Engineering

👋 Introduction

---

During the experimental part of our OptogenEYEsis project, we focused on testing four microbial opsin proteins already reported to have potential in vision restoration [1,2]: CatCh, ChrimsonR, Jaws and NpHR. Additionally, we also tested NpSRII for reasons we report in depth later on this page.

Our first goal was to functionally express the opsins in the Escherichia coli chassis, followed by developing a system that will enable a high throughput screening of opsin’s activity. Our final goal was the evolution of opsins and the isolation and identification of variants with enhanced sensitivity and wavelength tuning of their absorption spectrum.

This page describes our strategic choices, our designs and our experimental results. It is organized in several chapters that explain:

1. The characteristics of our 4+1 opsins of interest
2. The Evolution.T7 system in order to understand the structure of our opsins’ expressions cassettes
3. The different E. coli strains we used in order to clarify the choice of the chassis for different experiments
4. Opsins’ expression in E. coli and characterisation of their spectral properties
5. The channelrhodopsins screening systems
• 5.1 The Na⁺ riboswitch-based strategy that we designed for the detection of the activity of channelrhodopsins
• 5.2 The CoroNa™ Green-based strategy that we designed for the detection of the activity of channelrhodopsins
6. The halorhodopsins (and sensory rhodopsins) screening system
7. The evolution of a channelrhodopsin

🧶 1. Our 4+1 opsins of interest

---

Microbial opsins can be categorized in four families as following (Figure 1):

• Bacteriorhodopsins which are proton pumps exporting H⁺ from the cytoplasm to the extracellular medium.
• Channelrhodopsins which are pumps moving cations in or out of the cytoplasm.
• Halorhodopsins which are chloride pumps importing Cl^- from the extracellular medium into the cytoplasm.
• Sensory rhodopsins that do not exhibit a transport activity, but interact with a transducer protein (Htr) with has a histidine kinase domain and thus trigger intracellular processes by regulating phosphorylation of cellular factors.

Among them, only the channelrhodopsins (CR) and the halorhodopsins (HR) are adapted for optogenetic therapies for vision restoration [1,2], thanks to their cation and anion transport activity leading to a depolarising or hyperpolarizing effect, respectively.

As mentioned above, the experimental part of the OptogenEYEsis project focused on four microbial opsins with reported potential in vision restoration [2] CatCh, ChrimsonR, Jaws and NpHR, but also NpSRII.

CatCh (BBa_K4601000) is a channelrhodopsin derived from the ChR2 of Chlamydomonas reinhardtii [4], from which it only retains the first 309 amino acids containing the N-terminal transmembrane domain forming the channel. Compared to ChR2, it also contains one mutation L132C that modifies the photocurrent generated upon light irradiation. CatCh is a Ca²⁺ inward pump that is also capable of transporting, although to a lesser extent, Na⁺, K⁺ and Mg²⁺. It was successfully used in optogenetic therapies for visual restoration in primates [5].

ChrimsonR (BBa_K4601001) is a channelrhodopsin derived from the CR Chrimson of Chlamydomonas noctigama that is naturally red-shifted [6]. Compared to Chrimson, ChrimsonR contains one point mutation K176R that does not modify the absorption properties, but increases the photocurrent generated upon light irradiation. ChrimsonR is a Na⁺ inward pump that is also capable of transporting H⁺, K⁺ and Ca²⁺. It is one of the few microbial opsins in clinical trials for gene therapy for patients with documented diagnosis of non-syndromic Retinitis Pigmentosa [7].

Jaws (BBa_K4601002) is a halorhodopsin derived from Haloarcula (Halobacterium) salinarum (strain Shark) and is naturally red-shifted [8]. Compared to the wild-type HR (Halo57) it contains two mutations K200R and W214F that do not modify the absorption properties, but increase the photocurrent generated upon light irradiation.

NpHR (BBa_K4601003) is a halorhodopsin identified in the archaea Natronomonas pharaonis.

The fifth rhodopsin we used in our project is the NpSRII (BBa_K4601005) which is a sensory rhodopsin identified in the archaea N. pharaonis. We used it as a control for our synthetic constructs that aim to adapt the sensory rhodopsin signal transduction signal into a halorhodopsin sensing device.

2. The Evolution.T7 system 🧬

---

In 2021, the iGEM Evry Paris-Saclay team developed the Evolution.T7 project. As four of our team members also participated in the 2021 team: two students, Doriane Blaise and Georges Sainte-Rose, and two PIs, Ioana Popescu and Manish Kushwaha, we decided to use this powerful system for in vivo evolution of the selected microbial opsins.

Evolution.T7 is an in vivo directed evolution system that uses the orthogonal T7 RNA polymerase (T7RNAP) fused at the N-terminus with different cytosine or adenine deaminases (CD, AD), base deaminases (BD) for simplicity. T7RNAP is highly specific to its promoter sequence (TAATACGACTCACTATA), the strength of which is increased if it is followed by GGG at 3' end [9]. The T7RNAP initiates the transcription at the first guanidine of this stretch of three G and terminates it at specific terminator(s) [10]. When BD-T7RNAP fusion protein is expressed in E. coli, the sequence flanked by the T7 promoter and the T7 terminator(s) gets mutated as the BD deaminates randomly the nucleotides on the non template strand of the T7RNAP [11]. In an E. coli strain deficient in its natural DNA repairing system (Δung Δnfi), upon DNA replication, these deaminated bases lead to C->T or A->G transition mutations, depending on whether CD or AD was used.

Evolution.T7 also uses a mutated T7RNAP_CGG-R12-KIRV specific to an altered T7_CGG promoter sequence (TAATACcggTCACTATA) [12] which was placed in reverse orientation downstream of the region to be mutated in order to compensate for the above mentioned bias of deaminations occuring mainly on the non template strand. By targeting the opposite strand, T->C or G->A transition mutations may also be introduced in the gene of interest.

Five CD (AID, pmCDA1, rAPOBEC1, evoAPOBEC1-BE4max, evoCDA1-BE4max) and two AD (TadA^* and ABE8.20m) each linked to the T7RNAP and T7RNAP_CGG-R12-KIRV compose the Evolution.T7. They are carried by low copy plasmids (pSEVA221 and pSEVA471, respectively) to alleviate the burden of replication, reduce the toxicity of T7RNAP, and decrease the possibility of side mutational effects elsewhere in the E. coli genome. The expression is inducible by anhydrotetracycline for BD-T7RNAP and by L-arabinose for BD-T7RNAP_CGG-R12-KIRV allowing either a sequential or a concomitant mutagenesis on the two DNA strands.

The general architecture of the Evolution.T7 system is depicted in Figure 2 and all opsins expression cassettes follow it. Thus, both the expression of opsins in an E. coli DE3 strain is possible and their evolution.

🦠 3. E. coli strains used in this project

---

In the project we used various techniques and no single E. coli strain was suitable for all applications. Some strains were commercially obtained, other published strains were kindly provided by the authors, and some others were built by us. For a better understanding of our work and of our strain choices, here is their list followed by the description of their characteristics:

• NEB^® 5-alpha
• C43(DE3)
• MG1655^* Δflu ΔpyrF Δung Δnfi
• C43(DE3) ΔenvZ (ApraR)
• MG1655^* Δflu ΔpyrF Δung Δnfi ΔenvZ (ApraR)
• C43(DE3) ΔenvZ (ApraR) Δ(lacZ)M15
• MG1655^* Δflu ΔpyrF Δung Δnfi Δ(lacZ)M15

E. coli NEB^® 5-alpha is a commercial strain kindly provided by our sponsor New England Biolabs. It is genetically close to the widely known DH5𝛼 strain from which it was derived. We used NEB^® 5-alpha as a chassis for all our cloning experiments, but also for some experimental characterisations of our constructs when no particular genetic requirement was necessary.

E. coli C43(DE3) is a BL21(DE3) derivative selected for its increased capacity of expressing membrane proteins [13]. As all E. coli DE3 strains, it contains in its genome the T7 RNA polymerase (T7RNAP) gene under the control of the lacUV5 promoter and requires IPTG to induce expression. The T7RNAP is required for expression of genes driven by the T7 promoter. This strain is commercially available at Sigma-Aldrich/Merck and kindly provided by our host lab. We choose this strain as a chassis for the expression of opsins, as they are membrane proteins but also because their genes are under the control of the T7 promoter to allow their evolution using the Evolution.T7 system.

E. coli MG1655^* Δflu ΔpyrF Δung Δnfi [14] is a strain derived from the wild-type E. coli MG1655 in which several modifications have been introduced:

• ^*: the ^* denotes an insertion of a G in the rph gene after nucleotide 671 that corrects a frameshift observed in the MG1655 genome compared to other E. coli strains. This frameshift leads to the expression of a truncated version of the rph gene which lacks RNase PH activity. This premature translation stop impacts negatively the expression of the downstream gene pyrE expressed from the same polycistronic RNA [15]. The pyrE gene encodes the orotate phosphoribosyltransferase involved in the uracil metabolism. Its low expression in MG1655 leads to pyridine starvation and its higher expression in MG1655^* leads to an increased toxicity of 5’-fluorotic acid.
• Δflu: the flu gene encodes the antigen 43 (Ag43) which is an autotransporter protein localized in the outer membrane of E. coli and is involved in biofilm formation [16]. This locus was used for genome integration of synthetic constructs [14].
• ΔpyrF: the pyrF gene encodes the orotidine 5'-phosphate decarboxylase involved in the uracil metabolism [17]. Its deletion renders E. coli auxotrophic for uracil, but resistant to 5’-fluorotic acid.
• Δung: the ung gene encodes the uracil DNA N-glycosylase which is involved in the elimination of uracil from the DNA [18].
• Δnfi: the nfi gene encodes the exonuclease V of E. coli which is involved in the elimination of inosines from the DNA [19,20].

