**2.1 Reducing bottlenecks: protease activity**

The reduction in growth rate is particularly problematic, not only does it contribute to segregational plasmid instability, but severe growth rate perturbations at the onset of induced foreign protein synthesis have been shown to inhibit further expression of the desired protein [8]. Therefore, high levels of foreign protein expression are often unsustainable. Moreover, increased protease activity upon induction and overexpression of foreign protein generally leads to increased proteolysis, as described elsewhere [9–11]. These protease activities with uncharacterized specificity can be considered detrimental to the stability of the recombinant protein. Inefficient cell metabolism during overexpression, as indicated by acetate secretion of host cells, also results in lower protein expression [12]. These cell responses can greatly diminish the genetically-focused efforts to maximize both the final yield and concentration of recombinant proteins by increasing gene expression. In attempting to overcome these hurdles, cell dynamics during induced expression of chloramphenicol acetyl-transferase (CAT) expression have been examined and mathematically modeled in [13], suggesting that induction with an optimized amount of inducer (IPTG) at the onset of stationary phase can avoid growth rate suppression and achieve high expression. However, stimulated protease activity can be still observed. Intracellular proteases of recombinant *E. coli* have been differentiated by proteolytic activity and molecular weight and further characterized during the time course of protein overexpression [14]. Enhanced protease activity can respond quickly to induction, quicker than even the accumulation of the recombinant protein itself. To elicit and identify the proteases, transcriptional profiles of *E. coli* under stress of overexpression have been be mapped [15, 16]. Molecular chaperones (*groEL, ibpA*), lysis gene *mltB* and other DNA damage/bacteriophage associated genes (*recA, alpA, uvrB*) are all observed to be up-regulated along with proteases like *degP* and *ftsH*. It is also reported that cytoplasmic overexpression results in increased activity and expression of an outer membrane protease

**169**

*Repurposing E. coli by Engineering Quorum Sensing and Redox Genetic Circuits*

regulons, depending on the applied stress and the desired effects.

**2.2 Reducing bottlenecks: transcription factors**

non-antisense-producing cultures [27].

**2.3 Reducing bottlenecks: perspectives**

OmpT [17]. With this understanding, "cell-conditioning" by adding dithiothreitol (DTT) to alter the levels of the aforementioned host cell proteins prior to product (e.g., CAT) overexpression is capable of placing the cell in a particularly productive state, the result being a doubling of product level [15]. Other methods such as RNA interference (RNAi), and more recent CRISPR technologies can be exploited to downregulate bottlenecks, such as proteases, while ensuring maximal expression of the desired genes. These methods can be targeted to specific genes or even entire

Levels of the global heat shock transcription factor, σ32, for example, have been shown to increase rapidly during stress, including the stress associated with heterologous protein overexpression [18–23]. Indeed, a variety of cellular stresses induce the σ32-mediated stress response, including both ethanol and heat shock [19–22, 24]. While σ32 accumulation could be mediated by control of transcription and translation, its accumulation following production of recombinant protein is mainly due to an altering of its otherwise chaperone-sequestered state [19, 25]. To facilitate protein expression in recombinant *E. coli*, many have posited that simultaneous downregulation of global regulators (such as σ32) could simultaneously reduce the level of negative bottlenecks, such as the σ32-activated proteases. Noting that σ32-mutation is lethal at elevated temperatures [25, 26], methods such as RNAi were shown to transiently downregulate the σ32 stress response *in vivo* and these proved to be immensely advantageous. That is, using plasmids constructed with an antisense fragment of the σ32 gene, an early study showed that this successfully downregulated the expression of σ32 during the production of organophosphorus hydrolase (OPH), resulting increase specific OPH activity by six-fold compared to

Indeed, there have been countless studies demonstrating techniques to enhance the production of protein over the past 40+ years since recombinant DNA technology was first introduced. Besides choosing the right amount and type of inducer, optimal fermentation conditions have been developed to alleviate the reduction of growth rate during overexpression and enhance yield. Increasing stability of the protein product can also overcome the increased protease activity, this in addition to downregulation of protease-specific regulators. On top of the examples described above, an excellent review by Makrides [28] and a more recent review by Rosano [29] have discussed the various niches within which one can dig deeper in order to achieve higher yield and activity of the desired recombinant protein product.

