baculovirus vector for the deliveryof an NIS gene as a reporter to monitor the transplanted islets withtargeted gene expression.

2. Materials and Methods

2.1. Preparation of recombinant baculoviruses

A recombinant baculovirus plasmid expressing the NIS gene(pFBNIS) was constructed from the baculovirus vector pFBGFPR(received from theMolecular Biology Institute, Hong Kong University,Hong Kong, China) as previously described [13]. The baculovirusplasmids pFBGFPR or pFBNIS were transduced into competent DH10E. coli (Invitrogen, Carlsbad, CA, USA), which were cultured at 37°C for24–48 h. The bacmid plasmid DNAwas extracted and transfected intoSpodoptera frugiperda sf9 (Entomological Institute, Chinese Academyof Sciences, Shanghai, China) using the liposome method. Thesupernatants were collected after 5–6 days, and the viral titers weredetermined using plaque assays. For viral amplification, the sf9 cellswere incubated at a multiplicity of infection (MOI) of 0.1, thesupernatants were collected after 6–7 days, centrifuged at 4 °C(500 g, 20 min), filtered (0.45 μm; MILLEX-GP) and stored at 4 °C.The recombinant baculoviruses (Bac-GFP, Bac-NIS) titers wereapproximately both approximately 5 × 109 pfu/ml.

2.2. Isolation and identification of islets

Specific pathogen-free Sprague Dawley male rats (SD rat, 220–260 g) were raised in the Shanghai Slaccas Experiment AnimalCorporation, Shanghai Institute for Biological Science. The animalstudies were approved by the local Ethics Committee (Shanghai JiaoTong University, School of medicine) and performed according toethical principles of animal experimentation. The SD rats wereanesthetized with 1% pentobarbital sodium (40 mg/kg, i.p. (Sigma)).A medioventral laparotomy was performed, and the common bileduct was occluded at the opening of the duodenum and separated atthe proximal end. A tubelet (Anilab) was retrogradely inserted intothe common bile duct and fixed. The rats were sacrificed via openingof the chest and cardiac puncture. Collagenase P (8–10 ml, 0.5 mg/ml;Roche) was slowly injected through the tubelet, and the pipe wasremoved after complete filling of the tail of the pancreas andpancreatic body. The common bile duct was ligated, and the pancreaswas quickly excised. The pancreas was placed into 6 ml of Hanks’solution (Gibco), incubated in a water bath at 37 °C for 15 min, mixed,and 10 ml of bovine serum was added quickly after the pancreatictissue became sand-like. The volume was adjusted to 50 ml withHanks’ solution, and the tubes were centrifuged at 4 °C for 1 min(1000 rpm). The pellet was washed twice using Hanks’ solution,filtered through a 30-mesh stainless steel cellular sieve, and thevolume was adjusted to 50 ml with Hanks’ solution. The suspensionwas centrifuged at 1000 rpm for 3 min. The supernatant wereremoved. Then, 250 μl of BSA (Gibco) and 4 ml of 25% Ficoll400(GE) were added to the tube, mixed, and then 2 ml of Ficoll 400 withdensities of 23%, 20% and 11% and Hanks’ solution were added slowlyand successively before the tubes were centrifuged using a horizontalrotor (Shanghai Anting Instruments, Shanghai China; 15 min,2000 rpm). After centrifugation, the islets between the 23% and 20%

and 20% and 11% layers were placed in a 15 ml tube and centrifuged at1000 rpm for 5 min. The supernatant was removed, Hanks’ solutionwas added, and the mixture was transferred into a 10 ml culture dish.The islet cells were picked under a stereo microscope. Perfusion,washing and purification were all performed with pre-cooledsolutions (4 °C). The isolated islets were then incubated in completemedium (RPMI-1640 containing 10% FBS and 1% penicillin/strepto-mycin; all from Invitrogen) at 37 °C in 5% CO2.

To identify the islets, 10 mg of dithizone (DTZ; Gibco) wasdissolved in 10 ml of DMSO, diluted 1:100 with Hanks' solution (pH7.8), 0.22 μm filtered, and incubated with the islet preparation atroom temperature for 10 min. Islets with a diameter of 150 μm werecounted as one islet equivalent (IEQ); each IEQ contained approxi-mately 2000 cells [14].

