A Bireporter Vector System for Assaying Translational Activity of Regulatory Elements


 BackgroundMolecular biology has always shown some similarities with computer science. So, considering transient expression, one can see an analogy with a DDoS attack on a computing system. Like the DDoS attack, transient expression can carry a payload. In particular, analysis of the structure of cell mechanisms and signal amplification in the study of very subtle mechanisms of regulation.ResultsA new vector system for transient expression in plants is described; this system is intended for quantitative analysis of the contribution of regulatory elements to transcription and translation efficiencies. The proposed vector comprises two expression cassettes carrying reporter genes (of the Clostridium thermocellum thermostable lichenase and E. coli


Introduction
The recent advance in the technologies for obtaining omic data has allowed for accumulation of tremendous array of information. Correspondingly, molecular biologists frequently need experimental verification of the biomolecular data. A wide range of reporter systems working in most different organisms has been designed for this purpose. For plants, the most relevant method is transient expression-agroinfiltration; it consists in the transfer of a large number of copies of vectors via the infiltration of the agrobacteria (Agrobacterium tumefaciens) carrying these vectors to the mesophyll of a model plant. The apparent advantages of this approach are rapidness, simplicity, and availability. The vector systems optimized for the transient expression in plants commonly have the following specific features: the absence of a selective marker, the presence of silencing suppressor genes, and the fact that the reporter genes code for rapidly maturing proteins. This tool makes it possible to get precise data on the expression of target polypeptides in plants. Nonetheless, transient expression is the approach that depends on manifold parameters. A considerable set of factors, such as plant age and the conditions of plant growth, can add noise to the corresponding data [1]. In order to level the dependence of expression level on the uncontrolled conditions, we propose using the bireporter vector pLAUMe, carrying the genes of two reporter proteins, namely, Clostridium thermocellum thermostable lichenase (LicBM3) [2] and E. coli β-glucuronidase (GUS) [3,4], which have shown a good performance as reporter proteins with different sensitivities being simple in use. This approach allows for a concurrent assessment of the expression levels of both the reporter gene under the control of a tested element and the gene being a kind of internal control. Thus, the presence of internal control module allows for assessment of the expression level of a gene from the target cassette despite the changes in external conditions. In addition, this makes it possible to avoid co-transformation, which considerably simplifies the experiment. The main advantages of the proposed vector system include simplification of the analysis of regulatory elements (no need in co-transformation) and a direct quantification of reporter proteins.
In this work, we have designed and tested a vector system for assessing the efficiency of translation enhancers. Known and well-characterized enhancers were used for the analysis. In addition, we propose a new method for quantification of the effect of individual regulatory elements on expression based on assessment of enzyme activities of reporter proteins. We have demonstrated that it is feasible to measure not only a relative amount, but also an absolute amount of reporter proteins and, as a consequence, to quantify the contribution of each enhancer to transcription and translation. The general plot of the experiment is shown om the Figure 1.

Plant growing
Nicotiana benthamiana plants were used in the work; the plants were grown on mineral wool using Knop's solution according to the earlier described protocol [5]: photoperiod, 14 h of light/10 h of darkness and illumination, 130-150 µEm -2 s -1 . The plants aged 6 weeks were used for agroinfiltration.

Bacterial strains
The A. tumefaciens strain GV3101 transformed with individual constructs was grown for 48 h in the LB medium supplemented with 50 mg/L rifampin, 50 mg/L carbenicillin, and 100 mM of acetosyringone. In order to obtain a standardized lichenase preparation, E. coli strain XL1 Blue was transformed with the vector pQE30-LicBM3 (earlier developed by the team of the authors [6]).
Agroinfiltration Agroinfiltration followed the earlier described protocol [9]: A. tumefaciens cells of an overnight culture were centrifuged and suspended in the infiltration buffer (10 mM MES pH 5.5, 10 mM MgSO 4 , and 100 mM acetosyringone). For a typical assay, the leaves of greenhouse-grown N. benthamiana plants were infiltrated with the Agrobacterium mixture (50 mL/leaf) using a syringe without a needle. After the infiltration, the plants were further grown under greenhouse conditions. All experiments were performed in four to six replicates.

