A novel lentivirus for quantitative assessment of gene ... - Nature

Gene Therapy (2012) 19, 1123 - 1132 & 2012 Macmillan Publishers Limited All rights reserved 0969-7128/12 www.nature.com/gt

REVIEW

A novel lentivirus for quantitative assessment of gene knockdown in stem cell differentiation S Alimperti1, P Lei1, J Tian1,4 and ST Andreadis1,2,3 Loss of gene function is a valuable tool for screening genes in cellular processes including stem cell differentiation differentiation. However, the criteria for evaluating gene knockdown are usually based on end-point analysis and real-time, dynamic information is lacking. To overcome these limitations, we engineered a shRNA encoding LentiViral Dual Promoter vector (shLVDP) that enabled real-time monitoring of mesenchymal stem (MSC) differentiation and simultaneous gene knockdown. In this vector, the activity of the alpha-smooth muscle actin (aSMA) promoter was measured by the expression of a destabilized green fluorescent protein, and was used as an indicator of myogenic differentiation; constitutive expression of discosoma red fluorescent protein was used to measure transduction efficiency and to normalize aSMA promoter activity; and shRNA was encoded by a doxycycline (Dox)-regulatable H1 promoter. Importantly, the normalized promoter activity was independent of lentivirus titer allowing quantitative assessment of gene knockdown. Using this vector, we evaluated 11 genes in the TGF-b1 or Rho signaling pathway on SMC maturation and on MSC differentiation along the myogenic lineage. As expected, knockdown of genes such as Smad2/3 or RhoA inhibited myogenic differentiation, while knocking down the myogenic differentiation inhibitor, Klf4, increased aSMA promoter activity significantly. Notably, some genes for example, Smad7 or KLF4 showed differential regulation of myogenic differentiation in MSC from different anatomic locations such as bone marrow and hair follicles. Finally, Dox-regulatable shRNA expression enabled temporal control of gene knockdown and provided dynamic information on the effect of different genes on myogenic phenotype. Our data suggests that shLVDP may be ideal for development of lentiviral microarrays to decipher gene regulatory networks of complex biological processes such as stem cell differentiation or reprogramming. Gene Therapy (2012) 19, 1123--1132; doi:10.1038/gt.2011.208; published online 12 January 2012 Keywords: lentivirus arrays; dynamic gene expression; regulatable gene knockdown; stem cells; screening gene function INTRODUCTION Since its discovery, RNAi has become a powerful tool for understanding gene function through loss-of-function screening. For example, gene silencing was used to identify tumorsuppressor genes,1 genes essential for maintaining stemness2 or controlling cell differentiation.3,4 In addition to understanding biological processes, RNAi has been proposed as a novel strategy for cancer therapy, antiviral therapy, treatment of genetic or metabolic disorders as well as cell reprogramming for personalized regenerative medicine.5 In addition to single gene knockdown, investigators developed RNAi libraries to discover gene pathways controlling various cellular processes and facilitate global understanding of cellular behavior.6 - 9 This approach was used to identify genes that are involved in proteasome-mediated proteolysis,10 virus infection and replication,11 - 14 cytokinesis,8,9,15,16 endocytosis,17 stem cell self-renewal18 or cancer cell proliferation and survival.19,20 RNAi knockdown requires a good selection strategy to link genetic information to cellular phenotype. In many cases, such readout is based on cell proliferation or expression of marker proteins that are detected by antibody staining in order to score a ‘hit’. These procedures are usually destructive and therefore, they are employed as end-point analyses. In addition, quantitative comparison among samples suffers from differences in gene

transfer efficiency owing to differences in the titer of viral preparations. To overcome these limitations, we engineered a shRNA encoding LentiViral Dual Promoter (shLVDP) vector that contains three transcriptional units. The first encodes for unstable ZsGreen (ZsG-DR) under a promoter of interest and is used as readout to monitor the effect of gene knockdown on cellular phenotype. The second encodes for discosoma red fluorescent protein (DsRed)-Express under a constitutive promoter for example, human phosphoglycerate kinase promoter (hPGK) and is used to assess gene transfer efficiency as well as for data normalization to obtain quantitative data independent of the gene transfer efficiency.21,22 Finally, the third unit encodes for shRNA under a tetracycline-regulatable H1 promoter in the viral LTR. This vector allows shRNA expression and at the same time provides quantitative real-time monitoring of the effect of gene knockdown on cellular phenotype. Using the shLVDP vector, we measured the effects of several genes in the TGF-b1 and Rho pathways on maturation of human aortic smooth muscle cells (HASMC) and in myogenic differentiation of mesenchymal stem cells from bone marrow (BM-MSCs) and hair follicles (HF-MSC). TGF-b1 has been shown to upregulate the expression of alpha-smooth muscle actin (aSMA) in SMC23 and MSC,24,25 and RhoA is known to affect actin reorganization and contractile activity.26 Myogenic differentiation was monitored

1 Bioengineering Laboratory, Department of Chemical and Biological Engineering, University at Buffalo, State University of New York, Amherst, NY, USA; 2Department of Biomedical Engineering, University at Buffalo, State University of New York, Amherst, NY, USA and 3Center of Excellence in Bioinformatics and Life Sciences, Buffalo, NY, USA. Correspondence: Dr ST Andreadis, Bioengineering Laboratory, 908 Furnas Hall, Department of Chemical and Biological Engineering, University at Buffalo, State University of New York, Amherst, NY 14260-4200, USA. E-mail: [email protected] 4 Current address: Life Technologies, Grand Island, NY 14072, USA. Received 21 June 2011; revised 26 September 2011; accepted 10 October 2011; published online 12 January 2012

Gene function screening in stem cell differentiation S Alimperti et al

lentiviral RNA, the 30 LTR replaces the 50 LTR, thereby allowing expression of two shRNA copies for each integrated lentivirus genome. Overall, this modified vector expresses two genes and one shRNA and was termed shLVDP.

by the aSMA promoter activity and confirmed by western blot and immunostaining. Normalized promoter activity rendered the results independent of gene transfer efficiency and allowed comparison of the relative effect of each gene on myogenic differentiation of different cell types. Finally, Dox-inducible shRNA expression enabled temporal control of gene knockdown and prevented toxic effects associated with loss of genes that are important in cell spreading or cytoskeletal reorganization.

Signal normalization shows that promoter activity is independent of lentivirus titer The shLVDP allows measurement of promoter activity through changes in green fluorescent intensity (GFI), which depends on the number of infected cells and the number of transgene copies per cell. However, quantitative measurements of promoter activity may suffer by lack of consistency due to differences in gene transfer efficiency among different lentivirus preparations each expressing a different shRNA. As shLVDP also encodes for DsRed under a constitutive promoter we hypothesized that normalization of GFI by the red fluorescence intensity (RFI) may be independent of the gene transfer efficiency, thereby yielding quantitative data. To address this hypothesis, we employed shLVDP encoding for EGFP under the hPGK promoter, DsRed under the cytomegalovirus promoter (CMV) promoter and scramble shRNA under the H1 promoter to transduce HASMCs with two MOI that differed by 10-fold, MOI ¼ 1 and MOI ¼ 10. As expected, RFI and GFI increased with increasing virus concentration as measured by fluorescence microscopy (Figures 1b and c) or flow cytometry (Figures 1e and f), likely due to increased fraction of transduced cells and/or increased number of gene copies per cell. However, the normalized fluorescence intensity (GFI/RFI) remained the same regardless of the analysis method (Figures 1d and g), suggesting that the intrinsic promoter activity was independent of virus titer. This result suggests that internal signaling normalization can be used to eliminate disparity owing to differences in transduction efficiency, thereby enabling acquisition of quantitative data.