2.3. Evaluation of GFP expression in Bac-GFP-infected islets

Purified islets were seeded into 24-well plates (Corning) at200 IEQ/well, infected with Bac-GFP at MOIs of 0, 10, 20 and 40, andlater incubated in 500 μl of RPMI-1640 containing 11.1 mmol/Lglucose (Shanghai Biological Engineering, Shanghai China), 10% FBSand 1% penicillin/streptomycin at 37 °C in 5% CO2 for 4 h. Themediumwas removed, and the cells were washed twice with PBS followed byincubation in 500 μl of fresh complete medium for 24 h. GFPexpression was observed using an inverted fluorescence microscope(λex =488 nm, λem =520 nm). Because the islets were usuallypresent in clusters, the percentage of fluorescent islets in the totalislets was defined as the infection percentage of the islets.

To monitor the duration of GFP expression, purified islets wereseeded at 200 IEQ/well, infected with Bac-GFP at MOIs of 0 and 40 for4 h, and then incubated for 24 h as described above. The GFPexpression was observed under an inverted fluorescence microscopeat various time points (12 h, 1 day, 2 days, 5 days, 7 days and14 days).

The islet viability was measured with an MTT assay to evaluate thecytotoxicity of the baculovirus. For this assay, 50 IEQ/well seeded in a96-well plate were infected by Bac-GFP at an MOI = 0 or 40 asdescribed above. After 48 h incubation, 20 μl of 5 mg/mL MTTsolution was added to each well, and the medium was removedafter 4 h. Then, the formazan crystals were dissolved in 150 μl ofDMSO. The absorbance was measured at a 490 nm wavelength.

2.4. Measurement of iodine uptake in Bac-NIS-infected islets

Purified islets were in seeded in 24-well plates at 200 IEQ/well andinfected with Bac-NIS at different MOIs (0, 5, 10, 20, and 40) for 4 h,before incubation for 24 h as described above. Each well was washedtwice with bHBSS (HBSS solution adjusted with HEPES to pH 7.3),incubated with 0.5 ml of bHBSS containing 3.7 Bq of Na125I (ShanghaiHinko, Shanghai, China) and 1 μMNaI for 30 min. Thewells were thenwashed three times with ice-cold bHBSS and incubated with 0.5 ml ofice-cold NaOH for 15 min, and the radioactive counts per minute(cpm) of the NaOH solutions were measured using a well-typegamma counter (Shanghai Rihuan company, Shanghai, China).

For dynamic measurements, the purified islets were infected withBac-NIS at MOIs of 0 and 40 for 4 h and then incubated for 24 h,treated with 3.7 Bq of Na125I and 1 μM NaI for 15, 30, and 60 min,,washed three times with ice-cold bHBSS, and then incubated with0.5 ml of ice-cold NaOH for 15 min.

For the inhibition experiments, purified islets were infected withBac-NIS at an MOI of 40 for 4 h and then incubated for 24 h. Then,0.5 ml of bHBSS containing 0.1 μCi of Na125I, 1 μM NaI and variousconcentrations of NaClO4 (0 μMand 30 μM)was added into eachwell.Sixty minutes later, the islets were washed three times with PBS andthen incubated with ice-cold NaOH for 15 min. Then, cpm wasdetermined using a well-type gamma counter.

A

B

C

Fig. 1. (A) Islets viewed under a light microscope (×200). (B) All islets displayed greenfluorescence under fluorescence microscopy (×200). (C) Several fluorescent foci wereobserved in each islet under high power fluorescence microscopy (×400).

173S. Liu et al. / Nuclear Medicine and Biology 41 (2014) 171–178

2.5. Establishment of an in vivo islet transplantation model in mice

Specific pathogen-free NOD-SCID male mice (6–8 weeks) wereraised at the Shanghai Slaccas Experiment Animal Corporation,Shanghai Institute for Biological Science. The animal studies wereapproved by the local Ethics Committee (Shanghai Jiao TongUniversity, School of medicine) and performed according to theethical principles of animal experimentation.

Purified islets (2000 IEQ) were infected with Bac-GFP (MOI = 40)or Bac-NIS (MOI = 40) in 60 mm culture dishes with 4 ml of RPMI1640 containing 11.1 mmol/L glucose, 10% FBS and 1% penicillin/streptomycin for 4 h at 37 °C. The supernatant was removed; theislets werewashed twice with PBS and incubated overnight with freshglucose-free medium. The media and transplanted islets were placedinto 15 ml centrifuge tubes and centrifuged twice at 1000 rpm for2 min. Then, the islets were resuspended with 90 μl of culturemedium and 60 μl of Matrigel (BD Biosiences). The 150 μl suspensionwas injected into the left axillary cavity of NOD-SCID mice (n = 3).The mice were supplied with drinking water containing 5% glucose(Shanghai Chang Zheng Fumin Jinshan Pharmaceutical) to preventhypoglycemia. A control group (n = 3) was also established usingnon-infected islets.

