Potassium as a Macro-nutrient

Potassium is an essential plant macronutrients as it has numerous important roles in plants including osmoregulation, CO₂ regulation, starch synthesis and protein synthesis. Hence, the deficiency of K⁺ ion will result in abnormalities in plant growth and metabolism. It is fundamental to determine its concentration in soil in order to provide the proper amount of additional potassium that must be added to the plant by any particular fertilizers.

Our aim is to engineer a Potassium sensor that can detect a range of K⁺ concentration in the soil to ensure the suitable soil condition for the plant fitness. We utilized KdpFp, a promoter located upstream of KdpFABC operon in Escherichia coli which works under low [K⁺] condition. We put it upstream of a GFP reporter so as to characterize the promoter activity. For simplification, KdpFp will be called potassium promoter in the following context.

INSERT BADASS GRAPH HERE

---

Endogenous K sensing system

Badass graph please

The potassium ion uptake in Escherichia coli is regulated by several systems under different conditions. The potassium ion transporters, Trk and Kup are constitutively expressed (reference?) while KdpFABC, another transporter is activated under limited potassium ion concentration (reference?) and controlled by the KdpDE two-component system (TCS) (Polarek, 1992; Walderhaug, 1992). The TCS consists of KdpD< which is a membrane-bound sensor kinase and KdpE which is the cytoplasmic response regulator. The autophosphorylation of KdpD transfers a phosporyl group to the KdpE upon low potassium concentration (Voelkner, 1993; Puppe, 1996; Jung ,1997a; Jung, 2000). Phosphorylated KdpE activates kdpFABC operon.

KdpDE TCS is stimulated by both intracellular and extracellular potassium ion concentration (Jung ,2000; Jung, 2001; Roe, 2000; Yan, 2011a; Laermann, 2013). The intracellular and extracellular potassium ion concentration shows an inverse correlation to KdpDE phosphorylation. Under an increase in potassium ion concentration, KdpD phosphatase activity will be enhanced, causing a decrease in phospho-KdpE and kdpFABC expression. Decreasing potassium ion concentration shows the other way around (Zhang, 2014a; Laermann, 2013).

KdpFABC transporter, which depends on ATP, is a high affinity P-Type ATPase potassium ion uptake system (Siebers, 1988; Siebers 1989). The KdpFABC transporter is specific to potassium ion (Km= 10 μM), magnesium ion (Km= 80 μM), as well as other particular substrates (Siebers, 1988). The KdpFABC complex consists of kdpA, kdpB, kdpC, and kdpF subunits. The kdpA subunit controls the binding as well as transportation of substrate and kdpB subunit pairs the energy. The kdpC subunit keeps kdpA and kdpB together (Buurman95). In addition, the kdpF gene encodes a small polypeptide which stabilize the KdpFABC complex (Altendorf98, Gassel99).

---

Design of K+ sensing Device

As our potassium-sensing device, we adopt the promoter kdpFp from kdpFABC operon with -28 position of transcription start site relative to start the first gene, kdpF. The -10 and -35 box elements have been mapped which are TACCCT and TTGCGA respectively (Sugiura et al., 1992). The transcription factor, phosphorylated KdpE, binds to the kdpFp at binding site located from -71 to -55 site with reference to the transcription start site (Sugiura et al., 1992; Narayanan et al., 2012). We then combined the promoter with a downstream GFP generator, BBa_E0240, using BioBrick RFC 10 standard so that the promoter activity in different potassium level can be detected and characterized.

Illegal Site

In order to make our promoters, kdpFp, compatible with Biobrick RFC 10 standard, it should not contain any of the illegal sites (as they belong to the prefix and suffix). However, the promoter contains an EcoRI illegal site from -18 to -12 position. To make the potassium promoter readily accessible to the whole iGEM community, we removed that illegal site by mutating the thymine at -15 position to guanine, cytosine and adenine to give rise to 3 promoter mutants. Also, by removing the illegal site, different mutants of this promoter with different activity levels can be generated. We can choose the one with desirable activity to fit our need.

