a West China Medical Center Team for International Genetically Engineered Machine Competition 2017, West China School of Medicine / West China Hospital, Sichuan University, Chengdu 610041, China
b Department of Gastroenterology, West China Hospital, Sichuan University and Collaborative Innovation Center of Biotherapy, Chengdu 610041, China
c West China School of Stomatology, West China Hospital, Sichuan University, Chengdu 610041, China
d College of Physical Science and Technology, Sichuan University, Chengdu 610064, China # These authors contributed equally to this work
Hongbin Yu, Zheng Wang
Show less
History+
Received
Accepted
Published Online
2018-03-07
2018-09-29
2019-03-05
PDF
(1161KB)
Abstract
Measuring the absolute protein expression quantity for a specific promoter is necessary in the fields of both molecular biology and synthetic biology. The strength of a promoter is traditionally characterized by measuring the fluorescent intensity of the fluorescent protein downstream of the promoter. Until now, measurement of the absolute protein expression quantity for a promoter, however, has been unsuccessful in synthetic biology. The fact that the protein coding sequence influences the expression level for different proteins, and the inconvenience of measuring the absolute protein expression level, present a challenge to absolute quantitative measurement. Here, we introduce a new method that combines the insulator RiboJ with the standard fluorescence curve in order to measure the absolute protein expression quantity quickly; this method has been validated by modeling verification. Using this method, we successfully measured nine constitutive promoters in the Anderson promoter family. Our method provides data with higher accuracy for pathway design and is a straightforward way to standardize the strength of different promoters.
As shown in the enzyme kinetic equations, enzyme concentration plays an important role, as well as substrate concentration, in enzymatic reactions. In order to enhance the efficiency of an artificial metabolic pathway with a predictable substrate concentration, the expression of each enzyme in this pathway should be carefully considered. Overproduction of an enzyme may cause an excessive burden for the cell. On the other hand, insufficient protein expression can result in low efficiency of the metabolic flux. Thus, the promoter, which influences the absolute protein expression quantity of the enzyme, should be chosen appropriately. However, a method of measuring the absolute protein expression quantity of a specific promoter has not yet been reported. The characterization of a promoter is traditionally performed by measuring the fluorescent intensity or RT-PCR results [1–6] of the downstream protein or gene, which can only be compared in a relative manner. However, certain challenges affect the measurement of the absolute protein expression quantity of a specific promoter. The dynamics of gene expression are influenced by the protein coding sequences; this means that measuring a promoter by simply detecting its downstream protein expression level is not accurate [7]. In addition, it is inconvenient and expensive to measure the absolute protein expression quantity using western blotting, ELISA, or similar techniques.
RiboJ is a ribozyme that self cleaves, thereby removing the upstream region and, most importantly, the 5′ untranslated region. In the work of Lou et al. [7], the relative expression levels of two different variants of green fluorescence protein (GFP) (i.e., sfGFP and cl-sfGFP fusion) were not the same. However, when they added RiboJ before the coding region, the relative expression levels for the induction curves collapsed. This finding indicated that RiboJ can act as an insulator to homogenize the relative protein expression level for a given promoter, thereby making it possible to evaluate the protein expression ability of a specific promoter on a large scale, regardless of the coding sequence. To characterize the promoter in a quick and simple way, we attempted to determine the relationship between absolute protein quantity and enhanced green fluorescence protein (eGFP) fluorescence by making an eGFP fluorescence standard curve, and then using a protein expression system with a RiboJ design to determine the promoter strength. We chose nine promoters from the Anderson promoter family, which is an σ70 transcriptional promoter library of Escherichia coli (E. coli). The Anderson promoter family is widely used in the International Genetically Engineered Machine (iGEM) Competition, and has been characterized by expressing red fluorescence protein (RFP) and GFP [4,8]. Our design allowed us to successfully measure these nine promoters, and yielded higher-accuracy data for future pathway design. Our method also provides a straightforward way to standardize the strength of different promoters. Thus, this paper introduces a novel method to measure the absolute protein expression quantity of any constitutive promoter.
2. Materials and methods
2.1. Plasmids construction
The pET28a-eGFP plasmid was obtained from the State Key Laboratory of Biotherapy, in China. The eGFP gene was originally obtained from plasmid pEGFP-N1, and was then ligated into pET28a vector with a 6×His tag for purification. The measurement plasmid originated from BBa_J364001 in the pSB1C3 vector from the Registry of Standard Biological Parts. RiboJ was generated from synthetic oligos (TsingKe Biological Technology Co., Ltd., China) and was seamlessly added to the downstream of Anderson promoter sequence by means of Gibson assembly. The GFP was then replaced by eGFP by means of Gibson assembly in order to construct the final measurement device with a J23106 promoter. The other measurement devices, with different promoters, were generated by oligonucleotide-directed mutagenesis (KOD-Plus-Neo, TOYOBO Co., Ltd., Japan; DpnI, New England Biolabs Inc., USA). The sequences for the Anderson promoters are documented in the Registry of Standard Biological Parts [8]. The plasmids used in the study are listed in Table 1. The gene structure of the measurement device is shown in Fig. 1, and the sequence is documented in the Appendix A (Table S1).
