Replacing Standard Reporters from Molecular Cloning Plasmids with Chromoproteins for Positive Clone Selection

Cloning and expression plasmids are the workhorses of modern molecular biology. Despite the pathway paved by synthetic biology, laboratories around the globe still relay on standard cloning techniques using plasmids with reporter proteins for positive clone selection, such as β-galactosidase alpha peptide complementation for blue/white screening or ccdB, which encodes for a toxic DNA gyrase. These reporters, when interrupted, serve as a positive clone detection system. In the present report, we show that molecular cloning plasmids bearing the coding sequence for a 25.4 kDa protein, AmilCP, encoded by a 685 bp gene, that is well expressed in Escherichia coli, render blue-purple colonies. Using this reporter protein, we developed and tested a cloning system based on the constitutive expression of the non-toxic AmilCP protein, that once interrupted, the loss of purple color serves to facilitate positive clone selection. The main advantage of this system is that is less expensive than other systems since media do not contain chromogenic markers such as X-gal, which is both expensive and cumbersome to prepare and use, or inductors such as IPTG. We also designed an inducible expression plasmid suitable for recombinant protein expression that also contains AmilCP cloning selection marker, a feature not commonly found in protein expression plasmids. The use of chromogenic reporters opens an important avenue for its application in other organisms besides E. coli for clone selection or even for mutant selection.


Introduction
Plasmids have been used for several decades now as vehicles for cloning and sequencing, mutagenesis protocols, protein engineering and expression, and shuttles between organisms among many others [1]. Since the generation of the first recombinant molecule by Jackson and colleagues [2], giant steps towards the comprehension of living organism's fundamental processes have been made and the birth of biotechnology was made possible [3]. The more diverse and complex the plasmids are, the more applications and broader hosts we can genetically engineer. To expand our knowledge regarding gene function in specific organisms, plasmids are the most important tool for gene silencing experiments, protein overexpression and more recently, the generation of genome editing tools, rendering plasmids as the prime choice for modern biology research [3,4]. Still, the biology of these diverse molecules remains surprising for researchers due to their ecological role and the evolutionary pressure for the host [5]. 3 of 16 resistant to CcdB to generate enough plasmid for cloning, for example, One Shot ® ccdB Survival™ 2 T1R cells (Thermo Fisher Scientific, Waltham, MA, USA).
The selection of positive clones is still applied for cloning and sequencing, with cheaper procedures this technique can be applied even in diagnostics. One exceptional example is in the diagnosis of genetic mutations, such as described by Zhang [22], where the coupling of cloning and sequencing can lead to new deletion mutation in disease, accelerating and reducing costs for patients. Cheaper clone selection methods are needed for applications in diagnostic medicine.
Expression of proteins still requires tight control, relevant for toxic proteins, which is disadvantageous in induction-free plasmids [23].
Nevertheless, in nature we can find fluorescent proteins that have aided the generation of clone selection plasmids [13] as a good alternative. We found some difficulties with the use of fluorescent reporters: first, colonies can be screened with a source of light in the adequate wavelength that can excite the reporter protein without damaging the DNA inside the cells if the reporter is poorly expressed; second, UV or blue light is needed for the screening of positive clones when colonies are too small and third, the fluorescent signal should be strong (high extinction coefficient) or highly expressed. For these reasons, we explored the possibility of using chromoproteins available in nature. Chromoproteins are GFP-like proteins that have important spectroscopic characteristics (relevant in fluorescent labeling and their applications in imaging) related to their unique spectral, physical, and biochemical properties [24,25]. Studies like the one reported by Alieva and colleagues [24] have uncovered a wide diversity of chromoproteins with a diversity in excitation and emission spectra that can be applied to different reporter applications. We are interested in non-fluorescent purple proteins for two main reasons. First, non-fluorescent purple-blue proteins (called chromoproteins or pocilloporins, proteins that act as photoprotectant in excessive sunlight [25,26], are characterized by high molar extinction coefficient displaying virtually no fluorescence [27], the chromophore possesses another type, which is the isomerized version of the DsRed-like red chromophore [28], where the typical triad Gln-Tyr-Gly is identical to DsRed but the tyrosine ring is non-coplanar and in the trans configuration.
Second, these proteins have been shown to form tetramers [26,28], which render them stable but perhaps not highly useful for protein tagging and subcellular localization as shown for other monomeric fluorescent proteins giving rise to artifacts on protein localization [29].
Third, although fluorescent proteins can also be used as reporter for cloning plasmids when the protein exhibits a broad excitation and emission profile that can be detected by naked eye inspection [30], the confirmation of the phenotype using UV or blue light may be needed with smaller colony forming strains. Also, the simultaneous use of two fluorescent proteins for clone selection and generating a reporter plasmid can be troublesome, therefore, the combination of fluorescent proteins and chromoproteins in cloning plasmids can generate a more versatile system.
With the above information and the need of labs with lower budgets to economize to continue with their research, led us to adapt a marine organism chromoprotein into a cloning system for the following features: (i) the ease to design gene modules, with emphasis on codon usage optimization, transcription and translation elements openly available (http://parts.igem.org/Main_Page); (ii) the reduced costs of DNA synthesis and (iii) the evidence that chromoproteins can be efficiently expressed in E. coli (http://parts.igem.org/Main_Page). Here, we report a cloning system that uses the chromoprotein AmilCP as an efficient reporter for cloning plasmids and protein expression plasmids.