We note that the majority of these methodologies have targeted either cell-based genetic regulatory structures, the sequence space and alterations of the protein of interest, or the operating policies of the reactors used to cultivate the overproducing cells. These cells, in turn, have typically been monocultures of an optimized host. Rarely have methodologies appeared in which collectives of cells, either monocultures or controlled co-cultures or consortia, and the exogenous signaling thereof are used to produce products such as recombinant proteins. Particularly useful when the engineering of a particular host overburdens its natural regulatory circuitry, cell consortia or collectives provide an interesting alternative. Co-culture and small consortia concepts have recently emerged. Moreover, new methodologies for orthogonal stimulation of genetic circuits can minimize pleiotropic or off-target effects normally accompanying more common chemical inducers. In the sections

*DOI: http://dx.doi.org/10.5772/intechopen.81245*

*Repurposing E. coli by Engineering Quorum Sensing and Redox Genetic Circuits DOI: http://dx.doi.org/10.5772/intechopen.81245*

OmpT [17]. With this understanding, "cell-conditioning" by adding dithiothreitol (DTT) to alter the levels of the aforementioned host cell proteins prior to product (e.g., CAT) overexpression is capable of placing the cell in a particularly productive state, the result being a doubling of product level [15]. Other methods such as RNA interference (RNAi), and more recent CRISPR technologies can be exploited to downregulate bottlenecks, such as proteases, while ensuring maximal expression of the desired genes. These methods can be targeted to specific genes or even entire regulons, depending on the applied stress and the desired effects.

## **2.2 Reducing bottlenecks: transcription factors**

*Gene Expression and Control*

instance those of biosensors and bioelectric devices.

as well as the extracellular microenvironmental state in the vicinity of 'designer' production strains in order to program gene expression and behavior. These techniques incorporate the understanding of cell metabolism and the transcriptome, cell-cell communication (previously reviewed by [2, 3]), and biological redox reactions (previously reviewed by [4]). This chapter will mainly focus on recent advances in how actuation of genes is accomplished in *Escherichia coli* through methods that require only minimal genetic rewiring and the technologies developed on such platforms, for

**2. Optimizing protein expression: rational control of cell condition**

background pertaining to *E. coli* protein overexpression is presented.

The reduction in growth rate is particularly problematic, not only does it contribute to segregational plasmid instability, but severe growth rate perturbations at the onset of induced foreign protein synthesis have been shown to inhibit further expression of the desired protein [8]. Therefore, high levels of foreign protein expression are often unsustainable. Moreover, increased protease activity upon induction and overexpression of foreign protein generally leads to increased proteolysis, as described elsewhere [9–11]. These protease activities with uncharacterized specificity can be considered detrimental to the stability of the recombinant protein. Inefficient cell metabolism during overexpression, as indicated by acetate secretion of host cells, also results in lower protein expression [12]. These cell responses can greatly diminish the genetically-focused efforts to maximize both the final yield and concentration of recombinant proteins by increasing gene expression. In attempting to overcome these hurdles, cell dynamics during induced expression of chloramphenicol acetyl-transferase (CAT) expression have been examined and mathematically modeled in [13], suggesting that induction with an optimized amount of inducer (IPTG) at the onset of stationary phase can avoid growth rate suppression and achieve high expression. However, stimulated protease activity can be still observed. Intracellular proteases of recombinant *E. coli* have been differentiated by proteolytic activity and molecular weight and further characterized during the time course of protein overexpression [14]. Enhanced protease activity can respond quickly to induction, quicker than even the accumulation of the recombinant protein itself. To elicit and identify the proteases, transcriptional profiles of *E. coli* under stress of overexpression have been be mapped [15, 16]. Molecular chaperones (*groEL, ibpA*), lysis gene *mltB* and other DNA damage/bacteriophage associated genes (*recA, alpA, uvrB*) are all observed to be up-regulated along with proteases like *degP* and *ftsH*. It is also reported that cytoplasmic overexpression results in increased activity and expression of an outer membrane protease

