18607314, 2013, 1, Downloaded from https://analyticalsciencejournals.onlinelibrary.wiley.com/doi/10.1002/biot.201200120 by University Of Texas Libraries, Wiley Online Library on [13/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
`
`Biotechnology
`Journal
`
`Review
`
`DOI 10.1002/biot.201200120
`
`Biotechnol. J. 2013, 8, 46–58
`
`Promoter engineering: Recent advances in controlling
`transcription at the most fundamental level
`
`John Blazeck1 and Hal S. Alper1,2
`
`1 Department of Chemical Engineering, University of Texas at Austin, Austin, TX, USA
`2 Institute for Cellular and Molecular Biology, University of Texas at Austin, Austin, TX, USA
`
`Synthetic control of gene expression is critical for metabolic engineering efforts. Specifically, pre-
`cise control of key pathway enzymes (heterologous or native) can help maximize product forma-
`tion. The fundamental level of transcriptional control takes place at promoter elements that drive
`gene expression. Endogenous promoters are limited in that they do not fully sample the complete
`continuum of transcriptional control, and do not maximize the transcription levels achievable
`within an organism. To address this issue, several attempts at promoter engineering have shown
`great promise both in expanding the cell-wide transcriptional capacity of an organism and in
` enabling tunable levels of gene expression. Thus, this review highlights the recent advances and
`approaches for altering gene expression control at the promoter level. Furthermore, we propose
`that recent advances in the understanding of transcription factors and their DNA-binding sites will
`enable rational and predictive control of gene expression.
`
`Keywords: Bioengineering · Gene expression · Metabolic engineering · Synthetic biology
`
`Received 03 MAY 2012
`Revised 25 JUN 2012
`Accepted 17 JUL 2012
`
`1 Introduction
`
`Intracellular metabolic flux is regulated by a series of dis-
`tinct, yet interwoven, levels of regulatory control – occur-
`ring at the transcriptional, translational, and protein lev-
`els [1]. One of the fundamental access points to alter this
`metabolic flux is to control transcript production at the
`promoter level. Hence, metabolic engineering applica-
`tions have long relied on effective promoter discovery and
`characterization. Specifically, a wide array of expression
`capacities is required since the optimal expression level
`is likely gene specific and can vary by several orders of
`magnitude. While identifying an existing promoter repre-
`sents a path forward, the field of promoter engineering at-
`tempts to modulate promoter transcriptional capacity by
`mutating, enhancing, or otherwise altering promoter
`
`Correspondence: Dr. Hal Alper, University of Texas at Austin, 200 East
`Dean Keeton Street, C0400, Austin, TX 78712, USA
`E-mail: halper@che.utexas.edu
`
`Abbreviations: bp, base pair; Ep-PCR, error-prone polymerase chain reac-
`tion; PIC, pre-initiation complex; TFBS, transcription factor-binding site;
`TSS, transcription start site; UAS, upstream activation sequence
`
`DNA sequence. In doing so, promoter engineering can
`help generate the dynamic range necessary to enable
`fine-tuned gene expression for metabolic engineering ap-
`plications.
`Simple promoter-gene cassettes have been an essen-
`tial component of the metabolic engineering paradigm
`since the field was first described [2], and since that time
`promoters have become focal points as enabling “parts”
`for synthetic biology applications [3]. Promoters can ei-
`ther be isolated from endogenous sequences within the
`host organism or isolated from a virus that infects the
`host. Viral or phage-derived heterologous promoters can
`generate unregulated levels of transcription that are too
`high for many of the fine-tuned controls required in meta-
`bolic engineering applications. Thus, a number of en-
`dogenous promoters are typically employed as a set to en-
`able a range of gene expression. Endogenous promoter
`isolation and utilization is limited by a variety of difficul-
`ties, e.g. (i) promoter isolation and characterization can be
`tedious and genetic-context specific, (ii) isolated promot-
`ers only sample the continuum of gene expression at a few
`
`Colour online: See the article online to view figs. 2 + 3 in colour.