While the ^*, the Δflu and the ΔpyrF are modifications not necessary for our project, Δung and Δnfi are important for enhancing the mutation rate. We used this strain as a chassis for opsin evolution using the Evolution.T7 system developed by the iGEM Evry Paris-Saclay 2021 team.

This strain was kindly provided by Dr. Luis Ángel Fernández, at Centro Nacional de Biotecnología (CNB-CSIC) in Spain.

E. coli C43(DE3) ΔenvZ (ApraR) and E. coli MG1655^* Δflu ΔpyrF Δung Δnfi ΔenvZ (ApraR) were constructed by us from the above described strains C43(DE3) and MG1655^* Δflu ΔpyrF Δung Δnfi by replacing the envZ gene with an apramycin resistance cassette (ApraR).

The envZ gene encodes the sensor histidine kinase EnvZ which is part of the EnvZ-OmpR two component system [21] involved in sensing the variations in the osmolarity of external medium and the consequent regulation of the expression of OmpF and OmpC which are two outer membrane pores. EnvZ is a multidomain homodimeric protein. Notably, its cytoplasmic domain has a histidine kinase activity capable of autophosphorylation in trans within the EnvZ dimer and subsequent phosphorylation of the OmpR transcription factor. The phosphorylated OmpR dimerises and binds to specific sequences present in the promoter regions of the ompC and ompF genes thus regulating their expression.

This system had been widely adopted by the synthetic biology community, with a number of projects using the EnvZ histidine kinase domain fused to the protein of interest. Its activity was coupled to the expression of a reporter gene under the control of an OmpC promoter. Getting inspired by the work of the iGEM Tokyo-NoKoGen 2012 team, we decided to use this system for detecting the activity of halorhodopsins. But, to implement this, deleting the envZ gene from the E. coli genome is necessary.

We performed these deletions in above-described strains C43(DE3) and MG1655^* Δflu ΔpyrF Δung Δnfi using the 𝜆 Red system following the protocol described by Datsenko and Wanner [22]. For this, the ApraR cassette carried by the pEVL408 plasmid [23] (kindly provided by our host lab) was first amplified by PCR with primers containing the 50 nucleotides upstream and downstream of the envZ gene (5'-CCAGTATCTTATAGAAAGCAAAACGGGAGGCACCTTCGCCTCCCGTTTATAGAGCGGCCGCCACCG-3' and 5'-CCGTCTGGGGTCTGGGCTACGTCTTTGTACCGGACGGCTCTAAAGCATGAGCATATGCTGCGTGCATGCG-3'), then the PCR product was used to transform the E. coli strains carrying the pKD46 plasmid (kindly provided by our host lab) induced with 10 mM L-arabinose prior to transformation. ΔenvZ ApraR colonies were selected on apramycin and the correct insertion confirmed by colony PCR with external primers binding the pck gene (TCGATAAATACACCGACACC) and ompR gene (GAATATTCCGCAATGGAACG). The pKD46 plasmid was eliminated by growing the cells at 37°C.

E. coli C43(DE3) ΔenvZ (ApraR) Δ(lacZ)M15 and E. coli MG1655^* Δflu ΔpyrF Δung Δnfi Δ(lacZ)M15 were constructed by us from the above described strains C43(DE3) ΔenvZ (ApraR) and MG1655^* Δflu ΔpyrF Δung Δnfi by introducing a 93 nucleotides deletion in the 5’ region of the LacZ gene. This deletion leads to the expression of a non-functional LacZ protein that can be complemented in trans by the LacZ𝛼 fragment, a process known as the alpha complementation. Δ(lacZ)M15 is present in many E. coli strains constructed for cloning proposes (ex. DH5𝛼, TOP10, …), while the LacZ𝛼 fragment being smaller is encoded by many classical cloning vectors (ex. pUC19, pBlueScript, …). In our project we decided to use LacZ𝛼 as a reporter gene in order to be able to perform a colorimetric assay that will not use light for activation (as GFP-based measurements do) that may interfere with the activity of opsins, which are proteins activated by light too.

To introduce the Δ(lacZ)M15, we decided to take advantage of the possibility to do blue/white screening and use a 𝜆 Red-based scarless method that does not introduce an antibiotic resistance gene in the genome [24]. For this, we amplified a 1 kb DNA fragment flanking the Δ(lacZ)M15 from the E. coli NEB^® 5-alpha genome using the high fidelity DNA polymerase Q5^® and the primers 5’-TCCGGTCTCTTGGAATTACCCTGTTATCCCTAGATCAGATGGCGCTGGGCGCAATGC-3’ and 5’-TCCGGTCTCTCTGCATTACCCTGTTATCCCTAATGCCGCTCATCCGCCACATATCC-3’ carrying sites for the I-SceI restriction endonuclease. This fragment was cloned in the pSB1K3 backbone and was used to transform E. coli cells carrying the pACBSR plasmid. Transformed cells were grown in LB medium supplemented with 12.5 µg/mL kanamycin and 17.5 µg/mL chloramphenicol and, upon reaching the exponential growth phase, the 𝜆 Red system and the meganuclease I-SceI expression were induced with 15 mM L-arabinose. Recombinant cells were selected on LB agar plates containing 1 mM IPTG and 20 µg/mL X-gal. White colonies were isolated, and the introduction of the Δ(lacZ)M15 modification was first confirmed by colony PCR and then by sequencing of the PCR product.

4. Opsins’ expression in E. coli
🔬 & characterisation of their spectral properties

---

As mentioned above, during the experimental part of our OptogenEYEsis project, we focused our attention on four microbial opsins that have been reported to have potential in vision restoration [1,2]: CatCh, ChrimsonR, Jaws and NpHR, but also NpSRII. Our plan is to use these proteins as a control to compare them, in terms of light sensitivity and wavelength tuning of their absorption spectrum, with their mutated variants obtained by the Evolution.T7 system and to finally select the variants that show enhanced features with favorable mutations.

Design

To allow the expression of the five opsins, but also their evolution using the tool developed by the iGEM Evry Paris-Saclay 2021 team, we designed expression cassettes following the Evolution.T7 architecture (Figure 2). For this:

• We designed specific RBS libraries for each opsin using the De Novo DNA’s Library Calculator v2.0 [25–27] and selected the one predicted to have a Translation Initiation Rate (TIR) of about 10000.
• We placed this RBS+CDS under the control of the T7 promoter upstream and four T7 terminators downstream
• We also inserted the mutant T7_CGG promoter in reverse orientation between the end of the opsin’s coding sequence and the first of the four terminators to ensure the occurrence of mutations driven by the mutant T7RNAP_CGG-R12-KIRV linked with a base deaminase (BD) moving in reverse orientation. To stop the BD-T7RNAP_CGG-R12-KIRV four T7 terminators were placed in reverse orientation upstream of the T7 promoter

In total we designed six cassettes for the expression of CatCh (BBa_K4601200), ChrimsonR (BBa_K4601201), Jaws (BBa_K4601202), NpHR (BBa_K4601203), NpHR P240T F250Y (BBa_K4601204) and NpSRII (BBa_K4601205).

The E. coli C43(DE3) strain was chosen for opsin expression as it is a BL21(DE3) derivative, selected for its increased capacity of expressing membrane proteins [13]. This would be vital to concentrate the expressed opsins in the membrane and facilitate their spectroscopic analysis after membrane purification [28].

Knowing that different opsins have different absorption properties (Table 1), we expect to detect a peak in the absorption spectra of the membrane preparations characteristic to the opsin.

Opsin	Maximum absorption wavelength	Reference
CatCh	480 nm	[5]
ChrimsonR	590 nm	[6]
Jaws	632 nm	[8]
NpHR	589 nm	[29]
NpSRII	Around 500 nm	[30]

Build

The DNA sequences of all opsins mentioned above were synthesized, except that of NpSRII which was amplified directly from N. pharaonis genome (kindly provided by our host lab) with primers that introduce specific type IIS restriction sites (BsaI). The pSEVA721 backbone was also amplified by PCR to make it Golden Gate compatible. All of the expression cassettes (BBa_K4601200, BBa_K4601201, BBa_K4601202, BBa_K4601203, BBa_K4601204, BBa_K4601205) were assembled by Golden Gate in the low copy-number plasmid pSEVA721.

Test

Membranes from E. coli C43(DE3) cells expressing the opsin genes were extracted following a protocol adapted from [28]. Briefly, E. coli cells were first grown overnight at 37 °C at 200 rpm in LB (Lennox) supplemented with 10 µg/mL trimethoprim. The cells were then diluted by 100 times in 50 mL of the same media and, after 4 hours of incubation at 37°C at 200 rpm, the opsin expression was induced with 1 mM IPTG and 10 µg/mL all-trans-retinal. As all E. coli DE3 strains, the C43(DE3) contains inserted in its genome the T7 RNA polymerase gene required for expression from the T7 promoter that controls the opsin gene in the Evolution.T7 system. It is under the control of the lacUV5 promoter and requires IPTG to induce expression. Microbial opsins are type I opsins that require all-trans-retinal as co-factor for their activity. This compound is not naturally produced by E. coli, and for this reason we supplement it during culturing.

