Functional Characterization of Insulation Effect for Synthetic Gene

Subscriber access provided by READING UNIV

Article

Functional characterization of insulation effect for synthetic gene circuits in mammalian cells Weixi Liao, Bing Liu, Chih-Chun Chang, Ling-Jun Lin, Che Lin, Bor-Sen Chen, and Zhen Xie ACS Synth. Biol., Just Accepted Manuscript • DOI: 10.1021/acssynbio.7b00134 • Publication Date (Web): 01 Dec 2017 Downloaded from http://pubs.acs.org on December 3, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Synthetic Biology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Synthetic Biology

Functional characterization of insulation effect for synthetic gene circuits in mammalian cells Weixi Liao1#, Bing Liu2#, Chih-Chun Chang3, Ling-Jun Lin3, Che Lin3, Bor-Sen Chen3 and Zhen Xie1* 1

MOE Key Laboratory of Bioinformatics and Bioinformatics Division, Center for

Synthetic and System Biology, Department of Automation, Tsinghua National Lab for Information Science and Technology, Tsinghua University, Beijing 100084, China 2

Syngentech Inc., Zhongguancun Life Science Park, Changping District, Beijing

102206, China 3

Department of Electrical Engineering, National Tsing Hua University, Hsinchu,

30013, Taiwan #

These authors contribute equally to this work.

ABSTRACT Insulators are non-coding gene regulatory elements in eukaryotic genome, which function as chromatin partitioning boundaries, and block interference across different chromatin domains. To facilitate modular construction of synthetic gene circuit that is usually composed of multiple transcription cassettes, unwanted cross-regulations between different cassettes should be avoided. Here, we developed a quantitative method to characterize the functional effect of three insulators on the crossregulations of six promoters in mammalian cells. We showed that the unwanted crossregulations displayed a threshold-like effect, and the threshold position varied along with the context of promoters and insulators. We tested the function of insulators in both cascade and sensory switch circuits assembled in episomal plasmid vectors, and showed that the insulation effect was mainly revealed on the first regulatory layer of the cascade circuit. A deviation on the response curve of the sensory switch circuit with or without insulators was observed, but response intensity of some sensory switch circuits were not affected. Therefore, our results provided a general guidance on the selection of insulators with varying promoters in episomal synthetic gene

ACS Paragon Plus Environment

ACS Synthetic Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

circuits in mammalian cells, which may be useful to reduce the effect of the unwanted cross-regulations.

INTRODUCTION From an engineering perspective, synthetic gene circuit is an assembly of artificial gene parts to perform sophisticated information processing of multiple molecular signals and execute intended biological functions 1. Dozens of synthetic gene circuits have been constructed in mammalian cells 2,3. Some of these circuits are integrated in genome, actuating long-term functions, while the others are assembled in episomal plasmid vectors and transiently operating in host cells to reduce oncogenic risk from unexpected host genome integration. Because the capacity and capability of a single expression unit are limited, synthetic circuits usually consist of multiple expression cassettes. Assembling multiple modules on a single DNA vector alleviates the noise derived from copy number variation, but introduces unwanted cross-regulations between modules

4–6

. In addition, integrating all gene parts in one DNA vector can

facilitate the delivery of gene circuits for therapeutic purposes by using viral vehicles with a linear DNA genome, including adenovirus, adeno-associated virus and herpes simplex virus 7. Similar challenges exist in the development of electronic circuits, which can be solved by transferring signals in physically insulated wires. However, gene circuit operates in a noisy cellular environment with nearly no spatial isolation in the genome. Furthermore, different functional DNA elements in the gene circuit often interfere with each other, which may cause unexpected circuit behaviors from expectations. In prokaryotic cells, insulation from neighboring genetic context can be achieved by inserting cleavage site at the junction between promoters and genes to generate standardized genetic parts 8,9. However, this strategy is not applicable for gene circuits operating in mammalian cells. One of the solutions for insulating a transcription unit from its genomic environment in mammalian cells is to form the loop chromatin domain by recruiting various DNA binding to a class of genomic elements called “insulators”

10,11

. Insulators can be used to block the effect of neighboring enhancers

and serve as barriers of euchromatic and heterochromatic regions in mammalian genomes 10.