AmilCP as a Reporter for Cloning
The use of chromogenic proteins may be limited for protein tagging due to their oligomeric state. AmilCP expressing cells show a strong blue-purple color, which can be readily detected by eye inspection without the need for additional equipment as shown for other systems where high extinction  [30]. Sequence analysis showed that this protein is oligomeric, using osFP web server, a specialized software to predict oligomerization state of fluorescent proteins [29]. Also, AmilCP structure is closely related to the GFP family proteins [24] as depicted in Figure 1A, the protein is closely related to the GFP structure showing the predicted β-barrel surrounding the chromophore [24,31]. The residues related to GFP fluorescence (Ser65-Tyr70-Pro71), (green in Figure 1A). Nevertheless, the chromoprotein signature is Gln61, Tyr62 and Gly63 (red in Figure 1A) are present [32]. Also, osFP web server predicts as stated above to form oligomers, most likely tetramers, rendering this protein with limited applications for intracellular localization tagging but may be an excellent reporter protein.
To test the stability of this protein and its toxicity in E. coli cells, we expressed AmilCP from a commercial and IPTG inducible plasmid (pQE30, Qiagen, Hilden, Germany). In the cloning process, we used 0.5 mM IPTG in the plates when selecting the transformants. We obtained several blue-purple colonies, showing that the protein when is heavily expressed is non-toxic for E. coli. In Figure 1A, we show that cells expressing AmilCP exposed to different incubation temperatures for 1 h retain the blue-purple color, rendering suitable for even high-temperature screening if needed.
Molecules 2018, 23, x FOR PEER REVIEW 4 of 16 fluorescent proteins [29]. Also, AmilCP structure is closely related to the GFP family proteins [24] as depicted in Figure 1A, the protein is closely related to the GFP structure showing the predicted βbarrel surrounding the chromophore [24,31]. The residues related to GFP fluorescence (Ser65-Tyr70-Pro71), (green in Figure 1A). Nevertheless, the chromoprotein signature is Gln61, Tyr62 and Gly63 (red in Figure 1A) are present [32]. Also, osFP web server predicts as stated above to form oligomers, most likely tetramers, rendering this protein with limited applications for intracellular localization tagging but may be an excellent reporter protein.
To test the stability of this protein and its toxicity in E. coli cells, we expressed AmilCP from a commercial and IPTG inducible plasmid (pQE30, Qiagen, Hilden, Germany). In the cloning process, we used 0.5 mM IPTG in the plates when selecting the transformants. We obtained several bluepurple colonies, showing that the protein when is heavily expressed is non-toxic for E. coli. In Figure  1A, we show that cells expressing AmilCP exposed to different incubation temperatures for 1 h retain the blue-purple color, rendering suitable for even high-temperature screening if needed. AmilCP predicted structure and plasmid cassette designs. Panel (A), predicted structure of AmilCP, the known residues related to GFP fluorescence (Ser65-Tyr70-Pro71) are shown in green, which correspond to residues. The chromoprotein signature of this family of proteins is Gln61, Tyr62 and Gly63, indicated in red [25]. Bellow the predicted structure, cell pellets of E. coli expressing cells from an IPTG inducible plasmid (pQE30) incubated at different temperatures for 1 h, showing that the protein retains its color even in stressing temperatures. Panel (B), Cassettes designed for cloning and protein expression. All cassettes can be replaced using standard BP Clonase™ reactions.