RESULTS Generation of shLVDP vector Recently, our laboratory developed a LVDP that provides a high expression level of two genes from independent promoters in mammalian cells including MSCs.22 A constitutive promoter drives expression of one marker gene (for example, DsRed) and is used to measure transduction efficiency. The promoter of interest drives expression of another marker (for example, green fluorescent protein (EGFP)) and is used to monitor the biological process of interest. Polyadenylation, terminator and insulator sequences were inserted between the two transcription units to eliminate promoter interference yielding promoter activity that was similar to that exhibited by vectors containing a single transcription unit. This vector was used to generate lentiviral microarrays to monitor the dynamics of gene expression in real time in response to external stimuli.21 In this study, we modified this vector to also express shRNA from a tetracycline-regulatable H1 promoter, thereby allowing assessment of genes that may be involved in a process, which can be monitored in real time by the activity of a promoter or response element (RE) (Figure 1a). To this end, the 30 LTR of our LVDP vector was substituted by the 30 LTR of pLVTHM vector, which was engineered to express shRNA under a tetracyclineregulatable H1 promoter.27,28 During reverse transcription of the

5’LTR

sMAR8 TKpA

DsRed2

CMV

cHS4

Tactb

SPA

EGFP

hPGK

U3

tetO

H1

U5

scramble-shRNA Microscopy

GFI

RFI

2.E+04

1.E+04

8.E+04

4

6.E+04

3 GFI/RFI

3.E+04

4.E+04 *

*

MOI=10

MOI=1 250

200

200

150

150 GFI

250

100

MOI=10

0.5

*

0 MOI=1

MOI=10

1

Flow Cytometry

100 50

*

0

MOI=1

MOI=10

GFI/RFI

MOI=1

50

0

0.E+00

0.E+00

2 1

2.E+04

RFI

1124

0 MOI=1

MOI=10

MOI=1

MOI=10

Figure 1. Normalized promoter activity of shLVDP vector is independent of virus titer. (a) Schematic of shLVDP vector. (a) Red fluorescence intensity (RFI) and (c) green fluorescence intensity (GFI) of cells transduced with shLVDP vector as determined by microscopy. (d) The normalized fluorescence intensity was calculated as ratio of GFI/RFI. (e) Red fluorescence intensity (RFI) and (f) green fluorescence intensity (GFI) of cells transduced with shLVDP virus as determined by flow cytometry. (g) The normalized fluorescence intensity was calculated as ratio of GFI/RFI. *Significant difference from the MOI ¼ 10 (*Po0.05). cHS4, chicken hypersensitive site 4; H1, H1 promotersMAR8, synthetic MAR sequence; SPA, synthetic polyA; Tactb, transcriptional termination of human b-globin; tetO, tetracycline repressor-binding site; TKpA, herpes simplex virus thymidine kinase poly(A). Gene Therapy (2012) 1123 - 1132

& 2012 Macmillan Publishers Limited


shLVDP can be used to monitor smooth muscle cell maturation Next, we modified this vector to enable monitoring of gene expression from any promoter or RE. As REs usually require addition of a minimal CMV promoter to achieve high expression levels, we substituted the CMV promoter-driving DsRed with the hPGK promoter to avoid the potential of homologous recombination due to repeated sequences, as previously reported for lentiviral vectors.29 We also replaced EGFP with ZsG-DR, a brighter and destabilized (t1/2 ¼ 8 -- 12 h) green fluorescent protein to better capture the kinetics of gene expression. Finally, we cloned the promoter for the well-known early myogenic marker, aSMA upstream of ZsG-DR, to monitor smooth muscle cell maturation in response to TGF-b1 treatment. The resulting vector (Figure 2a) was used to produce recombinant lentivirus to transduce HASMCs. Three days after transduction, the cells were cultured either in growth medium (GM:SMGS) or differentiation medium (DM: SMDS þ 10 ng ml - 1 TGF-b1) in the presence or absence of TGF-b1 receptor inhibitor, SB431542 (SB4). Fluorescence intensity was measured 3 days later using fluorescence microscopy and flow cytometry. Under growth conditions, cells displayed basal aSMA promoter activity (normalized intensity, NI ¼ 0.94), which increased by 3-fold (NI ¼ 2.8) in the presence of TGF-b1 (Figure 2b). Addition of the TGF-b1 inhibitor, SB431542, significantly reduced the aSMA promoter activity by 16-fold (NI ¼ 0.17). Similarly, flow cytometry showed that the normalized intensity (GFI/RFI) increased by 2.5-fold (NI ¼ 2.45) upon treatment with TGF-b1, whereas addition of the TGF-b1 receptor inhibitor, SB431542, reduced GFI/RFI by 18-fold (NI ¼ 0.05) (Figure 2c). In addition to the normalized intensity, the percentage of ZsG-DR þ cells was 3.3±0.78, 26.1±1.56 or 0.17±0.06% in GM, DM or DM þ SB431542, respectively (Figure 2d). In agreement, western blot showed that aSMA protein level increased by 3.35±0.3-fold in DM (Figure 2e), which was further verified by immunostaining (Figure 2f). In particular, under GM aSMA expression was weak and diffuse but

5’LTR sMAR8 TKpA

DsRed2 hPGK

cHS4

upon TGF-b1 treatment, aSMA was organized in filaments--an indication of contractile myogenic phenotype (Figure 2g). Taken together, these results suggested that our shLVDP can be used to monitor TGF-b1-mediated myogenic differentiation. Normalized promoter activity is independent of the extent of gene silencing The ability to monitor cell differentiation using our lentiviral vector suggested that this system could be used to examine the effects of various genes/pathways in myogenesis using shRNA knockdown from the same vector as shown in Figure 3a. As the extent of gene silencing may vary with the transduction efficiency, we examined whether differential shRNA expression might affect the level of promoter activity. To address this issue, we expressed shRNAs-representing genes that are known to affect myogenic differentiation (Klf4 and Smad4) from the H1 promoter of the shLVDP vector shown in Figure 3a. The same vector also encodes for ZsG-DR under the aSMA promoter and DsRed under the hPGK promoter. As expected, higher virus titer resulted in higher knockdown of the Klf4 or Smad4 genes as shown by the decreased mRNA levels (Figure 3b). To evaluate promoter activity upon gene silencing, HASMC were transduced with the shLVDP at the indicated virus MOI and cultured in the presence of TGF-b1 to induce myogenic differentiation. Three days later, aSMA promoter activity was determined by fluorescence microscopy. As expected, higher virus concentrations resulted in higher transduction efficiency as indicated by higher RFI values (Figure 3c). Similarly, the level of GFI---an indicator of aSMA promoter activity---depended strongly on the transduction efficiency (Figure 3d). Specifically, for scramble shRNA that did not affect aSMA promoter activity, GFI increased with increasing lentivirus concentration. Knockdown of the myogenic inhibitor, Klf4 also enhanced GFI, whereas knockdown of pro-myogenic