Page 2 of 19

Page 3 of 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The first well-studied insulator was identified in the 5’ end of the chicken betaglobin locus (cHS4) that contains a 250-bp CCCTC-enriched region bound by the CCCTC binding factor (CTCF)

12–14

. Moreover, it has been shown that tDNA and

retrotransposon repeats can also function as insulators in mammalian genome, such as the short interspersed nuclear element B2 (SINE B2) in the growth hormone gene locus and the tDNA repeats upstream of the arachidonate lipoxygenase (ALOXE3) gene

15,16

. Interestingly, inserting the core cHS4 insulator in episomal plasmid and

lentivirus plasmid effectively suppresses transgene silencing and blocks the neighboring promoter interference 17,18. In addition, the insulator displays the blocking activity in the linearized vector only when the insulator is placed between the neighboring promoters and enhancers 19. In this study, we aimed at a comprehensive evaluation of the insulation effect of three insulators (cHS4, SINE B2 and ALOXE3) on six promoters (CAG, CAGop, CMV, hEF1α, UAS and TRE) that are commonly used in synthetic gene circuit in mammalian cells. We further evaluated the insulation effect on the cascade and sensory switch circuits assembled in episomal plasmid vectors, which provides a useful guidance to help reducing the internal interference in synthetic gene circuit with multiple gene expression units.

RESULTS AND DISCUSSION Characterization of the insulation effect on different promoters To measure the promoter/enhancer blocking efficiency of different insulators in mammalian cells, we assembled a TagBFP expression unit driven by varying promoters and a EYFP expression unit driven by a minimal CMV promoter into one linear DNA plasmid (Fig. 1A). The promoter/enhancer in either upstream or downstream of the EYFP expression unit may cross-activate the minimal CMV promoter. The EYFP expression unit was flanked with insulators at both sides to alleviate the interference of the promoter/enhancer (Fig. 1A). We measured the performances of three reported insulators (cHS4, ALOXE3 and SINE B2) in mammalian genome in either forward (labeled with a “-f”) or reversed (labeled with a “-r”) orientation, and analyzed the insulation activity against six promoters/enhancers



that are commonly used in synthetic gene circuits, including CAG, CAGop, CMV, hEF1α, UAS and TRE, located in either upstream (labeled with a “+”) or downstream (labeled with a “-”) sequences of the EYFP expression unit. We introduced individual synthetic circuit that contained the CAG, CAGop, CMV, or hEF1α promoters into HEK293 cells by transient transfection. Meanwhile, the UAS and TRE promoters are both synthetic promoters by combining the minimal CMV and repeating binding sites of the Gal4-VP16 and rtTA, respectively. For synthetic circuits with the UAS and TRE promoters, plasmid DNA constructs that expressed Gal4-VP16 and rtTA were respectively cotransfected into HEK293 cells. Then, we measured the EYFP and TagBFP levels by using flow cytometry. By comparing the mean value of the normalized EYFP levels among different combinations of promoters and insulators, we demonstrated that the EYFP expression unit was cross-activated by neighboring promoters and insulators generally alleviated the cross-activation significantly with a few exceptions like cHS4 used in reversed orientation (Fig. 1B). The promoters in the downstream of the EYFP expression unit led to higher EYFP levels than those in the upstream. Among these promoters assayed, the UAS and TRE promoters demonstrated a stronger cross-activation than the others. To evaluate the effect of copy number variation on the performance of synthetic gene circuit, we analyzed the flow cytometry data by dividing individual cell events into bins based on the mKate2 level that was used as an indication of copy numbers (Supplementary Fig. 1) 20. We observed that along with the increase of copy numbers, the insertion of insulators influenced simultaneously TagBFP and EYFP expression, and the correlation between the TagBFP and EYFP level was non-linear. These results suggested that strong cross-activation of EYFP we observed may be due to strong expression of TagBFP and the mean value of the EYFP or TagBFP may not be an appropriate indication to evaluate the insulator performance. Therefore, the evaluation approach need to be improved. To characterize the relation between TagBFP and EYFP values and normalize varying TagBFP levels among different combinations of promoters and insulators, we used kernel density estimation to obtain continuous density distribution of TagBFP and EYFP expression based on a reasonable assumption that only strong interference could overcome the insulation barriers (Fig. 2A). We fit the curves by using a threshold-like model with three parameters, including the base line (parameter a), the