Generation of Cloning Plasmids Using AmilCP Reporter
With the features displayed by AmilCP, we envision to design cloning plasmids that rely on the interruption of the ORF of amilCP gene and therefore generate a cloning/screening system independent of both IPTG and a chromogenic substrate such as S-gal or X-gal, thus reducing the expenses in cloning and screening. For such purpose, we designed two plasmid module variants for cloning and one variant for protein expression. In Figure 1B, we show that using intellectual property free parts we were able to assemble three functional genetic modules bearing the necessary elements for expressing AmilCP and cloning. Two cloning plasmids, one containing AmilCP as positive cloning selection marker were created and, to text the efficiency of the attP sequences attached to each cassette, another cloning plasmid with JuniperGFP reporter, a protein created by DNA 2.0 and available under the BioBrick Public Agreement (https://www.atum.bio/wpcontent/uploads/2013/07/Protein_Paintbox_SynBioPoster_072013.pdf) and intellectual property-free protein openly available (http://parts.igem.org/Part:BBa_J97001), as proof of concept for the use of , predicted structure of AmilCP, the known residues related to GFP fluorescence (Ser65-Tyr70-Pro71) are shown in green, which correspond to residues. The chromoprotein signature of this family of proteins is Gln61, Tyr62 and Gly63, indicated in red [25]. Bellow the predicted structure, cell pellets of E. coli expressing cells from an IPTG inducible plasmid (pQE30) incubated at different temperatures for 1 h, showing that the protein retains its color even in stressing temperatures. Panel (B), Cassettes designed for cloning and protein expression. All cassettes can be replaced using standard BP Clonase™ reactions.