`
`46
`
`© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
`
`LCY Biotechnology Holding, Inc.
`Ex. 1011
`Page 1 of 13
`
`

`

` 18607314, 2013, 1, Downloaded from https://analyticalsciencejournals.onlinelibrary.wiley.com/doi/10.1002/biot.201200120 by University Of Texas Libraries, Wiley Online Library on [13/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
`
`Biotechnol. J. 2013, 8, 46–58
`
`www.biotecvisions.com
`
`discrete points and may be plagued by disparate regula-
`tion patterns, and (iii) isolated endogenous promoters are
`unable to maximize the true transcriptional capacity at-
`tainable within the host. While high-strength heterolo-
`gous viral promoters alleviate this last concern in prokary-
`otes and metazoans, there is no such recourse in yeast.
`The field of promoter engineering has sought to overcome
`all of these above listed difficulties, and allow efficient op-
`timization of pathway flux for metabolic engineering uses.
`Other methods for fine-tuning gene and/or protein
` expression at the translational level have been successful
`at producing large ranges in gene expression, including
`design of synthetic ribosome-binding sites [4] and direct-
`ed RNase III cleavage of mRNA transcripts [5, 6], but we
`do not be discussed these here as they have been well-
`covered in other reviews.
`The purpose of this review is to provide an overview
`of promoter engineering techniques and recent success-
`es, especially in the area of designing promoters for meta-
`bolic engineering applications. This review focuses pre-
`dominantly on pertinent advances in the two most-com-
`monly utilized model organisms, Saccharomyces cere-
`visiae and Escherichia coli. First, we introduce promoter
`architecture and function and give examples of tradition-
`al promoter utilization in metabolic engineering. We then
`discuss recent advances in the field of promoter engi-
`neering and novel approaches to alter gene expression
`control at the promoter level. Finally, we highlight recent
`advances in the understanding of transcription factors
`and their DNA-binding sites, and propose how this
`
`knowledge sets the stage for de novo promoter design
`that could enable tunable gene expression in any system.
`
`2 Overview of promoter structure and
`function
`A promoter can be defined as any sequence of DNA that
`can independently facilitate the binding of transcription
`factors and enable transcription initiation. Such interac-
`tions between promoter DNA and transcription factors
`aid in the recruitment of the cellular machinery necessary
`for transcription of an open reading frame [7–9]. Consen-
`sus E. coli promoter structure includes a –35 “TTGACA”
`motif and a –10 “TATAAT” motif, relative to a +1 tran-
`scription start site (TSS) (Fig. 1A). These motifs are sepa-
`rated and surrounded by nucleotide spacer regions in
`which little or no nucleotide homology has been deduced;
`however, the nucleotide spacer region between the –35
`and –10 motifs has a consensus length of 17 base pairs
`(bp) [10, 11]. In prokaryotes, the σ factor of the RNA poly-
`merase is sufficient for promoter recognition and tran-
`scription initiation [12]. The α subunit of RNA polymerase
`can also recognize UP element DNA, a very rare A+T-rich
`upstream region of promoter consensus regions that can
`increase basal promoter transcription 1.5- to 90-fold [13]
`(Fig. 1A).
`Eukaryotic transcription initiation is far more com-
`plex, requiring DNA sequence-specific transcription fac-
`tors to bind within a promoter element and interact with
`transcriptional coactivators that help localize the basic
`
`Figure 1. Consensus promoter motifs and architecture for prokaryotic, yeast, and metazoan promoters. (A) A diagram of the rrnD P1 E. coli promoter
`modified from [13], illustrating prokaryotic promoter structure, including the very rare UP element. ‘Consensus’ –35 “TTGACA” and –10 “TATAAT” motifs
`are present and separated by a 17-bp nucleotide spacer region. (B) A diagram of the ‘consensus’ metazoan promoter, modified from [20], illustrating po-
`tential conserved elements including the TATA box, Inr (initiator), BREs (transcription factor for RNA polymerase IIB recognition element), MTE (motif ten
`element), DCE (downstream core element), DPE (downstream core promoter element), and XCPE1 (X core promoter element 1). (C) Understanding of
`yeast promoter structure is lacking for defined consensus motifs, although modularity has been demonstrated between the enhancer and core elements.
