Optimizing reporter constructs for in vivo bioluminescence imaging of interferon-γ stimulated mesenchymal stromal cells

Accepted Manuscript Optimizing reporter constructs for in vivo bioluminescence imaging of IFN-γ stimulated mesenchymal stromal cells Jorge Perez-Galarza, Françoise Carlotti, Martijn J. Rabelink, Steve Cramer, Rob C. Hoeben, Willem E. Fibbe, Melissa van Pel PII:

S0301-472X(14)00143-X

DOI:

10.1016/j.exphem.2014.04.004

Reference:

EXPHEM 3122

To appear in:

Experimental Hematology

Received Date: 28 November 2013 Revised Date:

26 March 2014

Accepted Date: 8 April 2014

Please cite this article as: Perez-Galarza J, Carlotti F, Rabelink MJ, Cramer S, Hoeben RC, Fibbe WE, van Pel M, Optimizing reporter constructs for in vivo bioluminescence imaging of IFN-γ stimulated mesenchymal stromal cells, Experimental Hematology (2014), doi: 10.1016/j.exphem.2014.04.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Optimizing reporter constructs for in vivo bioluminescence imaging of IFN-γ stimulated mesenchymal stromal cells Jorge Perez-Galarza1, Françoise Carlotti2,3, Martijn J. Rabelink2, Steve Cramer2, Rob

RI PT

C. Hoeben2, Willem E. Fibbe1 and Melissa van Pel1

Departments of 1Immunohematology and Bloodtransfusion, 2Molecular Cell Biology,

Corresponding author: Melissa van Pel, PhD

M AN U

SC

and 3Nephrology Leiden University Medical Center, Leiden, The Netherlands

Department of Immunohematology and Bloodtransfusion

TE D

Leiden University Medical Center

PO Box 9600, 2333AZ, Albinusdreef 2. Leiden, The Netherlands.

EP

[email protected] Tel: +31 71 526 5277

AC C

Fax: +31 71 526 5267

Category: stem cell biology Word count: 3661

1

ACCEPTED MANUSCRIPT ABSTRACT

Mesenchymal Stromal Cells (MSC) are a promising treatment modality for a variety of diseases. Strategies to investigate the fate of MSC in vivo are important to unravel

RI PT

their therapeutic mechanisms. However, currently available techniques are hampered by their low sensitivity. We therefore aimed to optimize in vivo

bioluminescence imaging of MSC. We compared MSC transduced with Firefly (Fluc)

SC

and transmembrane-bound Gaussia luciferase (TMGluc) driven by the human

cytomegalovirus (CMV), spleen focus-forming virus (SFFV) and elongation factor 1-

M AN U

alpha (EF1α) promoters.

Although CMV-TMGluc-transduced MSC showed the highest light intensity in vitro, the signal was almost undetectable in vivo. SFFV-Fluc-transduced MSC revealed a bright signal in vivo, but transgene expression was silenced upon in vitro stimulation

TE D

with IFN-γ. Therefore, the SFFV promoter was replaced by the EF1α promoter. Light emission of Fluc under the control of EF1α was similar to SFFV-Fluc. Although EF1α-Fluc light emission was 10-fold decreased in the presence of IFN-γ, when

EP

compared with unstimulated MSC, the bioluminescent signal could still be detected and was clearly distinguishable from untransduced MSC. Furthermore, stimulation of

AC C

MSC with TNF-α hardly affected transgene expression in EF1α-Fluc-transduced MSC. Thus, the use of the EF1α promoter partially overcomes silencing and allows in vivo BLI of IFN-γ stimulated MSC.

Keywords: mesenchymal stromal cells, bioluminescence imaging, luciferase, IFN-γ, animal models

2

ACCEPTED MANUSCRIPT INTRODUCTION Mesenchymal Stromal Cells (MSC) are multipotent cells that have the capacity to differentiate into cell types of the mesodermal lineage. MSC are considered to be important for maintenance and repair of tissues and organs [1,2]

RI PT

and have also been reported to have immunomodulatory capacities [3].

Currently, MSC are studied as a therapy in regenerative medicine [4-6],

immune diseases [7] and autoimmune diseases [8,9]. In addition, stimulation of MSC

SC

with IFN-γ has been shown to increase their immunomodulatory properties [10-12]. Despite their broad application in clinical studies, the mechanisms underlying the

M AN U

immunomodulatory and regenerative properties are not completely understood. In vivo tracking of MSC could lead to a better understanding of their in vivo fate, facilitate mechanistic studies and further improve the therapeutic use of MSC. Bioluminescence imaging (BLI) is a noninvasive molecular imaging technique

TE D

that permits tracking of light-emitting cells for a prolonged period of time in living subjects to study cellular kinetics and migration. Several luciferase reporter genes are available including Gaussia, Firefly, and Renilla, each using its own specific

EP

substrate [13]. Firefly luciferase (Fluc) is the most used enzyme in BLI. However,

AC C

recently, the signal induced by cells transduced with Gaussia luciferase (Gluc) has been reported to be 1,000 times brighter than Fluc-transduced cells [14]. In this study, we aim to optimize a lentiviral construct for in vivo imaging of MSC. To this end, we compared light emission of MSC transduced with either Gluc or Fluc, driven by the human cytomegalovirus (CMV), spleen focus-forming virus (SFFV) and elongation factor 1-alpha (EF1α) promoter in vitro and in vivo. Since IFN-γ may have a silencing effect on the promoters [15], we studied transgene expression in the presence and absence of IFN-γ stimulation. 3

ACCEPTED MANUSCRIPT MATERIALS AND METHODS

Animals Eight to twelve-week old male Balb/c mice (Charles River Maastricht, The

RI PT

Netherlands) were used in all experiments. The animals were fed commercial rodent chow and acidified water ad libitum and were maintained in the animal facilities of the Leiden University Medical Center under conventional conditions. All experimental

SC

protocols were approved by the institutional ethics committee on animal

M AN U

experiments.