Page 4 of 19

Page 5 of 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


turning point (parameter b) and the slope (parameter k). We demonstrated that in general the basal expression of EYFP fluctuated in a narrow range. However, the turning point shifted to the higher TagBFP level and the slope decreased when insulators were used (Supplementary Fig. 2). To generate a compound parameter to reflect the changes in the turning point and the slope, we calculated the Area Under Curve (AUC) to globally compare the influence on the EYFP by the TagBFP level to evaluate the insulation effect of different insulators (Fig. 2A). By using this method, the TagBFP levels of different promoters were normalized to the same level by the prediction of the threshold-like model, which allowed quantitative comparisons among different combinations of promoters and insulators. A few “ineffective” insulators shown in Fig. 1B were found to be effective in Fig. 2B by using the improved evaluation method, such as the cHS4 insulator used in reversed orientation. Those insulators with top insulation performance against cross-activation exerted by different promoters were marked in Fig. 2C and recommended for future episomal gene circuit engineering. Insulation effect on performance of synthetic gene circuits We constructed cascade circuits to investigate the influence of the unwanted cross-regulations on the intended regulations with or without insulations (Fig. 3A and 3C). Adding the small molecule doxycycline (Dox) induced the expression of both EYFP and TALER10. TALER10 repressed the expression of mKate2 and TALER21, while TALER21 repressed the expression of iRFP. The TALER10 and TALER21 are two orthogonal artificial repressors engineered in our previous study 6. The EYFP and mKate2 encoding genes were constructed into a single linear plasmid with no insulation, or with the cHS4-f that showed a medium insulation, or with the ALOXE3-r that displayed a strong insulation (Fig. 2C and Fig. 3A). The results demonstrated that the strong insulation effect protected the expression of mKate2 from the interference of the TRE promoter. However, the insulation effect on the expression of iRFP was not distinct, suggesting that the insulation effect was attenuated on the second regulatory layer in the cascade circuit (Fig. 3B). By using the binning analysis based on the TagBFP level, we observed a significant insulation effect on the response intensity of mKate2 especially when the plasmid copy number was high, while no significant insulation effect was observed on the response intensity of iRFP (Supplementary Fig. 3A). Similar results were observed when the modules



encoding TALER10 and TALER21 were exchanged in the cascade circuit (Fig. 3C, 3D and Supplementary Fig. 3B). Next, we assayed the insulation effect on the sensory switch circuits that we reported previously (Fig. 4) 6. TALER10 and TALER21 that mutually repressed with each other, co-expressed respectively with the EYFP and the mKate2 by using a selfcleaving 2A linker

21

. In addition, the shRNA-FF3 and shRNA-FF4 respectively

inhibited TALER21 and TALER10 (Fig. 4A). While we transfected the same amount of the sensory circuit into HEK293 cells, we gradually decreased the amount of shRNA-FF4 to 0 ng and then gradually increased the amount of shRNA-FF3 to the saturated level, which turned the sensory switch from the EYFP “OFF” and mKate2 “ON” state to the EYFP “ON” and mKate2 “OFF” state. We observed that a deviation on the response curve towards a higher leaky level of mKate2 at the EYFP “ON” state and a higher leaky level of EYFP at the mKate2 “ON” state (Fig. 4B). However, the response intensity of both mKate2 and EYFP were hardly changed (Fig. 4B). When we exchanged the modules encoding TALER10 and TALER21, EYFP response was improved but mKate2 response was reduced with the insertion of ALOXE3-r (Fig. 4C and 4D). We also tested sensory switch circuits with non-uniform enhancers by replacing one of the UAS enhancers with CMV, and the competence of insulators to simultaneously increase EYFP and mKate2 response was demonstrated (Fig. 4E-4H). We also evaluated the effect of copy number variation on the insulation performance by using the binning analysis based on the TagBFP level (Supplementary Fig. 3C-3F). In general, we found that sensory switch circuit with a mutual inhibition displayed a better tolerance of unwanted cross-regulation caused by neighboring gene expression units than cascade circuit when generating a switch-like behavior. Discussion In summary, we developed a threshold-like model to describe unwanted crossregulations occurred in the synthetic gene circuits assembled in episomal plasmid vectors and provided a general guideline for the selection of insulators when different promoters were used. We observed that unwanted cross-regulations were relatively weak compared to intended regulations (Fig. 2). However, the interference did reduce the response intensity of the first regulatory layer of the cascade circuit (Fig. 3). Interestingly, a deviation on the response curve was observed among sensory switch