Tactb SPA ZsG-DR

αSMA

U3

H1

tetO

U5

*

GM

4 B

M D

D

M

M

+S

M

* D

% ZsG-DR+ cells

4

30 25 20 15 10 5 0 G

G

+S M D

DM *

* M

4 B

M D

G

SMA protein level

4

*

M

**

3.5 3 2.5 2 1.5 1 0.5 0

+S B

GFI/RFI

*

D

3.5 3 2.5 2 1.5 1 0.5 0 M

GFI/RFI

scramble-shRNA

GM

DM

SMA

3 GAPDH 2 1 0 DM

GM

50μM

50μM

Figure 2. Optimization of smooth muscle cell differentiation system. (a) Schematic of smooth muscle-specific shLVDP vector carrying scramble shRNA (ZsG-DR: destabilized Green protein). Transduced HASMC were cultured in GM:SMGS, DM(SMDS þ 10 ng ml - 1 TGF-b1) or DM þ SB431542. (b) Normalized intensity (GFI/RFI) of transduced cells was determined by fluorescence microscopy. (c) Normalized intensity (GFI/RFI) of transduced cells was determined by flow cytometry. (d) The percent of positive ZsG-DR cells (%ZsG-DR þ ) was measured by flow cytometry. (e) Protein level of aSMA was measured by western blot. Immunostaining of HASMC for aSMA (green) after 3 days of culture in (f) GM; or (g) DM. Cell nuclei were counterstained with Hoechst dye (blue). Scale bar ¼ 50 mm. *,**Significant difference compared with GM (*Po0.05, **Po0.1). The color reproduction of this figure is available at the Gene Therapy journal online. & 2012 Macmillan Publishers Limited

Gene Therapy (2012) 1123 - 1132

1125


1126 5’LTR sMAR8 TKpA

DsRed2 hPGK

cHS4

Tactb SPA ZsG-DR αSMA

U3

tetO

H1

U5

scramble-shRNA SMAD4_2-shRNA KLF4-shRNA 100000 MOI=1

SMAD4_2 KLF4

1 MOI=0.5

0.5

10000 RFI

relative mRNA level expression

1.5

10

MOI=10

0

1 control

Low MOI High MOI

SMAD4_2

SCRAMBLE

KLF4

0.4

MOI=1 MOI=5 MOI=10

0.3 GFI/RFI

GFI

1500

MOI=10

100

MOI=1 MOI=5

2000

MOI=5

1000

1000 500

MOI=1 MOI=5 MOI=10

0.2 0.1

0 SMAD4_2

SCRAMBLE

KLF4

0 SMAD4_2

SCRAMBLE

KLF4

Figure 3. Normalized promoter activity is independent of the extent of gene silencing. (a) Schematic of lentiviral vector encoding for shRNA targeting Smad4, Klf4 genes or no gene (scramble). (b) Quantitative RT-PCR analysis of Klf4 and Smad4 mRNA expression in 293T cells transduced with lentivirus encoding for Klf4 or Smad4 shRNA, respectively, at the indicated MOI. (c) Red fluorescence intensity (RFI); (d) Green fluorescence intensity (GFI) and (e) normalized promoter activity (GFI/RFI) of HASMC transduced with lentivirus at the indicated MOI.

genes, for example, RhoA or Smad4 decreased GFI in a lentivirus concentration-dependent manner. In contrast to GFI, the normalized aSMA promoter activity (GFI/RFI) was independent of the lentivirus concentration for the range of MOI tested (Figure 3e), strongly suggesting that the normalized promoter activity was independent of the extent of gene silencing. Evaluating gene knockdown on TGF-b1-induced aSMA promoter activity We used TGF-b1 to induce SMC maturation as a model system to demonstrate the utility of the shLVDP vector in measuring promoter activity with simultaneous gene silencing. To this end, we employed the shLVDP vector encoding ZsG-DR under the aSMA promoter, DsRed under the PGK promoter and shRNA under the H1 promoter targeting the indicated genes. The genes that were selected for knockdown are known effectors of TGF-b1 signaling including Smad2, Smad3, Smad4 and Smad7; the RhoA pathway such as RhoA, Rock1, Mrtf-B/Mlk2, Pkn1; as well as other transcription factors that are known to be important in myogenic differentiation such as Srf, Klf4 and MyoCD. The knockdown efficiency of each shRNA sequence was first measured by real time-PCR in transiently transfected 293T cells and only shRNAs showing at least 60% knockdown were chosen for further experiments (Supplementary Figure S1; Supplementary Material). Next, HASMC were transduced with shLVDP each encoding for the indicated shRNA (Figure 4a). Three days later, modified cells were coaxed into a mature SMC phenotype by switching to DM and aSMA activity was measured using a Zeiss Observer.Z1 microscope (Carl Zeiss Inc., Thornwood, NY, USA) equipped with environmental room with temperature, CO2 and humidity control. Image analysis was performed using the Cell Profiler software (see Methods section) and GFI was normalized to RFI to report promoter activity independent of gene transfer efficiency. As shown in Figure 4b, silencing myogenic transcription factors (Srf, MyoCD), TGF-b or Rho pathway effectors (for example, Smad2, Smad3, Smad4, RhoA, Rock1, Mrtf-B/Mlk2, Pkn1) decreased aSMA promoter activity significantly as compared with control (scramble Gene Therapy (2012) 1123 - 1132

shRNA in DM). On the other hand, knocking down the myogenic differentiation inhibitor Klf4 increased aSMA promoter activity significantly by almost 2-fold. These measurements were verified by flow cytometry that showed a significant 3.40±0.41-fold increase in the fraction of ZsG-DR þ cells upon Klf4 knockdown (Figure 4c). In addition, immunostaining showed that aSMA promoter activity correlated with aSMA protein intensity and organization (Figure 5). Specifically, HASMC transduced by scramble or Klf4shRNA spread more and expressed higher level of aSMA, which was also organized in filaments, indicating mature myogenic phenotype. In contrast, RhoA knockdown induced cell rounding and abolished aSMA protein expression in agreement with the decline in aSMA promoter activity. Finally, cells that were modified with Smad4 shRNA became elongated with diffused aSMA staining and no filamentous organization. Taken together, these results demonstrate that the effect of shRNA on aSMA promoter activity correlated with protein expression and fibrous organization. shLVDP for monitoring MSC differentiation to the myogenic lineage To demonstrate the application of this approach in stem cell differentiation, we employed BM-MSCs and HF-MSC. The cells were transduced with the same group of lentiviruses, coaxed to differentiate into SMC in DM and 3 days later GFI and RFI were measured by fluorescence microscopy. Similar to HASMC, aSMA promoter activity decreased significantly by knocking down TGF-b1 or Rho pathway effectors and increased by knocking down Klf4 (Figures 6a and b). Interestingly, although knocking down Klf4 increased aSMA activity (GFI/RFI) in HASMC (1.68±0.07; Po0.05) and BM-MSC (2.03±0.42; Po0.05), it had no effect on HF-MSC (1.35±0.19; P40.05). On the other hand, Smad7 knockdown had a small effect on HASMC but decreased aSMA activity in BM-MSC (0.722±0.06; P ¼ 0.065) and to a much larger extent in HF-MSC (0.11±0.03; Po0.05). These results suggest that Klf4 and Smad7 may differentially regulate aSMA expression in HF-MSC compared with BM-MSC and HASMC. & 2012 Macmillan Publishers Limited