Bone marrow and adipose derived MSC isolation

Bone marrow-derived MSC (BM-MSC) are obtained and cultured as previously described [16]. Adipose tissue-derived MSC (AD-MSC) were obtained

TE D

from abdominal fat tissue. The tissue was fragmentized and incubated with Collagenase type IA (2 mg/ml; Sigma-Aldrich, Steinheim, Germany) for 1 hour at 37°C while rotating. Cells were filtered through a 70 µm cell strainer (BD

EP

Biosciences, Erembodegem, Belgium) and expanded. MSC from passage 1-2 were

AC C

used for lentiviral transduction and MSC of passages 4-7 were used in experiments. MSC differentiation was performed as previously described [16]. In indicated experiments, MSCs were cultured in the presence of recombinant murine IFN-γ (10 ng/ml) or recombinant murine TNF-α (10 ng/ml; boh R&D Systems, Abingdon, UK) for 7 days.

4

ACCEPTED MANUSCRIPT Generation of membrane-anchored Gaussia Luciferase Transmembrane Gluc (TMGluc) is obtained using the pDisplay vector system (Invitrogen). Proteins expressed from pDisplay are fused at the N-terminus to the murine Ig κ-chain leader sequence, directing the protein to the secretory pathway,

RI PT

and at the C-terminus to the platelet derived growth factor receptor (PDGFR)

transmembrane domain, which anchors the protein to the plasma membrane at the extracellular side.

SC

Humanized Gluc cDNA was isolated from the pCMV-Gluc1 vector using the following primers Fw: ctcggatccagccaccATGgg (insertion BamHI site before the start

M AN U

codon); Rv: gagCTGCAGgtcaccaccggcccccttg (deletion Stop codon and insertion PstI site). The PCR product was fused to the N-terminus of myc epitope/PDGFR-TM of the pDisplay vector using BamHI and PstI, deleting the N-terminal Ig κ-chain leader sequence and HA tag. The resulting product is a fusion Gluc-cMyc-tag-

TE D

PDGFR-TM

Generation of lentiviral constructs

EP

All lentiviral vectors used are derivatives of the 3rd generation self-inactivating

AC C

(SIN) vector pRRL-cPPT-X-PRE-SIN [17]. pRRL-CMV-TMGluc-IRES-Puro (CMVTMGluc-Puro) was generated by insertion of the TMGluc cDNA in pRRL-CMV-IRESPuro [18]. The TMGluc-IRES-Puro cassette was transferred from the pRRL-CMVTMGluc-IRES-Puro to the pRRL-SFFV lentivirus backbone, resulting in pRRL-SFFVTMGluc-IRES-Puro (SFFV-TMGluc-Puro). pRRL-SFFV was generated by replacing the CMV promoter by the SFFV promoter (generously provided by F.J.T. Staal, LUMC, Leiden, The Netherlands).

5

ACCEPTED MANUSCRIPT pRRL-CMV-DsRedT4-IRES-Luc+ (CMV-DsR-Fluc) was generated by inserting the DsRedT4 gene in the pRRL-CMV-IRES-Luc+. pRRL-CMV-IRES-Luc+ results from insertion of the firefly Luciferase Luc+ gene from pGL3 (Promega, Madison, WI, USA) under the control of an internal ribosomal entry site (IRES) sequence in the

RI PT

previously described pRRL-CMV [19]. The DsRedT4-IRES-Luc+ cassette was

transferred from the pRRL-CMV-DsRedT4-IRES-Luc+ to the pRRL-SFFV lentivirus backbone, resulting in pRRL-SFFV-DsRedT4-IRES-Luc+ (SFFV-DsR-Fluc).

SC

pRRL-EF1alpha-Luc+-IRES-GFP was generated by inserting the cassette Luc+-IRES-GFP in a pRRL vector containing the 1.1Kb Human EF1alpha promoter.

M AN U

The cassette Luc+-IRES-GFP encodes for the Firefly Luciferase Luc+ gene from pGL3 (Promega) with the GFP reporter gene under the control of IRES (figure 1).

Lentiviral particle production

TE D

Viral production was performed using the 293T cell line. Lentiviral transfer vector, viral envelope plasmid pMD2-VSVg, packaging backbone pRSV-REV, and gag-pol element pMDLg-pRRE were mixed with FuGene 6 Transfection Reagent

EP

(Roche Applied Science, Germany), added to IMDM and incubated at RT for 15 min.

AC C

DNA solution was added to the cells and incubated at 37° C, 5% CO 2. Viruses were harvested after 48 and 72 hours. Filtered supernatant was frozen at -80ºC or directly ultra-centrifuged using the SW28 rotor at 50,357 g for 14 hours at 4°C in a L-100 XP ultracentrifuge (Beckman Coulter, Fullerton, CA). Pellets were resuspended in alphaMEM and kept at -80°C. Virus was quantified by anti gen capture ELISA measuring HIV p24 levels (ZeptoMetrix Corporation, NY). Values were converted to an infectious titer using the approximation of 1 ng of p24 that equals 2,500 infectious

6

ACCEPTED MANUSCRIPT units. A functional titer was obtained by titration of transduced HeLa cells at different lentiviral concentrations.

Cell transduction

RI PT

Ten thousand MSC were transduced with a multiplicity of infection 30. Cells were incubated at 37°C and 5% CO 2 and media were refreshed completely after 24 hours of transduction. Transduction efficiency was evaluated by FACS and

SC

luminometry assay. Transduced MSC populations were ≥80% positive for the

M AN U

transgene expression either by direct transduction or by cell sorting.

Substrate preparation

D-luciferin (Synchem OHG, Felsberg-Altenburg, Germany) was diluted in PBS. Coelenterazine free base (Nanolight Technologies, Pinetop, USA) was

TE D

dissolved in acidified methanol (Merck, Darmstadt, Germany) and further diluted in

T cell proliferation

EP

PBS.

