Subscriber access provided by Northern Illinois University
Letter
Virus-enabled optimization and delivery of the genetic machinery for efficient unnatural amino acid mutagenesis in mammalian cells and tissues Yunan Zheng, Tommy L. Lewis, Peter Igo, Franck Polleux, and Abhishek Chatterjee ACS Synth. Biol., Just Accepted Manuscript • DOI: 10.1021/acssynbio.6b00092 • Publication Date (Web): 02 Aug 2016 Downloaded from http://pubs.acs.org on August 8, 2016
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 15
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
Virus-enabled optimization and delivery of the genetic machinery for efficient unnatural amino acid mutagenesis in mammalian cells and tissues
Yunan Zheng,1 Tommy L. Lewis Jr,2 Peter Igo,1 Franck Polleux,2 and Abhishek Chatterjee1*
1
2
*Department of Chemistry, Boston College, 2609 Beacon Street, Chestnut Hill, MA 02467, USA
Department of Neuroscience, Columbia University, 550 West 120th Street New York, NY 10027, USA
*email:
[email protected] 1 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
Abstract: Unnatural amino acid (UAA) mutagenesis of recombinant proteins in live mammalian cells requires co-expression of the mutant target, as well as an engineered tRNA/aminoacyl-tRNA synthetase pair. The ability to readily determine the optimal relative expression levels of these three genetic components for efficient expression of the UAA-modified target is highly desirable, but remains challenging to accomplish. Here we report a facile strategy to achieve this by taking advantage of the efficient gene-delivery by a baculovirus vector, which enables systematic variation of the expression level of each genetic component in a population-wide manner. Insights gained from this study led to the design of an optimal expression system, which can be delivered into mammalian cells by a single baculovirus vector to provide significantly improved UAA incorporation efficiency at a low virus load. Furthermore, this optimized baculovirus vector was shown to enable efficient UAA mutagenesis of proteins expressed in mouse brain tissue. Keywords: Mammalian genetic code expansion, unnatural amino acid incorporation, viral vector.
2 ACS Paragon Plus Environment
Page 2 of 15
Page 3 of 15
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
The ability to install novel unnatural amino acids (UAAs) into chosen sites in the proteome of a mammalian cell provides powerful new ways to probe and engineer its biology.1, 2 This can be achieved using the genetic code expansion approach, where an engineered orthogonal (i.e., does not cross-react with its endogenous counterparts) tRNA/aminoacyl-tRNA synthetase (aaRS) pair is used to specifically charge the desired UAA in response to a nonsense or frameshift codon. Co-expression of the nonsense mutant (typically TAG) of the target, the UAA-specific aaRS and its cognate nonsense-suppressor tRNA in mammalian cells has been shown to enable the expression of the target protein site-specifically incorporating UAAs.3-7 Various RNA polymerase II (pol II) recruiting promoters can be used to drive the expression of the aaRS and the target protein, while the tRNA-expression requires the use of RNA polymerase III (pol III) recruiting promoters.3-7 So far, the delivery of the genetic components necessary to express UAA modified proteins in mammalian cells has primarily relied upon transient transfection. Development of cell-lines stably expressing these components were also reported in recent years using either random genomic integration,8 or piggyBac mediated transposition of transfected DNA.9 However, clonal selection of these stable cells with the desired properties can be time consuming and technically demanding, and their development still relies upon transient transfection to deliver the DNA cargo to be integrated into the genome. The ability to deliver the UAA mutagenesis machinery into mammalian cells using a viral vector with broad host tropism offers a facile route to expand the scope of this technology. Attempts at generating a lentivirus vector for this purpose have been made, but its efficiency was compromised due to a limited cargo capacity of the vector, as well as the recombination-prone nature of its genome that precludes the use of more than one copy of tRNA per genome, which is insufficient to achieve high efficiency.10 A baculovirus vector was recently developed for efficient UAA mutagenesis in mammalian cells (Fig. S1).7 Incorporation of a mammalian virus coat protein (vesicular stomatitis virus G protein; VSVG) into its capsid enabled this engineered baculovirus to transduce a broad variety of mammalian cells with high efficiency, while its ultra-large cargo capacity (>30 kb) and high genome-stability allowed stable accommodation of the entire multicomponent genetic machinery into a single viral genome. In the absence