After an overnight incubation at 37°C at 200 rpm, cells were harvested by centrifugation (3000 g for 15 minutes at 4°C), washed twice in 20 mL of water and submitted to an osmotic lysis in 5 mL “Sweet” buffer containing 0.4 M sucrose, 75 mM TrisHCl pH 8.0 and 2 mM MgSO₄. After an hour of incubation at 37 °C at 200 rpm, the “Sweet” buffer was removed by centrifugation (3000 g for 15 minutes at 4°C) and the cells resuspended in a “Salt” buffer (0.8 M NaCl, 50 mM Tris pH 7.6, 10 mM MgSO₄). Cells were broken with 1 g of glass beads (1 mm diameter) by vortexing 5 times 1 minute at maximum speed interrupted by 1 minute on ice. The mix was transferred into microcentrifuge tubes and the membrane fraction was collected by centrifugation (15000 g for 20 minutes at 4°C).

After discarding the supernatant, the pellet (Figure 3) was solubilised using three different detergents, n-octyl-β-D-glucopyranoside, n-dodecyl-β-D-maltoside, or sodium cholate, each at a concentration of 3% in 10 mM PIPES-KOH pH 7 buffer in a total volume of 400 µL. After an overnight incubation at 4°C at 1400 rpm, the debris were removed by centrifugation (14000 g for 10 minutes at 4°C) and the supernatant collected.

Absorption spectra were recorded in an opaque wall 96-well polystyrene microplate (COSTAR 96, Corning) using a CLARIOstar (BMGLabtech) plate reader (Figure 5).

SDS-PAGE analysis (Figure 4) was performed on 5 µL of solubilised membranes that were mixed with Laemmli Buffer (final concentrations 20.83 mM Tris-HCl pH 6.8, 0.67% (w/v) SDS, 3.33% glycerol, 1.67% 2-mercaptoethanol, 0.5% bromophenol blue) and after 1 hour of incubation at room temperature they were loaded onto a 12% SDS-polyacrylamide gel for protein separation, using a Bio-Rad Protean mini-gel system. Electrophoresis was performed in the SDS-PAGE running buffer (3.03 g/L Tris base, 14.4 g/L Glycine, 1 g/L SDS, pH 8.3) at constant 150 V, until the dye migrated close to the bottom of the gel. The gel was then stained with Coomassie Blue R-250.

Learn

In order to assess the spectral properties of the different opsins, we need to purify the membrane fraction of E. coli C43(DE3) cells that should contain the expressed opsin proteins, according to a protocol adapted from [28]. Such an approach is reported to increase the signal due to the light absorption by opsins and reduce scattering. Moreover, it avoids the use of protein tags that might disrupt the assembly of opsin multimers in the membrane.

By looking at the color with the eye (Figure 3), all the pellets of the different opsins seem to be colorless similar to the negative control in which opsins are absent (First picture on the left), except that of ChrimsonR that shows a slight orange color. The slight orange color could be attributed to the expression of the opsins, but it is important to note that in addition to the expressed opsins, lipids, carotenoids and other colored membrane proteins can be present in the pellet [31]. Unlike what is expected [28], we did not observe a pellet and a darker film on top of it, but only saw a homogenous pellet. This might be the result of low expression levels of the opsins due to the use of pSEVA721 low copy number plasmid [32]. As a consequence, we proceeded to the solubilisation of the entire pellet using first the detergent n-octyl-β-D-glucopyranoside, then, in the follow-up experiments, also n-dodecyl-β-D-maltoside and sodium cholate.

In order to assess the protein expression, SDS-PAGE was performed for the solubilized membranes extracted from E. coli C43(DE3) cell pellets with the three different detergents (n-octyl-β-D-glucopyranoside, n-dodecyl-β-D-maltoside and sodium cholate, respectively). The visualization of the proteins extracted from the different pellets (Figure 4) shows similar patterns as that of the “No opsin” control regardless of the detergent used, suggesting that there is a similar level of membrane proteins expressed between the opsins/no opsins situations, which is logical as we are using the same strain and expression architecture. However, by looking at the gels we can't be certain that our opsins of interest are being expressed, although, we can see some bands with sizes similar to what is expected to obtain from our expression systems (CatCh: 34.31 kDa, ChrimsonR: 39.13 kDa, Jaws: 29.03 kDa, NpHR: 31.08 kDa and NpSRII: 23.93 kDa). Nonetheless, we can not neglect the possibility that these bands might correspond to other membrane proteins.

To check the profile of light absorbance by the different opsins we measured the absorption of the extracted membrane fractions (first with n-octyl-β-D-glucopyranoside as a detergent) from E. coli C43(DE3) carrying the opsins genes, within the range of visible light wavelengths (300 - 700 nm) (Figure 5). All the spectra obtained in this case (top left) show a peak at 412 nm which we consider as a reference peak as it is the most significant peak in the spectra of the different membranes [28]. Moreover, after normalizing the absorption data of the different membranes with “opsins'' to that of “no opsins'', we can spot a peak at around 560 nm in the case of Jaws, NpHR, NpHR-P240T-F250Y and NpSRII, while this peak did not exist in case of ChrimsonR and Catch. This peak at around 560 nm can be as a result of the absorption by opsins.

To investigate if the absence of the peak at position around 560 nm in case of ChrimsoR and CatCh is due to the influence of the detergent used, we repeated the experiment and replaced n-octyl-β-D-glucopyranoside with two other detergents (n-dodecyl-β-D-maltoside or sodium cholate). After subtracting the absorption values of “no opsins” spectra, we noticed slight peaks at around 560 nm for both CatCh and ChrimsonR when n-dodecyl-β-D-maltoside was used. It is obvious that with different detergents the shape of the spectra have changed including the peaks at 412 nm and 560 nm. As a result, this supports that the choice of the detergent has a significant impact on the absorption property.

Not noticing a clear and prevalent peak for the maximum absorbance of the various opsins at their respective wavelengths, as mentioned in Table 1, might be due to various factors such as; type of buffers (affect the pH which in turns alter the absorption) and/or detergents (can cause opsins to bleach) used for membranes purifications, low molar absorptivity (extinction coefficients) of opsins, interference of other light absorbing molecules in the membrane, and the possible low expression of opsins due to utilization of low copy number plasmid pSEVA721 [33,28,34,32].

In conclusion, the results obtained are promising but not sufficiently conclusive to confirm the production of opsins or identify their exact absorbance peaks in the spectra. Therefore, it would be interesting to explore the use of alternative buffers, detergents, and expression vectors to enhance expression capabilities and record the absorption spectra again.

5. Channelrhodopsins screening systems 🔎

---

Channelrhodopsins (CR) are the main class of microbial opsins used in optogenetic therapies for visual restoration [2]. They are cation pumps activated by light and have a depolarizing effect [35,36]. Among CR, several were already used in preliminary genetic therapeutic tests in animals and even humans and we choose to focus on two of the most promising ones CatCh and ChrimsonR.

To establish a screening system to detect their activity, we decided to concentrate our attention on the main cation they transport, which is the Na⁺ and implement in E. coli a Na⁺ biosensor who’s output would be connected to the light activation of opsins. Two different systems schematised in Figure 6 were conceived, tested and described here-after. The first one is based on a Na⁺ riboswitch and the second one on the fluorescent dye CoroNa™ Green.

5.1. Na⁺ riboswitch

---

A recently discovered class of riboswitches exhibits a remarkable ability to selectively sense and respond to sodium ions (Na⁺), and regulate the expression of genes relevant to sodium biology [37]. This riboswitch class, previously referred to as the 'DUF1646 motif', represents a novel addition to the diverse array of riboswitches found in bacteria [38,39]. The Na⁺ riboswitch was identified through bioinformatic analyses, revealing its presence in various bacterial phyla, including Firmicutes, Proteobacteria, Acidobacteria, and Verrucomicrobia [40].

It was predicted to form a distinctive secondary structure with two extended base-paired substructures, which together constitute the ligand-binding aptamer domain of the riboswitch (Figure 7). Experimental evidence demonstrated that the Na⁺ riboswitch selectively binds sodium ions while strongly rejecting other alkali and alkaline earth cations. It exhibits a dissociation constant (K_D) in the low millimolar range, emphasizing its specificity for sodium ions. This selectivity for Na⁺ was confirmed through various biochemical assays [37]. The Na⁺ riboswitch was found to function as a genetic ''ON'' switch in response to Na⁺ binding. It is typically associated with an intrinsic transcription terminator stem, and the binding of Na⁺ prevents the formation of this terminator stem, allowing transcription to proceed, leading to increased gene expression (Figure 7).

Furthermore, it was observed that the riboswitch's response to Na⁺ concentrations was more pronounced at higher pH levels. In alkaline environments, where the cellular concentration of Na⁺ is expected to be reduced due to Na⁺/H⁺ antiporters' action, the riboswitch plays a critical role in adjusting gene expression to maintain sodium homeostasis.