AC C

BALB/c splenocytes were stimulated with anti-CD3/CD28 microbeads (Life Technologies, Bleiswijk, The Netherlands) and cultured in the presence or absence of lentivirally transduced MSC for 5 days at 37°C and 5% CO 2. Next, the cultures were incubated for 16 hours with 3H-thymidine and incorporation was measured as a percentage of the control.

7

ACCEPTED MANUSCRIPT Luciferase activity of cultured cells To measure luciferase activity in vitro, cells were plated in a 96 well-plate and were allowed to adhere for 6 hours. Coelenterazine (BioLux Gluc Assay Kit, New England BioLabs) or D-luciferin potassium salt (5 mg per well; SynChem) was added

RI PT

to measure Gluc or Fluc activity respectively. Gluc activity was assessed directly and Fluc at 10 minutes after substrate addition using an IVIS Lumina II (Caliper Life

Science, Hopkinton, MA, USA). Results were analyzed using Living Image software

M AN U

Luciferase activity of lysed cells

SC

4.0 (Caliper).

MSC were plated in triplo in 96-well-plates and washed with PBS. Cells were lysed by luciferase cell lysis buffer (New England Biolabs) and analyzed using luciferin detection reagent (Promega) or coelenterazine (BioLux Gluc Assay Kit),

TE D

followed by immediate measurement using a bench top luminometer (Perkin Elmer).

Bioluminescence Imaging

EP

CMV-DsR-Fluc and CMV-TMGluc-Puro-transduced cells were injected

AC C

subcutaneously on the dorsal side of Balb/c recipients in 100 µl matrigel (BD Biosciences). Images were acquired at 10 minutes after D-Luciferin injection (150 mg/kg) or immediately after coelenterazine administration (7.7 mg/kg) using an IVIS Lumina II. All images were analyzed using Living Image 4.0 software (Caliper).When finishing the experiment, matrigel implants were isolated, fixed in 3.7% formaldehyde (Sigma-Aldrich) and embedded in paraffin. Three µm sections were stained with hematoxylin and eosin (H&E).

8

ACCEPTED MANUSCRIPT Flow Cytometry Analysis All antibodies used are described in supplemental table I. Cells were analyzed on a Canto II with Diva (BD Biosciences) and FlowJo software (Tree Star,

RI PT

USA).

Statistical Analysis

Statistical analysis was performed by the Student t-test using

SC

GraphPadPRISM (GraphPad Software, San Diego, CA). P values of <0.05 were

AC C

EP

TE D

M AN U

considered statistically significant.

9

ACCEPTED MANUSCRIPT

RESULTS

Lentiviral transduction does not alter MSC characteristics

RI PT

To generate MSC expressing the luciferase gene, BM -and AD-MSC were transduced with CMV-DsRed-Fluc, SFFV-DsRed-Fluc, CMV-TMGluc-Puro or EF1αFluc-GFP lentiviral vectors (figure 1). At least 80% of the cells expressed the

transgene. After puromycin selection of CMV-TMGluc-Puro transduced cells, a

SC

population of 100% transduced MSC was obtained (figure 2a).

M AN U

Since lentiviruses integrate randomly into the genome, the expression of cellular genes may be affected, resulting in changes in the biological properties of the MSC. Therefore, we compared growth kinetics, differentiation capacity and phenotype of untransduced and lentiviral vector-transduced MSC-lines obtained from the same parental line and at the same passage number. No differences in cell

TE D

numbers were observed for transduced and untransduced cells (data not shown). In addition, both transduced and untransduced MSC had a similar capacity to

EP

differentiate into the osteogenic and adipogenic lineages (figure 2b). Moreover, transduced MSC exhibit a surface-marker phenotype similar to untransduced cells

AC C

(figure 2c) and had the capacity to suppress T cell proliferation (figure 2d). Thus, transduction of MSC with these lentivirus vectors does not alter the growth kinetics, differentiation capacities and surface phenotype of MSC.

10

ACCEPTED MANUSCRIPT Transduction with Fluc results in a higher signal intensity in vivo compared to TMGluc We investigated the signal intensity of MSC transduced with either CMV-DsRFluc or CMV-TMGluc-Puro in vitro and in vivo. MSC were plated at serial dilutions

RI PT

and luciferase activity was assessed. CMV-TMGluc-Puro-transduced MSC exhibited a 32-fold increase in signal intensity compared to CMV-DsR-Fluc-transduced MSC (figure 3a). Since Fluc is retained intracellularly and TMGluc is presented on the cell

SC

surface, this may play a role in the difference in light-emitting capacity of the

luciferases. Therefore, CMV-DsR-Fluc or CMV-TMGluc-Puro-transduced MSC were

M AN U

lysed prior to substrate addition. Again, CMV-TMGluc-Puro-transduced MSC gave a higher signal compared to CMV-DsR-Fluc-transduced cells, although the difference is smaller following cell lysis (figure 3b).

Although MSC are largely trapped in the lungs upon intravenous

TE D

injection, a substantial number of MSC may spread throughout the recipient. As a consequence, following intravenous injection, the number of MSC at a certain location will be unknown. This will severely hamper the analysis of the

EP

sensitivity of the BLI constructs. Therefore, to compare the signal intensity of

AC C

CMV-DsR-Fluc -and CMV-TMGluc-Puro-transduced MSC in vivo, transduced MSC were loaded into matrigel and injected subcutaneously into Balb/c recipients. This allowed us to measure the bioluminescent signal that was generated by a fixed, and therefore exactly known, number of MSC. At 24 hours after cell injection, a minimum of 1x104 CMV-DsR-Fluc-transduced MSC were detected. In contrast, no specific signal from TMGluc-transduced MSC was observed (figure 3c). However, due to coelenterazine administration an autoluminescent signal

11

ACCEPTED MANUSCRIPT in the upper part of the abdomen was observed. This autoluminescence of coelenterazine was previously reported by others [20]. At 20 days after implantation, signals from CMV-DsR-Fluc-transduced MSC were still observed in one or two implants per mouse, indicating that over time some matrigel

RI PT

implants lost the cells, while others were able to retain the MSC (figure 3d). To

confirm the presence of MSC, the implants were harvested at day 20 and H&E

stainings were performed on paraffin-embedded sections. Indeed, implants that

SC

showed light activity at day 20 correlated with cell clusters in H&E-stained sections (figure 3d.1). No cells were observed in implants that did not show any light activity

M AN U

(figure 3d.2). In contrast, CMV-TMGluc-Puro-transduced MSC were neither detected by BLI in vivo nor in stained sections at day 20. In addition, when imaging the mice that received the matrigel implants, no luciferase signal was obtained ouside of the matrigel. This suggests that MSC did not migrate out of the

TE D

matrigel implants.