The Bac-NIS-infected islet transplanted models (n = 3) wereestablished in the same manner. However, the mice were adminis-tered 200 μl of sodium iodide solution orally for 2 days prior to islettransplantation to block thyroid uptake of radioiodine. Then, the micewere supplied drinking water containing 5% glucose and fed as usual.Non-infected islet cells were transplanted at the contralateral points.

2.6. Fluorescence imaging of Bac-GFP-infected islet transplantationin vivo

Mice inoculated with Bac-GFP-infected islets were anesthetizedwith 1% pentobarbital sodium (10 ml/kg, i.p.) 24 h after transplanta-tion and placed into a small animal fluorescent imaging system(Carestream In-vivo FXPRO) for GFP optical imaging (λex = 470 nm,λem = 535 nm). Molecular Imaging software (Carestream, version5.X users) was used for imaging and data processing to measure thephotons per square millimeter (Net P/s/mm, photons/mm2) of thetransplanted sites in the infected and noninfected groups.

2.7. 125I NanoSPECT/CT imaging of Bac-NIS-infected islet transplantationin vivo

Mice inoculated with Bac-NIS-infected islets were injected with200 μl of Na125I (18.5 MBq) via the tail vein 24 h after transplantation,anesthetized with 5% isoflurane (Baxter Healthcare Corporation,Puerto Rico) and placed into the NanoSPECT/CT (Bioscan) imagingbed at 30, 60, 120, 360 and 480 min. Anesthesia was maintainedduring imaging acquisition using 1%–2% isoflurane. CT images wereacquired (6 min; CTDI = 6.1 cGy), followed by NanoSPECT images(40 s/frame for systematic scans; total acquisition time, 26 min and40 s). The images were processed and reconstructed using Nuclearv1.02 software (Bioscan), HiSPECT 1.4.2 software (Bioscan) for imageacquisition and InVivoScope1.44 software (Bioscan) for imageanalysis. The region of interest (ROI) was determined in thereconstructed images. The radioactivity per volume unit (Conc) inthe ROIs was measured using InVivoScope 1.44 software (Bioscan).

2.8. Statistical analysis

Each experiment was repeated in triplicate. All data are repre-sented as themean ± SD andwere analyzedwith one-way analysis ofvariance or Student's t-test using GraphPad Prism 5.01(GraphPad, SanDiego, California USA). A P b 0.05 was considered significant.

3. Results

3.1. Expression of GFP in Bac-GFP-infected islets

After isolation and purification, all cells were observed to stain redunder a light microscope after DTZ staining, which confirmed that thecells were islets. After the islets were infected with Bac-GFP atdifferent MOIs, many islets expressed GFP with several foci of

174 S. Liu et al. / Nuclear Medicine and Biology 41 (2014) 171–178

fluorescence observed in each islet (Fig. 1). The percentage andfluorescence intensity of the Bac-GFP-infected islets graduallyincreased with the increase in MOI. The infection percentage reachedas high as 95% (MOI = 40).

Bac-GFP-infected islets (MOI = 40) were observed under aninverted fluorescence microscope at 12 h and 1 day, 3 days, 5 days,7 days and 14 days after infection. The islets began to display

B

C

A

Fig. 2. Islets viewed under fluorescence microscopy. (A) Islets displayed clearfluorescence after 24 h (×200). (B) The fluorescence intensity and number offluorescent-expressing significantly decreased after 7 days (×200). (C) Islets displayedweak fluorescence after 14 days (×200).

fluorescence at 12 h. Fluorescence intensity increased to the highestlevels at 24 h and significantly decreased after 7 days. The fluores-cence of some islets diminished after 14 days, with a very low level offluorescence expression (Fig. 2). Non-infected islets did not displayfluorescence at any time point.

3.2. Influence of Bac-GFP on islet survival

Islets infected with Bac-GFP at an MOI of 40 displayed nosignificant difference compared with the control group (MOI = 0)in the MTT assay, indicating that the recombinant baculovirus was notcytotoxic to the islets (data not shown).

Fig. 3. (A) Influence of MOIs on iodine intake by Bac-NIS-infected islets in vitro. Iodineuptake by Bac-NIS-infected islets increased with MOI increases. When MOI = 40,radioiodine uptake in infected groups was significantly higher (12-fold) than in non-infected groups (n = 3 p b 0.01); (B) Dynamics of iodide uptake by Bac-NIS-infectedislets in vitro. Radioiodine uptake gradually increased along with time, reached a peakapproximately 60 min and then gradually decreased with the passage of time; (C)Inhibition of iodide uptake by NaClO4 in Bac-NIS-infected islets in vitro. NaClO4

significantly inhibited 125I uptake in Bac-NIS-infected groups (n = 3; p b 0.01).