2.2. Purification of the eGFP
The expressing plasmid (pET28a-eGFP) was transformed into E. coli BL21(DE3) cells. Transformed cells were grown in 1 L lysogeny broth (LB) supplemented with 34 μg·mL−1 kanamycin at 37 °C, with agitation at 220 r·min−1 overnight. When the OD600 （optical density measured at a wavelength of 600 nm) reached 0.6–0.8, the cells were induced with a final concentration of 0.5 mmol·L−1 isopropyl β-D-1-thiogalactopyranoside (IPTG). After being cultured for a further 18 h at 16 °C, the cells were harvested by centrifugation at 4 °C and 4000 r·min−1 for 10 min. The harvested cells were suspended in 40 mL lysis buffer (PBS buffer, 140 mmol·L−1 NaCl, 2.7 mmol·L−1 KCl, 10 mmol·L−1 Na2HPO4, and 1.8 mmol·L−1 KH2PO4, at pH 7.5) and lysed by a high-pressure homogenization instrument (JN-02C, JNBIO Co., Ltd., China). The lysate was then subjected to centrifugation (4 °C, 15 000 r·min−1, 30 min; Thermo F21-8x50y rotor). The clear supernatant was collected and loaded onto a nickel-nitrilotriacetic acid (Ni-NTA) column pre-equilibrated with the lysis buffer at 4 °C, and the pellets were discarded. The column was washed three times with 3–5 column volume (CV) of washing buffer (10 mmol·L−1 imidazole dissolved in PBS buffer) and three times with 3–5 CV of elution buffer (200 mmol·L−1 imidazole dissolved in PBS buffer). The eluted protein containing eGFP was concentrated in a 15 mL concentrator (Amicon® Ultra-15, Merk Millipore, USA) before being loaded onto the Superdex 75 10/300 GL (GE Healthcare Conglomerate, USA). The single peak of eGFP was collected, and the eGFP was concentrated to 25 mg·mL−1. Part of the protein was analyzed immediately; the remaining portion was frozen using liquid nitrogen and kept in stock at −80 °C. We repeated the purification method five times and obtained five batches of purified eGFP for further measurement.
2.3. eGFP stability test
We measured the fluorescence intensity of each batch of eGFP immediately after purification in order to exclude the impact of fluorescence quenching. The eGFP sample was placed in a black 96-well plate fully covered with aluminum foil before measuring, and was then measured by means of a Spark™ 10 M (Tecan Group Ltd., Switzerland) provided by the West China School of Pharmacy at Sichuan University. The fluorescence intensity was measured with an excitation wavelength of 450 nm and an emission wavelength of 490 nm. The absolute protein concentration of the purified eGFP was measured using BCA Protein Assay Kit (Sangon Biotech (Shanghai) Co., Ltd., China). The experiment was repeated five times to test its repeatability.
The first batch of purified eGFP was stored at −80 °C; we then repeatedly measured the fluorescence intensity and absolute protein concentration 4, 8, 12, and 16 d after purification in order to test the stability of the eGFP and identify the impact of fluorescence quenching over time for our experiment.
2.4. Cell density and fluorescence measurement
Our measurement process followed the routine of the iGEM InterLab study to a large extent [9]. All the measurement devices were transformed into E. coli BL21 (DE3). Transformed cells were grown overnight at 37 °C and 220 r·min−1, in 5 mL LB medium with 25 μg·mL−1 chloramphenicol. At least three colonies per device were picked on the plate for measurement. The cultures were diluted to a target OD600 of 0.02 in 12 mL LB medium with 25 μg·mL−1 chloramphenicol in a 50 mL falcon tube, and incubated at 37 °C and 220 r·min−1. Three replicates, with 500 μL of LB medium each, were sampled from each tube for measurement after being incubated overnight (16–18 h). Cell density (OD) was measured at a wavelength of 600 nm, and the fluorescence intensity of the bacteria was measured with an excitation wavelength of 450 nm and an emission wavelength of 490 nm (Spark™ 10 M, Tecan Group Ltd., Switzerland).
2.5. Modeling and data analysis
This section first discusses the theory of protein production relative to promoter activity [10,11]. The velocity of the production per cell of eGFP mRNA molecules () is relative to the copy number (), promoter activity (), and mRNA degradation rate (), as described in Eq. (1). The amount of immature eGFP per cell () is dependent on the immature eGFP translation rate (), mature eGFP degradation rate (), and mature eGFP translation rate (), as shown in Eq. (2). The amount of mature eGFP per cell () is also affected by the degradation rate of mature eGFP (), as shown in Eq. (3).