`
`© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
`
`47
`
`LCY Biotechnology Holding, Inc.
`Ex. 1011
`Page 2 of 13
`
`

`

` 18607314, 2013, 1, Downloaded from https://analyticalsciencejournals.onlinelibrary.wiley.com/doi/10.1002/biot.201200120 by University Of Texas Libraries, Wiley Online Library on [13/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
`
`Biotechnology
`Journal
`
`www.biotechnology-journal.com
`
`transcriptional machinery [14]. For the most common
`form of RNA polymerase (Pol II), up to 30 protein-based
` elements comprising the five general transcription factors
`are required to assemble RNA Polymerase II into the pre-
`initiation complex (PIC) at the core promoter element [15,
`16]. Transcription initiation is then triggered as promoter
`DNA is unwound around the PIC. The PIC undergoes a
`conformation change into the open complex [17], scans
`for a suitable TSS, and initiates active elongation [15, 18].
`Routinely, eukaryotic promoters are thought to con-
`tain two distinct regions: (i) a core element and (ii) an up-
`stream enhancer element. Transcriptional direction and
`start site are determined by the core promoter element,
`and the upstream enhancer element helps determine
`transcriptional frequency, or promoter strength. The core
`promoter is the minimal promoter region required to initi-
`ate transcription [14–16], and is typically a succinct
`stretch of less than 80 nucleotides, extending roughly
`35 bp upstream or downstream from the +1 site [16, 19].
`Metazoan core promoter structure has been recently
` reviewed [19, 20] and may contain a variety of distinct
`DNA motifs that modulate core promoter activity
`(Fig. 1B). Metazoan core elements are typically not con-
`served in S. cerevisiae or E. coli. In S. cerevisiae, only
`~20% of genes contain a TATA box, and little is known
`about other potential consensus motifs (Fig. 1C) [21]. In
`TATA-box-containing promoters in S. cerevisiae, RNA
`polymerase II initiates transcription at a site 45–120
` nucleotides downstream from the TATA element [22].
`Zhang et al. [23] aligned the flanking sequences of 4637
`TSSs to identify the consensus A(Arich)5NPyA(A/T) NN
`(Arich)6 pattern, confirming and expanding the previously
`reported PyA(A/T)Pu sequence [22, 24]. Yeast promoter
`structure, promoter function, and transcriptional regula-
`tion have also recently been thoroughly reviewed [14]. As
`the core promoter’s function is to enable basal transcrip-
`tion, a suitable TSS and the capacity to recruit PIC com-
`ponents are equally essential.
`Eukaryotic upstream enhancer elements (similar to
`prokaryotic UP elements) localize trans-acting regulatory
`elements (transcription factors) as a means of controlling
`transcriptional frequency or imparting regulation to a core
`promoter. Within the enhancer element, concise and spe-
`cific DNA sequences serve as transcription factor-bind-
`ing sites (TFBSs) or “docking points” for transcriptional
`activators or repressors [25–30]. Promoter-bound tran-
`scription factors interact with one another locally and
`with the basal transcriptional machinery to establish pro-
`moter regulation and promoter strength, or frequency of
`PIC formation and subsequent transcription initiation [9,
`31, 32]. DNA regions prone to increase transcriptional
` frequency of a core promoter are commonly referred to as
`upstream activation sequences (UAS). Similarly, an
` upstream repressive sequence localizes transcription fac-
`tors that reduce transcription rate [14]. Promoter regula-
`tion (induction or repression dependent on varying con-
`
`Biotechnol. J. 2013, 8, 46–58
`
`ditions) is also a result of transcription factor-mediated
` interactions in the enhancer element [33–36]. Promoter
`engineering techniques alter both core and enhancer ele-
`ments to modulate overall promoter expression capacity.