175S. Liu et al. / Nuclear Medicine and Biology 41 (2014) 171–178

3.3. Functional iodine uptake by Bac-NIS-infected islets

Iodine uptake by Bac-NIS-infected islets increased with MOI. At anMOI of 40, the uptake of radioactive 125I was approximately 12-foldhigher than in the control group (MOI = 0; Fig. 3A).

125I uptake gradually increased in Bac-NIS-infected islets overtime. The uptake was highest at 60 min and then gradually reducedover time because of the export of iodine. However, Bac-NIS-infectedislets maintained a higher iodine uptake between 15 and 120 mincompared with non-infected groups (Fig. 3B), indicating that NISexpressed in Bac-NIS-infected islets could take in iodine. However,iodine levels could not be maintained as islets cannot effectivelyprocess and store iodine effectively.

When the cells were treated with 30 μM NaClO4, the rate of 125Iuptake in Bac-NIS-infected islets was significantly lower than in thenon-infected group (Fig. 3C), indicating that radioiodine uptakemediated byNISwas inhibited by theNa+-K+-ATPase inhibitor NaClO4.

3.4. Fluorescence imaging in animal model of islet transplantation

The expression of GFP in the Bac-GFP-infected islet transplantationmodel was observed using a small animal fluorescence imaging system.The optical signal at the left axillary cavitywas significantly higher in theinfectedgroups than in thenon-treated group (Fig. 4A). Thephotonspersquare millimeter (Net P/s/mm2) at the transplant sites of the Bac-GFP-infected group were significantly higher than those in the non-infectedgroup (Fig. 4B). These results indicate that the recombinant GFP genewas expressed after the Bac-GFP-infected islets implanted into mice,

infectedcontrol

A

B0

Fig. 4. (A) Fluorescence imaging of Bac-GFP-infected islets in vivo using a small animalmodel of islet transplantation.Theoptical signal at the left axillary cavityof themice injectedwith Bac-GFP-infected islets was significantly higher than that of non-infected groups; (B)Quantitativefluorescence imagingof Bac-GFP-infected islets invivo. Thephotonspersquaremillimeter for the ROI (Net P/s/mm2) of the transplanted Bac-GFP-infected islets weresignificantly higher than those of non-treated groups (n = 3, P b 0.05).

which provides a foundation for baculovirus-mediated monitoring ofislet transplantation using the NIS reporter gene.

3.5. 125I SPECT/CT imaging in animal model of islet transplantation

Twenty-four hours after the mice were inoculated with Bac-NIS-infected islet cells, Na125I (18.5 MBq) was injected into the tail vein forSPECT/CT imagingatdifferent timepoints. Significant radioactive uptakewas observed at the transplant sites within 30 min. Other organs, suchas the heart, kidney, stomach and bladder, also displayed significantuptake of 125I at this time (Fig. 5A). The transplant sites and heart hadclear radioactivity at 60 min, whereas the kidney had decreasedradioactivity (Fig. 5B.). The radioactivity of the transplant sites slightlydecreased at 120 min, when radioactivity was barely detectable in thekidney and radioactivity increased in the stomach (Fig. 5C). Theradioactivity of the transplant sites decreased gradually between 360and 480 min and displayed low radioactive uptake at 480 min. Thestomach displayed high radioactivity at 360 min, which decreased after480 min. Cardiac radioactivity significantly reduced after 480 min, andthe bladder displayed massive radioactive retention throughout theimaging process (Fig. 5D&E). The transplanted non-infected islets at theright axillary cavity displayed no radioactive uptake.

SPECT/CT tomography scanning, performed at 60 min after Na125Iadministration, allowed transplanted infected islets to be located fromthe coronal, sagittal and transverse planes accurately (Fig. 6).

ROIs were created by CT positioning during SPECT imaging todescribe the sites of transplanted islets, heart, liver, kidney andmuscle. Their Conc values were obtained at various time points(Fig. 7). The Conc value of graft was highest at 60 min after injectionof Na125I and then slightly decreased at 360 min. The cardiac Concvalue was highest also at 60 min, remarkably decreased to a long-term plateau between 120 and 360 min and then continued todecrease. The cardiac Conc value was conspicuously lower than that ofthe graft between 120 and 360 min. The Conc value of the kidney washighest at 30 min and then decreased rapidly until reaching its lowestvalue at 120 min. The Conc values of the liver and muscle remainedlow at all time points, though the values in the liver were slightlyhigher than in the muscle. The Conc value of the transplanted siteswas about three-fold higher than the muscle time points.