Apart from the wild-type kdpFp (containing illegal site), We designed another 3 mutants which have one base-pair at -15 position with reference to the transcription start site, changed from thymine (T) to either cytosine (C), guanine (G) or adenine (A). For convenience, we denote them as kdpFp(A)/A-mutant, kdpFp(G)/G-mutant and kdpFp(C)/C-mutant respectively. All the mutants, thereby, have their illegal site removed. However, we are expecting different activity from the four promoters due to their difference in binding energy between the promoter and RNA polymerase because of the base-pair changes (Brewster, 2012). Therefore, we characterized all of the promoters later to compare their strength by the mean of fluorescence level and RPU measurements so as to obtain a more comprehensive knowledge in the activity and working range of the 4 kdpFp promoters.

Interference from other Endogenous Systems

Another concern is the masking of activity of kdpFp by other native constitutive potassium transport systems of E.coli. Other than the inducible KdpFABC system(Km=2uM), E. coli has Trk and Kup systems which are constitutively expressed, low-affinity potassium transport systems (Km=1.5mM) (reference?). A study has found that the expression level of KdpFABC is generally higher in a strain without Trk and Kup system (Laimins et al., 1981). It is due to the fact that E. coli uses the two saturable transporters, Trk and Kup, under normal physiological conditions to uptake extracellular potassium ion. Therefore, only at a very low concentration of intracellular potassium, these two systems can no longer satisfy the need of potassium ion of E. coli. Hence, the kdpFp will be activated and drive the expression of KdpFABC complex to uptake more K⁺. As the result, the activity of kdpFp may be masked. The activity of this promoter cannot be truly reflected over a range of K+ concentration in the presence of these 2 systems (Laermann et al., 2013).

We decided to measure the activity of kdpFp in E.coli TK2240 (kdp+ Δtrk Δkup) strain, which is a strain defectve in Trk and Kup system. Such that we are able to characterize kdpFp promoter in a more accurate manner.

---

Measurement and Characterization

In order to assemble a device that can be widely used by all iGEM community, we characterized kdpFp promoter by using Relative Promoter Unit (RPU). Additionally, we intended to find the comparison of the activities between different promoters, thus, we also measured Relative Fluorescence Unit (RFU) of the 4 potassium promoters.

Relative promoter Unit measurement

badass rpu graph

Figure . Relative promoter unit (RPU) of kdpFp[-15,T>G] across different concentration of K⁺. The measurement of the activity of G mutant was carried out using fluorescence-activated cell sorting. The strength of the promoter is presented in relative to the strength of standard reference promoter. Error bar are represented as SEM.

From the lowest concentration of potassium ion to 0.025 mM, the strength of the G mutant promoter was found to be about 0.5 RPU and decreasing when concentration increased. After the concentration of potassium ion at 0.025 mM, there was no significant change on the RPU values. Strength of G mutant kdpFp at the lowest concentration was about 1.5 times higher than those after 0.025 mM.

High concentration of potassium ion could repress the expression of kdpFp, due to satisfaction of potassium ion by constitutively expressed low-affinity K⁺ transporter system, Trk and Kup(Laermann et al., 2013). Therefore the activity of kdpFp promoter, that is represented by the relative fluorescence intensity, at high concentration of potassium ion, should be low and it is shown in our experiments.

Relative fluorescence measurement

graph graph graph

PAIRED graph to be put here

---

Achievement

Our team has finished characterizing all the constructs, including the wild type promoter and 3 mutants controlling the expression of GFP. We also contemplated the activity of the promoters over a varying range of K⁺ concentration. We had discovered a comparison between different promoters and a dynamic relationship between K⁺ concentration and the promoters’ activity. Upon different concentration of K⁺, Potassium sensor will show different fluorescence level due to distinctive effect of K⁺ ion to kdpFp.