`
`3 Examples of promoter selection
`in metabolic engineering applications
`
`Promoter selection is often a key component of the meta-
`bolic engineering design cycle. Often, E. coli is selected
`as a host for the overexpression of heterologous proteins.
`In such applications, high strength and tightly controlled
`promoters are generally required to maximize protein pro-
`duction and reduce toxicity during growth phase [37].
`The end result is that only a few promoters are used for
`protein production despite the hundreds of E. coli pro-
`moter sequences that have been elucidated [38]. Among
`these select promoters are two very high strength phage-
`derived promoter systems based on the T7 RNA poly-
`merase and the PL temperature-regulated phage promot-
`er systems [37, 39–41]. The abnormally high transcrip-
`tional capacity of these systems creates an excessive
`metabolic load on the E. coli host that decreases product
`formation in other metabolic engineering applications
`[42]. Hence, several lower strength, but still strong pro-
`moters, such as the lac, tac, trc, PBAD, or rhaPBAD promot-
`er systems are more commonly utilized to maximize prod-
`uct formation [43–46]. Blending gene overexpressions
`with these promoters into E. coli’s natural metabolism
` enabled the production of 1,4-butanediol [47], 1-butanol
`[48], isobutanol and other branched-chain higher alcohols
`[49], and polylactic acid and its copolymers [50]. Synthet-
`ic promoter regulation has been imparted to E. coli pro-
`moters utilizing the tetA promoter/operator and tetR
` repressor system [34, 35, 51]. Reviews of these and other
`advances have been recently published [52–54].
`In yeast, strong endogenous constitutive promoters
`(including PTEF [55], PHXT7 [56], and PGPD [57, 58]) or galac-
`tose-inducible promoters [33, 59] are typically employed
`for metabolic engineering purposes [60, 61]. Similar to the
`case of E. coli discussed above, these promoters have en-
`abled metabolic engineering successes in yeast. Consti-
`tutive overexpression of the pentose-phosphate pathway
`enzymes transketolase and transaldolase enabled yeast
`fermentation of xylose [62]. In separate studies, ethanol
`yield from xylose was further increased by modulating
`overexpression
`level of these and another enzyme
`through multiple-gene-promoter shuffling [63], and the
`bacterial L-arabinose degradation pathway was over -
`expressed
`to enable arabinose
`fermentation
`[64].
` Inducible promoters offer a complementary method for
` recombinant protein expression in yeast and, as such, the
`GAL promoters have been widely employed in pathway
`engineering applications including the production of
`arteminisic acid [65], isoprenoids [66, 67], and n-butanol
`
`48
`
`© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
`
`LCY Biotechnology Holding, Inc.
`Ex. 1011
`Page 3 of 13
`
`

`

` 18607314, 2013, 1, Downloaded from https://analyticalsciencejournals.onlinelibrary.wiley.com/doi/10.1002/biot.201200120 by University Of Texas Libraries, Wiley Online Library on [13/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
`
`Biotechnol. J. 2013, 8, 46–58
`
`[68], and various other metabolites and products [61].
`Thorough reviews of promoter utilization for metabolic
`engineering purposes have been recently published
`[60, 61].
`
`4 Promoter engineering strategies
`
`As described above, strong overexpression is not always
`optimal for a given gene, thus a range of promoter
`strengths is necessary. Promoter engineering is becom-
`ing an enabling technology to facilitate this optimization
`and provide more synthetic promoter elements. The basis
`for much of this engineering is related to promoter archi-
`tecture. As examples, prokaryotic promoters contain two
`essential motifs surrounded by variable spacer regions of
`DNA. Thus, introducing variation into these spacer re-
`gions can modulate expression potential in these promot-
`ers. Visualizing eukaryotic promoters as a fusion of core
`and enhancer facilitates engineering hybrid promoters, in
`which modular enhancer and core combinations can
` determine promoter regulation and transcriptional capac-
`ity. At the more basic level, both enhancer and core ele-
`ments possess specific TFBSs that determine overall pro-
`moter
`function. Randomized promoter mutagenesis
`through error-prone PCR (Ep-PCR) alters TFBSs, thus
` altering promoter strength. Since binding site mutations
`are far more likely to reduce transcription factor inter -
`actions, these random mutagenesis approaches often
`produce promoter variants with lower strengths than the
`template sequence. In this section, we discuss each of
`these promoter engineering strategies in detail and give
`examples of their effectiveness and utility. Instances of
`each of these strategies are summarized in Table 1.