These results demonstrated that, after Bac-NIS-infected isletstransplantation, the expression of the NIS gene effectively mediatedthe uptake of radioiodine, indicating that monitoring transplantedBac-NIS-infected islets in vivo by SPECT is feasible.

4. Discussion

Islet transplantation can potentially bring new hope for patientswith diabetes; however, many problems still need to be solved toimprove the efficacy of this technique, including the optimal cellnumber of transplantation and monitoring of immune rejection afterislet transplantation. Notably, the methods for non-invasively anddynamically in vivo monitoring of the distribution, survival andmigration of transplanted islets require further research.

There is unique advantage in molecular imaging in monitoring celltransplantation. Lu et al. [15] demonstrated that adenoviral- andlentiviral-mediated luciferase genes transduced to islets could beused for monitoring islet transplantation in vivo; a recombinanttransfected lentivirus enabled long-term monitoring. Virostko et al.[16] generated a transgenic mouse expressing luciferase under thecontrol of the mouse insulin I promoter (mouse insulin promoter-luciferase-Vanderbilt University (MIP-Luc-VU)) and performed afluorescence quantitative imaging study of transplanted islet β-cells,which could reflect the number and function of transplanted cells.However, fluorescence was only expressed for a short period, and thedeeporgans couldnot be imaged easily.MRIhas a high spatial resolutionand provides clear images. MRI imaging of SPIO (superparamagnetic

A B C D E

Fig. 5. Dynamic 125I SPECT/CT imaging of Bac-NIS-infected islets in vivo. (A) Radioactive uptake significantly increased at the graft sites in infected groups at 30 min after Na125I(18.5 MBq) injection into the tail vein of the mice (white arrow). Other organs such as the heart, kidney, stomach and bladder also had significant radioactivity at 30 min; (B) Thetransplanted infected islets and heart had clear radioactivity at 60 min, whereas the radioactivity in the kidney decreased; (C) The radioactivity of the transplanted-treated cellsslightly decreased at 120 min, and the kidney radioactivity was unclear, whereas the radioactivity of the stomach increased; (D, E) The radioactivity of the infected-islets graduallydecreased between 360 and 480 min, and low radioiodine uptake was observed at 480 min; the stomach had obvious radioactivity at 360 min, which decreased after 480 min;cardiac radioactivity was significantly reduced after 480 min, and the bladder had massive radioactive retention throughout the imaging process; The transplanted non-infectedislets in the right axillary of the same mouse displayed no radioiodine uptake.

176 S. Liu et al. / Nuclear Medicine and Biology 41 (2014) 171–178

iron oxide)-labeled transplanted islets significantly improved thesensitivity of detection. However, dead SPIO-labeled cells can also bedetected, even when engulfed by macrophages, MRI imaging cannotdistinguish the dead transplanted islets from the surviving islets [17].

Radionuclide-mediated reporter gene imaging could potentially beused as an indirect non-invasive method to monitor the migration,distribution and survival of transplanted cells by detecting the uptake ofradionuclide to monitor and evaluate transplanted cells. Lu et al. [18]performed PET imaging of transplanted islets transfected with arecombinant adenovirus expressing mutant thymidine kinase (sr39tk)gene. OnMicroPET imaging, the signal intensity of the transplanted sitesreflected the number of transplanted islet cells. Subsequently, Lu et al.[19] attempted long-term monitoring using a lentivirus vector-mediated sr39tk gene. A persistent and very stable signal could bedetected by PET imaging in the transplanted islets for over 90 days.

However, there are some defects using adenovirus and lentivirusas vectors, such as the strong immunogenicity of adenovirus andseemingly random genome insertion into the host chromosomeduring gene transfer via lentivirus vectors. In contrast, baculovirusvectors are novel and easily prepared gene carriers without anycytotoxicity. Ma et al. infected human and murine islet cells withbaculoviruses expressing GFP or Lac-Z in vitro and observed that therecombinant genes were successfully expressed in the islets, and nocytotoxicity was observed even at an MOI of 1000 [20]. In this study,purified rat islets were infected with baculoviruses expressing GFP orNIS in vitro. The percentage of Bac-GFP-infected islets reached up to95%,even at a relatively low MOI (MOI = 40), indicating that baculoviruscan mediate reporter gene expression in islets without cytotoxicity. Inaddition, the purified rat islet cells infected with Bac-NIS displayediodine uptake that could be inhibited by NaClO4, which is consistentwith the reported function of NIS as an iodide symporter [20].