---

Mechanism

In our project, we use the native potassium ion transport system in Escherichia coli (E. coli), Kdp system as our potassium sensing part. The Kdp system is composed of two major parts, KdpFABC, a high-affinity potassium transporter as well as two types of regulatory proteins, a sensory kinase KdpD and a response regulator KdpE. KdpD and KdpE works together as a two-component system, tracking and responding to the intra and extracellular potassium level then interacting with the KdpFABC encoding operon. kdpFABC operon is up-regulated under low potassium ion concentration and is inhibited under high concentration.

KdpD, which is a trans-membrane protein, auto-phosphorylates itself, also phosphorylates and dephosphorylates KdpE. Low concentration of potassium ions favors the phosphorylation of KdpE, which then gives rise to the enhancement of the level of phosphorylated KdpE, and as a result, triggers the up-regulation of kdpFABC operon.

As our potassium-sensing device, we adopt the promoter kdpFp from kdpFABC operon. The sequence was obtained by oligos, we then combine the promoter with the downstream GPF generator using biobrick RFC 10 so that the change of the promoter activity in different potassium level can be detected and characterized.

---

Limitations

There are two major limitations in making use of the Kdp system as our potassium-sensing module. The first main concern is that the promoter kdpFp contains the EcoRI illegal site. While the second concern is about the background noise contributed by other native constitutive potassium transport systems of E. coli, including trk and Kup systems, which are potassium ions influx systems and are expected to lower the activity of our promoter kdpFp.

---

Solutions to the limitations

We have come up with solutions to tackle the aforementioned limitations of Kdp system. For the EcoRI illegal site inside the promoter, we ordered 4 different versions of kdpFp, one of them is the wild-type promoter; for the other three, they have one base-pair at -15 site, where the illegal site locate, changes from thymine (T) to cytosine (C), guanine (G) and adenine (A) respectively. This make the three promoters into three different mutants, we denote them as A-mutant, G-mutant and C-mutant respectively. All the mutants, thereby, have their illegal site removed.

---

Result obtained

FACS

Figure . Relative promoter unit (RPU) of kdpFp[-15,T>G] across different concentration of K+. The measurement of the activity of G mutant was carried out using fluorescence-activated cell sorting. The strength of the promoter is presented in relative to the strength of standard reference promoter. Error bar are represented as SEM.

From the lowest concentration of potassium ion to 0.025 mM, the strength of the G mutant promoter was found to be about 0.5 RPU and decreasing when concentration increased. After the concentration of potassium ion at 0.025 mM, there was no significant change on the RPU values. Strength of G mutant kdpFp at the lowest concentration was about 1.5 times higher than those after 0.025 mM.

High concentration of potassium ion could repress the expression of kdpFp, due to satisfaction of potassium ion by constitutively expressed low-affinity K⁺ transporter system, Trk and Kup(Laermann et al., 2013). Therefore the activity of kdpFp promoter, that is represented by the relative fluorescence intensity, at high concentration of potassium ion, should be low and it is shown in our experiments.

GFP synthesis rate and RFU

Figure . Activity of kdpFp in E. coli DH10B in different K⁺ concentrations.A, T C and G represent A mutant, wild type promoter, C mutant and G mutant respectively. Error bar are represented as SEM.

Both C and G mutants similarly expressed higher fluorescence intensity compared to the wild type and the A mutant. A higher level of potassium ion inside K minimal medium repressed the expression level of kdpFp, which was shown by lower fluorescence intensity of all promoters at higher concentration of potassium ion. The expression level of both C and G mutant promoters were significantly higher than the wild type one, while A mutant promoter was always the lowest.

The fluorescence intensity decreases over the concentration of potassium ion in K minimal medium. After several trials, we found the dynamic range of our promoters. It is between 0 to 0.1 mM of potassium ion in K minimal medium. Thus, we characterized our promoters at those range of concentrations (0 to 0.2 mM).