`
`4.1 Ep-PCR
`
`Ep-PCR introduces random mutations into a DNA
` sequence of choice (Fig. 2A). When applied to an entire
`promoter region, mutations occur throughout the consen-
`sus and spacer regions, and lead to disparate function.
`This approach is guaranteed to yield novel promoter vari-
`ants (with sufficient library sizes), and proper selection
`techniques allow the isolation of promoters driving a wide
`variation in gene expression. For example, mutagenesis
`of the bacteriophage-derived PL-λ constitutive E. coli pro-
`moter yielded a library of engineered promoters of varying
`strengths spanning a 196-fold range with identical regu-
`lation [69]. Individual promoters displayed uniform ex-
`pression on the single-cell level, and library application
`enabled identification of optimal expression levels of
`phosphoenolpyruvate carboxylase (ppc) and deoxy-xylu-
`lose-P synthase (dxs) to maximize the desired growth and
`lycopene-production phenotypes, respectively. Moreover,
`optimal expression levels were revealed to be dependent
`on strain genetic background, thus demonstrating the ne-
`
`www.biotecvisions.com
`
`cessity of promoter libraries to sample ranges of expres-
`sion [69]. Ep-PCR of the strong constitutive S. cerevisiae
`TEF1 promoter generated a similarly diversified promoter
`library spanning a 15-fold range [69, 70]. Utilizing these
`promoters in knock-in promoter replacement cassettes
`revealed a linear relationship between glycerol-3-phos-
`phate dehydrogenase (GPD1) expression on glycerol
`yield, which saturated at the higher activity of the mutant
`TEF1 series [70]. Additionally, a recent application of the
`TEF1 promoter series enabled a graded dominant mutant
`approach that provided novel insight into the catalytic
`function of global yeast GCN5p [71]. Finally, random mu-
`tagenesis of the oxygen-responsive S. cerevisiae DAN1
`promoter enabled the isolation of two mutants induced
`under less-stringent anaerobiosis than the wild-type pro-
`moter, enabling induction of gene expression simply by
`oxygen depletion during cell growth [72].
`Ep-PCR of an endogenous constitutive promoter is
`likely to decrease promoter activity by mutating TFBSs
`and reducing transcription factor-promoter affinity [69,
`70, 73]. In this manner, it is possible to discover and char-
`acterize TFBSs through random Ep-PCR-mediated TFBS
`[74]. Promoter mutagenesis and characterization allowed
`identification of functionally important mutations with
`the PL-λ promoter in E. coli and in sugar cane cells [73, 74].
`Through the same mechanism, Ep-PCR of a highly regu-
`lated promoter is likely to temper strict regulation by de-
`creasing promoter-transcription factor affinity [72]. Inter-
`estingly, successive mutation of a randomly chosen inac-
`tive eukaryotic DNA sequence from the HeLa genome
`through four rounds of Ep-PCR generated a strong E. coli
`promoter [75], indicating that it is relatively easy to gen-
`erate or improve a prokaryotic promoter. Likely, this is due
`to the simplicity and conservation of prokaryotic promot-
`er function, allowing highly selective targeting for pro-
`moter function.
`
`4.2 Saturation mutagenesis of nucleotide spacer
`regions
`
`More directed promoter engineering efforts focus on
` retaining consensus regions of the promoter structure
`while mutating only variable regions. Prokaryotic promot-
`ers contain consensus –35 and –10 motifs separated and
`surrounded by variable nucleotide spacer regions [10]
`(Fig. 1A). Hence, the saturation mutagenesis of these nu-
`cleotide spacer regions (while keeping consensus motifs
`intact) represents a somewhat rational methodology to
`modify prokaryotic promoter strength (Fig. 2B). Jensen et
`al. [76] demonstrated that saturation mutagenesis of a
`Lactococcus lactis promoter drastically modulates ex-
`pression, generating a synthetic promoter library span-
`ning a 400-fold range in expression. Mutations within the
`consensus motifs or alterations of spacer length greatly
`reduced promoter function, complimenting and increas-
`ing library coverage to a range of three to four logs [76].