Generation of Cloning Plasmids Using AmilCP Reporter
With the features displayed by AmilCP, we envision to design cloning plasmids that rely on the interruption of the ORF of amilCP gene and therefore generate a cloning/screening system independent of both IPTG and a chromogenic substrate such as S-gal or X-gal, thus reducing the expenses in cloning and screening. For such purpose, we designed two plasmid module variants for cloning and one variant for protein expression. In Figure 1B, we show that using intellectual property free parts we were able to assemble three functional genetic modules bearing the necessary elements for expressing AmilCP and cloning. Two cloning plasmids, one containing AmilCP as positive cloning selection marker were created and, to text the efficiency of the attP sequences attached to each cassette, another cloning plasmid with JuniperGFP reporter, a protein created by DNA 2.0 and available under the BioBrick Public Agreement (https://www.atum.bio/wp-content/uploads/2013/ 07/Protein_Paintbox_SynBioPoster_072013.pdf) and intellectual property-free protein openly available (http://parts.igem.org/Part:BBa_J97001), as proof of concept for the use of reporter proteins that can be detected both by naked eye inspection or by UV-light irradiation (Supplementary Figure S1). Using this plasmid, we showed that the Gateway®system is functional using these sequences in our plasmid cassettes (Supplementary Figure S1).
The cassettes were synthesized by ATUM and provided them as cloned fragments, then we removed part of the multiple cloning sites of different plasmids (see Table 1) and replace them with the synthesized cassettes. Upon arrival, all plasmids when transformed provided the resulting colonies with the expected phenotype, colonies showed an intense blue-purple color (AmilCP containing cassettes).
The first plasmid we characterized was the pDeepPurple1.0 plasmid designed for cloning applications, bearing the pTZ19R backbone (Fermentas, Burlington, Ontario, Canada) and fully removed the original multiple cloning site (depicted in Figure 2A). The plasmid showed in the agarose gel is undigested and restriction analyzed with two restriction enzymes recognizing two sites in the multiple cloning site showing that the full vector is still under 5 kb, making this plasmid amenable for any cloning applications. In Figure 2B,C we show that the resulting colonies form the ligation with two fragments obtained from the 1kb DNA ladder (Invitrogen) rendered both white and blue-purple colonies, that remained blue-purple once picked and grown again in LB containing only the selection marker for this plasmid (ampicillin). The screening was done on 12 independent colonies of each color and results of three colonies, two for the 2 kb band and one for the 1 kb band is shown in Figure 2D. The success rate was comparable to pTZ19R plasmid standard cloning and screening (data not shown). In the case of pDeepPurple1.0, no false positives were observed, while using pTZ19R when plating 12 colonies that looked white, 2 turned blue and had no insert. The same cassette was cloned into two other vectors, pEXT20 [33] and pGEM T-easy (pDeepPurple1.1 and pDeepPurple1.2 respectively). reporter proteins that can be detected both by naked eye inspection or by UV-light irradiation (Supplementary Figure S1). Using this plasmid, we showed that the Gateway® system is functional using these sequences in our plasmid cassettes (Supplementary Figure S1). The cassettes were synthesized by ATUM and provided them as cloned fragments, then we removed part of the multiple cloning sites of different plasmids (see Table 1) and replace them with the synthesized cassettes. Upon arrival, all plasmids when transformed provided the resulting colonies with the expected phenotype, colonies showed an intense blue-purple color (AmilCP containing cassettes).
The first plasmid we characterized was the pDeepPurple1.0 plasmid designed for cloning applications, bearing the pTZ19R backbone (Fermentas, Burlington, Ontario, Canada) and fully removed the original multiple cloning site (depicted in Figure 2A). The plasmid showed in the agarose gel is undigested and restriction analyzed with two restriction enzymes recognizing two sites in the multiple cloning site showing that the full vector is still under 5 kb, making this plasmid amenable for any cloning applications. In Figure 2B,C we show that the resulting colonies form the ligation with two fragments obtained from the 1kb DNA ladder (Invitrogen) rendered both white and blue-purple colonies, that remained blue-purple once picked and grown again in LB containing only the selection marker for this plasmid (ampicillin). The screening was done on 12 independent colonies of each color and results of three colonies, two for the 2 kb band and one for the 1 kb band is shown in Figure 2D. The success rate was comparable to pTZ19R plasmid standard cloning and screening (data not shown). In the case of pDeepPurple1.0, no false positives were observed, while using pTZ19R when plating 12 colonies that looked white, 2 turned blue and had no insert. The same cassette was cloned into two other vectors, pEXT20 [33] and pGEM T-easy (pDeepPurple1.1 and pDeepPurple1.2 respectively). Usually, the selection of positive clones using the lacZ disruption can render false positives due to some lesser extent mutations (both in the lacZa fragment or the chromosomal fragment of lacZ) or Usually, the selection of positive clones using the lacZ disruption can render false positives due to some lesser extent mutations (both in the lacZa fragment or the chromosomal fragment of lacZ) or instability of the plasmid generating a non-functional β-galactosidase enzyme [1,32,34]. We explored the stability of the blue-purple coloration in the cells after containing the pDeepurple1.0 and associated plasmids after 24 h in the absence of selective pressure. In Figure 3, the spot analysis revealed that even in the absence of antibiotic selection, pDeepPurple1.0 and pDeepPurple1.1 showed that most of the cells retained the plasmid when plated again in the presence of the antibiotic. Also, the color of the colonies remained blue-purple, showing a more intense coloration in the high copy number backbone pTZ19R (pDeepPurple1.0). pGEM T-easy backbone seemed to be the lest stable, which may be related to either segregationally loss of the plasmid and more sensitive to be lost during starvation [34] Thus, this system proves to be stable and the remaining cells containing the plasmid expresses correctly AmilCP. instability of the plasmid generating a non-functional β-galactosidase enzyme [1,32,34]. We explored the stability of the blue-purple coloration in the cells after containing the pDeepurple1.0 and associated plasmids after 24 h in the absence of selective pressure. In Figure 3, the spot analysis revealed that even in the absence of antibiotic selection, pDeepPurple1.0 and pDeepPurple1.1 showed that most of the cells retained the plasmid when plated again in the presence of the antibiotic. Also, the color of the colonies remained blue-purple, showing a more intense coloration in the high copy number backbone pTZ19R (pDeepPurple1.0). pGEM T-easy backbone seemed to be the lest stable, which may be related to either segregationally loss of the plasmid and more sensitive to be lost during starvation [34] Thus, this system proves to be stable and the remaining cells containing the plasmid expresses correctly AmilCP.