Thus, despite the high levels of light emission in vitro, CMV-TMGluc-Purotransduced MSC show a 5-100-fold reduction in signal intensity in vivo compared to

AC C

EP

CMV-DsR-Fluc-transduced MSC.

The SFFV promoter increases the luciferase signal compared to the CMV promoter

It has been reported that the SFFV promoter may both improve the efficiency of transduction and the level of transgene expression compared to the CMV promoter [21,22]. To further increase the BLI signal, the CMV promoter was replaced by SFFV in CMV-DsR-Fluc and CMV-TMGluc-Puro lentiviral vectors (figure 1). Next,

12

ACCEPTED MANUSCRIPT the expression levels of the CMV-DsR-Fluc, SFFV-DsR-Fluc, CMV-TMGluc-Puro and SFFV-TMGluc-Puro constructs were compared in vitro and in vivo. In transduced MSC, replacement of the CMV promoter with the SFFV promoter enhanced the expression of DsRed and TMGluc (figure 4a and 5a). BLI of

RI PT

SFFV-DsR-Fluc-transduced MSC increases light emission 2 to 9-fold compared to CMV-DsR-Fluc-transduced MSC (figures 4b and 4c). Similar results were obtained for SFFV-TMGluc-Puro-transduced MSC in comparison with CMV-TMGluc-Puro-

SC

transduced cells (figure 5b and 5c). Furthermore, a 12-fold increase in light

production was found in lysates obtained from SFFV-DsR-Fluc-transduced MSC

M AN U

compared to CMV-DsR-Fluc-transduced cells (figure 4d). Similarly, a 5.7-fold increase was observed for SFFV-TMGluc-Puro-transduced MSC compared to CMVTMGluc-Puro-transduced cells (figure 5d). Thus, light emission was significantly increased when the luciferase gene was driven by the SFFV promoter instead of the

TE D

CMV promoter. Furthermore, in vitro, 50 SFFV-DsR-Fluc-transduced MSC were detected, compared to 500 CMV-DsR-Fluc-transduced MSC (figure 4c). A similar higher expression was found when SFFV-TMGluc-Puro-transduced MSC were

EP

compared with CMV-TMGluc-Puro-transduced cells (Figure 5c).

AC C

To compare the signal intensity of MSC transduced with CMV-DsR-Fluc, SFFV-DsR-Fluc, CMV-TMGluc-Puro or SFFV-TMGluc-Puro in vivo, a titration of MSC loaded into matrigel was analyzed. Implants containing a minimum of 5,000 SFFV-DsR-Fluc-transduced cells were detected. In contrast, light emission generated by 5,000 CMV-DsR-Fluc-transduced MSC was at levels comparable to PBS implants (figures 4e and 4f). Moreover, SFFV-TMGluc-Puro-transduced MSC were only distinguishable from background levels at the highest cell numbers, while all CMV-TMGluc-Puro implants were at background levels (figure 5e). All MSC 13

ACCEPTED MANUSCRIPT implants were monitored for 20 days. In time, some implants retained the luciferase signal while others lost signal in time. Additionally, some implants with a low signal at day 1, exhibited a higher signal at day 20 suggesting in vivo proliferation of MSC where conditions were favorable (figure 4g).

RI PT

In conclusion, the SFFV promoter increases Fluc and TMGluc signals

compared to the CMV promoter. Despite the high signal generated by the TMGluc constructs in vitro, the signal is strongly decreased in vivo using either SFFV or CMV

SC

promoters.

M AN U

Silencing of the SFFV promoter upon IFN-γ stimulation is partially overcome by the EF1α promoter

Previously, it has been suggested that IFN-γ-stimulated MSC have an enhanced immunomodulatory effect in vivo [3,10-12,23]. Moreover, IFN-γ stimulation

TE D

silences the transcriptional activity of the CMV promoter [15]. To investigate the effect of IFN-γ stimulation on transgene expression, MSC were stimulated with IFN-γ for one week. Stimulation with IFN-γ upregulated the expression of MHC class I,

EP

CD166, CD105 and Sca-1 and to a lesser extent of CD29, CD90, CD44, CXCR4 and

AC C

MHC class II compared to unstimulated cells (figure 6a). Furthermore, after one week of culture in the presence of IFN-γ, DsRed expression was downregulated for 2 logs compared to unstimulated transduced MSC (figure 6e), suggesting that the SFFV promoter was silenced upon IFN-γ stimulation. No difference was observed in transgene expression following stimulation with higher doses of IFN-γ (supplemental figure 1). The silencing of SFFV constructs by IFN-γ was confirmed by the analysis of light emission in vitro (data not shown) and by studying light emission by cell lysates obtained from transduced MSC (figure 6b). In all in vitro 14

ACCEPTED MANUSCRIPT studies, stimulated SFFV-DsR-Fluc-transduced MSC showed transgene silencing in comparison with transduced and unstimulated cells. In vivo, at least 1.0x105 cells were required to produce a signal that was distinguishable from background levels when SFFV-DsR-Fluc-transduced MSC were cultured in the presence of IFN-γ and

RI PT

subsequently implanted in vivo (figures 6c and 6d).