of a better strategy, the expression system encoded in this vector (pAcBac2; Figure 1A) was designed to express tRNA, aaRS and the reporter protein using some of the strongest promoters available. Even though pAcBac2 was capable of eliciting strong, population-wide expression of UAA-modified reporter in mammalian cells, the use of a high virus concentration was essential (>400 virus per cell),7 alluding to the suboptimal nature of this expression system design. An improved viral vector that enables efficient UAA-mutagenesis at a low virus concentration can extend the application of this powerful technology to challenging systems, such as mammalian tissues and model animals, but would require further optimization of the expression system encoded in its genome. Previous efforts to optimize the UAA mutagenesis machinery have involved the construction of a few different expression systems, varying in their promoter usage and tRNA copy number, and evaluating the associated UAA incorporation efficiency using transient transfection.5, 6 Improved performance in these reports were primarily associated with constructs that express the suppressor tRNA at high levels, using a combination of optimal pol III promoters and increased tRNA copy number, indicating tRNA expression could be a limiting factor.5, 6 However, comparing the performance of a small number of distinct constructs fails to show how systematically altering the expression levels of the tRNA, aaRS and the mutant target in a population wide manner influence the overall performance of this genetic machinery – 3 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
insights that are critical for rationally redesigning it for improved efficiency. Another concern arises from the use of transient transfection in such optimization experiments. Previous studies have shown that standard transient-transfection conditions result in a large intracellular plasmid accumulation (>104 per cell; >103 per nucleus), and that a subpopulation of transfected cells exhibits unusually high plasmid accumulation, as well as transgene expression.11-13 Such high plasmid accumulation was also shown to result in the expression of the encoded transgene in near-saturated levels.12, 14 Decreasing input plasmid levels led to non-linear reductions in nuclear accumulation,11-13 but it also sharply decreased the number of cells successfully expressing the transgene,13 suggesting high plasmid accumulation is necessary for successful transgene expression when using transient transfection. The inability to control the number of plasmids delivered per cell and the expression levels of the encoded genes renders the use of transient transfection suboptimal for our optimization experiments. Furthermore, high plasmid accumulation and associated overexpression of all the encoded transgenes in successfully transfected cells may mask subtle intrinsic differences between different expression system designs. Results and Discussion We delivered a wild-type EGFP gene into the HEK293T cells using either polyethyleneimine (PEI) assisted transient transfection,15 or our baculovirus vector to characterize the associated expression pattern (Figure S2). Adding increasing quantities of preformed DNA:PEI complex was associated with a nonlinear increase in EGFP fluorescence (Figure S2A). In contrast, addition of increasing numbers of baculovirus per cell resulted in linear increase in EGFP expression over a large range of virus-to-cell ratios. We identified a condition for each delivery strategy that leads to similar levels of total EGFP expression, and analyzed the associated expression pattern using fluorescence microscopy, as well as fluorescence-activated cell sorting (FACS). Fluorescence imaging revealed a small fraction of highly fluorescent cells among the transfected population (Figure S2B, right), consistent with previous observations.11-13 FACS analysis further confirmed that the majority of the transfected cells express little to no EGFP (Figure S2D; majority of the population are either overlapping with, or marginally more fluorescent than the negative control, Figure S2C), while a small subpopulation exhibits very high fluorescence levels. In contrast, fluorescence imaging of the baculovirus infected cells shows a more evenly distributed fluorescence across the population (Figure S2B, left). FACS analysis further reveals that the majority of this population exhibit moderate to high fluorescence (Figure S2E). Addition of as few as 50 baculovirus particles per cell led to EGFP expression in over 88% of the cells (Figure S2F), and the expression level increased linearly with increasing virus concentration (Figure S2A), suggesting that the number of genomes delivered per cell and the associated transgene expression can be controlled significantly better using viral delivery, relative to transient