Protein Expression Plasmids Based on AmilCP Cloning Selection
Following the same design, we envisioned for cloning plasmids, we generated a plasmid for protein expression by adding a 6X his tag. We also replaced the constitutive strong promoter used in the previous designs with the hybrid pTac promoter, which is inducible with IPTG [35]. The original design for the protein expression plasmid relied on the use of an in-frame attP1 sequence that could express the reporter AmilCP protein and once interrupted, generate white colonies but also maintaining the reading frame for the protein of interest ( Figure 4A). Such endeavor required a modified attP1 sequence that prevented the ribosome slippage [36]. Upon arrival, we transformed the pJ215 plasmid from ATUM (termed pDeepPurple2.0) containing the designed cassette depicted

Protein Expression Plasmids Based on AmilCP Cloning Selection
Following the same design, we envisioned for cloning plasmids, we generated a plasmid for protein expression by adding a 6X his tag. We also replaced the constitutive strong promoter used in the previous designs with the hybrid pTac promoter, which is inducible with IPTG [35]. The original design for the protein expression plasmid relied on the use of an in-frame attP1 sequence that could express the reporter AmilCP protein and once interrupted, generate white colonies but also maintaining the reading frame for the protein of interest ( Figure 4A). Such endeavor required a modified attP1 sequence that prevented the ribosome slippage [36]. Upon arrival, we transformed the pJ215 plasmid from ATUM (termed pDeepPurple2.0) containing the designed cassette depicted in Figure 4A, to our surprise, the colonies were all white. We modified the plasmid to generate a functional version that showed intense blue-purple color to achieve clone selection easier. We hypothesized that the attP1 sequence affected the oligomeric state of AmilCP, rendering white colonies. Therefore, we replaced the attP1 sequence with a NotI restriction site, this modification resulted in blue-purple colonies, even in the presence of IPTG (Figures 4B and 5B). Thus, the absence of color was related to the attP1 sequence and not to a mutation associated to AmilCP. This plasmid was further modified by adding an in-frame multiple cloning site (pDeepPurple2.1, Figure 4B) which also resulted in purple colonies. in Figure 4A, to our surprise, the colonies were all white. We modified the plasmid to generate a functional version that showed intense blue-purple color to achieve clone selection easier. We hypothesized that the attP1 sequence affected the oligomeric state of AmilCP, rendering white colonies. Therefore, we replaced the attP1 sequence with a NotI restriction site, this modification resulted in blue-purple colonies, even in the presence of IPTG (Figures 4B and 5B). Thus, the absence of color was related to the attP1 sequence and not to a mutation associated to AmilCP. This plasmid was further modified by adding an in-frame multiple cloning site (pDeepPurple2.1, Figure 4B) which also resulted in purple colonies. With the toxicity problem bypassed, we generated two plasmids, one using the pTZ19R backbone (high copy number plasmid, termed pDeepPurple3.0, Figure 5A) and the pACT3 low copy number plasmid [33] termed pDeepPurple3.1 for the expression of toxic proteins.
In Figure 5A we show the map of the pDeepPurple3.0 plasmid and the response to IPTG for the expression of the reporter AmilCP protein after 2 h induction, observing some leaky expression of the AmilCP reporter. In Figure 5B we show that the resulting plasmids also showed stability and expression of the reporter AmilCP protein in the presence of IPTG after 24 h growth in the absence of the selective agent. In the low-copy-number plasmid pACT3, the color is not so evident by eye inspection. Nevertheless, this low-copy-number plasmid can also be used for toxic protein expression. Our results suggest that the best candidate for overexpressing proteins could be the pDeepPurple3.0 plasmid. To test this, we cloned the encoding sequence for the β-galactosidase enzyme, a 110 kDa protein. We first selected the transformant cells in LB supplemented with ampicillin, X-gal, and IPTG to screen the cells. Then, cells were analyzed in the absence or presence of IPTG. The β-galactosidase activity in the plate was barely detectable in the absence of IPTG, while with the addition of IPTG, the activity was evident ( Figure 6A). With the toxicity problem bypassed, we generated two plasmids, one using the pTZ19R backbone (high copy number plasmid, termed pDeepPurple3.0, Figure 5A) and the pACT3 low copy number plasmid [33] termed pDeepPurple3.1 for the expression of toxic proteins.
In Figure 5A we show the map of the pDeepPurple3.0 plasmid and the response to IPTG for the expression of the reporter AmilCP protein after 2 h induction, observing some leaky expression of the AmilCP reporter. In Figure 5B we show that the resulting plasmids also showed stability and expression of the reporter AmilCP protein in the presence of IPTG after 24 h growth in the absence of the selective agent. In the low-copy-number plasmid pACT3, the color is not so evident by eye inspection. Nevertheless, this low-copy-number plasmid can also be used for toxic protein expression. Our results suggest that the best candidate for overexpressing proteins could be the pDeepPurple3.0 plasmid. To test this, we cloned the encoding sequence for the β-galactosidase enzyme, a 110 kDa protein. We first selected the transformant cells in LB supplemented with ampicillin, X-gal, and IPTG to screen the cells. Then, cells were analyzed in the absence or presence of IPTG. The β-galactosidase activity in the plate was barely detectable in the absence of IPTG, while with the addition of IPTG, the activity was evident ( Figure 6A).  Induction was confirmed by SDS-PAGE, showing that β-galactosidase protein was readily detectable after 2 h induction ( Figure 6B), while the control condition, AmilCP is detectable even in  Induction was confirmed by SDS-PAGE, showing that β-galactosidase protein was readily detectable after 2 h induction ( Figure 6B), while the control condition, AmilCP is detectable even in Induction was confirmed by SDS-PAGE, showing that β-galactosidase protein was readily detectable after 2 h induction ( Figure 6B), while the control condition, AmilCP is detectable even in the absence of IPTG, suggesting that AmilCP is accumulated in the cell from the leaky expression of the pTac promoter.