Design

Two distinct versions of the Na⁺ riboswitch, based on the 'DUF1646 motif' present upstream of the kefB gene of Clostridium acetobutylicum [37], were designed and employed to investigate the activity of the channelrhodopsins (Figure 6):

• The Na⁺ riboswitch v1 (BBa_K4601021) includes the key elements of the Na⁺ riboswitch: the ligand-binding aptamer domain and the sequence forming the intrinsic transcription terminator stem (Figure 7). To assess how this riboswitch influences transcription termination and gene expression in response to Na⁺ ions, we placed this sequence upstream of a reporter gene and its corresponding RBS and downstream of the medium strength constitutive promoter (BBa_J23110) for transcription initiation.
• The Na⁺ riboswitch v2 (BBa_K4601022) includes all the v1’s sequence and an additional 49 nucleotides present upstream of it in the genome of C. acetobutylicum. It was this sequence that was tested in vivo in Bacillus subtilis to drive the expression of the lacZ gene encoding the full length sequence of the β-galactosidase [37] and thus assess how the Na⁺ riboswitch influences gene expression in a cellular context in response to environmental factors and cellular conditions. It is to be noted that no B. subtilis promoter sequence was placed upstream of lacZ gene. Thus, as a control of the Na⁺ riboswitch v1, we placed no E. coli promoter either.

Build

The expression cassettes of the two versions of the Na⁺ riboswitch were assembled in the pSB3T5 backbone with one of the reporter genes of sfGFP (BBa_K4601221 and BBa_K4601222), LacZ𝛼 (BBa_K4601231 and BBa_K4601232), the ampicillin resistance gene AmpR (BBa_K4601241 and BBa_K4601242) or chloramphenicol resistance gene CmR (BBa_K4601251 and BBa_K4601252). Thise constructs, when bound to the Na⁺, should allow for the transcription of those reporter genes, while, when there is no Na⁺ binding, no transcription should occur.

Test

The first functional tests were performed in E. coli NEB^® 5-alpha cells carrying the plasmids, in the pSB3T5 backbone, containing the Na⁺ riboswitches v1 and v2 followed by two different reporter genes: sfGFP or chloramphenicol resistance gene CmR.

When the reporter gene was sfGFP, E. coli cells were grown overnight at 37 °C at 200 rpm in 96-deep-well plates with 1 mL of LB (Lennox) supplemented with 10 µg/mL tetracycline. The cells were then diluted by 40 times in LB NaCl-free media with 10 µg/mL tetracycline and, after 4 hours of incubation at 37°C at 200 rpm, they were further diluted by 20 times in media containing 10 µg/mL tetracycline and increasing concentrations of NaCl in an opaque wall 96-well polystyrene microplate (COSTAR 96, Corning). This final media was either LB NaCl-free as such or buffered at pH 6, 7, 8 or 9 with 100 mM MES, PIPES, TAPS or AMPSO, respectively (all buffers were chosen based on the literature [42] and prepared with KOH, not NaOH). We also tested the minimal salts (MS) media composed of 50 mM K₂HPO₄, 20 mM NH₄Cl, 4 mM citric acid, 1 mM MgSO₄, 0.2% glucose, 0.01 µM nitrilotriacetic acid, 3 µM CaCl₂, 3 µM FeCl₃, 1 µM MnCl₂, 0.3 µM ZnCl₂, 0.3 µM H₃BO₃, 0.3 µM CrCl₃, 0.3 µM CoCl₂, 0.3 µM CuCl₂, 0.3 µM NiCl₂, 0.3 µM Na₂MoO₄, 0.3 µM Na₂SeO₃ pH 7.2. The plate was then incubated at 37°C at 200 rpm and the sfGFP fluorescence (λ_excitation 488 nm and λ_emission 530 nm) and the optical density at 600 nm (OD600) were measured every 10 minutes for 24 hours, in a CLARIOstar (BMGLabtech) plate reader. Fluorescence values were normalized by OD600.

When the reporter gene was CmR, E. coli cells were grown overnight at 37 °C at 200 rpm in LB (Lennox) supplemented with 10 µg/mL tetracycline. The cells were then diluted to an OD_600nm of 1, 0.1 and 0.01 in LB NaCl-free and 5 µL of each suspension was spotted onto by LB agar plates containing 5 µg/mL tetracycline and increasing concentrations of chloramphenicol, followed by an overnight incubation at 37°C. Three different types of LB were used: LB NaCl-free, LB Lennox which contains 5 g/L NaCl and LB Luria-Miller which contains 10 g/L NaCl. When indicated, the LB agar was buffered at pH 6, 7, 8 or 9 with 100 mM MES, PIPES, TAPS or AMPSO, respectively (all buffers were chosen based on the literature [42] and prepared with KOH, not NaOH).

Based on the results of these experiments (described below), we decided not to conduct any tests with LacZ𝛼 and AmpR as a reporter genes.

Learn

In the assessment of the Na⁺ riboswitches v1 and v2, we conducted a series of experiments to understand their functionality in E. coli NEB^® 5-alpha cells.

First, we examined the expression of sfGFP, a fluorescent reporter gene, under the control of these riboswitches in different media and at varying pH levels, while gradually increasing the concentrations of NaCl (Figure 8). Surprisingly, we observed no significant changes in sfGFP fluorescence upon the addition of NaCl. This result suggests that the Na⁺ riboswitches v1 and v2 may not have a substantial impact on sfGFP expression under these conditions.

Additionally, we investigated the growth of E. coli NEB^® 5-alpha cells carrying the chloramphenicol acetyltransferase gene under the control of the Na⁺ riboswitches v1 and v2 in the presence of increasing concentrations of chloramphenicol (Figure 9). Notably, all cells appeared to grow similarly, regardless of the riboswitches' influence, indicating that the presence of Na⁺ riboswitches v1 and v2 is not significantly affecting chloramphenicol resistance.

However, it's important to note that our assessment could not be fully explored as, at pH 9, we observed no bacterial growth in liquid cultures, and growth was limited on plates. A similar trend was observed at pH 8, where bacterial growth remained very low. This suggests that the alkaline pH conditions may not be conducive to robust bacterial growth, potentially impacting the assessment of riboswitch functionality. The evaluation of riboswitch functionality at pH levels above 8 was challenging because E. coli exhibited difficulties in growing under such conditions. This limitation prevented a comprehensive assessment of riboswitch behavior at highly alkaline pH values in E. coli.

Moreover, in order to better understand the not significant changes in the behavior of both sfGFP and CmR reporter genes when the Na⁺ riboswitches v1 and v2 were present upstream or not, at pH values lower than 8 or in non buffered LB conditions, we took a closer look at the Na⁺ riboswitches v1 and v2 sequences and analyzed them with state of the art available tools (Figure 10).

In summary, our experiments shed light on the behavior of Na⁺ riboswitches v1 and v2 in E. coli NEB^® 5-alpha cells. While sfGFP fluorescence and chloramphenicol resistance did not appear to be significantly affected by the riboswitches under the tested conditions, the presence of internal promoters and the absence of terminators within the riboswitch sequences may explain the absence of function. Furthermore, challenges associated with bacterial growth at high pH levels expose the experimental limits of testing riboswitches in high pH conditions.

5.2 CoroNa™ Green-based detection of channelrhodopsins’ activity

---

CoroNa™ Green is a chemical compound developed by the company Molecular Probes to be used as a Na⁺ indicator. According to its supplier, “The CoroNa™ Green dye is an improved green-fluorescent sodium (Na⁺) indicator that exhibits an increase in fluorescence emission intensity upon binding Na⁺, with little shift in wavelength”.

For these reasons, we choose to use CoroNa™ Green to evaluate the activity of channelrhodopsins (CR), some of which are light-driven Na⁺ pumps importing sodium ions inside the cell.

Among our opsins of interest, two are CR: CatCh and ChrimsonR.

Design

CoroNa™ Green dye is commercially available in two versions: one which is cell-permeable and another one which is not.

The cell-impermeable version is the one capable of binding the Na⁺ ions and emitting a fluorescence signal, but it cannot penetrate cells which make this version not suitable for our goal of detecting the channelrhodopsin’s activity by evaluating the intracellular Na⁺ content. Moreover, the cell-impermeable version may lead to an increased background due to detection of extracellular Na⁺ ions.

In contrast, the cell-permeable version of CoroNa™ Green dye does not have the same capacity of binding the Na⁺ ions. However, once inside the cell, it can be hydrolysed by cellular esterases and converted into the impermeable version. Thus, fluorescence will be emitted only by intracellular CoroNa™ Green molecules activated by intracellular Na⁺ ions. Release in the environment will only happen upon cell death, which is prevented in the staining protocol by using E. coli growth media throughout the procedure.

CoroNa™ Green is mainly used in the literature in eukaryotic cells and the supplier was unable to provide us with guidance for using it in bacteria. Nevertheless, we found a few publications, all from the same lab, that used CoroNa™ Green in prokaryotes [45,46].

Build

No special genetic constructs were necessary for these tests. We used the above described plasmids expressing the various opsin genes under the control of T7 promoter in the Evolution.T7 system in the pSEVA721 backbone (BBa_K4601200, BBa_K4601201, BBa_K4601202, BBa_K4601203, BBa_K4601204, BBa_K4601205).

Test

The CoroNa™ Green staining was performed according to a protocol adapted from the one described by Morimoto et al. [46]. Briefly, E. coli C43(DE3) cells were grown overnight at 37 °C at 200 rpm in 12 mL tubes with 3 mL of LB (Lennox) supplemented with 10 µg/mL trimethoprim. The cells were then diluted by 100 times in 20 mL of the same media and after 4 hours of incubation at 37°C at 200 rpm, they were induced with 1 mM IPTG and 10 µg/mL all-trans-retinal.

The culture was then split into two and one tube was incubated overnight at 37°C at 200 rpm in dark (covered in aluminum foil) while the other was placed in a shaking incubator equipped with LEDs producing white light with an intensity of 2000 lux at 37°C at 200 rpm. After these overnight incubations, 400 µL of cells were harvested by centrifugation (6000 g for 2 minutes), washed twice in 1 mL of T-broth (1% Bacto tryptone, 10 mM potassium phosphate, pH 7.0) and resuspended in 100 μL of T-broth containing 40 μM CoroNa™ Green (cell permeant version, Molecular Probes C36676) and 10 mM EDTA-KOH pH 8.0.