transfection. Even though there was significant heterogeneity of reporter expression across the population, as may be expected due to intrinsic cell-to-cell variability, as well as varying numbers of viruses invading different cells, we anticipated that the ability of this vector to achieve population-wide gene expression in a tunable manner – simply by modulating the amount of the virus added – can be utilized to investigate how the interplay of the three necessary genetic components influence the efficiency of UAA incorporation in mammalian cells. We started with the pAcBac2 vector encoding an EGFP reporter harboring a TAG nonsense codon (EGFP-39-TAG; EGFP*) at a permissive site, as well as an E. coli tyrosyl-tRNA synthetase/tRNA derived suppression system, which allows the use of a variety of UAAs due to its relaxed substrate specificity (Figure 1A).7 Since viral delivery would allow tuning the population-wide expression of the 4 ACS Paragon Plus Environment
Page 4 of 15
Page 5 of 15
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
encoded components just by altering the number of virus particles added, unlike transient transfection, the performance of an expression system can be easily evaluated at varying expression levels of the associated machinery. When HEK293T cells were infected with this virus at different multiplicities of infection (MOI; or number of infective virus per cell) in the presence of 1 mM O-methyltyrosine (OMeY, Figure 4F), robust reporter expression was only observed beyond a MOI of 300 (Figure 1B). In contrast, when a pAcBac1 virus was used to deliver a wild-type EGFP reporter (pAcBac1.EGFP; Figure S3), EGFP-expression increased linearly with the increasing MOI from 25 to 200 (Figure 1B). The latter observation suggests that the lack of reporter expression from pAcBac2 at lower MOI is not a result of inefficient delivery by the virus, but likely due to the suboptimal nature of our first-generation expression system design. To determine if insufficient expression of one or more genetic components contributes to the poor reporter expression from pAcBac2 at low MOI, we individually encoded the tRNA, aaRS and the EGFP* genes into three separate baculovirus vectors (Figure S3), and used them to separately supplement a low MOI (100) pAcBac2 infection of HEK293T cells. While pAcBac2 alone was unable to express the EGFP* reporter at this low MOI, supplementing with additional pAcBac1.tRNA virus led to a dosedependent increase in full-length EGFP expression (Figure 1C). Supplementing pAcBac2 with pAcBac1.aaRS had no effect on reporter expression, whereas the use of additional pAcBac1.EGFP* led to a weak increase (Figure 1C). These results identify tRNA as the component limiting the performance of the pAcBac2 expression system at low MOI. This is not surprising, given the tRNA is bacterial in origin and may not have optimal interaction with the eukaryotic translational machinery. Even at the highest MOI (400) of pAcBac1.tRNA virus added, rescue of EGFP* expression from pAcBac2 was incomplete, but the use of additional virus was precluded due to associated cytopathic effects on the cells. To further evaluate how changing the cellular abundance of the tRNA, aaRS and the reporter influence the efficiency of UAA incorporation, we took advantage of the three aforementioned singlecomponent baculovirus vectors (Figure S3). HEK293T cells were infected with a fixed MOI (100) of one of these three vectors, while the ratio of the two other vectors was systematically altered, keeping the total MOI constant at 400 (Figure 2A). When the pAcBac1.tRNA or pAcBac1.EGFP* virus was kept constant at a MOI of 100, the highest reporter expression was associated with the use of the lowest ratio of the pAcBac1.aaRS virus (MOI of ~30). These observations indicated that the aaRS is required at a significantly lower level than the other two components, and that the use of the strong CMV promoter to drive aaRS expression from pAcBac2 may not be ideal. In a related experiment, when the aaRS-only virus was kept at a constant MOI of 100, use of either the pAcBac1.tRNA or pAcBac1.EGFP* virus at a significant excess over the other led to poor reporter expression, indicating the need to express both components at high levels (Figure 2A). Taken together, these findings suggest two potential alterations that can lead to improved UAA incorporation efficiency relative to pAcBac2: 1) augmenting the expression level of the tRNA; 2) reduction in aaRS expression levels. We generated three variants of pAcBac2, by replacing CMV with three weaker promoters for aaRS expression: SV40, human-PGK (hPGK), and UbiC (Figure 2B).16 Virus produced from each vector was used to infect HEK293T cells at a fixed MOI of 200, and the expression of EGFP* was monitored in the presence of 1 mM OMeY. All three vectors exhibited enhanced reporter expression relative to pAcBac2, and the use of UbiC was associated with the highest improvement (Figure 2C). We introduced a C-terminal myc-tag into the aaRS to analyze its expression from pAcBac2 and pAcBac2-UbiC by western blot, which revealed that the enhanced EGFP* expression from pAcBac2UbiC is indeed associated with a substantial decrease in aaRS levels relative to pAcBac2 (Figure 2D). A 5 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