Page 6 of 19

Page 7 of 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


circuits with or without insulators, but the performance of sensory switch circuits was hardly affected (Fig. 4). It will be interesting to study the deviation propagation in gene circuits with different topologies. On the other hand, further efforts are necessary to understand the interference between the host system and synthetic gene circuits, such as the epigenetic regulations, deregulation of transcription and competition for limited cellular resources. Besides, delicate elucidation of natural insulation system will also be another interesting direction to pursue for reducing the interference of unwanted cross-regulations in synthetic gene circuit.

METHODS Reagents and enzymes Restriction endonuclease, polynucleotide kinase (PNK), T4 DNA ligase and Quick DNA ligase (New England BioLabs) were used in cloning. Phusion HighFidelity DNA polymerase (New England BioLabs) were used in PCR amplification. Oligonucleotides were synthesized by Genewiz. Dox was purchased from Clontech. The Gateway LR reaction (Life Technologies) were performed by following the manufacturer’s protocols. Linear DNA vector pJAZZ was purchased from Lucigen. Plasmid DNA constructs Insulators used in Fig. 1 and 2 were synthesized or amplified by PCR from the genome DNA, and inserted into the donor vectors of Gateway LR reaction. Promoters used in Fig. 1 and 2 were synthesized or amplified by PCR from previous plasmids (Xie et al., 2011), and inserted into the entry vectors. TALER and fluorescent proteins used in Fig. 3 and 4 were amplified by PCR from previous plasmids (Li et al., 2015), and inserted into the entry vectors. Modules were constructed by Gateway LR reaction with respective entry and donor vectors. Multiple modules were assembled together as described (Guye et al., 2013). Briefly, each module was pooled at 7 fmol and digested with I-SceI, and was then added into a Gibson reaction buffer with an integrative carrier vector and an adaptor vector. Cell culture and transfection



HEK293 (293-H) cell line was purchased from Invitrogen. HEK293 cells was cultured in high-glucose DMEM complete medium (Dulbecco’s modified Eagle’s medium (DMEM), 4.5 g/L glucose, 0.045 units/mL of penicillin and 0.045 g/mL streptomycin and 10% FBS (Invitrogen)) at 37 °C, 100% humidity and 5% CO2. In transfection experiments, ~6 × 104 HEK293 cells in 0.5 mL of high-glucose DMEM complete medium were seeded into each well of 24-well plastic plates (Falcon) and grown for ~24 h. Shortly before transfection, the medium was replaced with fresh DMEM complete medium. Lipofectamine 3000 (Life Technologies) was used by following the manufacturer’s protocol. pDT7004 (pUBI-linker-NOS) containing a maize ubiquitin promoter (UBI) followed by a NOS terminator with no proteincoding sequences between UBI and NOS was used to ensure an equal amount of plasmid DNA (Xie et al., 2011). The amount of plasmid DNA and the final concentration of Dox added to each well are listed in the Supplementary Table 1. Cells were cultured for 48 h before flow cytometry analysis. FACS measurement Cells were trypsinized 48 h after transfection and were then centrifuged at 250 g for 10 min at 4 °C. The supernatant was removed, and the cells were resuspended in 1× PBS free of calcium or magnesium. Fortessa flow analyzers (BD Biosciences) were used for fluorescence-activated cell sorting (FACS) analysis with the following settings. EBFP2 and TagBFP were measured using a 405-nm laser and a 450/50 filter with a photomultiplier tube (PMT) 260 V. EYFP was measured with a 488-nm laser and a 530/30 filter using a PMT 260 V. mKate2 was measured with a 561-nm laser and a 670/30 filter using a PMT 400 V. iRFP was measured using a 640-nm laser and a 780/60 filter with a PMT 450 V. For each sample, ~1 × 104 to ~1 × 105 cell events were collected. In parallel, Rainbow Calibration Particles (Spherotech Inc.) were measured in order to standardize the data (described in the section of Data analysis). Data analysis We used Rainbow Calibration Particles to convert relative fluorescent units to standardized units and the normalized fluorescence was calculated as described (Li et al., 2015). Briefly, we plotted the intensity of brightest four peaks of each channel we used (MEFL for EYFP channel, MECY for mKate2 channel, MEBFP for TagBFP channel and MEAPCY7 for iRFP channel) in units of absolute fluorescence versus