Tubes were covered in aluminum foil (to keep them in the dark) and after one hour of incubation at room temperature in a tube rotator at 5 rpm, cells were harvested by centrifugation (6000 g for 2 minutes) and washed three times in 1 mL of MS media (composition indicated in chapter n°5.1 on this page). Finally, cells were resuspended in 200 μL MS media and the suspension transferred in an opaque wall 96-well polystryrene microplate (COSTAR 96, Corning). The CoroNa™ Green fluorescence (λ_excitation 488 nm and λ_emission 530 nm) and optical density at 600 nm (OD600) were measured in a CLARIOstar (BMGLabtech) plate reader before and after the addition of NaCl at a concentration of 100 mM. Fluorescence values were normalized by OD600.

Learn

The CoroNa™ Green staining tests were performed in E. coli C43(DE3) carrying cells the various opsins genes under the control of T7 promoter in the Evolution.T7 system in the pSEVA721 backbone (BBa_K4601200, BBa_K4601201, BBa_K4601202, BBa_K4601203, BBa_K4601204, BBa_K4601205) or as control an empty backbone (pSEVA721). The opsins’ expression was induced with IPTG and all-trans-retinal was supplemented into the growth media to allow the formation of a functional protein. Cells were grown in the light or, as control, in the dark, followed by CoroNa™ Green staining.

The results presented in Figure 11, show an increased fluorescent output for E. coli cells expressing the ChrimsonR channelrhodopsin when cells were cultured in the light compared to when they were kept in the dark, the dark values being in the same range of values obtained with the empty backbone as negative control. These results provide clear evidence of the light-dependent Na⁺ import by cells expressing ChrimsonR.

ChrimsonR is the only opsin for which this experiment showed Na⁺ import activity, which is not unexpected. Indeed, Jaws, NpHR and NpHR P240T, F250Y are halorhodopsins that transport Cl^- ions into the cell, while NpSRII is a sensory rhodopsin that does not have an ion transport activity. For these 4 proteins, the obtained results are as expected: not different compared to the negative control. However, for CatCh we were expecting to obtain some fluorescence output, as this channelrhodopsin is also capable of importing Na⁺ ions. Our results indicate that either CatCh was not significantly expressed or that its activity is too low to be detected by this method.

6. Halorhodopsins (and sensory rhodopsins) screening system 🔍

---

After channelrhodopsins, the halorhodopsins (HR) are the second class of microbial opsins used in optogenetic therapies for visual restoration [2]. They are chloride pumps activated by light and have a hyperpolarizing effect [47]. Among HR, two were already used in preliminary genetic therapeutic tests in animals: NpHR and Jaws.

As mentioned above (chapter n°1 on this page), we also used in our project the NpSRII (BBa_K4601004) of N. pharaonis. Sensory rhodopsins are not pumps, but photoreceptors that respond to light and transmit the signal into the cell via a signaling cascade. For this, in N. pharaonis, NpSRI and NpSRII interact with the transducer proteins NpHtrI and NpHtrII, respectively [48]. NpHtrI and NpHtrII are homodimeric membrane proteins that have also a cytoplasmic histidine kinase domain responsible for initiating the signaling cascade that mediates chemotaxis. The interaction between the SR and their cognate Htr transducers are highly specific. However, a mutant halorhodopsin (NpHR P240T, F250Y, BBa_K4601004) was identified that is able to interact with NpHtrII [49].

With the above considerations in mind, and getting inspired by the work of the iGEM Tokyo-NoKoGen 2012 team, we decided to establish a halorhodopsin screening system.

Design

Our halorhodopsin screening system uses fusion proteins composed of the halorhodopsin of interest, the NpHtrII transmembrane domain and the E. coli EnvZ histidine kinase domain (Figure 12). This design is based on several considerations:

• The HR is covalently linked to the NpHtrII transmembrane domain (via a short linker sequence) to ensure their colocalization
• The natural histidine kinase domain of the NpHtrII was removed as it does not specifically interact with an E. coli signaling cascade. It was replaced by an endogenous E. coli histidine kinase domain.
• The E. coli EnvZ was chosen as a histidine kinase domain as it is part of the widely known EnvZ-OmpR two component system [21] (described above on this page in chapter n°3, but also in the Contribution page on this wiki).

In this system (Figure 13), it is expected that, in the presence of light, the halorhodopsin activation is transmitted via protein-protein interaction to the NpHtrII transmembrane domain, which in turn leads to the autophosphorylation of the EnvZ histidine kinase domain and subsequently the phosphorylation of the OmpR transcription factor. The phosphorylated OmpR dimerises and binds to specific sequences present in the promoter regions of the ompC and ompF genes thus upregulating or downregulating their expression, respectively.

Based on this property, we designed several reporters in which the pOmpC promoter controls the expression of sfGFP (BBa_K4601223), LacZ𝛼 (BBa_K4601233), the ampicillin resistance gene AmpR (BBa_K4601243) or chloramphenicol resistance gene CmR (BBa_K4601253).

Opsin activation should lead to the expression of the reporter gene and the detection of either a fluorescent output, a colorimetric one, or the capacity to grow in the presence of ampicillin or chloramphenicol. The latter systems would constitute a good selection tool for the identification of halorhodopsins with improved properties in different conditions of light.

To be functional in E. coli this system requires the knockout of the natural envZ gene. For this we constructed two E. coli ΔenvZ strains as described above on this page in chapter n°3.

To allow the expression of these fusion proteins, but also their evolution using the tool developed by the iGEM Evry Paris-Saclay 2021 team, we designed expression cassettes following the Evolution.T7 architecture (Figure 14) and the same strategy described above (chapter n°4) for individual opsin expression. Thus, five fusion proteins were designed as follows: for the three halorhodopsins used in optogenetic therapies for visual restoration Jaws-NpHtrII-EnvZ (BBa_K4601007), NpHR-NpHtrII-EnvZ (BBa_K4601008) and its mutant NpHR-P240T-F250Y-NpHtrII-EnvZ (BBa_K4601009). As positive control, we designed the NpSRII-NpHtrII-EnvZ (BBa_K4601010) and as negative control the “no opsin” version NpHtrII-EnvZ (BBa_K4601006).

Build

The expression cassettes of the different opsins-NpHtrII-EnvZ fusions (BBa_K4601007, BBa_K4601008, BBa_K4601009, BBa_K4601010) as well as the negative control the “no opsin” version NpHtrII-EnvZ (BBa_K4601006) were assembled by Golden Gate in the pSEVA721 backbone. This backbone was chosen as it has a very low copy-number [32] compatible with the evolution strategy (if a high copy plasmid is employed, many different mutants will coexist in the same cell which will introduce heterogeneity in the selection process).

The different opsin sequences synthesized for the above described experiments, were PCR amplified with primers allowing the introduction of specific type IIS restriction sites (BsaI). The NpSRII and the NpHtrII sequences were recovered by PCR from the genome of N. pharaonis (kindly provided by our host lab) and the EnvZ histidine kinase domain from the E. coli genome.

In parallel, four reporter plasmids were assembled by Golden Gate in the pSB3T5 backbone. They contain the pOmpC promoter followed by different reporter genes: sfGFP (BBa_K4601223), LacZ𝛼 (BBa_K4601233), the ampicillin resistance gene AmpR (BBa_K4601243) or chloramphenicol resistance gene CmR (BBa_K4601253).

Test

The first functional tests were performed in E. coli C43(DE3) or C43(DE3) ΔenvZ (ApraR) cells that were co-transformed with two plasmids. The first plasmid carries either the different opsins-NpHtrII-EnvZ fusions in the pSEVA721 backbone (BBa_K4601007, BBa_K4601008, BBa_K4601009, BBa_K4601010), or as controls either an empty backbone (pSEVA721) or an NpHtrII-EnvZ fusion without the N-terminal opsin (BBa_K4601206). The second (reporter) plasmid, on the pSB3T5 backbone, contains the pOmpC promoter followed by different reporter genes: sfGFP (BBa_K4601223), LacZ𝛼 (BBa_K4601233), the ampicillin resistance gene AmpR (BBa_K4601243) or chloramphenicol resistance gene CmR (BBa_K4601253).

Cells were then grown overnight at 37 °C at 200 rpm in 96-deep-well plates with 1 mL of LB (Lennox) supplemented with 5 µg/mL tetracycline and 5 µg/mL trimethoprim. The cells were then diluted by 40 times in the same media and after 4 hours of incubation at 37°C at 200 rpm, they were further diluted by 20 times in media containing also 10 µg/mL all-trans-retinal and 1 mM IPTG in three opaque wall 96-well polystryrene microplates (COSTAR 96, Corning).

One plate was incubated at 37°C at 200 rpm and the sfGFP fluorescence (λ_excitation 488 nm and λ_emission 530 nm) and optical density at 600 nm (OD600) were measured every 10 minutes for 24 hours, in a CLARIOstar (BMGLabtech) plate reader. The second plate was incubated at 37°C at 200 rpm in dark (covered in aluminum foil) and a third plate also at 37°C at 200 rpm but in a shaking incubator equipped with LEDs producing white light with an intensity of 2000 lux.