Analyzing reporter gene transcription
The transcription of reporter genes was assessed using quantitative PCR in one experiment in five biological replicates and tree technical replicates.
Total RNA from the majority of samples was extracted using TRIzol reagent (Evrogen, Russia) according to the manufacturer's protocol. Prior to cDNA synthesis, RNA was treated with RNase-free DNase I (Thermo Scientific, United States) according to the manufacturer's protocol to ensure no DNA contamination; then, the first-strand cDNA synthesis was carried out with approximately 2 µg RNA using a Maxima H Minus First Strand cDNA Synthesis Kit (Thermo Scientific, United States) and oilgo-dT primers according to the manufacturer's protocol. The primers were designed using PrimerBLAST (Table 1) with melting temperatures of 60 C and amplicon lengths of 159 and 143 bp, respectively.
RT-qPCR was conducted in an Applied Biosystems QuantStudio 5 (Thermo Scientific, United States) using qPCRmix-HS SYBR (Evrogen, Russia). The reactions were performed in a total volume of 20 µl of the reaction mixture containing 1µl of the template, 5 µl of 5× SYBR mix, 1 µl of each specific primer to a final concentration of 200 nM under the following conditions: initial denaturation at of 95 C for 180 s followed by two-step thermal cycling profile of denaturation at 95 C for 15 s, and 40 cycles of combined primer annealing/extension at 60 C for 30 s. Notemplate controls were included for each primer pair and each PCR reaction was 1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60  61  62  63  64  65 completed in triplicate. To verify the specificity of the amplicon for each primer pair, a melting curve analysis was performed ranging from 60 to 95 Cw i t ht h e temperature increasing steps of 1.6 C /s at the end of each PCR run.
Constructing standard curves for copy number determination and absolute quantification The pLAUMe (described above) vector carrying the uid A and licBM3 genes was used as a standard. The standard sample was tenfold diluted to cover the concentration range of 0.2 to 200 ng/15 µL. The absolute quantitative assay was performed using the Design Analysis Software v. 2.5.1 and Standard Curve v. 1.5.1 (Thermo Scientific, United States).

Preparing lichenase standard
An overnight E. coli strain XL1-Blue (Stratagene, United States) culture carrying the earlier produced vector pQE-LicBM3 [6] was diluted (1 : 50) with LB medium (Amresco, United States) and grown at 37 C to an OD 600 of 0.5. Then, the gene expression was induced with 1 mM isopropyl-β-D-1-thiogalactopyranoside (IPTG) to grow the culture at 28 C for 48 h. The cells were separated from the medium by centrifugation for 15 min at 3160×g, washed twice with 50 mM Tris-HCl buffer pH 8.0, and suspended in the buffer containing 50 mM Tris-HCl pH 8.0, 10 mM EDTA, 0.1% Triton X-100, 5 mM DTT, 0.01% SDS, and 10 mM NaCl. The cells were incubated at 65 C for 30 min and clarified by centrifugation for 30 min at 16 000×g and 4 C. The supernatant was purified on a HisTrap HP column according to the manufacturer's protocol (GE Healthcare, 17-5247-01). The eluted proteins were dialyzed against 5 mM Tris-HCl pH 8.0 at 4 C and the purified thermostable lichenase protein was diluted in the buffer containing 20 mM Tris-HCl pH 7.4, 0.1 mM EDTA, 1 mM DTT, 200 µg/mL BSA, 50% glycerol, and 100 mM KCl to a final concentration of 1 µg/µL to use in the further experiments. The protein amount in preparations was determined using bicinchoninic acid (Sigma, United States) [10]. The proteins were separated by 12% SDS-PAGE according to Laemmli [11]. The molecular weight of proteins was determined using a Thermo Scientific PageRuler Unstained Protein Ladder (Thermo Fisher Scientific, Inc., United States). See Supplementary Materials ( Figure S1) for electrophoretic pattern.
Preparing plant protein lysates The leaves sampled from N. benthamiana plants on days 4-7 after agroinfiltration were pulverized in liquid nitrogen to a fine powder. Each powdered sample was suspended in three volumes of the 1× PBS containing 0.5% Triton X-100 and incubated for 15 min at 4 C and for 15 min at 50 C. Cell debris was removed by centrifuging twice for 5 min at 16 000×g. The concentration of the samples was adjusted with 1× reaction buffer. Translational activities of the reporter genes were measured in two independent experiments (eight to ten biological replicates each).