Page 8 of 19

Page 9 of 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


measured peak intensity for a given set of machine settings, and established a linear regression of standardized units versus mean fluorescence on a logarithmic scale. To compensate for transfection efficiency variation, non-transfected samples (NT) were measured and a constitutively expressed fluorescence reporter was used as an internal control (IC). The normalized fluorescence (NFL) was calculated from the mean value in the measured cell population as follows. NFL= Fitting the results with non-linear models in Fig. 2 was processed in four steps. Firstly, we characterized the linear calibration curves of machine units of a given set of machine settings versus absolute fluorescence intensity of Rainbow Calibration Particles and translated the measured values of EYFP and TagBFP as MEFL and MEBFP following the calibration curves. Secondly, kernel density estimation was used to obtain a continuous probability density distribution. Thirdly, it was converted into marginal probability density distribution to obtain the corresponding peaks and for visualization. Lastly, a threshold-like model was used to fit the peaks. We found that three parameters were adequate to fully describe the model. They were the base line (parameter a), the turning point (parameter b) and the increasing rate (parameter k).

The Area Under Curve (AUC) was calculated to globally evaluate the relation between MEBFP and MEFL.

where xL and yL indicated the minimum values on the X and Y axes, while xH indicated the maximum value on the X axis. All data were processed using Matlab (Mathworks). The response intensity (RI) was calculated from the maximum (Vmax) and minimum (Vmin) values among multiple experimental groups as follows.



ASSOCIATED CONTENT Supporting Information Supplementary figures, DNA sequences of promoters and insulators, and transfection configurations.

AUTHOR INFORMATION Corresponding Author * Email: [email protected] (Z.X.). Author Contributions Z.X., C.L. and B.C. conceived of the ideas implemented in this work. W.L., B.L., performed most of the experiments. C.C, L.L. made plasmid constructs in Fig. 1 and Fig. 2. Z.X., W.L., B.L., C.L. and B.C. performed data analysis. Z.X. supervised the project. Z.X. and W.L. wrote the paper. Notes The authors declare no financial interests in the work described in this manuscript.

ACKNOWLEDGEMENTS We thank members of Xie lab for helpful discussions. We thank Yun Jiang and Huiya Huang for technical support. We thank Xiaowo Wang for insightful discussions. The research was supported by the Cross-strait Tsinghua Foundation (Z.X.), the National Key Basic Research Program of China (2014CB745200), the Basic Research Program of Tsinghua National Lab for Information Science and Technology, and the National Science Council (B.C., NSC 98-221-E-007-113-MY3).