When the reporter gene was sfGFP, after the overnight culture, the fluorescence and the OD600 were recorded with the CLARIOstar plate reader. Fluorescence values were normalized by OD600.

When the reporter gene was LacZ𝛼, after the overnight culture, a Miller assay was performed following the single-step method for β-galactosidase assays in E. coli using a 96-well plate reader [50,51]. For this 80 µL of the overnight cultures were mixed with 120 µL of β-gal mix containing 60 mM Na₂HPO₄, 40 mM NaH₂PO₄, 10 mM KCl, 1 mM MgSO₄, 50 mM 2-mercaptoethanol, 0.2 mg/mL lysozyme, 1.1 mg/mL o-nitrophenyl-β-D-galactoside (ONPG), 6.7% PopCultures Reagent pH 7.0. The plate was incubated at 37°C at 200 rpm and the optical densities at 600 nm (OD600) and 420 nm (OD420) were measured every 2 minutes for 2 hours, in a CLARIOstar (BMGLabtech) plate reader with the path correction option turned on. The slope of the variation of OD420 over time (OD420/min) was determined and to calculate the Miller Units, it was multiplied by 1000 and divided by the OD600 and the culture volume used in the assay (80 µL).

When the reporter gene was AmpR or CmR, after the overnight culture, the cell suspension was diluted 100 fold in LB (Lennox) media and 5 µL of the diluted and non diluted culture were deposited on LB (Lennox) agar plates containing 100 µg/mL ampicillin or 35 µg/mL of chloramphenicol respectively. As control, 5 µL of the diluted and non diluted culture were also deposited on plates containing 5 µg/mL tetracycline and 5 µg/mL trimethoprim.

Learn

To assess the functionality of our halorhodopsin screening system, we first evaluated the expression of the different reporter genes in E. coli cells expressing the various opsin-NpHtrII-EnvZ fusions (and the corresponding controls). We chose to perform these tests in the C43(DE3) ΔenvZ (ApraR) strain for several reasons:

• it has increased capacity of expressing membrane proteins [13],
• possess the T7RNAP in its genome, required for the expression of our opsin fusion proteins that are placed under the control of the T7 promoter,
• It has a knockout of the envZ gene. Thus the pOmpC promoter controlling the expression of our reporter genes is not activated by changes in the medium osmolarity as it happens in wild type E. coli strains.

When using sfGFP as a reporter (Figure 15), we observed generally increased Fluorescence/OD600 values when bacteria were cultivated in the dark compared to the cells grown in the light or in the plate reader. These results are unexpected, and we suspect this is most probably due to bleaching of sfGFP by the light. They are thus not conclusive as we are unable to assess the effect of light on the opsins’ activation and the subsequent expression of our reporter gene.

For this reason, we turned to other reporters to eliminate the direct influence of light on the output.

When using LacZ𝛼 as a reporter and the process of in vivo alpha complementation followed by a colorimetric Miller assay, the results presented in Figure 16 show no significant difference between the negative control (the empty backbone) and the cells carrying the various opsins fusions and this regardless of the presence or absence of light. As a positive control of our assay, we also evaluated the β-galactosidase activity using a LacZ𝛼 constitutive expression cassette (BBa_K4601234). The values were higher, but not as high as those obtained with the the C43(DE3) strain expressing the wild type β-galactosidase, indicating sub effective alpha complementation in the C43(DE3) ΔenvZ (ApraR) Δ(lacZ)M15 cells.

When using AmpR or CmR as reporters, the growth of E. coli cells in the presence or absence of the corresponding antibiotic was evaluated. Again, we observed no difference in the behavior of the different constructs. As shown in Figure 17, all constructs, including the positive and negative controls were able to grow. A leaky expression, even at very low levels, of the pOmpC promoter could lead to the expression of the AmpR and CmR resistance genes. As both the resistance mechanisms involve modification of the antibiotic (and thus its inactivation) by the enzyme encoded by the resistance gene, even a low expression could lead to the appearance of a resistance phenotype to the corresponding antibiotic.

As control, we also tested the growth of E. coli cells carrying the other reporters (sfGFP or LacZ𝛼) under the control of the pOmpR promoter. No growth was observed in these cases (data not shown).

Based on these four experiments, and notably on the alpha complementation assay, we conclude that the halorhodopsin screening system is not functional in E. coli as such. This may be due to the low expression of the fusion proteins (as described above in chapter n°4 also for the opsins expression alone), most probably due to the very low copy number of the backbone. The RBS strength may also be responsible and increasing it is an alternative for boosting the expression.

In addition, these preliminary tests allowed us to eliminate sfGFP, AmpR, and CmR systems from the list of potential selection markers of opsin activity because of light sensitivity (as in the case of sfGFP) or because their leaky expression is enough to lead to antibiotic resistance (as in the case of AmpR and CmR).

🎲 7. Evolution of ChrimsonR

---

Having established that the CoroNa™ Green staining is a good tool for detecting a light-dependent increase in the sodium ions content of a cell carrying the ChrimsonR expression cassette, we moved forward and started the experiments for the evolution of ChrimsonR using the Evolution.T7 in vivo directed evolution tool developed by the iGEM Evry Paris-Saclay 2021 team.

Design

Evolution of opsin sequences with the aim to select enhanced variants is the objective of our project. With this goal in mind, all our expressing cassettes were designed from the start to express opsins in a genetic configuration compatible with the Evolution.T7 tool (Figure 2).

For the evolution of ChrimsonR, the MG1655^* Δflu ΔpyrF Δung Δnfi mutagenesis dedicated strain was transformed with the plasmid carrying the ChrimsonR expression cassette and an equimolar mix of all BD-T7RNAP and BD-T7RNAP_CGG-R12-KIRV mutagenic plasmids of the Evolution.T7. This choice was made in order to increase the chances of obtaining enhanced variants. Indeed, in 2021, the iGEM Evry Paris-Saclay team established that some base deaminases are more potent then others. While one may be tempted to use the strongest mutator domains, a too high mutagenic rate may quickly lead only to inactive variants. For this reason, we decided to use also only the forward or only the reverse mutators: BD-T7RNAP or BD-T7RNAP_CGG-R12-KIRV, respectively. This strategy also limits the burden imposed on the cell by the presence of three plasmids and eliminates the risk of collision between the two polymerases moving in opposite directions on the DNA.

After the in vivo generation of ChrimsonR mutants with the Evolution.T7, cells were stained with CoroNa™ Green as before, and sorted using Fluorescence-Activated Cell Sorting (FACS), a technique that allows isolation of cells having the desired fluorescence intensity.

In our case, we expect that the enhanced ChrimsonR variants would be able to have an increased activity in low light intensity conditions compared to the parental protein, thus importing more Na⁺ ions and, as a consequence, display increased fluorescence upon CoroNa™ Green staining.

Build

No special genetic constructs were necessary for the tests. We used the above described plasmid expressing the ChrimsonR gene under the control of T7 promoter in the Evolution.T7 system in the pSEVA721 backbone (BBa_K4601201).

The Evolution.T7 mutagenic plasmids were kindly provided by the iGEM Evry Paris-Saclay 2021 team.

Test

E. coli MG1655^* Δflu ΔpyrF Δung Δnfi were first co-transformed with the plasmid expressing the ChrimsonR gene under the control of T7 promoter in the Evolution.T7 system in the pSEVA721 backbone (BBa_K4601201) and with a mix of all the BD-T7RNAP and BD-T7RNAP_CGG-R12-KIRV mutagenic plasmids of the Evolution.T7 in pSEVA221 and pSEVA471 backbones respectively. Control transformations were performed with the plasmids expressing the T7RNAP and the T7RNAP_CGG-R12-KIRV non mutagenic plasmids or the pSEVA221 and pSEVA471 empty backbones.

Transformed cells were selected on LB (Luria-Miller) media containing 5 µg/mL trimethoprim, 12.5 µg/mL kanamycin and 25 µg/mL spectinomycin, then grown overnight at 37 °C at 200 rpm in 12 mL tubes with 3 mL of LB (Luria-Miller) supplemented with the same antibiotics. The cells were then diluted by 100 times in 20 mL of the same media and after 4 hours of incubation at 37°C at 200 rpm, they were induced with 200 ng/µL anhydrotetracycline, 1.5 mM L-arabinose and 10 µg/mL all-trans-retinal. The culture was then split into two and one tube was incubated overnight at 37°C at 200 rpm in dark (covered in aluminum foil) while the other was placed in a shaking incubator equipped with LEDs producing white light with an intensity of 2000 lux at 37°C at 200 rpm. After these overnight incubations, 400 µL of cells were stained with CoroNa™ Green as described above (in chapter n°5.2 on this page).

Flow cytometry analysis was performed on 50 µL of the cell suspension diluted 100 fold in MS media (composition indicated in chapter n°5.1 on this page) supplemented or not with 100 mM NaCl using the MACSQuant^® Analyzer 16 (Miltenyi Biotec).

FACS experiments were performed on 750 µL of the cell suspension diluted 100 fold in MS media supplemented with 100 mM NaCl using the MACSQuant^® Tyto^® cell sorter (Miltenyi Biotec) using a single laser operating at 488 nm for excitation and two bandpass filters PE-H 585/40 nm and FITC-H 525/50 nm. The selection was triggered by fluorescence intensity.