number of factors could contribute to the detrimental effect of aaRS-overexpression on UAAincorporation efficiency under tRNA-limiting conditions, such as a sequestration of the active tRNA population by the excess aaRS, or an unnecessary exhaustion of cellular resources due to aaRS overproduction. Since pAcBac2 employs two of the most efficient pol III promoters (U6 and H1)17 to express tRNA, incorporation of additional copies of the tRNA gene was the only way to further increase its production. Presence of such repeating sequence in a viral genome is typically unstable due to its propensity for recombination. However, the baculoviral genome naturally harbors multiple homologous repeat elements – believed to be early transcriptional hot-spots18 – indicating possible tolerance for additional tRNAexpression cassettes. We inserted 16 additional tandem copies of the tRNA-gene into pAcBac2-UbiC to generate a new vector pAcBac3 (Figure 3A). Packaging this vector in Sf9 cells led to successful production of baculovirus, at titers similar to pAcBac2. pAcBac3 introduces >15 kb of genetic cargo encoding 24 individual genes into a single virus genome, underscoring the unique advantages of our baculovirus delivery vector. When we compared the activities of pAcBac3 and pAcBac2 at various MOI, the former was capable of expressing EGFP* at a significantly lower virus concentration (Figure 3B-C). At MOI 100 and 200, pAcBac3 enables roughly 58 and 32 fold more reporter expression, respectively, relative to pAcBac2. A comparison of pAcBac2, pAcBac2-UbiC, and pAcBac3 revealed that the improvement associated with pAcBac3 arises from a combination of enhanced tRNA copy number and reduced aaRS expression (Figure S4). Expression level of EGFP* from pAcBac3 was comparable with wild type EGFP expression from pAcBac1.EGFP (no nonsense codon) at a MOI of 200 (Figure 3B). FACS analysis of the resulting cells in both cases illustrated similar distribution, where >98% pAcBac1.EGFP infected cells were fluorescent relative to >96% for pAcBac3 (Figure S5). MS analysis of EGFP-39-OMeY isolated from HEK293T cells using Ni-NTA affinity chrmoatography show a homogeneous species with the expected mass, confirming the incorporation of the UAA with high fidelity (Figure S6). Interestingly, when pAcBac2, pAcBac2-UbiC, and pAcBac3 vectors were transfected into HEK293T cells using PEI, instead of delivery mediated by baculovirus, comparable expression of EGFP* was observed from all three vectors (Figure S7). This further demonstrates how the lack of copy number control associated with transient transfection – where the transfected cells receive and express the encoded genes at high levels – can significantly blur the intrinsic differences in performance among different expression systems. The enhanced performance of pAcBac3, coupled with its ability to infect a broad variety of different mammalian cells7 suggests that it may be possible to use this optimized viral vector to apply the UAAmutagenesis technology to live mammalian tissues as well as model mammals. A previous report used in utero electrporation to deliver the UAA incorporation machinery into mouse brain tissue with some success.19 The ability to use viral delivery may provide a more efficient and facile approach. To determine if this is indeed feasible, we took advantage of organotypic ex vivo slice cultures of mouse brain which preserves the endogenous tissue architecture, while allowing convenient experimental access to these cells.20-22 Long-term ex vivo slice-culture represents a powerful model system to study the molecular and cellular processes underlying neuronal development, as well as neuropathological conditions such as Alzheimer’s disease.23 We first infected monolayer cultures of dissociated primary mouse brain tissue with our delivery vector in the presence or absence of 1 mM p-acetylphenylalanine (pAcF, Figure 4F). Such cultures are mainly composed of neurons and glial cells, where the abundance of the latter is significantly higher.24 We observed robust EGFP expression 48 hr post-infection in the 6 ACS Paragon Plus Environment
Page 6 of 15
Page 7 of 15
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
presence of the UAA in the media, but not in its absence (Figure 4A, 4C). Selective staining with a neuron-specific marker (MAP2) revealed that the reporter expression is predominantly associated with glial cells (Figure S8) with