REFERENCES


Page 10 of 19

Page 11 of 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(1) Xie, Z., Wroblewska, L., and Weiss, R. (2014) RNAi Synthetic Logic Circuits for Sensing, Information Processing, and Actuation, in Encyclopedia of Molecular Cell Biology and Molecular Medicine. (2) Wieland, M., and Fussenegger, M. (2012) Engineering Molecular Circuits Using Synthetic Biology in Mammalian Cells 209–234. (3) Lienert, F., Lohmueller, J. J., Garg, A., and Silver, P. A. (2014) Synthetic biology in mammalian cells : next generation research tools and therapeutics. Nat. Publ. Gr. 15, 95–107. (4) Bleris, L., Xie, Z., Glass, D., Adadey, A., Sontag, E., and Benenson, Y. (2011) Synthetic incoherent feedforward circuits show adaptation to the amount of their genetic template. Mol. Syst. Biol. 7, 1–12. (5) Weiss, R. (2014) Accurate Predictions of Genetic Circuit Behavior from Part Characterization and Modular Composition. (6) Li, Y., Jiang, Y., Chen, H., Liao, W., Li, Z., Weiss, R., and Xie, Z. (2015) Modular construction of mammalian gene circuits using TALE transcriptional repressors. Nat. Chem. Biol. 1–28. (7) Miest, T. S., and Cattaneo, R. (2014) New viruses for cancer therapy: meeting clinical needs. Nat. Rev. Microbiol. 12, 23–34. (8) Lou, C., Stanton, B., Chen, Y., Munsky, B., and Voigt, C. A. (2012) Ribozymebased insulator parts buffer synthetic circuits from genetic context. (9) Qi, L., Haurwitz, R. E., Shao, W., Doudna, J. a, and Arkin, A. P. (2012) RNA processing enables predictable programming of gene expression. Nat. Biotechnol. 30, 1002–1006. (10) Amouyal, M. (2010) Gene insulation. Part II: natural strategies in vertebrates. Biochem. cell Biol. 88, 885–898. (11) Yang, J., and Corces, V. G. (2012) Insulators, long-range interactions, and genome function. Curr. Opin. Genet. Dev. 22, 86–92. (12) Chung, J. H., Bell, a C., and Felsenfeld, G. (1997) Characterization of the chicken beta-globin insulator. Proc. Natl. Acad. Sci. U. S. A. 94, 575–580. (13) Bell, A. C., West, A. G., and Felsenfeld, G. (1999) The protein CTCF is required for the enhancer blocking activity of vertebrate insulators. Cell 98, 387–396. (14) Yusufzai, T. M., and Felsenfeld, G. (2004) The 5’-HS4 chicken beta-globin insulator is a CTCF-dependent nuclear matrix-associated element. Proc. Natl. Acad. Sci. U. S. A. 101, 8620–4. (15) Lunyak, V. V, Prefontaine, G. G., Núñez, E., Cramer, T., Ju, B.-G., Ohgi, K. A., Hutt, K., Roy, R., García-Díaz, A., Zhu, X., Yung, Y., Montoliu, L., Glass, C. K., and Rosenfeld, M. G. (2007) Developmentally regulated activation of a SINE B2 repeat as a domain boundary in organogenesis. Science 317, 248–51. (16) Raab, J. R., Chiu, J., Zhu, J., Katzman, S., Kurukuti, S., Wade, P. A., Haussler,



D., and Kamakaka, R. T. (2012) Human tRNA genes function as chromatin insulators. EMBO J. 31, 330–350. (17) Chen, Z.-Y., Riu, E., He, C.-Y., Xu, H., and Kay, M. a. (2008) Silencing of episomal transgene expression in liver by plasmid bacterial backbone DNA is independent of CpG methylation. Mol. Ther. 16, 548–556. (18) Uchida, N., Hanawa, H., Yamamoto, M., and Shimada, T. (2013) The Chicken Hypersensitivity Site 4 Core Insulator Blocks Promoter Interference in Lentiviral Vectors. Hum. Gene Ther. Methods 24, 117–124. (19) Recillas-Targa, F., Bell, a C., and Felsenfeld, G. (1999) Positional enhancerblocking activity of the chicken beta-globin insulator in transiently transfected cells. Proc. Natl. Acad. Sci. U. S. A. 96, 14354–14359. (20) Yuan, Y., Liu, B., Xie, P., Zhang, M. Q., Li, Y., Xie, Z., and Wang, X. (2015) Model-guided quantitative analysis of microRNA-mediated regulation on competing endogenous RNAs using a synthetic gene circuit. Proc. Natl. Acad. Sci. 112, 3158– 3163. (21) Szymczak, A. L., Workman, C. J., Wang, Y., Vignali, K. M., Dilioglou, S., Vanin, E. F., and Vignali, D. A. A. (2004) Correction of multi-gene deficiency in vivo using a single “self-cleaving” 2A peptide-based retroviral vector. Nat. Biotechnol. 22, 589–594.

Figure 1. Characterization of the insulation effect on different promoters (A) Schematic representation of circuits for characterization of unwanted crossregulations. Promoter/enhancer +/- indicates the promoter/enhancer in the upstream/downstream sequences of EYFP expression unit respectively. Insulator indicates a pair of insulators on the flank of the output unit. Insulator “–f” / “-r” indicates forward /reversed orientations. Constitutively expressed mKate2 was used as the internal control. The rtTA encoding plasmid was optional when the TRE promoter was characterized and the GAL4-VP16 encoding plasmid was optional when the UAS promoter was characterized. (B) Characterization of indicated insulators and interference promoters. Each bar shows mean ± SD from 3 independent replicates of EYFP (MEFL) divided by mKate2 (MECY) normalized by the negative control (no interference promoter/enhancer). The one-sided Student’s t-test was used. Top lines indicate P-value < 0.05. Solid/dashed lines indicate significant difference between “promoter-” / “promoter+” groups with and without indicated insulators.