To further improve the constructs to track IFN-γ-stimulated MSC in vivo, the SFFV promoter was replaced by the EF1α promoter and FLuc was placed as a direct

SC

transgene (figure 1). In vitro, stimulation of EF1α-Fluc-GFP-transduced MSC with IFN-γ partially reduces GFP expression (figure 6e). In vivo, IFN-γ-stimulated MSC

M AN U

transduced with the EF1α construct allowed detection of a cell cluster consisting of 1x104 MSC, while already 5x103 unstimulated EF1α-transduced MSC were detected (figure 6c and 6d). Moreover, the silencing effect of IFN-γ in both SFFV-DsR-Flucand EF1α-Fluc-GFP-transduced MSC partially recovered following a week without

TE D

IFN-γ stimulation (figure 6e).

The immunosuppressive function of MSC is further enhanced when IFN-γ is combined with other proinflammatory cytokines, including TNF-α [12,24]. To test

EP

whether transgene expression is affected by stimulation with TNF-α alone or with the

AC C

combination of IFN-γ and TNF-α, MSC were cultured in the presence of IFN-γ and/or TNF-α for one week. Stimulation of EF1α-Fluc-GFP-transduced MSC with TNF-α resulted in a minor reduction of GFP expression compared to IFN-γ stimulation (figure 6f). When EF1α-Fluc-GFP-transduced MSC were stimulated with both IFN-γ and TNF-α, GFP expression was slightly more decreased compared to IFN-γ stimulation alone.

15

ACCEPTED MANUSCRIPT In conclusion, IFN-γ, but not TNF-α, stimulation silences transgene expression of SFFV constructs, while the expression of EF1α-driven constructs is only partially

AC C

EP

TE D

M AN U

SC

RI PT

repressed.

16

ACCEPTED MANUSCRIPT DISCUSSION To investigate the migratory capacities of MSC upon in vivo administration, bioluminescence is an appropriate method that allows longitudinal monitoring of MSC in vivo for extended periods of time. Furthermore, BLI can also be applied as

RI PT

an ex vivo application to localize and subsequently isolate cells from tissues and organs of sacrificed animals.

Transduction of MSC may induce changes in the genome, which in turn may alter

SC

the biological properties of the transduced cells. Therefore, it is important to verify that transduced MSC have retained their main characteristics such as cell growth,

M AN U

differentiation capacity, cell surface phenotype and the capacity to inhibit T cell proliferation. We did not observe any differences between transduced and untransduced cells, suggesting that transduction did not modify the main characteristics of the MSC. When testing the different reporter constructs,

TE D

remarkable differences were found in vitro and in vivo. In vitro, TMGluc-transduced MSC showed higher light emission than Fluc-transduced MSC. However, in vivo was a markedly reduction of light emission of TMGluc-transduced MSC. As a

EP

consequence, implants containing TMGluc-transduced MSC could hardly be distinguished from background levels, while SFFV-TMGluc-Puro-transduced MSC

AC C

could only be detected in clusters of 5x105 MSC. Due to the short wavelength of the light emitted by Gluc, the signal is absorbed in vivo by surrounding tissues [25,26]. Thus the damping of the Gluc signal could not be overcome by the increase in brightness of Gluc. This limitation hampers the application of Gluc as a reporter gene to image low numbers of cells in vivo. Moreover, following a prolonged period in vivo, no TMGluc-transduced MSC could be detected, while Fluc-transduced MSC were still present at the site of injection. This points towards a specific effect on 17

ACCEPTED MANUSCRIPT TMGluc-transduced MSC. It is therefore conceivable that TMGluc-transduced cells are immunogenic in vivo. In addition, no luciferase signal was obtained outside of the matrigel, suggesting that MSC did not migrate out of the matrigel implants.

RI PT

Another advantage of Fluc is the substrate that is used to visualize

transduced cells. Luciferin is easy to use and has an extended half-life (30 minutes), compared to coelenterazine, the substrate for Gluc. Furthermore, coelenterazine

SC

shows auto-oxidation which leads to increased background levels, especially in the liver [20]. Taken together, we conclude that Fluc is a more suitable reporter enzyme

M AN U

than Gluc to study MSC migration by BLI.

Besides the type of reporter enzyme and the cell number, the promoter of the reporter construct exerts influence on the sensitivity of BLI. We were able to increase the sensitivity of BLI by replacing the CMV promoter with SFFV. This promoter

TE D

change allowed us to detect 5,000 cells in vivo.

Stimulation of cells with IFN-γ downregulates transgene expression. Strong decreases in signal intensity were observed for MSC transduced with constructs

EP

containing either the CMV or the SFFV promoters. Silencing and subsequent

AC C

methylation of the CMV promoter upon IFN-γ stimulation has already been reported and it is conceivable that the same silencing mechanism is applicable for the SFFV promoter [15]. The silencing effect was transient, since transgene expression was partially restored after withdrawal of IFN-γ from the culture. This is fully consistent with a mechanism involving transcriptional silencing of transgene expression followed by a DNA methylation-dependent epigenetic lock that freezes the transgene cassette in an inactive state [27].

18

ACCEPTED MANUSCRIPT To overcome the silencing effect of IFN-γ on the CMV and SFFV promoters and to increase the sensitivity of the BLI, Fluc was placed under the control of the cellular EF1α promoter. Previously, a DNA-methylation-mediated silencing effect on SFFV and EF1α was described, illustrating the epigenetic regulation of these

RI PT

promoters [28]. However, in our hands, the EF1α promoter expression was stable for at least 6 cell passages (data not shown). The use of the EF1α promoter resulted in an increase of the signal of the transgene by 1 log both in vitro and in vivo and this is

SC

only partially decreased by IFN-γ. Despite this partial reduction of the transgene expression by IFN-γ, a cluster of 10,000 MSC could still be detected in vivo.

M AN U

Moreover, in vitro stimulation of EF1α-Fluc-GFP-transduced MSC with TNF-α resulted in a minor reduction of GFP expression in comparison with IFN-γ stimulation. This allows imaging of MSC that are stimulated with TNF-α or a combination of both.