**2.1 Reducing bottlenecks: protease activity**

There is no doubt that among the myriads of systems available for heterologous protein expression, the Gram-negative bacterium *Escherichia coli* remains one of the most popular owing to its relative simplicity, its inexpensive and fast high-density cultivation, its well-known genetics, and the large number of cloning vectors and mutant host strains that are commonly available. Though not every gene can be efficiently and fully expressed in this system, much progress has been made to improve the performance and versatility of this workhorse microbe. One of the most sought after outcomes is the overexpression of high quality target proteins, however difficulties such as stimulated protease activity and reduced growth rate, as pointed out decades ago, often arise accompanying overexpression [5–7]. In this section, a brief review of the general

**168**

Levels of the global heat shock transcription factor, σ32, for example, have been shown to increase rapidly during stress, including the stress associated with heterologous protein overexpression [18–23]. Indeed, a variety of cellular stresses induce the σ32-mediated stress response, including both ethanol and heat shock [19–22, 24]. While σ32 accumulation could be mediated by control of transcription and translation, its accumulation following production of recombinant protein is mainly due to an altering of its otherwise chaperone-sequestered state [19, 25]. To facilitate protein expression in recombinant *E. coli*, many have posited that simultaneous downregulation of global regulators (such as σ32) could simultaneously reduce the level of negative bottlenecks, such as the σ32-activated proteases. Noting that σ32-mutation is lethal at elevated temperatures [25, 26], methods such as RNAi were shown to transiently downregulate the σ32 stress response *in vivo* and these proved to be immensely advantageous. That is, using plasmids constructed with an antisense fragment of the σ32 gene, an early study showed that this successfully downregulated the expression of σ32 during the production of organophosphorus hydrolase (OPH), resulting increase specific OPH activity by six-fold compared to non-antisense-producing cultures [27].

#### **2.3 Reducing bottlenecks: perspectives**

Indeed, there have been countless studies demonstrating techniques to enhance the production of protein over the past 40+ years since recombinant DNA technology was first introduced. Besides choosing the right amount and type of inducer, optimal fermentation conditions have been developed to alleviate the reduction of growth rate during overexpression and enhance yield. Increasing stability of the protein product can also overcome the increased protease activity, this in addition to downregulation of protease-specific regulators. On top of the examples described above, an excellent review by Makrides [28] and a more recent review by Rosano [29] have discussed the various niches within which one can dig deeper in order to achieve higher yield and activity of the desired recombinant protein product.

We note that the majority of these methodologies have targeted either cell-based genetic regulatory structures, the sequence space and alterations of the protein of interest, or the operating policies of the reactors used to cultivate the overproducing cells. These cells, in turn, have typically been monocultures of an optimized host. Rarely have methodologies appeared in which collectives of cells, either monocultures or controlled co-cultures or consortia, and the exogenous signaling thereof are used to produce products such as recombinant proteins. Particularly useful when the engineering of a particular host overburdens its natural regulatory circuitry, cell consortia or collectives provide an interesting alternative. Co-culture and small consortia concepts have recently emerged. Moreover, new methodologies for orthogonal stimulation of genetic circuits can minimize pleiotropic or off-target effects normally accompanying more common chemical inducers. In the sections

that follow, we describe efforts to minimally alter the native bacterial signaling processes of quorum sensing and oxidative stress to repurpose *E. coli* for application in new platforms.