`
`© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
`
`49
`
`LCY Biotechnology Holding, Inc.
`Ex. 1011
`Page 4 of 13
`
`

`

` 18607314, 2013, 1, Downloaded from https://analyticalsciencejournals.onlinelibrary.wiley.com/doi/10.1002/biot.201200120 by University Of Texas Libraries, Wiley Online Library on [13/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
`
`Biotechnology
`Journal
`
`www.biotechnology-journal.com
`
`Biotechnol. J. 2013, 8, 46–58
`
`[105]
`
`[104]
`
`[102]
`
`[100]
`
`[100]
`
`[99]
`
`[97]
`
`[87]
`
`[83]
`
`[82]
`
`[79]
`
`[78]
`
`[76]
`
`[70]
`
`[69]
`
`Decreasedb)
`
`N/A
`
`Increased
`
`N/A
`
`Increased
`
`Increased
`
`Increased
`
`Decreased
`
`N/A
`
`Decreased
`
`Decreased
`
`N/A
`
`N/A
`
`Decreased
`
`Decreased
`
`Yes
`
`No
`
`No
`
`Yes
`
`Yes
`
`Yes
`
`No
`
`Yes
`
`Yes
`
`Yes
`
`Yes
`
`Yes
`
`Yes
`
`Yes
`
`Yes
`
`promoter strength
`increased/decreased
`
`Reference
`
`Predominantly
`
`range?
`sampled
`
`Fully
`
`28
`
`~40
`
`975
`
`50
`
`90
`
`400
`
`32
`
`9
`
`5286
`
`10
`
`160
`
`349
`
`6833
`
`15
`
`196
`
`(fold)
`range
`
`Expression
`
`Promoter or biological parts utilized
`
`Table 1. Improvement/diversification of gene expression afforded by the examples discusseda)
`
`50
`
`promoters PLEUM, PGPD, PTEF, PCYC, and PCYC158
`S. cerevisiae UAS elementsUASCLB, UASCIT, and UASTEF, and core
`TEFcore promoter series
`Y. lipolytica UAS1B enhancer, LEU2minimal core promoter,
`Y. lipolytica UAS1B enhancer and LEU2minimal core promoter
`26 conserved and 48 random nucleotides
`S. cerevisiaePFY1pcore promoter – Reb1p TFBS, poly-dT element,
`and 83 random nucleotides
`S. cerevisiaecore promoter – two CT boxes, twp RPG boxes, one TATA box,
`
`Mammalian ‘Jet’ promoter – 61 conserved and 69 random nucleotides
`
`(W, R and D) and 20 random nucleotides
`Lb. plantarum‘consensus’ promoter – 16 conserved, 3 semi-conserved
`and 20 random nucleotides
`E. coli‘consensus’ promoter – 24 conserved, 13 semi-conserved (W, R and D)
`and 22 random nucleotides (Fig. 1A)
`L. lactis‘consensus’ promoter – 31 conserved, 2 semi-conserved (W or R)
`S. cerevisiae TEF1promoter
`Phage PL-λ
`
`Hybrid promoter engineering
`
`Hybrid promoter engineering
`
`Hybrid promoter engineering
`
`spacer regions
`Saturation mutagenesis of
`
`spacer regions
`Saturation mutagenesis of
`
`spacer regions
`Saturation mutagenesis of
`
`spacer regions
`Saturation mutagenesis of
`
`spacer regions
`Saturation mutagenesis of
`
`spacer regions
`Saturation mutagenesis of
`
`Ep-PCR
`
`Ep-PCR
`
`strategy
`Promoter engineering
`
`a)Several initial promoters employed for saturation mutagenesis of spacer regions were composite ‘consensus’ promoters derived from available literature and sequence data. Thus, there is no reference promoter to de-
`
`b)Several modified AOX1promoters displayed increased activity compared to wild type, but the majority did not.