TE D

Here, we show that reporter constructs can be silenced by cytokines that are present in the environment of the transduced cells in vivo. Therefore it is important to test how the expression of each reporter construct is affected by cytokines as this

EP

may lead to incorrect interpretations of the results. In transplantation research, the

AC C

absence of reporter gene expression may be interpreted as the absence of the grafted cells. Alternatively, such observation may be explained by silencing of the promoter due to the cytokine milieu in vivo. Here, we show that an appropriate reporter construct can circumvent this pitfall. Transduction of MSC with the EF1αFluc-GFP construct partially overcomes promoter silencing by IFN-γ and allows in vivo imaging of IFN-γ stimulated MSC in murine recipients. In future studies, we will apply the EF1α-Fluc-GFP construct to study MSC migration in disease models with an inflammatory environment. 19

ACCEPTED MANUSCRIPT

Acknowledgements This work received financial support from the Netherlands Organization for Scientific Research (NWO) ZonMW Translational Adult Stem Cell (TAS) program nr

RI PT

11.600.1016 and from the Wijnand M. Pon foundation.

Disclosure statement

AC C

EP

TE D

M AN U

SC

No competing financial interests exist.

20

ACCEPTED MANUSCRIPT References

Gaebel R, Furlani D, Sorg H et al. Cell origin of human mesenchymal stem cells determines a different healing performance in cardiac regeneration. PLoS ONE. 2011;6:e15652-

2.

van Velthoven CT, Kavelaars A, van Bel F, Heijnen CJ. Mesenchymal stem cell treatment after neonatal hypoxic-ischemic brain injury improves behavioral outcome and induces neuronal and oligodendrocyte regeneration. Brain Behav Immun. 2010;24:387-393.

3.

Bernardo ME and Fibbe WE. Mesenchymal stromal cells: sensors and switchers of inflammation. Cell Stem Cell. 2013;13:392-402.

4.

Lee JK, Jin HK, Endo S et al. Intracerebral transplantation of bone marrow-derived mesenchymal stem cells reduces amyloid-beta deposition and rescues memory deficits in Alzheimer's disease mice by modulation of immune responses. Stem Cells. 2010;28:329-343.

5.

Loffredo FS, Steinhauser ML, Gannon J, Lee RT. Bone marrow-derived cell therapy stimulates endogenous cardiomyocyte progenitors and promotes cardiac repair. Cell Stem Cell. 2011;8:389-398.

6.

Stamm C, Westphal B, Kleine HD et al. Autologous bone-marrow stem-cell transplantation for myocardial regeneration. Lancet. 2003;361:45-46.

7.

Le Blanc K, Frassoni F, Ball L et al. Mesenchymal stem cells for treatment of steroid-resistant, severe, acute graft-versus-host disease: a phase II study. Lancet. 2008;371:1579-1586.

8.

Connick P, Kolappan M, Patani R et al. The mesenchymal stem cells in multiple sclerosis (MSCIMS) trial protocol and baseline cohort characteristics: an open-label pre-test: post-test study with blinded outcome assessments. Trials. 2011;12:62-

9.

Duijvestein M, Vos AC, Roelofs H et al. Autologous bone marrow-derived mesenchymal stromal cell treatment for refractory luminal Crohn's disease: results of a phase I study. Gut. 2010;59:1662-1669.

10.

11.

AC C

EP

TE D

M AN U

SC

RI PT

1.

Duijvestein M, Wildenberg ME, Welling MM et al. Pretreatment with interferon-gamma enhances the therapeutic activity of mesenchymal stromal cells in animal models of colitis. Stem Cells. 2011;29:1549-1558. Polchert D, Sobinsky J, Douglas G et al. IFN-gamma activation of mesenchymal stem cells for treatment and prevention of graft versus host disease. Eur J Immunol. 2008;38:1745-1755.

21

ACCEPTED MANUSCRIPT Ren G, Zhang L, Zhao X et al. Mesenchymal stem cell-mediated immunosuppression occurs via concerted action of chemokines and nitric oxide. Cell Stem Cell. 2008;2:141-150.

13.

Prescher JA and Contag CH. Guided by the light: visualizing biomolecular processes in living animals with bioluminescence. Curr Opin Chem Biol. 2010;14:80-89.

14.

Tannous BA, Kim DE, Fernandez JL, Weissleder R, Breakefield XO. Codon-optimized Gaussia luciferase cDNA for mammalian gene expression in culture and in vivo. Mol Ther. 2005;11:435-443.

15.

Harms JS and Splitter GA. Interferon-gamma inhibits transgene expression driven by SV40 or CMV promoters but augments expression driven by the mammalian MHC I promoter. Hum Gene Ther. 1995;6:12911297.

16.

Nauta AJ, Westerhuis G, Kruisselbrink AB et al. Donor-derived mesenchymal stem cells are immunogenic in an allogeneic host and stimulate donor graft rejection in a nonmyeloablative setting. Blood. 2006;108:2114-2120.

17.

Carlotti F, Bazuine M, Kekarainen T et al. Lentiviral vectors efficiently transduce quiescent mature 3T3-L1 adipocytes. Mol Ther. 2004;9:209217.

18.

Uil TG, de Vrij J, Vellinga J et al. A lentiviral vector-based adenovirus fiberpseudotyping approach for expedited functional assessment of candidate retargeted fibers. J Gene Med. 2009;11:990-1004.

19.

Carlotti F, Zaldumbide A, Martin P et al. Development of an inducible suicide gene system based on human caspase 8. Cancer Gene Ther. 2005;12:627-639.

20.

Kimura T, Hiraoka K, Kasahara N, Logg CR. Optimization of enzymesubstrate pairing for bioluminescence imaging of gene transfer using Renilla and Gaussia luciferases. J Gene Med. 2010;12:528-537.

SC

M AN U

TE D

EP

AC C

21.

RI PT

12.