Quantification of LicBM3 lichenase
For this purpose, lichenan at a concentration of 125 µg/mL (if not stated otherwise) and Congo red solution at a final concentration of 0.005% were used. The fluorescence was assessed in a Synergy H1 (BioTek, United States) multimode microplate reader using 96-well microtiter plates [12].
Statistical data processing and analysis For this purpose, we wrote a special Python [13] script using several libraries, namely, pandas [14] for table data; NumPy [15] for data arrays; SciPy [16] for statistical data processing; math for mathematical functions; and seaborn and matplotlib [17,18] for data visualization. Data processing comprises the following stages:

Model of vector system
Two reporter systems were united in the designed vector, namely, the uid A gene under the control of arabidopsis actin promoter, which acts as the internal control module, and the licBM3 gene under the control of enhanced CaMV 35S RNA promoter, the test module for assessing the contribution of the studied regulatory elements ( Figure 2). In addition, the vector carries the cassette with the tombusvirus p19 silencing suppressor under the control of arabidopsis translationally controlled tumor protein promoter [19]. Being united in one vector molecule, these structural modules make it possible to avoid the co-transformation with individual vectors. The castor bean catalase intron and potato ST-LS1 intron were integrated into the uid A and licBM3 genes, respectively, to prevent unauthorized expression in the intermediate prokaryotic systems. See Supplementary materials ( Figure S2) for the detailed map of the vector. The vector thus constructed was tested in N. benthamiana (the corresponding data are shown below with the results of testing of the remaining vectors). To verify the designed test system, we decided to use the already described and characterized enhancers (listed in table). Note that   1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60  61  62  63  64  65 for better coverage, we selected four short plant enhancers contrasting in their expression levels (deletion variants AT30, At65, AT100, and AT208: AT5G46430, AT1G07260, AT1G67090, AT1G58420 of A. thaliana, respectively [19]), one long plant enhancer (GGR, geranyl-geranyl reductase enchander) [20], and two synthetic enhancers (SynJ and SynM (MsynJ and SynM) [21]. The enhancers were integrated using CPEC method [22], which allowed for a seamless integration of target fragments alone. Thus, the following expression vectors were obtained using the pLAUMe vector as the major component: pLAUMe-AT30, pLAUMe-AT65, pLAUMe-AT100, pLAUMe-AT208, pLAUMe-SynJ, pLAUMe-SynM, and pLAUMe-GGR, respectively. In these vectors, the sequence of a tested enhancer is located upstream of the licBM3 reporter gene. Each of the constructed vectors was used to transform A. tumefaciens strain GV3101, which were further used to agroinfiltrate N. benthamiana plants.
A comparative analysis of the functional activity of translation enhancers Total mRNA and total soluble protein were isolated in parallel from the agroinfiltrated leaf fragments to assess the expression of the reporter gene under the control of a tested regulatory element at the stages of transcription and translation. The transcription of reporter genes was assessed by qPCR in one experiment. The translational activities of reporter genes were analyzed according to the enzyme activities of reporter proteins. Using the above-described (Materials and methods) pipeline for data processing, we determined the regression slopes to construct the rating of the tested enhancers ( Table 2). The column Normalized lists the ratios of reporter proteins (LicBM 3/GU S) normalized by the ratio RN A enchanser /RN A pGLR ,w h e r e RN A enchanser is the ratio of licBM3/uid A mRNAs for a selected enhancer and RN A pGLR is the control variant without any translation enhancer, pGLR. Analysis of the transcriptional (RT-qPCR; Figure 3.1) and translational (enzyme activities of reporter proteins; Figure 3.2) activities of the tested enhancers showed a linear dependence between the transcription and translation levels of the reporter genes. Table 2 lists the comparative data and statistical tests.
The numerical data listed in table and visualized in plots suggest the following inference: the distribution of measurements corresponds to the proposed model described below (see Discussion). The absolute contribution of a tested enhancer to the translation level (Q abs ) is determined by the ratios of coefficients A (slope angle of regression line) from the equation f (x)=Ax: Q abs = A protein /A RN A ; f (x), a dependent variable, is the level of lichenase under the control of a tested element (protein or mRNA) and x, an independent variable, is the level of β-glucuronidase. The relative contribution requires normalization by the selected internal control, i.e., the enhancer playing the role of standard, but does not require the precise amounts of enzymes in samples to be calculated.