1127 SMAD3 SMAD 4_1 SMAD 4_2 SMAD2

SMAD7 RHOA

ROCK1

2

KLF4

MRTFB

SRF

PKN1

MYOCD

Scr/DM

5 *

* scramble/GM

*

scramble/ DM

*

MYOCD

*

*

scramble/DM+SB4

*

SRF

*

PKN1

*

KLF4

*

RHOA

*

MRTFB/MLK2

*

*

SMAD7

0

ROCK1

1

SMAD2

scramble/DM

scramble/GM

MYOCD

scramble/ DM+SB4

SRF

PKN1

KLF4

2

*

* MLK2/MRTFB

RHOA

ROCK1

*

* SMAD7

SMAD3

SMAD4_1

0

*

*

3

SMAD4_2

* *

SMAD4_2

*

*

*

SMAD3

*

*

0.5

4

SMAD4 _1

Normalized % of ZsG-DR+ cells

1

SMAD2

GFI/RFI

1.5

Figure 4. shLVDP cell array to assess the effects of the indicated genes on SMC maturation in response to TGF-b1. (a) Schematic of HASMC cell array. Each well was infected with lentivirus encoding for shRNA targeting the indicated gene. (b) Normalized aSMA promoter activity (GFI/RFI) when each of the indicated genes was knocked down. (c) The percentage of ZsG-DR þ cells was measured by flow cytometry and normalized to that of control (scramble/DM). *Significant difference as compared with scramble control (*Po0.05).

50μm

200μm

200μm

200μm

200μm

200μm

200μm

200μm

200μm

50μm

50μm

50μm

Figure 5. aSMA promoter activity correlates with protein expression. HASMC were transduced with shRNA targeting: (a, e, i) scramble; (b, f, j) RhoA; (c, g, k) Smad4; and (d, g, l) Klf4 lentiviral vector and cultured in DM. (a -- d) RFI; (e -- h) GFI; (i -- l) Immunostaining for aSMA (green) and counterstaining with the nuclear dye Hoeschst (blue). (a -- h): Bar ¼ 200 mm; (i -- l): Bar ¼ 50 mm. The color reproduction of this figure is available at the Gene Therapy journal online.

We also verified MSC differentiation into the SMC lineage using a gel compaction assay, an indicator of force generation ability, the quintessential property of SMC. To this end, HF-MSCs or & 2012 Macmillan Publishers Limited

BM-MCS were induced to differentiate into SMC in DM and then embedded into fibrin hydrogels (106 cells ml - 1). After polymerization was complete (1 h), the gels were released from Gene Therapy (2012) 1123 - 1132


1128 2

2 1

MYOCD

GFI/RFI

* scramble/DM

* PKN1

1.5 * SRF

ROCK1

MRTFB/MLK2

* KLF4

* RHOA

SMAD4_2

* SMAD7

*

SMAD2

* SMAD4_1

0

* * SMAD3

GFI/RFI

SCRAMBLE SMAD4_2 ROCK1 KLF4

HF-MSC array

3

1

0.5

BM-MSC array

3

-72hr *

scramble/DM

MYOCD

PKN1

SRF

MRTB/MLK2

12

24

36

Time (hr)

Figure 6. shLVDP cell array to assess the effects of the indicated genes on MSC differentiation into SMC. (a) HF-MSC or (b) BM-MSCs were transduced with shLVDP encoding for the indicated shRNA. Normalized fluorescence intensity (GFI/RFI) indicates the effect of gene knockdown on aSMA promoter activity. *Significant difference as compared with scramble control (*Po0.05).

the wells and allowed to contract. At 48 h, the gels were photographed and the gel area was measured using Image J software. As shown in Supplementary Figure S2A, HF-MSCs compacted the hydrogels by B70% in DM but only by B20% in GM and compaction was diminished in the presence of SB431542. Similar results were observed with BM-MSCs (Supplementary Figure S2B). In agreement with aSMA activity, this data suggested that treatment with DM-induced HF-MSC and BM-MSC differentiation into contractile SMCs. Dox-inducible shRNA expression during MSC differentiation To study MSC differentiation with regulated shRNA expression, HF-MSC were transduced with a tet-pLKO-puro lentivirus encoding for Tet repressor (TetR), selected with puromycin and expanded. These cells were then transduced with shLVDP encoding for shRNA against Rock1, Smad4 or Klf4 and coaxed to differentiate toward SMC lineage in the presence of DM. Three days later, Dox was applied to trigger shRNA expression and aSMA activity was measured as a function of time following addition of Dox. As expected, knockdown of genes that are important in SMC contractility decreased promoter activity as compared with control cells (scramble-shRNA) (Figure 7). Specifically, the level of aSMA activity was diminished within 12 h after the expression of Rock1 shRNA and decreased more slowly upon expression of Smad4 shRNA---down to 40% and 25% after 12 and 36 h, respectively. In contrast, Klf4 shRNA had no effect on aSMA promoter activity. These data demonstrated that shLVDP can capture the kinetics of promoter activity following Dox-inducible gene knockdown.

DISCUSSION In this study, we sought to develop a strategy to monitor cellular activity in real time and at the same time examine how this Gene Therapy (2012) 1123 - 1132

0

*

* *

KLF4

* ROCK1

*

* RHOA

SMAD4_1

SMAD3

0

*

SMAD7

* *

SMAD2

1 SMAD4_2

GFI/RFI

* 2

DM

+Dox

Figure 7. Kinetics of aSMA promoter activity in response to Dox-induced gene knockdown. TetR expressing HF-MSCs were transduced with shLVDP vector encoding for Klf4, Smad4, Rock1 or scramble shRNA and coaxed to differentiate with TGF-b1 (10 ng ml - 1). Three days later, Dox (10 mg ml - 1) was added in the medium and the normalized aSMA promoter activity (GFI/RFI) was measured as a function of time using fluorescence microscopy.

activity is affected by certain genes or pathways. To this end, we generated a lentiviral vector with three transcriptional units to enable quantitative monitoring of gene expression with concomitant gene knockdown. Using this vector, we generated a lentiviral array, whereby each virus encoded for a different shRNA, and examined myogenic differentiation by monitoring aSMA promoter activity in response to gene knockdown. Our strategy yielded quantitative information of gene expression in human primary cells and stem cells, thereby providing a useful platform for deciphering gene regulatory networks of complex biological processes such as stem cell differentiation. Total fluorescent intensity of each spot on the array depends on the number of infected cells and the number of transgene copies per cell. These factors, however, might change significantly owing to variations in the viral titer making it difficult to obtain truly quantitative data. This may be particularly important for highthroughput experiments that require a large number of lentiviral vectors because it is very difficult to obtain lentivirus preparations of equal potency. As a result, the difference in signal intensity may reflect differences in gene transfer efficiency rather than the effect of gene knockdown. Although, in transfection methods co-transfection with a second plasmid constitutively encoding for a second reporter gene, for example, Renilla luc has been used for signal normalization, it is difficult to ensure similar transfection efficiency for both plasmids. In the present case, this strategy would be further complicated by the requirement for three plasmid co-transfection. On the other hand, expressing two genes and shRNA from the same vector provides an internal control for signal normalization enabling acquisition of quantitative data. Indeed, our results showed that the novel shLVDP vector addressed this issue as the normalized fluorescence intensity (GFI/RFI) was independent of the gene transfer efficiency, for the range of viral titers tested in this study. Specifically, although increasing the viral titer increased the efficiency of gene knockdown as well as GFI and RFI, the normalized aSMA promoter activity (GFI/RFI) remained unaffected. As expected, knockdown of genes that promote contractility, for example, RhoA decreased normalized promoter activity significantly, whereas knockdown of myogenic & 2012 Macmillan Publishers Limited