Baculovirus can be inactivated in vivo by the complement system[21], which may limit their direct application in vivo. However, theislets were infected with the recombinant baculovirus in vitro andthen transplanted into animals, which might avoid the problemcaused by the complement system. By establishing a Bac-GFP-infectedislet transplantation model, we confirmed that high GFP fluorescencesignal could be detected at the transplanted sites by in vivo imaging,which indicated that the exogenous baculovirus-encoded recombi-nant gene could be expressed in vivo.

125I SPECT/CT imaging for baculovirus-mediated NIS gene expres-sion in transplanted islets was performed in vivo; the transplantedislets could be clearly observed up to 8 h. The radioactivity washighest at 30–60 min after injection of 125I. The results confirmedthat the recombinant Bac-NIS-infected islets were successfullytransplanted into mice and also that exogenous NIS couldeffectively mediate uptake of radioiodine, demonstrating thatusing a baculovirus-mediated NIS gene to monitor islet transplan-tation is feasible. It might be helpful to have a clearer imaging ofthe graft site if increasing the number of islet cells implanted.However, because of the difficulties in separation and purificationof islet cells, the main purpose of this study was to observewhether the baculovirus-mediated reporter gene could be adoptedin vivo to monitor the islet cell transplantation. In this case, thepresent research was only conducted with a relatively modestnumber of islet cells for transplantation.

As 125I is mainly excreted via the kidneys, the radioactivity of thekidneys was clear at 30 min after injection of Na125I, which wassignificantly reduced at 60 min, whereas the bladder maintained ahigh level of radioactivity. Similarly, high levels of 125I weremaintained in the blood and heart. Unlike thyroid cells, whereaccumulated 125I could be effectively maintained, exogenous NIS-mediated 125I uptake could not be maintained in baculovirus-infectedislets. However, 125I circulated for a long period in the blood.Therefore, the islets expressing NIS constantly took up radioiodineand could be clearly imaged over a long period of time, which enabledthe convenient monitoring of islet transplantation.

As the thyroid glands of the NOD-SCID mice were blocked withLugol's Solution before establishing the Bac-NIS-infected islet trans-plantation model, the thyroid was not observed to be radioactive.Thyroxine synthesized in the thyroid is mainly metabolized in theliver; however, as the thyroid could not take up radioiodine, the liverwas not observed to be radioactive. As the normal gastric mucosaexpresses NIS, the gastric mucosa could continuously take up 125Ifrom circulation, which led to obvious radioactivity in the stomach.Similarly, as the stomach could take up 125I but not effectivelymaintain it, the intestine was not observed to be radioactive.

NIS is a glycoprotein located on the thyroid cell membrane, whichcan transport one I− molecule and two Na+ molecules into thyroidcells via an outer membrane Na+ electrical potential difference, and

D E

F

B CA

Fig. 6. 125I SPECT/CT imaging of Bac-NIS-infected islets in vivo at 60 min after injection of Na125I (18.5 MBq) into the tail vein. The upper row shows the SPECT image (A), CT image(B) and SPECT/CT fusion image (C) at 60 min. The implanted sites and various organs can be clearly located by CT. The radioactive uptake significantly increased at the transplantedsite (white arrow). The lower row displays the tomography of the graft in the coronal (D), sagittal (E), and transverse (F) planes, which clearly indicates the location, size and shapeof the transplanted islets.

177S. Liu et al. / Nuclear Medicine and Biology 41 (2014) 171–178

the energy for this process mainly comes from Na+-K+-ATPase [22].Therefore NIS, which mediates uptake of 125I−, 123I−, 131I− and also99mTcO4

−, is considered a good reporter gene for imaging. These

radiopharmaceuticals are commonly used clinically, are non-immu-nogenic, conveniently sourced, and can be imaged using routineSPECT techniques. The positron radionuclide 124I produced by particleaccelerators can be used for PET imaging and has a long half-life,which may improve the spatial resolution, sensitivity and long-termmonitoring of islet transplantation. 124I also avoids the disadvantages

of F-18-labeled drugs, such as a short half-life and complex synthesis.Therefore, the monitoring of islet transplantation using Bac-NIS andradioiodine ions has potential clinical value.