Page 12 of 19

Page 13 of 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2. Characterization of cross-regulations by using a threshold-like model (A) The schematic of the FACS data processing (See details in Methods). (B) Characterization of Area Under Curve (AUC) in the threshold-like model of indicated combination of insulators and promoters. Each bar shows mean ± SD from 3 independent replicates of AUC. The one-sided Student’s t-test was used. Top lines indicate P-value ≤ 0.05. Solid/dashed lines indicate significant difference between “promoter-” / “promoter+” groups with and without indicated insulators respectively. (C) Rank of insulator performance towards different promoters by using AUC characterization. The one-sided Student’s t-test was used, and insulators with red flag markers are not significantly weaker than the top rank towards the respective promoter (P-value > 0.05).

Figure 3. Functional assay of insulation effect on synthetic cascade circuits (A) Schematic representation of TALER cascade circuits. Insulators were placed at the flanking region of the mKate2 expression unit. (B) Characterization of (A) in response to varying amount of Dox. Mean values ± SD from 3 independent replicates of mKate2 (MECY) or iRFP (MEAPCY7) divided by TagBFP (MEBFP) normalized by the sample result without Dox induction are shown on the left. The one-sided Student’s t-test was used. “*” symbol indicates the P-values (≤0.05) that reflected the differences between circuits with and without indicated insulators. Mean ± SD from 3 independent replicates of response intensity of mKate2 and iRFP are showed on the right. The one-sided Student’s t-test was used to evaluate the differences between circuits with and without insulators. “n” indicates the P-value >0.05, and “*” indicates the P-value ≤0.05. (C) Schematic representation of cascade circuits which exchanged TALER10 and TALER21 encoding modules in (A). (D) Characterization of (C) in response to varying amount of Dox. Mean values ± SD from 3 independent replicates of EYFP (MEFL) or iRFP (MEAPCY7) divided by TagBFP (MEBFP) normalized by the sample result without Dox induction are showed on the left. The one-sided Student’s t-test was used. “*” symbol indicates the P-values (≤0.05). Mean values ± SD from 3 independent replicates of response intensity of EYFP and iRFP are showed



on the right. The one-sided Student’s t-test was used. “n” indicates the P-value >0.05, and “*” indicates the P-value ≤0.05.

Figure 4. Functional assay of insulation effect on sensory switch circuits (A) Schematic representation of sensory switch circuits for insulator functional assay. Insulators are on the flank of the mKate2 expression unit. (C) Schematic representation of sensory switch circuit which exchanged TALER10 and TALER21 encoding modules in (A). (E) Schematic representation of sensory switch circuits in which one of the UAS enhancers in (A) was replaced with CMV. (G) Schematic representation of sensory switch circuits in which one of the UAS enhancers in (C) was replaced with CMV. (B, D, F and H) Characterization of (A, C, E and G) in response to varying amount of shRNAs. Mean values ± SD from 3 independent replicates of EYFP (MEFL) or mKate2 (MECY) divided by TagBFP (MEBFP) from minimum to maximum are showed on the left. Mean ± SD from 3 independent replicates of response intensity of EYFP (MEFL) and mKate2 (MECY) are showed on the right. The one-sided Student’s t-test is used. “n” indicates the P-value >0.05, and “*” indicates the P-value ≤0.05.


Page 14 of 19

Page 15 of 19


A

phEF1α pCAG pCAG

B

rtTA

TagBFP

for TRE

Gal4-VP16

for UAS

PromoterPromoter+ None cHS4-f cHS4-r

EYFP (log10MEFL, AU)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

miniCMV

Promoter+

mKate2

EYFP

Insulator-f(orward)/-r(everse) TagBFP EYFP PromoterminiCMV

significant difference with/without insulators ALOXE3-f ALOXE3-r SINE B2-f SINE B2-r

1.2 0.6 0

-0.6

CAG

CAGop

CMV

hEF1α


UAS

TRE


A

EYFP (MEFL, AU)