The equation system can be solved with Mathematica (Wolfram Research, USA):
where is a function of the , , , , , , and . For one type of promoter, we regard , and so on as a constant, referred to as A. Next, after analyzing the data of fluorescence from the fluorescence experiments using linear regression, the coefficient () between concentration of (CeGFP) and fluorescence intensity () can be obtained.
Finally, we performed a series of experiments to obtain the relationship between the fluorescence and the concentration, and thus created a scale for the promoters. The molarity of the eGFP () can be written as follows, and it was found to be 31 534.35 Da:
where is the concentration of 1 mol eGFP, is the concentration of cells, and is a constant with a value of 8 × 108[12]. Thus, the promoter activity A can be written as follows:
colonies were picked. We measured the fluorescence intensity and cell density times for each colony and then measured the fluorescence intensity and OD times for the control (LB medium). The equation can be written as follows:
3. Results
3.1. eGFP purification results
We were able to successfully obtain the purified eGFP. The gel filtration result showed a single peak in the sample (Fig. 2(a)). Samples analyzed by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) revealed an about 30 kDa induction band (Fig. 2(b)). The size of this band was consistent with the predicted mass of the recombinant enzyme.
3.2. eGFP stability test
We analyzed by linear regression the results of five independent experiments in which the eGFP was freshly purified. Our findings indicated a strong linear correlation between the fluorescence intensity and the absolute protein concentration (C) measured by BCA assays in our experiment range (r2 = 0.9733, Fig. 3). The results show that the experiment possesses excellent repeatability.
Next, we collected all the data from the fluorescent quenching experiment, as shown in Fig. 4. The data collected by continuous measurement over 16 d was analyzed by linear regression. The results showed that the fluorescence intensity of the eGFP and the absolute protein concentration show a strong linear correlation (r2 = 0.9936); this means that the purified eGFP has a level of high stability and can resist quenching over time, because the measurement results from each time were similar and formed a linear correlation.
3.3. Data analysis
After measuring the OD600 and fluorescence intensity for the measurement device using the method described above, the data were collected and the results were calculated. The original measurement results are documented in the Appendix A (Tables S2 and S3). The absolute protein expression quantity and relative activity of the promoters are shown in Table 2 and Fig. 5.
It is notable that the time-invariance character of the promoter can be demonstrated by the modeling data. According to Table 2, Eq. (4), and the experimental results, the rate of protein production can be determined, as shown in Fig. 6. The proportion of accumulated protein expressed by different promoters over different times is the same. Fig. 6(b–d) show the protein expression proportion based on the experimental result at a given time.
4. Discussion
Absolute protein expression quantity is important for quantitative pathway construction, which has rarely been reported to date. The quantitative measurement of promoter activity that was attained in this study is useful in promoter selection. To verify the usefulness of our method, we selected a three-enzyme metabolic pathway as an example.
We set as the total concentration of the enzyme, as enzyme concentration, as the substrate concentration, as the concentration of the enzyme-substrate complex, as the product concentration, as the equilibrium constant of the first positive reaction, as the equilibrium constant of the first reverse reaction, and k3 as the equilibrium constant of the second positive reaction [13].
The chemical equation can be written as follows:
Assuming that each part of the reaction has reached equilibrium, the production and degradation rates of are equal.
Rearranging this equation gives the following:
Assuming that
Eq. (12) become the following:
Note that the reaction of and was ignored, even though will influence the equilibrium, which will affect the production rate. Therefore, as time passes, Eq. (16) may become less accurate. Nevertheless, generally speaking, it is able to describe this process.
Assuming the pathway through which compound A becomes C includes two reactions with two enzymes required, P1 and P2:
The reaction rates of these two reactions are represented by v1 and v2, and the concentrations of compound A, B, and C are , , and .
Without considering spontaneous degradation, the net reaction rate of one compound (A, B, or C) is equal to its generation rate minus its consumption rate, assuming that the proportions of each reaction are 1:1.
Whether it is a first-order reaction or a zero-order reaction, the rate is only determined by the concentration of enzyme, when the type of enzyme and the initial concentration of the substrate are given (as shown in Eq. (16)). It is apparent that the higher the concentration of enzyme is, the faster the reaction will be. If there is a major difference between the enzyme's maximum rate, , of two reactions, then the product of one of the reactions cannot be produced, which limits the metabolic rate. Therefore, there is a global best solution to optimize the reactions for multi-step enzyme catalysis (Fig. 7).