inhibitors, fpr example, Klf4 increased promoter activity as compared with scramble control. However, in all cases the normalized activity remained independent of the efficiency of gene knockdown owing to variations in viral titer, suggesting that shLVDP enabled quantitative assessment of gene knockdown on the activity of a promoter that can be used to monitor a biological process, for example, stem cell differentiation. We chose myogenic differentiation as an example to obtain proof-of-concept that the shLVDP can be used to examine how different pathways may affect MSC differentiation. To this end, we employed aSMA promoter as an indicator of myogenic phenotype30 and selected several genes that are known to regulate contractile function, through the Rho or TGF-b1 pathways. In agreement with previous studies,31,32 TGF-b1 induced MSC differentiation toward the myogenic lineage as shown by enhanced aSMA promoter activity and verified at the protein level by western blot and immunostaining. Blocking TGF-b1 receptor activity or knocking down the TGF-b1 mediators Smad2, Smad3 or Smad4 significantly decreased aSMA activity. In addition, non-Smad pathways that are induced by TGF-b1 such as RhoA and its downstream effector Rock1(refs 33,34) are important for myogenic differentiation. Similarly, MyoCD and myocardin related transcription factors, Mrtf-A and Mrtf-B are necessary for development of contractile phenotype through their role as SRF co-activators.35 - 38 As a result, knocking down any one of those factors severely impaired aSMA activity. On the other hand, knockdown of the MyoCD antagonist, Klf4 increased aSMA activity.39 These results are in agreement with the literature and suggest that the normalized aSMA promoter activity accurately reflects the effect of well-known pathways on myogenic differentiation. Surprisingly, we found that Smad7 knockdown had no significant effect in HASMC and BM-MSC but significantly decreased aSMA activity in HF-MSC, suggesting that Smad7 may differentially regulate myogenesis in different cell types. As inhibitor of Smad2/3, Smad7 is expected to decrease aSMA activity. However, Smad7 has also been implicated in myogenicpromoting pathways. For example, Smad3 was shown to interact with SRF,40 whereas Smad7 activated the small GTPase Cdc4241 and interacted with myocardin in the nucleus-promoting myogenesis.42 Therefore, our results may suggest that in HASMCs and BM-MSCs, Smad7 may participate in both anti- and pro-myogenic interactions, so that their overall action balances out the influence of these pathways. On the other hand, in HF-MSC, SMAD7 may act predominantly in a pro-myogenic fashion, for example, by interacting with MyoCD, so that when it is knocked down, aSMA activity is reduced significantly. Although the mechanism remains elusive, our results suggest that gene knockdown with concomitant monitoring of promoter activity may reveal interesting differences between stem cells of different tissue origin as they differentiate along a common pathway. The shLVDP can be adapted to a chemically-inducible shRNA expression system by introducing a second vector (tet-pLKO-puro) encoding for TetR, which binds to the TRE and inhibits shRNA expression. Upon addition of Dox, transcription is resumed as Dox binds to TetR releasing it from the promoter.43 This system enables temporal control of shRNA expression at preferred times during the experiment, which is essential for studying genes that may have adverse effect on cell viability or other biological processes, for example, cytoskeletal re-organization.44 For example, during myogenic differentiation, we observed a considerable level of cytotoxicity in cells actively expressing Rock1 shRNA (data not shown), making data interpretation difficult. Use of the inducible shLVDP enabled TGF-b1-induced differentiation before shRNA expression, thereby allowing us to assess the effect of Rock1 on aSMA promoter activity instead of viability. This system may be useful in tissue engineering where temporal control of shRNA expression may enable engineering of tissues lacking & 2012 Macmillan Publishers Limited

expression of genes that may be essential during a particular stage of tissue development. For the inducible shRNA experiments, the cells were transduced with two lentiviruses: shLVDP and a virus encoding for TetR, raising the concern that any variability in the transduction efficiency of the TetR-encoding lentivirus might affect data normalization due to potential variability in shRNA induction. To minimize this possibility, HF-MSCs were first transduced with TetR lentivirus and selected with puromycin before they were transduced with the indicated shLVDP viruses. Most important, any variability in TetR concentration should be eliminated when the Dox concentration is high enough to block all the available repressor, thereby rendering shRNA induction solely dependent on the H1 promoter of the shLVDP vector. Indeed, preliminary experiments employing Smad4 shRNA showed that at 10 mg ml - 1 of Dox the aSMA promoter response reached steady state, thereby suggesting that at that concentration Dox eliminated any variability in shRNA induction between cells (data not shown). For this reason, all experiments in this study were conducted with Dox at 10 mg ml - 1. The shLVDP vector may be ideal for generating shRNA libraries to discover new genes that are necessary for commitment of adult, embryonic or induced pluripotent stem cells to a particular lineage. Although several lentiviral constructs such as pLKO.1, pENTR or pLVTHM have been applied for generation of RNAi libraries,8,45,46 the shLVDP provides a means for monitoring differentiation in real time, thereby facilitating selection of live cells. Moreover, by enabling quantitative activity measurements shLVDP libraries may enable development of quantitative criteria to evaluate the relative importance of different genes or pathways in stem cell differentiation and select cells that meet these criteria. This system would be useful in evaluating differentiation of embryonic or induced pluripotent stem cells, which are limited by low differentiation efficiency resulting in a heterogeneous mixture of differentiated, partially differentiated and undifferentiated cells. In conclusion, the shLVDP lentivirus can be used to evaluate the effects of gene networks on complex biological processes such as stem cell differentiation, in live cells and in real time. Libraries of shLVDP vectors containing carefully selected transcription factorbinding sites (REs) and shRNAs may provide quantitative data that may be amenable to mathematical modeling for reverse engineering the gene networks controlling stem cell fate decisions or other complex biological processes.

MATERIALS AND METHODS Cell culture 293T/17 cells (ATCC, Manassas, VA, USA) were cultured in Dulbecco’s Modified Eagle Medium (DMEM; GIBCO BRL, Grand Island, NY, USA) supplemented with 10% (v/v) fetal bovine serum (GIBCO) and 1% (v/v) antibiotic-antimycotic (GIBCO). BM-MSCs were purchased from Stem Cell Technologies (Vancouver, Canada) and HF-MSCs were isolated as described previously.47,48 Both cells were cultured in GM constituted of DMEM supplemented with 5% (v/v) fetal bovine serum and 1% (v/v) antibiotic-antimycotic and 1.0 ng ml - 1 basic fibroblast growth factor (BD Biosciences, San Jose, CA, USA). The GM for HASMCs (GIBCO) constituted of 231 Medium (GIBCO) that was supplemented with smooth muscle growth supplement (SMGS; GIBCO).