Cells having a fluorescence above average were isolated and a fraction was subsequently re-analyzed by flow cytometry, while the rest were spread on LB agar plates containing 10 µg/mL trimethoprim. A total of 48 isolated colonies that have grown on these plates were selected, 12 for each of the four FACS experiments (Table 2). Afterwards, the plasmid DNA was extracted and transformed into E. coli C43(DE3) cells for clonal selection and ChrimsonR expressing. A colony was further selected, and cells were subject to CoroNa™ Green staining as described above (in chapter n°5.2 on this page). The plasmid DNA contained in "interesting" colonies was extracted from the C43(DE3) cells and sequenced.

Learn

Prior to FACS, a flow cytometry analysis was performed in order to analyze at single cell level the cell suspensions of ChrimsonR expressing the variants evolved in different conditions of presence or absence of the different Evolution.T7 mutagenic plasmids.

Bacteria were identified by their size and granularity based on the side scatter (SSC) and the forward scatter (SSC) values as illustrated in Figure 18 A & C, then the fluorescent cells were counted based on the B1(FITC) values (Figure 18 B & D).

The results of these quantifications are presented in Figure 19. The higher number of fluorescent cells were observed in the positive control containing the wild type ChrimsonR protein expressed by the T7RNAP_CGG-R12-KIRV, followed by the population of variants obtained in the corresponding conditions with the BD-T7RNAP_CGG-R12-KIRV mutagenic plasmids. The T7RNAP_CGG-R12-KIRV is the mutant T7RNAP specific to the mutant T7_CGG promoter that is placed in reverse orientation compared to the ChrimsonR gene. However, it retains partially its capacity to recognise the wild type T7 promoter [12] and this can explain the expression of ChrimsonR. We also observe that the wild type T7RNAP is not efficient to drive the expression of ChrimsonR, the number of fluorescent cells being low, in the same range as the negative controls. This difference may be attributed to different promoter strengths that control the two RNA polymerases (pTetA for T7RNAP and pBad for T7RNAP_CGG-R12-KIRV), to the efficiency of induction of their expression (by anhydrotetracycline and L-arabinose respectively), but also by the copy number of the plasmids encoding them 1-3 copies / cell for pSEVA221 and 3-5 copies / cell for pSEVA471 [32].

When comparing the light versus dark growing conditions, we observe a higher number of fluorescent cells in the light conditions for the wild type ChrimsonR. These results are in line with what we observed previously when we stained with CoroNa™ Green the C43(DE3) cells expressing various opsins and measured the fluorescence/OD600 values (Figure 11, chapter n°5.2 on this page). These results provide clear evidence of the light-dependent Na⁺ import by cells expressing ChrimsonR.

This same trend was not observed upon analysis of the library of ChrimsonR variants produced with the Evolution.T7 tool. A drop in the fluorescent output is nevertheless not unexpected as a mutation is generally detrimental for the protein activity, while gain of function mutants are rare.

In order to find these rare gain of function mutants, fluorescence-activated cell sorting (FACS) was used on the cell suspensions of ChrimsonR mutant libraries having the highest number of intense fluorescent cells (Figure 20 and Table 2).

Cells having a fluorescence above average were isolated and subsequently re-analyzed by flow cytometry (Figure 21) and plated on LB agar plates containing 10 µg/mL trimethoprim. Moreover, a fraction of the sorted cells were subject to a second FACS for further enrichment.

As a control, we used an E. coli culture constitutively expressing sfGFP in the pSB3T5 backbone (BBa_K2675056). In this case the percentage of fluorescent cells determined by flow cytometry was very high reaching 99.4%, confirming the accuracy of our experimental setup. It’s worth noting that the ChrimsonR selected cells reached a fluorescence level compared to the one of sfGFP expressing E. coli cells.

	E. coli cells carrying the ChrimsonR variants
	ChrimsonR + BD-T7RNAPCGG-R12-KIRV (dark) = code TS-D	ChrimsonR + BD-T7RNAPCGG-R12-KIRV (light) = code TS-L	ChrimsonR + BD-T7RNAPCGG-R12-KIRV (resorted) = code TS-DL	ChrimsonR + BD-T7RNAP + BD-T7RNAPCGG-R12-KIRV (light) = code TKS-L	sfGFP
Triggered events	8049000	11020000	707300	7442000	1749000
Gated events	17654	12840	2292	5578916	423900
Sorted events	5931	10850	773	2190	163300
Aborted events	108	733	5	25	8617
Gated / Triggered events (%)	0.22%	0.12%	0.32%	74.97%	24.24%
Sorted / Triggered events (%)	0.07%	0.10%	0.11%	0.03%	9.34%
Sorted / Gated events (%)	33.60%	84.50%	33.73%	0.04%	38.52%

In the next step, we selected a total of 48 clones (12 from each FACS experiment listed in Table 2) to individually characterize their activity based on CoroNa™ Green staining followed by Fluorescence/OD600 measurements in a plate reader. For this, first the mutant plasmids were transferred in C43(DE3) cells to allow on one hand a better expression of the opsins than in the K12 strains, but also to achieve a better clonal isolation. Indeed, the mutants are carried by the pSEVA721 backbone which has a very low copy number with between 1 and maximum 3 copies per cell [32], still inside a cell selected by FACS, the possibility of having a mix of different mutated plasmids per cell exist.

CoroNa™ Green staining of these 48 clones (Figure 22), did not reveal any obvious gain of function mutant. In addition, for unexplained reasons, we observed an increased fluorescent output for E. coli cells expressing the wild-type ChrimsonR channelrhodopsin when cells were cultured in the dark compared to when they were kept in the light, contrary to our first results presented above in Figure 11. This trend was also observed for 34 out of the 48 selected clones, while 14 clones showed the profile we were expecting, meaning an increased fluorescence in the presence of light. Nevertheless, in all cases (light or dark) the presence of Na⁺ led to an increased fluorescence/OD600 only when the ChrimsonR channelrhodopsin was expressed, in line with its activity of Na⁺ inward pump.

Further, we selected 20 clones having either an increased fluorescence/OD600 compared to wild-type ChrimsonR or an increased fluorescence/OD600 in the presence of light compared to dark. Their sequence analysis revealed that 16 of them contained the gene coding for a wild-type ChrimsonR channelrhodopsin albeit sometimes with synonymous mutations. One clone contained a non sens mutation, thus explaining the very low fluorescence/OD600 values in the same range of those obtained with the empty backbone as negative control. Finally, three clones expressed ChrimsonR variants with up to 3 mutated amino acids. Additional analyses are required to address the effect of these mutations on ChrimsonR activity.

Conclusions

---

We have successfully produced microbial opsins in our E. coli chassis and analyzed their activity by several techniques.

In parallel, we designed three independent approaches with the aim to develop a high throughput screening system for channelrhodopsins, halorhodopsins and sensory rhodopsins light-driven activity.

We demonstrate the efficiency of the CoroNa™ Green fluorescent dye to stain E. coli cells expressing the ChrimsonR channelrhodopsin when grown under low light intensity conditions. Moreover, we successfully used the Evolution.T7 tool originally developed by the iGEM Evry Paris-Saclay 2021 team to generate a library of ChrimsonR variants to which we applied fluorescence-activated cell sorting (FACS) to isolate cells exhibiting increased fluorescence upon CoroNa™ Green staining, an indicator of increased channelrhodopsin activity.

This paves the way for the discovery of novel microbial opsin sequences with enhanced sensitivity, and the desired wavelength shift in their absorption spectrum. This will overcome the limitations of existing bacterial opsins that tend to exhibit suboptimal sensitivity, narrow absorption spectra, and limited efficiency in converting light to an electrical signal for proper visual processing.