Discussion
Model of the contribution of regulatory elements First and foremost, it is necessary to consider the main question potentially arising when reading this paper: What is the purpose of the internal control? Since the vector contains two genes of reporter proteins at once, the ratio of the copies 1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60  61  62  63  64  65 (licBM3/uid A) in T-DNA will always be 1 : 1. In the expression under controlled conditions (since the reporter genes are under the control of constitutive promoters), the ratio of expression levels (licBM3/uid A) at the stage of transcription expressed as the mRNA copy number is also constant and reflects a certain linear dependence (determined by the strength of promoters). It is reasonable to assume that, other conditions being equal, the same ratio will be observable for the target proteins (LicBM3/GUS). However, if one of the reporters is placed under the control of a regulatory element that influences translation process, this ratio (LicBM3/GUS) will change in the direction of the effect of this regulatory element. With a set of assumptions, this dependence may be considered linear. Than the ratio of the slopes of lines, which reflect the ratios of mRNAs of reporters and their protein products, will show the contributions of a given regulatory element to the control of translation. We propose to assess the contribution of particular translation enhancers (both known and potential) using the following scheme: (i) Analysis of the transcription of reporter genes with qPCR for determination of the absolute copy number of the corresponding mRNAs; (ii) Analysis of the enzyme activities of reporter proteins using the calibration curves constructed according to purified standardized enzymes for determination of the absolute copy numbers of the corresponding enzymes; and (iii) Statistical analysis of the data and comparison of the slopes of the lines reflecting the ratios of mRNAs (licBM3/uid A) and proteins (LicBM3/GUS). Thus, the system of internal control physically linked to the main reporter gene and a tested regulatory element provides the possibility to more precisely measure the expression level of the reporter under the control of this element as well as to considerably accelerate and simplify the very process of testing thanks to avoiding the stage of co-transformation. As for the currently existing vectors with more than one reporter, bicistronic systems have been widely used recently. These systems have been applied to localize pathogens [23][24][25]; study enzyme activities [26,27], heat shock proteins [28], and transcriptional and posttranscriptional modifications [29,30]; and assess the activity of bacterial promoters [31] and the cell cycle [32]. Of special interest is the role of bicistronic systems in IRES analysis [33].
However, a bicistronic system is not completely adequate for analyzing the cisregulatory elements.
It is reasonable to turn back to the stage of qPCR analysis of transcription; this stage is necessary since several studies [34, 35] demonstrate that some functional elements of promoter can be located downstream of the transcription start site; correspondingly, these elements reside in the 5'UTR. The opposite situation is equally true: if a 5'UTR in the distal region contains the motifs similar to functional elements of promoter, such 5'UTRs can be involved in the regulation not only of translation, but also of transcription. Thus, we believe that a more precise characterization of a tested translation enhancer requires analysis of its role at the stage of transcription as well.
In this work, we propose to perform a relative analysis of the role of translation enhancers by arbitrarily selecting one of them as a control. It is necessary to take into account that the 5'UTRs frequently contain proximal promoter regulatory elements (boxes) [34,35], thereby also influencing the process of transcription. Correspondingly, we propose to assess the changes in the ratio of the products of the reporter genes at the stages of both transcription and translation to more accurately determine the contribution of translation enhancers to the expression level. We assume that the enzyme activity (U e ) is associated with the number of protein molecules (N e ) as N e = k e × U e ( k e , coefficient); thus, the ratio of the normalized amounts of reporters in absolute units (moles or the number of molecules in a unit volume or a unit weight of a sample) is expressed as Then, assuming that K = k Lic k Gus and normalizing by the contribution of enhancer at the stage of translation (R abs = RN A Lic RN A Gus ) we get Thus, a mere comparison of the normalized ratios of reporter enzyme activities allows us to make the conclusions on a relative contribution of enhancers at the stage of translation.
It is reasonable to separately discuss the possibilities of an absolute estimation of the contribution of a tested regulatory element. In our view, the processes of transient expression are rather variable; correspondingly, it is not realistic to assess the changes in the amount of protein expressed under the control of a certain enhancer in terms of weight units, i.e., relative estimates are sufficient. Nonetheless, an absolute estimate is not too complex in terms of methodology provided that the k Lic and k Gus coefficients are determined from Eq. (1).