However, if is adjusted, the product will not accumulate (Fig. 8). Therefore, appropriate promoters must be chosen in order to make of each of the three reactions equal. The method described in this paper is an appropriate method of doing so. Given the promoter activity obtained earlier, it is possible to accurately calculate the concentration of protein (), which is equal to .
The modeling results indicate that plays an important role in the metabolic pathway construction. Thus, we have introduced a novel and quick method for absolute protein expression quantitative measurement that will be useful in future applications. Furthermore, the promoter data and gene structure provided by our method can be used directly for metabolic pathway construction in the future.
In addition, we have provided a straightforward way to standardize the strength of different promoters, since the evaluation standard based on the absolute protein quantity can be used directly for all promoters. We have introduced this standardized method and the RiboJ design in order to encourage researchers to characterize promoters using universally comparable parameters (e.g., absolute protein quantity), and to employ the insulator RiboJ to promote the repeatability and transplantability of gene circuit design in different research studies. Furthermore, the ranking of the relative strength of promoters that was performed in this study differs from the ranking that is based on simple RFP fluorescence intensity measurement, as documented in the Registry of Standard Biological Parts [14]. The difference may be caused by the RiboJ design in our research, which can significantly influence the expression quantity [7,15]. Thus, our method provides greater accuracy in measurement results and gene structure, which will assist pathway construction in the future.
5. Conclusion
In this study, we introduced a novel method to measure the absolute protein expression quantity of a specific promoter. We then utilized our method to successfully measure nine constitutive promoters in the Anderson promoter family, and achieved results that differ from the fluorescent intensity measurement results reported in the literature. Our method provides data with greater accuracy, and offers a straightforward way to standardize the strength of different promoters.
Compliance with ethics guidelines
Hongbin Yu, Zheng Wang, Hanyue Xu, Jiusi Guo, Qingge Ma, Xiangxu Mu, and Yunzi Luo declare that they have no conflict of interest or financial conflicts to disclose.
MaedaM., ShimadaT., IshihamaA.. Strength and regulation of seven rRNA promoters in Escherichia coli. PLoS ONE. 2015; 10(12): e0144697.
[2]
RudgeT.J., BrownJ.R., FedericiF., DalchauN., PhillipsA., AjiokaJ.W., . Characterization of intrinsic properties of promoters. ACS Synth Biol. 2016; 5(1): 89-98.
[3]
PedrazaJ.M., Van OudenaardenA.. Noise propagation in gene networks. Science. 2005; 307(5717): 1965-1969.
[4]
ChappellJ., FreemontP.. In vivo and in vitro characterization of σ70 constitutive promoters by real-time PCR and fluorescent measurements. In: editor. Totowa: Humana Press; 2013. p. 61-74.
[5]
KellyJ.R., RubinA.J., DavisJ.H., Ajo-FranklinC.M., CumbersJ., CzarM.J., . Measuring the activity of BioBrick promoters using an in vivo reference standard. J Biol Eng. 2009; 3(1): 4.
[6]
LuoY., ZhangL., BartonK.W., ZhaoH.. Systematic identification of a panel of strong constitutive promoters from Streptomyces albus. ACS Synth Biol. 2015; 4(9): 1001-1010.
[7]
LouC., StantonB., ChenY., MunskyB., VoigtC.A.. Ribozyme-based insulator parts buffer synthetic circuits from genetic context. Nat Biotechnol. 2012; 30(11): 1137-1142.
[8]
Anderson JC. Anderson promoter collection [Internet]. 2018 [cited 2018 Mar 6]. Available from: http://parts.igem.org/Promoters/Catalog/Anderson.
[9]
Beal J, Haddock-Angelli T, Gershater M, De Mora K,Lizarazo M, Hollenhorst J, et al. Reproducibility of fluorescent expression from engineered biological constructs in E. coli. PLOS One 2016;11(3):e0150182.
[10]
SongK.. Introduction to synthetic biology. Chinese
[11]
Deft TU. Modeling [Internet]. 2018 [cited 2018 Mar 6]. Available from: http://2015.igem.org/Team:TU_Delft/Modeling.
[12]
Agilent. E. coli cell culture concentration from OD600 calculator [Internet]. 2018 [cited 2018 Mar 6]. Available from: https://www.genomics.agilent.com/biocalculators/calcODBacterial.jsp.
[13]
In: editor. Biochemistry: a short course. 3rd ed. New York: W. H. Freeman and Company; 2012.
[14]
Anderson JC. Strength of Anderson Promoter [Internet]. 2018 [cited 2018 Mar 6]. Available from: http://parts.igem.org/wiki/index.php/Part:BBa_J23100.
[15]
William and Mary Team. RiboJ [Internet]. 2018 [cited 2018 Apr 19]. Available from: http://2016.igem.org/Team:William_and_Mary/RiboJ.
PDF
(1161KB)