Demaison C, Parsley K, Brouns G et al. High-level transduction and gene expression in hematopoietic repopulating cells using a human immunodeficiency [correction of imunodeficiency] virus type 1-based lentiviral vector containing an internal spleen focus forming virus promoter. Hum Gene Ther. 2002;13:803-813.

22.

Yam PY, Li S, Wu J et al. Design of HIV vectors for efficient gene delivery into human hematopoietic cells. Mol Ther. 2002;5:479-484.

23.

English K, Barry FP, Field-Corbett CP, Mahon BP. IFN-gamma and TNFalpha differentially regulate immunomodulation by murine mesenchymal stem cells. Immunol Lett. 2007;110:91-100. 22

ACCEPTED MANUSCRIPT Ren G, Su J, Zhang L et al. Species variation in the mechanisms of mesenchymal stem cell-mediated immunosuppression. Stem Cells. 2009;27:1954-1962.

25.

Bovenberg MS, Degeling MH, Tannous BA. Enhanced Gaussia luciferase blood assay for monitoring of in vivo biological processes. Anal Chem. 2012;84:1189-1192.

26.

Tannous BA. Gaussia luciferase reporter assay for monitoring biological processes in culture and in vivo. Nat Protoc. 2009;4:582-591.

27.

Hoeben RC, Migchielsen AA, van der Jagt RC, van OH, van der Eb AJ. Inactivation of the Moloney murine leukemia virus long terminal repeat in murine fibroblast cell lines is associated with methylation and dependent on its chromosomal position. J Virol. 1991;65:904-912.

28.

Zhang F, Frost AR, Blundell MP et al. A ubiquitous chromatin opening element (UCOE) confers resistance to DNA methylation-mediated silencing of lentiviral vectors. Mol Ther. 2010;18:1640-1649.

AC C

EP

TE D

M AN U

SC

RI PT

24.

23

ACCEPTED MANUSCRIPT

Figures Legends

Figure 1. Scheme of lentiviral constructs used to transduce MSC. CMV,

RI PT

Cytomegalovirus promoter. SFFV, Spleen focus-forming virus promoter. EF1α, elongation factor 1-alpha promoter. IRES, internal ribosome entry site. Fluc, Firefly luciferase. TMGluc, Trans-membrane Gaussia luciferase. DsRed, Discosoma red

SC

fluorescent protein. GFP, green fluorescent protein. Puro, puromycin.

M AN U

Figure 2. Generation of lentiviral-transduced BM-derived MSC. (a) MSC transduced with different constructs show similar transgene expression. BM-MSC were transduced with either CMV-DsR-Fluc (magenta), SFFV-DsR-Fluc (red), CMVTMGluc-Puro (purple) or EF1α-Fluc-GFP (green). Untransduced MSC are depicted as filled gray in all histograms (b) EF1α-Fluc-GFP-transduced (EF1αFluc), SFFV-

TE D

DsR-Fluc-transduced (SFFVDsR) and untransduced BM-MSC (untrans) have the capacity to differentiate into the adipogenic and osteogenic lineages. (c) Phenotypic

EP

analysis of CMV-DsR-Fluc-transduced, SFFV-DsR-Fluc-transduced, EF1α-FlucGFP-transduced and untransduced BM-MSC. (d) EF1α-Fluc-GFP-transduced MSC

AC C

have the capacity to suppress T cell proliferation. BALB/c splenocytes were stimulated with anti-CD3/CD28 microbeads and cultured in the presence or absence of lentivirally transduced MSC for 5 days. Next, the cultures were incubated for 16 hours with 3H-thymidine and incorporation was measured as a percentage of the control. Results of 2 experiments (each performed in triplicate), using 2 different MSC lines are shown as mean ± SEM.

24

ACCEPTED MANUSCRIPT

Figure 3. In vitro and in vivo comparison of MSC that are transduced with CMVTMGluc-Puro or CMV-DsR-Fluc constructs. (a) Comparison of light emission by CMV-TMGluc-Puro- and CMV-DsR-Fluc-transduced MSC cultured at serial dilutions

RI PT

in a 96-well-plate. Data from one of three independent experiments (n=3 per

experiment) are shown as means ± SD. (b) Serial cell dilutions of lysed CMV-

TMGluc-Puro- and CMV-DsR-Fluc-transduced MSC were analyzed by luminometry.

SC

Data from one of three independent experiments (n=3 per experiment) with similar results are shown as means ± SD. RLU, relative light units. (c) In vivo BLI of CMV-

M AN U

DsR-Fluc- and CMV-TMGluc-Puro-transduced MSC at 24 hours after implantation. Cell numbers range from 5 x105-5 x103. A representative mouse out of 3 for each group is shown. (d) H&E staining of matrigel implants isolated at day 20 from representative mice receiving either CMV-DsR-Fluc BM-MSC (1) or CMV-TMGluc-

TE D

Puro BM-MSC (2).

Figure 4. Fluc expression under the control of the SFFV promoter is higher

EP

compared to the CMV promoter. (a) DsRed expression of SFFV-DsR-Fluc-

AC C

transduced MSC (red) in comparison with CMV-DsR-Fluc-transduced MSC (magenta). (b) Luciferase activity of a titration of SFFV-DsR-Fluc- and CMV-DsRFluc-transduced MSC in a 96-well-plate. A representative image of light emission is shown. (c) Comparison of light emission of SFFV-DsR-Fluc- and CMV-DsR-Fluctransduced MSC at serial cell dilutions [analysis of (b)] Data from one of three independent experiments (n=3 per experiment) with similar results are shown as means ± SD. (d) Serial cell dilution of lysed SFFV-DsR-Fluc- and CMV-DsR-Fluctransduced MSC is analyzed by luminometry. Data from one of three independent 25

ACCEPTED MANUSCRIPT experiments (n=3 per experiment) with similar results are shown as means ± SD. RLU, relative light units. (e) In vivo BLI of CMV-DsR-Fluc-transduced and SFFVDsR-Fluc-transduced MSC at 24 hours after implantation. Cell numbers range from 5 x105-5 x103. A representative mouse out of 3 for each group is shown. (f) Direct

RI PT

comparison of SFFV-DsR-Fluc- and CMV-DsR-Fluc-transduced MSC at 24 hours after implantation [analysis of (e)] (n=3. means ± SD are shown). (g) In vivo BLI of CMV-DsR-Fluc- and SFFV-DsR-Fluc-transduced MSC at 20 days after implantation

SC

in matrigel. Cell numbers range from 5 x105-5 x103.