Expanding the Application of pDeepPurple1.2 Plasmid
The plasmid pDeepPurple1.2 was further modified by adding the OriV replication origin at SstI restriction site of pGEM T-easy (in close vicinity to the integrated cassette but without disturbing other plasmid elements, Figure 7A). With the resulting plasmid, we transformed K. pneuomoniae and P. aeruginosa and selected by ampicillin resistance and blue-purple colonies. Two selection passes were done to assess retention of the plasmid. The analysis of the untransformed and transformed strains by light microscopy is shown in Figure 7B. First, the analysis of E. coli expressing AmilCP from the pQE30 plasmid showed that the cells under the microscope showed a blue-purple color ( Figure 7B). This observation is in concordance with the biochemical analysis currently conducted in our labs, AmilCP has been purified to near homogeneity from the soluble fraction (unpublished data). Second, the cells expressing AmilCP in K. pneumoniae or P. aeruginosa strains showed a similar color pattern. Third, the strong constitutive promoter used to generate this plasmid is functional in K. pneumonaie and P. aeruginosa cells.
Molecules 2018, 23, x FOR PEER REVIEW 9 of 16 the absence of IPTG, suggesting that AmilCP is accumulated in the cell from the leaky expression of the pTac promoter.

Expanding the Application of pDeepPurple1.2 Plasmid
The plasmid pDeepPurple1.2 was further modified by adding the OriV replication origin at SstI restriction site of pGEM T-easy (in close vicinity to the integrated cassette but without disturbing other plasmid elements, Figure 7A). With the resulting plasmid, we transformed K. pneuomoniae and P. aeruginosa and selected by ampicillin resistance and blue-purple colonies. Two selection passes were done to assess retention of the plasmid. The analysis of the untransformed and transformed strains by light microscopy is shown in Figure 7B. First, the analysis of E. coli expressing AmilCP from the pQE30 plasmid showed that the cells under the microscope showed a blue-purple color ( Figure 7B). This observation is in concordance with the biochemical analysis currently conducted in our labs, AmilCP has been purified to near homogeneity from the soluble fraction (unpublished data). Second, the cells expressing AmilCP in K. pneumoniae or P. aeruginosa strains showed a similar color pattern. Third, the strong constitutive promoter used to generate this plasmid is functional in K. pneumonaie and P. aeruginosa cells. With the cumulative data presented here, AmilCP renders an excellent reporter protein for bacterial organisms but requires further research to understand its behavior inside bacterial cells. With the cumulative data presented here, AmilCP renders an excellent reporter protein for bacterial organisms but requires further research to understand its behavior inside bacterial cells.