`
`termine if mutant promoters had increased or decreased expression capacities. N/A, not available.
`
`Pichia pastoriaAOX1promoter
`CArG-binding factor A (CBF-A), and a TATA Box
`Mammalian TFBS for activating protein-1 (AP-1), nuclear factor κB (NF-κB),
`SV40, and FIX core promoters
`
`TFBS modification
`
`TFBS modification
`
`Hybrid promoter engineeringMammalian apoE and ABP enhancers and the ADH6, hAAT, CYP,
`
`TFBS modification
`Hybrid promoter engineering/S. cerevisiae Gal4p TFBSs and PLEUMcore promoter
`
`© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
`
`LCY Biotechnology Holding, Inc.
`Ex. 1011
`Page 5 of 13
`
`

`

` 18607314, 2013, 1, Downloaded from https://analyticalsciencejournals.onlinelibrary.wiley.com/doi/10.1002/biot.201200120 by University Of Texas Libraries, Wiley Online Library on [13/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
`
`Biotechnol. J. 2013, 8, 46–58
`
`www.biotecvisions.com
`
`Figure 2. Overview of promoter engineering methods. (A) Ep-PCR introduces random mutations (depicted as red stars) into a wild-type promoter element
`that alter promoter sequence, modified from [69]. Large-scale characterization of mutated promoters facilitates isolation of a promoter library that retains
`endogenous regulation but spans large expression ranges. (B) Saturation mutagenesis of nucleotide spacer regions diversifies non-consensus nucleotides
`within a promoter to enable wide ranges in promoter library strength. A schematic of the saturation mutagenesis of three nucleotide spacer regions within
`four conserver regions of a consensus L. lactis promoter has been modified from [76]. Underlined ‘G’ denotes the +1 TSS. (C) Hybrid promoter engineering
`utilizes tandem upstream activations sequences to modulate core promoter expression to construct synthetic hybrid promoters with novel strength or reg-
`ulation, modified from [100]. (D) Directed introduction, deletion, or modification of TFBSs rationally alters promoter strength of regulation. Addition of one
`to three distinct Gal4p TFBSs to a constitutive core promoter enabled tunable galactose-induction over a 50-fold range, as shown in [100].
`
`© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
`
`51
`
`LCY Biotechnology Holding, Inc.
`Ex. 1011
`Page 6 of 13
`
`

`

` 18607314, 2013, 1, Downloaded from https://analyticalsciencejournals.onlinelibrary.wiley.com/doi/10.1002/biot.201200120 by University Of Texas Libraries, Wiley Online Library on [13/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
`
`Biotechnology
`Journal
`
`www.biotechnology-journal.com
`
`Application of this synthetic library successfully modulat-
`ed individual and multi-gene operon expression levels
`[77], and utilizing the promoter library in E. coli revealed
`that promoter strength is not conserved between prokary-
`otic hosts [76]. Interestingly, the drastic decrease in pro-
`moter activity due to mutation of the consensus spacer
`length was not observed in E. coli [76], a finding con-
`firmed in a separate study [78]. Accordingly, elongation of
`the 16-bp spacer region of the E. coli hybrid tac promoter
`by one or two bp (creating the trc and tic promoters, re-
`spectively) decreased in vivo activity only slightly; the trc
`promoter exhibited 90% tac in vivo activity, and the tic,
`65% [44]. Application of spacer region mutagenesis to
`E. coli and Lactobacillus (Lb.) plantarum successfully cre-
`ated high-coverage synthetic promoter libraries [78, 79].
`Utilization of the Lb. plantarum promoter library enabled
`production of PepN protein at approximately 10–15% of
`total cellular protein, and the promoters possessed the
`same activity level in another Lactobacillus species [79].