M AN U

Figure 5. TMGluc expression under the control of the SFFV promoter is higher compared to the CMV promoter. (a) FACS analysis of SFFV-TMGluc-Purotransduced MSC (blue) in comparison with CMV-TMGluc-Puro-transduced MSC (purple). (b) Luciferase activity of a titration of SFFV-TMGluc-Puro- and CMV-

TE D

TMGluc-Puro-transduced MSC in a 96-well-plate. A representative image of light emission is shown. (c) Comparison of light emission of SFFV-TMGluc-Puro- and CMV-TMGluc-Puro-transduced MSC at serial cell dilutions [analysis of (b)]. Data

EP

from one of three independent experiments (n=3 per experiment) with similar results

AC C

are shown as means ± SD. (d) Serial cell dilution of lysed SFFV-TMGluc-Puro- and CMV-TMGluc-Puro-transduced MSC is analyzed by luminometry. Data from one of three independent experiments (n=3 per experiment) with similar results are shown as means ± SD. RLU, relative light units. (e) In vivo BLI of SFFV-TMGluc-Puro- and CMV-TMGluc-Puro-transduced MSC at 24 hours after implantation in matrigel. Cell numbers range from 5 x105-5 x103. A representative mouse out of 3 for each group is shown.

26

ACCEPTED MANUSCRIPT Figure 6. Silencing of the SFFV promoter upon IFN-γ stimulation is partially overcome by the EF1α promoter. (a) Representative phenotypic analysis of EF1αFluc-GFP-transduced MSC that are cultured in the absence (green line) and presence of IFN-γ (blue line). Unstained MSC are shown in grey. Similar data were

RI PT

obtained for SFFV-DsR-Fluc-transduced MSC. (b) Comparison of light emission of EF1α-Fluc-GFP -and SFFV-DsR-Fluc -transduced MSC, cultured at serial cell

dilutions in the presence and absence of IFN-γ (n=3; means ± SD). RLU, relative

SC

light units. (c) In vivo BLI of EF1α-Fluc-GFP -and SFFV-DsR-Fluc- transduced MSC, previously cultured in the absence (left panel) or presence of IFN-γ (right panel).

M AN U

Mice are imaged at 24 hours after implantation of the MSC. Cell numbers range from 5 x105-5 x103. A representative mouse out of 3 for each group is shown. (d) Direct comparison of SFFV-DsR-Fluc- and EF1α-Fluc-GFP-transduced MSC at 24 hours after implantation [analysis of (d)] (n=3. means ± SD are shown). (e) Withdrawal of

TE D

IFN-γ from the SFFV-DsR-Fluc-transduced MSC culture partially restores SFFVdriven transgene expression. Unstimulated MSC (red line), IFN-γ stimulated MSC (green line); MSC cultured for 1 week in the presence of IFN-γ followed by 1 week in

EP

the absence of IFN-γ (brown line), untransduced and unstimulated MSC are shown

AC C

in grey. (f) Stimulation with TNF-α leads to a minor decrease in GFP expression. Unstimulated MSC (green line), IFN-γ and TNF-α stimulated (black line), IFN-γ stimulated (blue line) and TNF-α stimulated EF1α-Fluc-GFP-transduced MSC were analyzed for GFP expression using flowcytometry.

Supplemental figure 1. Transgene expression in SFFV-DsR-Fluc –and EF1αFluc-GFP-transduced MSC is not altered following culturing in higher doses of IFN-γ. SFFV-DsR-Fluc –transduced MSC (left panel) and EF1α-Fluc-GFP27

ACCEPTED MANUSCRIPT transduced MSC (right panel) were cultured for one week in the absence (red line) or presence of 10 ng/ml (blue line) and 56 ng/ml (orange line) of IFN-γ.

AC C

EP

TE D

M AN U

SC

RI PT

Next FACS analysis is performed to measure transgene expression.

28

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT Company

BD Biosciences BD Biosciences BD Biosciences BD Biosciences BD Biosciences BD Biosciences

Miltenyi Biotec R&D systems

BD Biosciences BD Biosciences BD Biosciences BD Biosciences BD Biosciences BD Biosciences BD Biosciences

RI PT

BD Biosciences

BD Biosciences BD Biosciences BD Biosciences BD Biosciences

SC

Clone HMb1-1 MEC 13.3 RAM34 IM7 104 53-2.1 53-2.1 MJ7/18 429 2B8 APA5 ME-9F1 Polyclonal 2B11 TER-119 D7 D7 AF6-88.5 M5/114.15.2 AF6-120.1 -

M AN U

Label PE-Cy7 biotine Pacific Blue APC APC-eFluor 780 eFluor 450 biotine Pacific Blue Alexa Fluor 647 APC-eFluor 780 APC biotine biotine biotine APC-eFluor 780 PerCP-Cy5.5 biotine biotine eFluor 450 Biotine PE-Cy7 eFluor 450 APC-eFluor 780 Alexa Fluor 488 Alexa Fluor 647

TE D

Antibody CD29 CD31 CD34 CD44 CD45.2 CD90.2 CD90.2 CD105 CD106 CD117 CD140a CD146 CD166 CD184 TER119 Sca-1 Sca-1 b H-2K I-A/I-E b I-A Gaussia luciferase Streptavidin Streptavidin Streptavidin Goat anti-rabbit Goat anti-rabbit

Nanolight Technologies

eBiosciences eBiosciences eBiosciences Molecular Probes

Molecular Probes

AC C

EP

Table 1. Overview of the antibodies that are used thoughout the study.