Lentiviral vector construction strategy and lentiviral vector preparation The lentiviral vectors used in this study were generated as follows. First, we used the shRNA cassette from pLVTHM (Addgene, Cambridge, MA, USA), which encodes for shRNA under the Dox-regulatable H1 promoter between MluI and ClaI restriction sites in the 30 LTR of the vector. An EcoRV site was added immediately downstream of MluI for subsequent cloning. Afterwards, the 30 LTR of pLVTHM was amplified by PCR with Gene Therapy (2012) 1123 - 1132

1129


1130

primers THM-for: 50 -TCTTCCGCGTCTTCGCCTTC-30 and THM-rev: 50 -AATTCT AGATGCTGCTAGAG-30 . The PCR product was used to replace the 30 LTR in the pCSCG (Addgene) between the KpnI and PmeI sites. The resulting vector was digested with PpuMI and Bsu36I to remove CMV-GFP and replace it with the dual promoter cassettes: sMAR8_TKpA_DRE2_CMV_ cHS4_Tactb_SPA_EGFP_hPGK, or sMAR8_TKpA_DRE2_hPGK_cHS4_Tactb_ SPA_ZsG-DR_MCS from our previously reported LVDP vectors in which DsRed2 was replaced by DsRed-Express2 (Clontech, Mountain View, CA, USA). These sequences were cloned in trans to the viral LTRs to avoid premature termination of viral mRNA by the inserted polyAs during virus production in packaging cells as reported previously.22 The resulting shRNA encoding LentiViral Dual Promoter vectors were termed shLVDP vectors. shRNAs were cloned into this vector between the EcoRV and ClaI sites immediately downstream of the H1 promoter in the 30 LTR. Briefly, the two complementary oligos for each shRNA were mixed at 1:1 ratio to a final concentration of 50 mM and annealed under the following conditions: 951C for 30 s, 721C for 2 min, 37 1C for 2 min and 25 1C for 2 min. All cloned sequences were confirmed by sequencing with ABI PRISM 3130XL Genetic Analyzer (Applied Biosystems, Foster City, CA, USA). All siRNA sequences are listed in Table 1. Finally, the aSMA promoter sequence was cloned into the multiple cloning site (MCS) to drive ZsG-DR expression. To this end, the promoter was amplified from human genomic DNA by PCR with primers: hACTA2F: 50 -ACAACAGTTAACCTTCCTGTCAAGGAGGTTAG-30 and hACTA2R: 50 -ACAAC AACCGGTCGGGTAATTAAAAGAGCCAC-30 . Phusion High-Fidelity DNA Polymerase (New England Biolabs, Ipswich, MA, USA) was used for PCR according to the manufacturer’s protocol and the PCR product was ligated into the MCS between the HpaI and AgeI sites. For lentivirus production, 293T/17 cells were transfected with three plasmids: (16.8 mg each lentiviral vector, 15 mg psPAX2, and 5 mg pMD.G), using the standard calcium phosphate precipitation method. Virus was harvested 24 h post transfection, filtered through 0.45 mm filter (Millipore, Bedford, MA, USA), pelleted by ultracentrifugation (50 000 g at 4 1C for 2 h) and resuspended in fresh medium. The virus titer was determined using HASMCs cells and was used for subsequent experiments at an MOIB5.

Transduced HAMCS cells were cultured in DM for 3 days. Afterwards, the cells were washed three times with phosphate-buffered saline (PBS), and fixed in a mixture of methanol and acetone (75:25, v/v) for 10 min at room temperature (RT). After fixation, the cells were washed with PBS twice and permeabilized with 0.1% (v/v) Triton X 100 in PBS for 10 min at RT. Afterwards, the cells were blocked with 0.01% (v/v) Triton X 100 and 1% (w/v) bovine serum albumin (EMD Chemicals, San Diego, CA, USA) in PBS for 1 h at RT. Thereafter, the cells were incubated with primary antibody (1:50 dilution, mouse anti-human aSMA; Serotec, Raleigh, NC, USA) in blocking buffer overnight at 4 1C. The cells were washed for three times with PBS and Alexa 488-conjugated goat-anti-mouse IgG (20 mg ml - 1, Molecular Probes, Grand Island, NY, USA) in blocking buffer was applied for 1 h at RT. The cells were washed three times with PBS and the cell nuclei were counterstained with Hoechst 33258 (25 mg ml - 1 in TNE buffer: 10 mmol l - 1 Tris, 2 mol l - 1 NaCl, 1 mmol l - 1 EDTA, pH 7.4; Molecular Probes) for 10 min at RT. Finally, the cells were washed for three times and fluorescent images were obtained using an inverted fluorescence microscope Zeiss Observer.Z1.

Infection and differentiation protocol

Compaction of fibrin hydrogels

HASMCs (12 000 cells per well), BM-MSCs (12 000 cells per well) or HF-MSC (4000 cells per well) were plated in 96-well tissue culture treated plates (NUNC, Rochester, NY, USA). The cells were transduced with different viruses carrying different shRNAs in the presence of 8 mg ml - 1 polybrene under GM. Viruses were removed 2 h after infection and fresh cell culture medium was added. After 3 days, the cells were induced to differentiate by treatment with 231 medium supplemented with SMDS (GIBCO), 5% (v/v) fetal bovine serum and 10 ng ml - 1 TGF-b1 (Biolegends, San Diego, CA, USA), (DM). DM for BM-MSCs or HF-MSCs constituted of DMEM containing 5% (v/v) fetal bovine serum, 30 mg ml - 1 heparin (App Pharmaceuticals, Schaumburg, IL, USA) and 10 ng ml - 1 TGF-b1. After 3 days, the cells were

Compaction of three-dimensional fibrin hydrogels was described elsewhere.47 - 49 Briefly, fibrin gels were prepared with HF-MSCs or BM-MCS cells (106 cells ml - 1) and allowed to polymerize each in a well of a 24-well

Table 1.

Cloning of different shRNAs

Accession number NM_005359 NM_005359 NM_005901 NM_005904 NM_001664 NM_005406 NM_003131 NM_153604 NM_014048 NM_00274 NM_005902 NM_004235

Gene

Sense sequence (50 -- 30 )

Scramble SMAD4_2 SMAD4_1 SMAD2 SMAD7 RHOA ROCK1 SRF MYOCD MRTFB/MLK2 PKN1 SMAD3 KLF4

CAACAAGATGAAGAGCACCAA GTGGCTGGTCGGAAAGGATTT CCCAGGATCAGTAGGTGGAAT CTCTTCTGGCTCAGTCTGTTA AGTCAAGAGGCTGTGTTGCTGTGAA TGTTTGAGAACTATGTGGCAGATAT CCTTCAAGCTCGAATTACATCTTTA GCGTGAAGATCAAGATGGAGTTCAT CCGACTTGGTTAATATGCACATACT GACTCAGCCTCAGATAGCAACTGCT CACTGTGCTTAAGCTGGATAACACA GAGAAACCAGTGACCACCAGATGAA GATGGAAATTCGCCCGCTCAGATGA

Gene Therapy (2012) 1123 - 1132

imaged and 2 days later they were procesed for flow cytometry as described previously.22

Cell transfection and gene knockdown 293T cells (8 105 per well) were seeded in 6-well tissue culture-treated plates and transfection was done using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA), following the manufacturer’s instructions. Transfection was performed for 24 h and cells were replenished with fresh medium. The cells were cultured for 3 days and gene-knockdown efficiency was assessed by qRT-PCR. Briefly, total mRNA was isolated from the cells with RNeasy Mini Kit (Qiagen, Valencia, CA, USA) and cDNA was synthesized from the isolated mRNA using Superscript III cDNA Synthesis Kit (Invitrogen). The cDNA was subjected to real-time PCR using SYBR Green Kit (Bio-Rad, Hercules, CA, USA). Primers for real-time PCR were listed in Table 2.

Immunostaining

Table 2.