Discussion
In this work, we generated a set of plasmids designed for cloning and protein expression using a reporter protein that is stable and well expressed in bacteria. Also, the generated plasmid for expressing desired proteins, bearing this reporter protein can facilitate cloning, a feature not commonly found in protein expression vectors. AmilCP protein showed to be a stable and easy to detect. Our results indicate that the cloning plasmids generated using the pTZ19R backbone showed to be the most stable and provided the best results for cloning and positive clone selection. Also, as shown in Figure 7, this system is also applicable to other microorganisms, expanding the molecular tool repertoire for molecular microbiology studies.
With the design presented here, these plasmids can be used to generate other derivatives for distinct applications by either removing parts of the cassettes or completely remove the reporter coding sequence using the att sequences, this feature is true for all the plasmids generated and be further custom modified using the atPL sequences flanking the designed cassettes. This is an easy task since the combination of the color phenotype of the transformants, along with the selection markers, make the system flexible. By coupling robotic stations and Gateway ® cloning, analysis of several samples can be accelerated and even translated into clinical samples.
To our knowledge, plasmids relaying on chromoproteins have been reported, using AmilCP in the entry vector, for the Golden-Gate method [37]. Nevertheless, standard cloning that is still used around the globe requires plasmid vectors that are user-friendly with cheaper selection markers. Also, the use of chromogenic markers can lead to novel molecular tools not only for plasmid clone selection but for mutant selection along with other good and reliable positive cloning selection with fluorescent proteins that are readily detected with the naked eye and can be coupled to high throughput strategies [36][37][38]. Taken together, the strategy presented here, which can reduce cloning costs, taking advantage of the also reduced costs for sequencing, can lead to a new cheaper personalized diagnostic medicine [22]. Using the described strategies here, we envision that synthetic biology can benefit all kinds of research and nature has provided all kinds of tools that can be adapted to novel molecular tools.

Synthetic Cassettes and Plasmid Constructions
The approach to generate the cloning and expression cassettes consisted in using the standards for cloning BioBricks™ ( [39] and http://openwetware.org/wiki/The_BioBricks_Foundation:RFC), but we replaced the need for restriction enzymes due to the 8-bp scar between assembled fragments. To avoid this, we designed the cassettes using iGEM standard parts, eliminated the BioBricks prefix and suffix (in those parts that had them) and assembled them in sillico using NEB Builder ® (New England Biolabs, Ipswich, MA, USA) and Gene Designer ® (DNA 2.0, now ATUM, Newark, CA, USA) softwares. In Table 2 we provide the complete list of parts used and in Supplementary File 1 the complete sequence file of the plasmids generated in this work. In Figure 1B we depict (not in scale) the cassette modules designed, which were thought to be used in either traditional cloning using restriction enzymes or Gateway ® recombination cloning. The coding sequences for AmilCP was codon optimized (Integrated DNA Technologies Codon Optimization tool, http://www.idtdna.com/CodonOpt) for E. coli but also with a CAI suitable for other bacterial organisms, assessed with GenScript Rare Codon Analysis Tool (https://www.genscript.com/tools/rare-codon-analysis). All cassettes were tested first in the ATUM vector (either pIJ201 or pJ215). Then, all cassettes were PCR amplified using the pQE30 sequencing primers and cloned either in pTZ19R, pACT3 or pGEM T-easy plasmids. In the case of pTZ19R, the multiple cloning site was removed as EcoRI-HindIII digestion, filled in and blunt-ended cloned. In the case of pACT3 plasmid, we mutagenized a SspI restriction site contained in the cat gene to remove the pTac promoter and multiple cloning site as SspI restriction and blunt-ended cloning of the desired cassette. Finally, pGEM T-easy was digested with SphI and NdeI, blunt ended with Klenow fragment and then cloned also as blunt-ended the desired cassette. For further characterization of AmilCP, we generated an independent construct by cloning the coding sequence of AmilCP into the pQE30 plasmid (Qiagen) as a BamHI-HindIII fragment. In Table 3 we listed the primers used in this study. All PCR reactions were conducted using Herculase II fusion DNA polymerase (Agilent Technologies, USA, Ceedar Creek, TX, USA) using the standard amplification protocol: 94 • C for 2 min, 30 cycles of 94 • C for 20 s, 50 • C for 30 s and 72 • C for 1 min due to the sizes of the cassettes used in this study.
The modified plasmid for other hosts, besides E. coli, was created by PCR amplifying the OriV [40] from a synthesized gBlock based on the BBa_K125340 sequence. SstI pTZ19R bearing the AmilCP cloning cassette (named pDeppPurple1.0) was blunt ended with Klenow and ligated to the blunt-ended OriV PCR fragment. Transformants were selected in E. coli and PCR verified.