Though the common sites used for islet transplantation includethe kidneys [23,24], liver [23,25], spleen [26,27], muscle [28,29] andother tissues and organs, considering the advantages and disadvan-tages of these sites, there is still a controversy as to which site is mostoptimal for islet transplantation [30,31]. While, axillary cavity hasalso been chosen as the islet transplantation site [18,19] to focus on

Fig. 7. Dynamic quantitative analysis of radioiodine uptake in various organs of mice.ROIs for infected-islets transplantation site, heart, liver, kidney and muscle werepositioned by CT for SPECT imaging, the Conc values (μCi/mm3) of the organs wereplotted along with time. The Conc value of the infected islet transplantation site washighest at 60 min and then slightly reduced at approximately 360 min. The cardiacConc value reached a peak at 60 min and then significantly decreased. The valuemaintained a long-term plateau between 120 and 360 min and then apparentlydecreased. Between 120 and 360 min, the cardiac Conc value was obviously lower thanthat site of infected islet transplantation. The Conc value of the kidney was highest at30 min, then significantly decreased and was lowest at 120 min. The Conc value of theliver and muscle remained low during the whole imaging process, whereas the Concvalue of the liver was slightly higher than that of muscle.

178 S. Liu et al. / Nuclear Medicine and Biology 41 (2014) 171–178

observing iodine uptake by graft and avoid the potential influence ofradioactive drugs onmetabolism. Our study of 125I SPECT/CT imagingrevealed low and stable radioactivity in liver and muscle tissue over48 h, with slightly higher levels in liver, indicating that baculovirus-mediated monitoring of islet transplantation into the liver or muscletissue using NIS may be available in theory. At the same time, theresults also demonstrated that baculovirus-mediated monitoring ofislet transplantation into the kidney using the NIS reporter gene isnot feasible, as 125I is mainly excreted via the kidneys.

Although the monitoring effect using baculovirus is not as longlasting as when using lentivirus, it demonstrated its advantage indetecting the cell number after islet transplantation because of itshigh infection efficiency and low cytotoxicity. In the present study,we focused on the feasibility of baculovirus-mediated gene transferas a non-invasive approach to monitoring the islet cells aftertransplantation in vivo. Further study will be necessary toinvestigate the correlation between the cell number and itsradioactive counting and to also evaluate the therapeutic effectafter islet cell transplantation.

5. Conclusions

In conclusion, this study confirms that baculovirus-mediatedNIS reporter gene monitoring of islet transplantation is feasible invivo. This technique provides a foundation for molecular imaging-based monitoring of the survival and distribution of islets after invivo transplantation.

Acknowledgments

This work was supported by grants from the National NaturalScience Foundation of China (No. 81171367, 30570525) and theInnovation Program of Shanghai Municipal Education Commission(No. 12YZ041).

References

[1] Shapiro AM, Lakey JR, Ryan EA, Korbutt GS, Toth E, Warnock GL, et al. Islettransplantation in seven patients with type 1 diabetes mellitus using a glucocorti-coid-free immunosuppressive regimen. N Engl J Med 2000;343(4):230–8.

[2] Low G, Hussein N, Owen RJ, Toso C, Patel VH, Bhargava R, et al. Role of imaging inclinical islet transplantation. Radiographics 2010;30(2):353–66.

[3] Atkinson MA, Maclaren NK. The pathogenesis of insulin-dependent diabetesmellitus. N Eng J Med 1994;331(21):1428–36.

[4] Wang P, Medarova Z, Moore A. Molecular imaging: a promising tool to monitorislet transplantation. J Transplant 2011;2011:202915.

[5] Bangari DS, Mittal SK. Current strategies and future directions for eludingadenoviral vector immunity. Curr Gene Ther 2006;6(2):215–26.

[6] Yang Y, Nunes FA, Berencsi K, Furth EE, Gönczöl E, Wilson JM. Cellular immunity toviral antigens limits E1-deleted adenoviruses for gene therapy. Proc Natl Acad SciU S A 1994;91(10):4407–11.

[7] Hu YC. Baculoviral vectors for gene delivery: a review. Curr Gene Ther 2008;8(1):54–65.

[8] Jarvis DL, Garcia Jr A. Long-term stability of baculoviruses stored under variousconditions. Biotechniques 1994;16(3):508–13.

[9] Airenne KJ, Peltomaa E, Hytonen VP, Laitinen OH, Ylä-Herttuala S. Improvedgeneration of recombinant baculovirus genomes in Escherichia coli. Nucleic AcidsRes 2003;31(17):e101.

[10] Airenne KJ, Laitinen OH, Mähönen AJ. Safe, simple and high-capacity gene deliveryinto insect and vertebrate cells by recombinant baculoviruses. In: Friedmann T,Rossi J, editors. Gene transfer: delivery and expression of DNA and RNA. New York:Cold Spring Harbor laboratory Press; 2007. p. 313–25.