Without insulators

Marginal probability density distribution

Probability density distribution

Normalized scatter plot

Threshold-like model

k

TagBFP (MEBFP, AU)

EYFP (MEFL, AU)

With insulators

a

AUC b

TagBFP (MEBFP, AU)

B Area Under Curve (AU)

PromoterPromoter+ None cHS4-f cHS4-r

significant difference with/without insulators ALOXE3-f ALOXE3-r SINE B2-f SINE B2-r

12 11 10

8

CAGop -f -r -f -r -f -r

TRE Rank 1 2 3 4 5 6

-

+

-

+

-

+

-

+


-

+

-

+ TRE

SINE B2

UAS

UAS

ALOXE3

hEF1α

hEF1α

cHS4

CMV

CMV

CAG

CAGop

7

CAG

C

9

: not significantly weaker than the top rank

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 19

TALER10

Insulator

Dox

mKate2

EYFP

iRFP

phEF1a

T10

UAS T21

2A EYFP TALER10

Insulator

Dox concentration (ng/mL)

D

0

0

100 101 102

0


n

None cHS4-f ALOXE3-r 0

** *

-0.5

2A mKate2 TALER21 pTRE

101 102

-1 0

100 101 102



1

0.5

0

0

100 101 102


0.9

*n *

0.6 0.3 0

iRFP

miniCMV

100

0.2

EYFP

Dox

miniCMV 2A 2A UAS rtTA TagBFP Gal4-VP16 iRFP

-1 0

*

0.4

ALOXE-r

C

2A

0.5

ALOXE-r

UAS

EYFP pTRE

**

iRFP

T10

2A mKate2 TALER21

*

-0.5

0.6

cHS4-f

miniCMV

n

*n n

cHS4-f

Dox

*

0.8

None

T21

1

None

miniCMV 2A 2A UAS rtTA TagBFP Gal4-VP16 iRFP

None cHS4-f ALOXE3-r 0

Response intensity (AU)

phEF1a

B


iRFP

iRFP (log10MEAPCY7, AU)

mKate2

iRFP (log10MEAPCY7, AU)

EYFP

Dox

mKate2 (log10MECY, AU)

A

mKate2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53


EYFP (log10MEFL, AU)

Page 17 of 19

Insulator

Gal4-VP16

2A

TagBFP

UAS T10

shRNA-FF3

EYFP

mKate2

shRNA-FF4 2A EYFP TALER10

miniCMV CMV T21

F

0.5 -0.5 -1.5 -2.5 -1.5

Insulator

2A

EYFP

mKate2

shRNA-FF3 2A mKate2 TALER21

miniCMV CMV T10

H

1

0.5

-0.5 0.5 EYFP (log10MEFL, AU)

0

* ** n

-0.5 -1

None cHS4-f ALOXE3-r

-1.5 -2

-2.5 -2

Insulator


ALOXE3-r

cHS4-f

1.5

-1

0 1 2 EYFP (log10MEFL, AU)

3 2

mKate2


miniCMV UAS T21

TagBFP

shRNA-FF3

2

** n *

EYFP

Gal4-VP16

shRNA-FF4


G pCAG


0

mKate2


miniCMV

shRNA-FF4


pCAG

0 1 EYFP (log10MEFL, AU)

EYFP

E

-1

ALOXE3-r

-3.5 -2

1

ALOXE3-r

UAS T10

-2.5

2

1 0

ALOXE3-r


miniCMV

-1.5

3

cHS4-f

UAS T21

mKate2


n n

mKate2


miniCMV

EYFP

-0.5

*n

cHS4-f

TagBFP

D

EYFP

Gal4-VP16

2A

shRNA-FF3

0

cHS4-f

shRNA-FF4


C pCAG

1 2 EYFP (log10MEFL, AU)

None

Insulator

0

1

None

-3 -1

2

None

-2

3

None

UAS T21

-1



miniCMV

0



UAS T10

mKate2

1

Page 18 of 19

mKate2


miniCMV

EYFP

B


TagBFP

shRNA-FF3


Gal4-VP16

2A

shRNA-FF4


A pCAG

n n n n

EYFP

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53


Page 19 of 19 1 2 3 4



Functional Characterization of Insulation Effect for Synthetic Gene

Recommend Documents