Chromogenic Plasmid Stability Analysis
To determine qualitatively the number of remaining cells carrying the plasmids generated after selective agent removal for 24 h, we conducted a spot test. First, cells were grown in the presence of the selective agent overnight. Second, cells were diluted to an OD 600 of 0.2 in LB without a selective agent and incubated for 24 h. The control condition was done equally but in the presence of the selective agent. Finally, cells were serial diluted as indicated in the figures in LB medium. Three microliters of these cell suspensions were spotted on LB agar plates and incubated for 16 h at 37 • C, plate image was recorded using an Image Station 2000R (KODAK, Rochester, NY, USA). Experiments were performed in triplicate.

Cloning Test
To demonstrate that the cloning system is effective, the pDeepPurple1 plasmid was digested with EcoRV. 1 kb ladder (Invitrogen) size marker bands (2 and 1 kb) were gel purified and blunt-ended with Klenow DNA polymerase. The fragments were ligated and then transformed. Screening of colonies was performed by blue-purple/white discrimination. Standard miniprep plasmid purification was carried and BamHI-HindIII restriction analysis was done to confirm the presences of either the 2 kb or 1 kb band.

Protein Expression Test in pDeepPurple 3.0
To test protein overexpression, we chose the β-galactosidase enzyme as proof of concept. This is a 110 kDa protein that we can also trace for activity. We cloned as a BamHI and HindIII fragment the whole coding sequence into pDeepPurple3.0 plasmid and screened the transformants with X-gal and IPTG containing plates. Expression was tested in LB with 0.5 mM IPTG for 2, 4 and 6 h. Cells were collected by centrifugation by taking 1 mL of cell culture, resuspended in PBS 1X. Cells were lysed in 2X loading buffer, boiled for 5 min. and loaded into a 12% SDS page to compare both AmilCP (25.4 kDa) and β-galactosidase (110 kDa) enzyme.

Plasmid Mutagenesis
The protein expression cassette was mutagenized using the QuikChange II Site-Directed Mutagenesis Kit (Agilent) following the manufacturer's instructions and using the primers listed in Table 3. First, the protein expression cassette was mutagenized to eliminate attP1 sequences and replacing it with a NotI restriction site. Then, the multiple cloning site was added by aligning the primers listed in Table 3 MCS1 and MCS2 that once aligned left 5' and 3' ends compatible to NotI and regenerated the same restriction site. The pACT3 was mutated to remove the SspI restriction site at the cat gene using the same strategy and primers mutsspIpact3 and mutsspIpact3compl (Table 3).

Protein Sequence and Structural Analysis
AmilCP protein sequence was analyzed using osFP web server for oligomerization prediction [30]. Protein structure was modeled using the Phyre2 web server. Protein surface was estimated using DeepView-Swiss PDB viewer [41] to calculate distances and hydrophobicity. Also, using Phyre2 webserver we estimated the conservation of protein models [42,43], visualization was done with PyMol software version 1.8 [44].

K. preumoniae and P. aeruginosa Transformation
Both strains were tested for ampicillin resistance. K. pneuomniae JH1 strain showed sensitivity to 100 µg/ml, while the P. aeruginosa PA01 strain showed sensitivity at 300 µg/mL. Cells were electroporated according to the protocols of the Gene Pulser Xcell™ electroporation system (Bio-Rad, USA, Hercules, CA, USA) without any modification. Electroporation was carefully done on each strain using 2 µg of purified plasmid pDeppPurple1.2V, bearing the OriV replication origin. pGEM'T-easy backbone was the only vector in our set that permitted us to do this cloning. Transformed cells were selected in LB media with the corresponding amount of ampicillin on each plate. Two replica plating was used to isolate blue-purple colonies and analyzed by light microscopy (Axioskop 40, ZEISS, USA, Thornwood, NY, USA) using 100× amplification with immersion oil.