`These examples demonstrate that semi-rational promoter
`diversification via saturation mutagenesis repeatedly pro-
`duces robust synthetic promoter libraries in prokaryotes
`[80, 81]. Such libraries represent ideal tools for fully exam-
`ining how gene expression levels alter intracellular flux
`and product accumulation.
`Saturation mutagenesis of nucleotide spacer regions
`has also been successful applied to metazoan and yeast
`promoter architectures for synthetic library construction
`[82, 83]. Mutagenesis of the hybrid Jet promoter yielded a
`mammalian promoter library with a 10-fold range, in
`which all mutants displayed reduced activity compared
`to the original Jet hybrid [82]. As S. cerevisiae lacks a
`strict consensus structure, Jeppsson et al. [83] pieced to-
`gether a hybrid promoter containing two Gcr1p TFBSs,
`two Rap1p TFBSs, and a TATA box, separated by degen-
`erate nucleotide spacer regions following spacial guide-
`lines elucidated in other promoter studies [84–86]. Varia-
`tions in the spacer regions allowed isolation of 37 syn-
`thetic promoters covering a range of three orders of mag-
`nitude. Utilizing this synthetic promoter
`library to
`down-regulate native ZWF1 expression increased ethanol
`production by 16% and decreased xylitol accumulation by
`55% in yeast xylose fermentation [83]. Spacer region di-
`versification also enabled construction of a yeast synthet-
`ic promoter library based on the PFY1 promoter scaffold
`[87]. Specifically, introduction of tetR regulation and bind-
`ing sites for customized transcription activator-like effec-
`tors [88] enabled orthogonal regulation of this promoter for
`synthetic biology purposes [87]. As a final application in
`yeast, spacer-region diversification, combined with tet-
`based and UASGAL-based [36] control, enable model-guid-
`ed construction of a synthetic gene network that con-
`trolled yeast-sedimentation timing [89].
`As demonstrated in this section, directed diversifica-
`tion of non-consensus promoter sequences has enabled
`promoter library constructions across species, most often
`
`Biotechnol. J. 2013, 8, 46–58
`
`in prokaryotes. These libraries benefit from very large ex-
`pression ranges and represent an enabling technology for
`pathway
`flux optimization. Saturation mutagenesis
` efforts have often utilized synthetic, composite promoter
`scaffolds that are stitched together using motifs and bind-
`ing sites from disparate promoters. As a result, establish-
`ing a base line is difficult as there is no reference promot-
`er to determine whether the diversified promoter libraries
`tend to reduce or increase expression capacity.
`
`4.3 Hybrid promoter engineering
`
`A hybrid promoter engineering approach entails the
` assembly of enhancer element–core promoter fusions to
`rationally enhance basal core transcriptional capacity or
`enable novel promoter regulation (Fig. 2C). Basic hybrid
`promoter work in E. coli led to the formation of many com-
`monly utilized promoters, including the tac promoter (a
`fusion derived from the trp and lac promoters) [43], and
`the rhaPBAD, a tightly regulated arabinose and rhamnose
`promoter fusion [46]. Traditionally, hybrid promoters have
`been utilized to dissect promoter function and regulation
`in S. cerevisiae [43, 90–95]. In this light, essential DNA
` sequences are identified in part due to the modularity of
`hybrid promoter core and enhancer elements. Upstream
`enhancer elements contain TFBSs that enable native reg-
`ulation and expression activation or repression to be
`maintained independently of core promoter region. Mini-
`mal regions of these DNA regions can identify specific
` regions essential for transcriptional activation or induc-
`tion control (i.e. upstream activation sequences). Con-
`structing hybrid promoters composed of tandem repeat-
`ing UAS elements can radically increase core promoter
`expression capacity, as each additional UAS increases
`overall hybrid promoter strength. Thus, hybrid promoter
`engineering present the dual advantage of (i) generating
`large-coverage promoter libraries, and (ii) enhancing the
`transcriptional capacity of even the strongest endoge-
`nous core promoters.
`The oleaginous yeast, Yarrowia lipolytica, has serve