Primers for qRT-PCR

Accession number

Gene

Sense sequence (50 -- 30 )

NM_005359

SMAD4

NM_005901

SMAD2

NM_005904

SMAD7

NM_005902

SMAD3

NM_005406

ROCK1

NM_001664

RHOA

NM_003131

SRF

NM_153604

MYOCD

NM_014048

MRTFB/MLK2

NM_00274

PKN1

NM_004235

KLF4

For: CTTAGTGACCACGCGGTCTT Rev: GCTGCTGCATCTGTCGATGA For:CCGAAATGCCACGGTAGA Rev: GAATTTTACACACTGTTGCAGG For: TCTCAGGCATTCCTCGGAAG Rev: GCAGAGTCGGCTAAGGTGAT For: CTTCGCAGAGTGCCTCAGTG Rev: CTGGAACAGCGGATGCTTGG For: GTGCCACAGAGATCACTTAG Rev: CCAGATGGTGGATTCTTAGG For: TCTTCAGCAAGGACCAGTTC Rev: GCCTCAGGCGATCATAATCT For: AGCCTGAGCGAGATGGAGAT Rev: TGGAGAGTCTGGCGAGTTGA For: TCCAACGGCTTCTACCACTT Rev: AGCGGCTTCTTCTCTGAACA For: CATCGTGGCAGTGTCATCAT Rev: AGAACTGTGAAGGCGACATC For: CAGCGTGCTAGGCAGATGAA Rev: CAGAGCCTTGATGGCGAACA For: GTCAGCGTCAGCCTCCTCTT Rev: GAGACCGCCTCCTGCTTGAT



1131 plate to form gel with final concentration of fibrinogen and thrombin at 2.5 mg ml - 1 and 2.5 U ml - 1, respectively. After 1 h, the gels were released from the walls and incubated in GM, DM or DM plus SB431542 and 48 h the gels were photographed at a fixed distance using a gel documentation imaging system (UVP). The area of each gel was measured using Image J and normalized to the initial area (t ¼ 0 h).

Induced regulated shRNA expression protocol HF-MSCs (5 104 cells per well) were plated in 6-well plate. The cells were transduced with tet-pLKO-puro (Addgene) lentivirus encoding for the Tetracycline Repressor (TetR) in the presence of 8 mg ml - 1 polybrene. Thereafter, transduced cells were selected with 0.5 mg ml - 1 puromycin (Sigma, St Louis, MO, USA) for 5days. At the end of selection, HF-MSC (4 103 cells per well) were plated in 96-well tissue culture-treated plates. Cells in each well were transduced with virus carrying the indicated shRNA in the presence of 8 mg ml - 1 polybrene. The virus was removed at 2 h post infection and replaced with fresh cell culture medium. After 3 days, the cells were induced to differentiate in DM. Three days later, the cells were treated with Dox (10 mg ml - 1; BD Biosciences, Palo Alto, CA, USA) and cell images were taken every 12 h.

Image acquisition and analysis The image-analysis protocol was described previously.2 Briefly, fluorescence images of the cells were acquired at 3 days after induction of cell differentiation using a Zeiss Observer.Z1 inverted fluorescence microscope. The integral total fluorescence intensity of cells in each well was determined using Cellprofiler (http://www.cellprofiler.org/).

Statistical analysis Statistical analysis of the data was performed using a two-tailed Student’s t-test (a ¼ 0.05) in Microsoft Excel (Microsoft, Redwood, CA, USA). The sample size was n ¼ 3 and each experiment was repeated two to three times unless indicated otherwise.

CONFLICT OF INTEREST The authors declare no conflict of interest.

ACKNOWLEDGEMENTS This work was supported by grants from the National Heart and Lung Institute (R01 HL086582), the National Science Foundation (BES-0853993) and the New York Stem Cell Science Fund (NYSTEM, Contract# C024315) to STA.

REFERENCES 1 Zhou LL, Zhao Y, Ringrose A, DeGeer D, Kennah E, Lin AE et al. AHI-1 interacts with BCR-ABL and modulates BCR-ABL transforming activity and imatinib response of CML stem/progenitor cells. J Exp Med 2008; 205: 2657 - 2671. 2 Velkey JM, O’Shea KS. Oct4 RNA interference induces trophectoderm differentiation in mouse embryonic stem cells. Genesis 2003; 37: 18 - 24. 3 Zou GM, Chan RJ, Shelley WC, Yoder MC. Reduction of Shp-2 expression by small interfering RNA reduces murine embryonic stem cell-derived in vitro hematopoietic differentiation. Stem Cells 2006; 24: 587 - 594. 4 Zou GM, Luo MH, Reed A, Kelley MR, Yoder MC. Ape1 regulates hematopoietic differentiation of embryonic stem cells through its redox functional domain. Blood 2007; 109: 1917 - 1922. 5 Angaji SA, Hedayati SS, Poor RH, Madani S, Poor SS, Panahi S. Application of RNA interference in treating human diseases. J Genet 2010; 89: 527 - 537. 6 Paddison PJ, Silva JM, Conklin DS, Schlabach M, Li M, Aruleba S et al. A resource for large-scale RNA-interference-based screens in mammals. Nature 2004; 428: 427 - 431. 7 Foley E, O’Farrell PH. Functional dissection of an innate immune response by a genome-wide RNAi screen. PLoS Biol 2004; 2: E203. 8 Moffat J, Grueneberg DA, Yang X, Kim SY, Kloepfer AM, Hinkle G et al. A lentiviral RNAi library for human and mouse genes applied to an arrayed viral high-content screen. Cell 2006; 124: 1283 - 1298. 9 Kittler R, Surendranath V, Heninger AK, Slabicki M, Theis M, Putz G et al. Genome-wide resources of endoribonuclease-prepared short interfering RNAs for specific loss-of-function studies. Nat Methods 2007; 4: 337 - 344.