References

---

• [1] Simon C-J, Sahel J-A, Duebel J, Herlitze S, Dalkara D. Opsins for vision restoration. Biochemical and Biophysical Research Communications (2020) 527: 325–330.
• [2] Sakai D, Tomita H, Maeda A. Optogenetic therapy for visual restoration. International Journal of Molecular Sciences (2022) 23: 15041.
• [3] Zhang F, Vierock J, Yizhar O, Fenno LE, Tsunoda S, Kianianmomeni A, Prigge M, Berndt A, Cushman J, Polle J, Magnuson J, Hegemann P, Deisseroth K. The microbial opsin family of optogenetic tools. Cell (2011) 147: 1446–1457.
• [4] Kleinlogel S, Feldbauer K, Dempski RE, Fotis H, Wood PG, Bamann C, Bamberg E. Ultra light-sensitive and fast neuronal activation with the Ca²⁺-permeable channelrhodopsin CatCh. Nature Neuroscience (2011) 14: 513–518.
• [5] Chaffiol A, Caplette R, Jaillard C, Brazhnikova E, Desrosiers M, Dubus E, Duhamel L, Macé E, Marre O, Benoit P, Hantraye P, Bemelmans A-P, Bamberg E, Duebel J, Sahel J-A, Picaud S, Dalkara D. A new promoter allows optogenetic vision restoration with enhanced sensitivity in macaque retina. Molecular Therapy: The Journal of the American Society of Gene Therapy (2017) 25: 2546–2560.
• [6] Klapoetke NC, Murata Y, Kim SS, Pulver SR, Birdsey-Benson A, Cho YK, Morimoto TK, Chuong AS, Carpenter EJ, Tian Z, Wang J, Xie Y, Yan Z, Zhang Y, Chow BY, Surek B, Melkonian M, Jayaraman V, Constantine-Paton M, Wong GK-S, Boyden ES. Independent optical excitation of distinct neural populations. Nature Methods (2014) 11: 338–346.
• [7] GenSight Biologics. A phase 1/2a, open-label, non-randomized, dose-escalation study to evaluate the safety and tolerability of GS030 in subjects with Retinitis Pigmentosa. clinicaltrials.gov (2022).
• [8] Chuong AS, Miri ML, Busskamp V, Matthews GAC, Acker LC, Sørensen AT, Young A, Klapoetke NC, Henninger MA, Kodandaramaiah SB, Ogawa M, Ramanlal SB, Bandler RC, Allen BD, Forest CR, Chow BY, Han X, Lin Y, Tye KM, Roska B, Cardin JA, Boyden ES. Noninvasive optical inhibition with a red-shifted microbial rhodopsin. Nature Neuroscience (2014) 17: 1123–1129.
• [9] Imburgio D, Rong M, Ma K, McAllister WT. Studies of promoter recognition and start site selection by T7 RNA polymerase using a comprehensive collection of promoter variants. Biochemistry (2000) 39: 10419–10430.
• [10] Temme K, Hill R, Segall-Shapiro TH, Moser F, Voigt CA. Modular control of multiple pathways using engineered orthogonal T7 polymerases. Nucleic Acids Research (2012) 40: 8773–8781.
• [11] Moore CL, Papa LJ, Shoulders MD. A processive protein chimera introduces mutations across defined DNA regions in vivo. Journal of the American Chemical Society (2018) 140: 11560–11564.
• [12] Meyer AJ, Ellefson JW, Ellington AD. Directed evolution of a panel of orthogonal T7 RNA polymerase variants for in vivo or in vitro synthetic circuitry. ACS synthetic biology (2015) 4: 1070–1076.
• [13] Miroux B, Walker JE. Over-production of proteins in Escherichia coli mutant hosts that allow synthesis of some membrane proteins and globular proteins at high levels. Journal of Molecular Biology (1996) 260: 289–298.
• [14] Álvarez B, Mencía M, de Lorenzo V, Fernández LÁ. In vivo diversification of target genomic sites using processive base deaminase fusions blocked by dCas9. Nature Communications (2020) 11: 6436.
• [15] Jensen KF. The Escherichia coli K-12 “wild types” W3110 and MG1655 have an rph frameshift mutation that leads to pyrimidine starvation due to low pyrE expression levels. Journal of Bacteriology (1993) 175: 3401–3407.
• [16] van der Woude MW, Henderson IR. Regulation and function of Ag43 (flu). Annual Review of Microbiology (2008) 62: 153–169.
• [17] Lee JK, Houk KN. A proficient enzyme revisited: the predicted mechanism for orotidine monophosphate decarboxylase. Science (New York, N.Y.) (1997) 276: 942–945.
• [18] Pearl LH. Structure and function in the uracil-DNA glycosylase superfamily. Mutation Research (2000) 460: 165–181.
• [19] Guo G, Ding Y, Weiss B. nfi, the gene for endonuclease V in Escherichia coli K-12. Journal of Bacteriology (1997) 179: 310–316.
• [20] Vik ES, Nawaz MS, Strøm Andersen P, Fladeby C, Bjørås M, Dalhus B, Alseth I. Endonuclease V cleaves at inosines in RNA. Nature Communications (2013) 4: 2271.
• [21] Mizuno T, Mizushima S. Signal transduction and gene regulation through the phosphorylation of two regulatory components: the molecular basis for the osmotic regulation of the porin genes. Molecular Microbiology (1990) 4: 1077–1082.
• [22] Datsenko KA, Wanner BL. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proceedings of the National Academy of Sciences of the United States of America (2000) 97: 6640–6645.
• [23] Trottier K. Exploration d’acides ribonucléiques catalytiques et implémentation d’une fonction essentielle dans Escherichia coli. (2019).
• [24] Herring CD, Glasner JD, Blattner FR. Gene replacement without selection: regulated suppression of amber mutations in Escherichia coli. Gene (2003) 311: 153–163.
• [25] Reis AC, Salis HM. An automated model test system for systematic development and improvement of gene expression models. ACS synthetic biology (2020) 9: 3145–3156.
• [26] Farasat I, Kushwaha M, Collens J, Easterbrook M, Guido M, Salis HM. Efficient search, mapping, and optimization of multi-protein genetic systems in diverse bacteria. Molecular Systems Biology (2014) 10: 731.
• [27] Ng CY, Farasat I, Maranas CD, Salis HM. Rational design of a synthetic Entner-Doudoroff pathway for improved and controllable NADPH regeneration. Metabolic Engineering (2015) 29: 86–96.
• [28] Maresca JA, Keffer JL, Miller KJ. Biochemical analysis of microbial rhodopsins. Current Protocols in Microbiology (2016) 41: 1F.4.1-1F.4.18.
• [29] Scharf B, Engelhard M. Blue halorhodopsin from Natronobacterium pharaonis wavelength regulation by anions. Biochemistry (1994) 33: 6387–6393.
• [30] Ren L, Martin CH, Wise KJ, Gillespie NB, Luecke H, Lanyi JK, Spudich JL, Birge RR. Molecular mechanism of spectral tuning in sensory rhodopsin II. Biochemistry (2001) 40: 13906–13914.
• [31] Keffer JL, Hahn MW, Maresca JA. Characterization of an unconventional rhodopsin from the freshwater actinobacterium Rhodoluna lacicola. Journal of Bacteriology (2015) 197: 2704–2712.
• [32] Jahn M, Vorpahl C, Hübschmann T, Harms H, Müller S. Copy number variability of expression plasmids determined by cell sorting and Droplet Digital PCR. Microbial Cell Factories (2016) 15: 211.
• [33] Snodderly DM. Reversible and irreversible bleaching of rhodopsin in detergent solutions. Proceedings of the National Academy of Sciences of the United States of America (1967) 57: 1356–1362.
• [34] McCarren J, DeLong EF. Proteorhodopsin photosystem gene clusters exhibit co-evolutionary trends and shared ancestry among diverse marine microbial phyla. Environmental Microbiology (2007) 9: 846–858.
• [35] Deisseroth K, Hegemann P. The form and function of channelrhodopsin. Science (New York, N.Y.) (2017) 357: eaan5544.
• [36] Govorunova EG, Sineshchekov OA, Spudich JL. Emerging diversity of channelrhodopsins and their structure-function relationships. Frontiers in Cellular Neuroscience (2021) 15: 800313.
• [37] White N, Sadeeshkumar H, Sun A, Sudarsan N, Breaker RR. Na⁺ riboswitches regulate genes for diverse physiological processes in bacteria. Nature Chemical Biology (2022) 18: 878–885.
• [38] McCown PJ, Corbino KA, Stav S, Sherlock ME, Breaker RR. Riboswitch diversity and distribution. RNA (New York, N.Y.) (2017) 23: 995–1011.
• [39] Breaker RR. The biochemical landscape of riboswitch ligands. Biochemistry (2022) 61: 137–149.
• [40] Weinberg Z, Lünse CE, Corbino KA, Ames TD, Nelson JW, Roth A, Perkins KR, Sherlock ME, Breaker RR. Detection of 224 candidate structured RNAs by comparative analysis of specific subsets of intergenic regions. Nucleic Acids Research (2017) 45: 10811–10823.
• [41] Kerpedjiev P, Hammer S, Hofacker IL. Forna (force-directed RNA): Simple and effective online RNA secondary structure diagrams. Bioinformatics (Oxford, England) (2015) 31: 3377–3379.
• [42] Stancik Lauren M., Stancik Dawn M., Schmidt Brian, Barnhart D. Michael, Yoncheva Yuliya N., Slonczewski Joan L. pH-Dependent expression of periplasmic proteins and amino acid catabolism in Escherichia coli. Journal of Bacteriology (2002) 184: 4246–4258.
• [43] LaFleur TL, Hossain A, Salis HM. Automated model-predictive design of synthetic promoters to control transcriptional profiles in bacteria. Nature Communications (2022) 13: 5159.
• [44] Naville M, Ghuillot-Gaudeffroy A, Marchais A, Gautheret D. ARNold: a web tool for the prediction of Rho-independent transcription terminators. RNA biology (2011) 8: 11–13.
• [45] Minamino T, Morimoto YV, Hara N, Aldridge PD, Namba K. The bacterial flagellar type III export gate complex is a dual fuel engine that can use both H⁺ and Na⁺ for flagellar protein export. PLoS pathogens (2016) 12: e1005495.
• [46] Morimoto YV, Namba K, Minamino T. Bacterial intracellular sodium ion measurement using CoroNa Green. Bio-Protocol (2017) 7: e2092.
• [47] Engelhard C, Chizhov I, Siebert F, Engelhard M. Microbial halorhodopsins: light-driven chloride pumps. Chemical Reviews (2018) 118: 10629–10645.
• [48] Jung KH, Spudich EN, Trivedi VD, Spudich JL. An archaeal photosignal-transducing module mediates phototaxis in Escherichia coli. Journal of Bacteriology (2001) 183: 6365–6371.
• [49] Sudo Y, Yamabi M, Kato S, Hasegawa C, Iwamoto M, Shimono K, Kamo N. Importance of specific hydrogen bonds of archaeal rhodopsins for the binding to the transducer protein. Journal of Molecular Biology (2006) 357: 1274–1282.
• [50] Schaefer J, Jovanovic G, Kotta-Loizou I, Buck M. Single-step method for β-galactosidase assays in Escherichia coli using a 96-well microplate reader. Analytical Biochemistry (2016) 503: 56–57.
• [51] Schaefer J, Jovanovic G, Kotta-Loizou I, Buck M. A data comparison between a traditional and the single-step β-galactosidase assay. Data in Brief (2016) 8: 350–352.