[11] Laitinen OH, Airenne KJ, Hytonen VP, Peltomaa E, Mähönen AJ, Wirth T, et al. Amultipurpose vector system for the screening of libraries in bacteria, insect andmammalian cells and expression in vivo. Nucleic Acids Res 2005;33(4):e42.

[12] Ma L, Tamarina N, Wang S, Kuznetsov A, Patel N, Kending C, et al. Baculovirus-mediated gene transfer into pancreatic islet cells. Diabetes 2000;49(12):1986–91.

[13] Zhou X, Li B, Wang J, Yin H, Zhang Y. The feasibility of using a baculovirus vector todeliver the sodium-iodide symporter gene as a reporter. Nucl Med Biol2010;37(3):299–308.

[14] Bonner-Weir S. Anatomy of the islet of Langerhans. In: Samols E, editor. Theendocrine pancreas. New York: Raven Press; 1991. p. 15–7.

[15] Lu YX, Dang H, Middleton B, Zhang Z, Washburn L, Campbell-Thompson M, et al.Bioluminescent monitoring of islet graft survival after transplantation. Mol Ther2004;9(3):428–35.

[16] Virostko J, Radhika A, Poffenberger G, Chen Z, Brissova M, Gilchrist J, et al.Bioluminescence imaging in mouse models quantifies β cell mass in the pancreasand after islet transplantation. Mol Imaging Biol 2010;12(1):42–53.

[17] Biancone L, Crich SG, Cantaluppi V, Romanazzi GM, Russo S, Scalabrino E, et al.Magnetic resonance imaging of gadolinium-labeled pancreatic islets for experi-mental transplantation. NMR Biomed 2007;20(1):40–8.

[18] Lu YX, Dang H, Middleton B, Zhang Z, Washburn L, Stout DB, et al. Noninvasiveimaging of islet grafts using positron-emission tomography. Proc Natl Acad Sci U SA 2006;103(30):11294–9.

[19] Lu Y, Dang H, Middleton B, Campbell-Thompson M, Atkinson MA, Gambhir SS,et al. Long-term monitoring of transplanted islets using positron emissiontomography. Mol Ther 2006;14(6):851–6.

[20] Kogai T, Schultz JJ, Johnson LS, Huang M, Brent GA. Retinoic acid inducessodiumyiodide symporter gene expression and radioiodide uptake in the MCF-7breast cancer cell line. Proc Natl Acad Sci U S A 2000;97(15):8519–24.

[21] Hofmann C, Sandig V, Jennings G, Rudolph M, Schlag P, Strauss M. Efficient genetransfer into human hepatocytes by baculovirus vector. Proc Natl Acad Sci U S A1995;92(22):10099–103.

[22] Dohán O, De la Vieja A, Paroder V, Riedel C, Artani M, Reed M, et al. Thesodium/iodide symporter (NIS): characterization, regulation, and medicalsignificance. Endocr Rev 2003;24(1):48–77.

[23] Mellgren A, Schnell Landström AH, Petersson B, Andersson A. The renalsubcapsular site offers better growth conditions for transplanted mousepancreatic islet cells than the liver or spleen. Diabetologia 1986;29(9):670–2.

[24] Hiller WF, Klempnauer J, Lück R, Steiniger B. Progressive deterioration ofendocrine function after intraportal but not kidney subcapsular rat islettransplantation. Diabetes 1991;40(1):134–40.

[25] Kemp CB, Knight MJ, Scharp DW, Ballinger WF, Lacy PE. Effect of transplantationsite on the results of pancreatic islet isografts in diabetic rats. Diabetologia1973;9(6):486–91.

[26] Kaufman DB, Morel P, Field MJ, Munn SR, Sutherland DE. Purified canine isletautografts: functional outcome as influenced by islet number and implantationsite. Transplantation 1990;50(3):385 391385 391.

[27] Gray DW. Islet isolation and transplantation techniques in the primate. SurgGynecol Obstet 1990;170:225–32.

[28] Axen KV, Pi-Sunyer FX. Long-term reversal of streptozotocin-induced diabetes inrats by intramuscular islet implantation. Transplantation 1981;31(6):439–41.

[29] Juang JH, Hsu BR, Kuo CH. Islet transplantation at subcutaneous and intramuscularsites. Transplant Proc 2005;37(8):3479–81.

[30] van der Windt DJ, Echeverri GJ, Ijzermans JN, Cooper DK. The choice of anatomicalsite for islet transplantation. Cell Transplant 2008;17(9):1005–14.

[31] Merani S, Toso C, Emamaullee J, Shapiro AM. Optimal implantation site forpancreatic islet transplantation. Br J Surg 2008;95(12):1449–61.