10 Silva JM, Mizuno H, Brady A, Lucito R, Hannon GJ. RNA interference microarrays: high-throughput loss-of-function genetics in mammalian cells. Proc Natl Acad Sci USA 2004; 101: 6548 - 6552. 11 Brass AL, Dykxhoorn DM, Benita Y, Yan N, Engelman A, Xavier RJ et al. Identification of host proteins required for HIV infection through a functional genomic screen. Science 2008; 319: 921 - 926. 12 Konig R, Zhou Y, Elleder D, Diamond TL, Bonamy GM, Irelan JT et al. Global analysis of host-pathogen interactions that regulate early-stage HIV-1 replication. Cell 2008; 135: 49 - 60. 13 Krishnan MN, Ng A, Sukumaran B, Gilfoy FD, Uchil PD, Sultana H et al. RNA interference screen for human genes associated with West Nile virus infection. Nature 2008; 455: 242 - 245. 14 Yeung ML, Houzet L, Yedavalli VS, Jeang KT. A genome-wide short hairpin RNA screening of jurkat T-cells for human proteins contributing to productive HIV-1 replication. J Biol Chem 2009; 284: 19463 - 19473. 15 Echard A, Hickson GR, Foley E, O0 Farrell PH. Terminal cytokinesis events uncovered after an RNAi screen. Curr Biol 2004; 14: 1685 - 1693. 16 Mukherji M, Bell R, Supekova L, Wang Y, Orth AP, Batalov S et al. Genome-wide functional analysis of human cell-cycle regulators. Proc Natl Acad Sci USA 2006; 103: 14819 - 14824. 17 Collinet C, Stoter M, Bradshaw CR, Samusik N, Rink JC, Kenski D et al. Systems survey of endocytosis by multiparametric image analysis. Nature 2010; 464: 243 - 249. 18 Zhang JZ, Gao W, Yang HB, Zhang B, Zhu ZY, Xue YF. Screening for genes essential for mouse embryonic stem cell self-renewal using a subtractive RNA interference library. Stem Cells 2006; 24: 2661 - 2668. 19 Ngo VN, Davis RE, Lamy L, Yu X, Zhao H, Lenz G et al. A loss-of-function RNA interference screen for molecular targets in cancer. Nature 2006; 441: 106 - 110. 20 Sheng Z, Li L, Zhu LJ, Smith TW, Demers A, Ross AH et al. A genome-wide RNA interference screen reveals an essential CREB3L2-ATF5-MCL1 survival pathway in malignant glioma with therapeutic implications. Nat Med 2010; 16: 671 - 677. 21 Tian J, Alimperti S, Lei P, Andreadis ST. Lentiviral microarrays for real-time monitoring of gene expression dynamics. Lab Chip 2010; 10: 1967 - 1975. 22 Tian J, Andreadis ST. Independent and high-level dual-gene expression in adult stem-progenitor cells from a single lentiviral vector. Gene Therapy 2009; 16: 874 - 884. 23 Hautmann MB, Madsen CS, Owens GK. A transforming growth factor beta (TGFbeta) control element drives TGFbeta-induced stimulation of smooth muscle alpha-actin gene expression in concert with two CArG elements. J Biol Chem 1997; 272: 10948 - 10956. 24 Kinner B, Zaleskas JM, Spector M. Regulation of smooth muscle actin expression and contraction in adult human mesenchymal stem cells. Exp Cell Res 2002; 278: 72 - 83. 25 Wang D, Park JS, Chu JS, Krakowski A, Luo K, Chen DJ et al. Proteomic profiling of bone marrow mesenchymal stem cells upon transforming growth factor beta1 stimulation. J Biol Chem 2004; 279: 43725 - 43734. 26 Mack CP, Somlyo AV, Hautmann M, Somlyo AP, Owens GK. Smooth muscle differentiation marker gene expression is regulated by RhoA-mediated actin polymerization. J Biol Chem 2001; 276: 341 - 347. 27 Szulc J, Wiznerowicz M, Sauvain MO, Trono D, Aebischer P. A versatile tool for conditional gene expression and knockdown. Nat Methods 2006; 3: 109 - 116. 28 Wiznerowicz M, Trono D. Conditional suppression of cellular genes: lentivirus vector-mediated drug-inducible RNA interference. J Virol 2003; 77: 8957 - 8961. 29 ter Brake O, t Hooft K, Liu YP, Centlivre M, von Eije KJ, Berkhout B. Lentiviral vector design for multiple shRNA expression and durable HIV-1 inhibition. Mol Ther 2008; 16: 557 - 564. 30 Owens GK, Kumar MS, Wamhoff BR. Molecular regulation of vascular smooth muscle cell differentiation in development and disease. Physiol Rev 2004; 84: 767 - 801. 31 Massague J. Receptors for the TGF-beta family. Cell 1992; 69: 1067 - 1070. 32 Ng F, Boucher S, Koh S, Sastry KS, Chase L, Lakshmipathy U et al. PDGF, TGF-beta, and FGF signaling is important for differentiation and growth of mesenchymal stem cells (MSCs): transcriptional profiling can identify markers and signaling pathways important in differentiation of MSCs into adipogenic, chondrogenic, and osteogenic lineages. Blood 2008; 112: 295 - 307. 33 Mack CP, Somlyo AV, Hautmann M, Somlyo AP, Owens GK. Smooth muscle differentiation marker gene expression is regulated by RhoA-mediated actin polymerization. J Biol Chem 2001; 276: 341 - 347. 34 Zhang YE. Non-Smad pathways in TGF-beta signaling. Cell Res 2009; 19: 128 - 139. 35 Larsson E, Zhou X, Akyurek LM. RhoA-dependent vascular smooth muscle cell-specific transcription: adding diaphanous formins to the puzzle. Arterioscler Thromb Vasc Biol 2007; 27: 448 - 449. 36 Wang D, Chang PS, Wang Z, Sutherland L, Richardson JA, Small E et al. Activation of cardiac gene expression by myocardin, a transcriptional cofactor for serum response factor. Cell 2001; 105: 851 - 862.

Gene Therapy (2012) 1123 - 1132


1132 37 Wang DZ, Li S, Hockemeyer D, Sutherland L, Wang Z, Schratt G et al. Potentiation of serum response factor activity by a family of myocardin-related transcription factors. Proc Natl Acad Sci USA 2002; 99: 14855 - 14860. 38 Wang DZ, Olson EN. Control of smooth muscle development by the myocardin family of transcriptional coactivators. Curr Opin Genet Dev 2004; 14: 558 - 566. 39 Liu Y, Sinha S, McDonald OG, Shang Y, Hoofnagle MH, Owens GK. Kruppel-like factor 4 abrogates myocardin-induced activation of smooth muscle gene expression. J Biol Chem 2005; 280: 9719 - 9727. 40 Qiu P, Feng XH, Li L. Interaction of Smad3 and SRF-associated complex mediates TGF-beta1 signals to regulate SM22 transcription during myofibroblast differentiation. J Mol Cell Cardiol 2003; 35: 1407 - 1420. 41 Edlund S, Landstrom M, Heldin CH, Aspenstrom P. Smad7 is required for TGF-betainduced activation of the small GTPase Cdc42. J cell sci 2004; 117: 1835 - 1847. 42 Miyake T, Alli NS, McDermott JC. Nuclear function of Smad7 promotes myogenesis. Mol cell biol 2010; 30: 722 - 735. 43 Wiederschain D, Wee S, Chen L, Loo A, Yang G, Huang A et al. Single-vector inducible lentiviral RNAi system for oncology target validation. Cell Cycle 2009; 8: 498 - 504.

44 Dickins RA, Hemann MT, Zilfou JT, Simpson DR, Ibarra I, Hannon GJ et al. Probing tumor phenotypes using stable and regulated synthetic microRNA precursors. Nat Genet 2005; 37: 1289 - 1295. 45 Beronja S, Livshits G, Williams S, Fuchs E. Rapid functional dissection of genetic networks via tissue-specific transduction and RNAi in mouse embryos. Nat Med 2010; 16: 821 - 827. 46 Pongratz C, Yazdanpanah B, Kashkar H, Lehmann MJ, Krausslich HG, Kronke M. Selection of potent non-toxic inhibitory sequences from a randomized HIV-1 specific lentiviral short hairpin RNA library. PloS one 2010; 5: e13172. 47 Liu JY, Peng HF, Andreadis ST. Contractile smooth muscle cells derived from hair-follicle stem cells. Cardiovasc Res 2008; 79: 24 - 33. 48 Liu JY, Peng HF, Gopinath S, Tian J, Andreadis ST. Derivation of functional smooth muscle cells from multipotent human hair follicle mesenchymal stem cells. Tissue Eng Part A 2010; 16: 2553 - 2564. 49 Han J, Liu JY, Swartz DD, Andreadis ST. Molecular and functional effects of organismal ageing on smooth muscle cells derived from bone marrow mesenchymal stem cells. Cardiovasc Res 2010; 87: 147 - 155.

Supplementary Information accompanies the paper on Gene Therapy website (http://www.nature.com/gt)

Gene Therapy (2012) 1123 - 1132