We observed that the fluorescence intensity expressed by C and G mutants are always at around similar values since the binding affinity might be the same with reference to the energy matrix (reference?). Moreover, both of them are always higher than the one expressed by the wild type promoter. Meanwhile, the fluorescence intensity expressed by A mutant is lower than the the one expressed by the wild-type promoter. Also, the relationship of relative fluorescence intensity and K⁺ concentration of all the promoters are coherent with the previous RPU measurement of G mutant.

Figure . The activities of kdpFp in the medium containing no potassium ion.A fluorescence/absorbance plot was obtained from measuring the relative fluorescence level exhibited by the promoters, kdpFp and its 3 mutants (A, C and G), in DH10B cells in 0 mM K minimal medium. The expression level of both C and G mutants are significantly higher than the expression level of wild-type promoter and A mutant promoter. There was no significant difference between the activity of G mutant and C mutant promoters, as well as between wild type promoter and A mutant. Error bars are represented as SEM.

We also would like to make a clear comparison between the expression level of the different mutant promoters. We tried to make a comparison of the expression of all the promoters at the lowest level of potassium ion, that is no potassium ion in K minimal medium. From the graph that we obtained, we observed that the expression level of both C and G mutants are about 2 times higher than the wild type promoter and A mutant. These results demonstrate our former experiments that the relative fluorescence intensity level of both C and G mutants are always higher than both wild type promoter and A mutant.

Figure . Comparison between the activities of kdpFp[-15, T>G] in DH10B and TK2240 strain. A fluorescence/absorbance plot was obtained from measuring the relative fluorescence level exhibited by the G mutant pormoter in DH10B and TK2240 cells. At the concentration of potassium ion lower than 0.0125 mM, the fluorescence intensity expressed by G mutant in DH10B was significantly greater than the one in TK2240 strain. The fluorescence intensity expressed by G mutant in DH10B strain decreased over the concentration, while the one in TK2240 remained the same. Starting from the concentration of potassium ion at 0.0125 mM onwards, the difference of GFP expressed by both strains was not significant. After passing the concentration of potassium ion at 0.05 mM, the expression level of kdpFp in TK2240 strain exceeded the one in the DH10B strain. Error bars are represented as SEM.

We did another relative fluorescnece measurement using TK2240 to observe a more accurate activity of kdpFp. The TK2240 strain is defective in trk and kup gene. The only gene which could respond to potassium ion in TK2240 strain is only kdp gene. Therefore the expression of kdp gene would not be masked by the saturable low-affinity K⁺ transporter system. We were expecting that the activity of kdpFp in TK2240 are higher than that in DH10B. The fact mentioned earlier corresponds to the graph at higher concentration of potassium ion, although the expression level in DH10B is higher than in TK2240 at low concentration of potassium ion.

We could observe the difference of the fluorescence intensity expressed by both strains at the concentration of potassium ion lower than 0.0125 mM. The fluorescence intensity expressed by G mutant in DH10B strain decreases over the concentration, while the one in TK2240 remains the same. Since TK2240 has a greater need of kdpFABC complex to help scavenge the external K+, so what might have caused this would be TK2240 spend too much energy at that stage to make kdpFABC complex and thus spare less energy on producing GFP.

At higher concentration of potassium ion, starting from 0.0125 mM onwards, the difference of fluorescence intensity expressed by both strains is not significant anymore. Both strains show decrease in fluorescence intensity over the increasing concentration of potassium ion in K minimal medium.

After passing the concentration of potassium ion at 0.05 mM, the fluorescence intensity expressed by TK2240 strain exceeds the one in the DH10B strain due to the constant need of potassium ion of the cells and in TK2240, there is only KdpFABC transporter system that can satisfy this need of E. coli.

Figure . The GFP synthesis rate of kdpFp mutants in DH10B. A GFP synthesis rate plot was obtained from measuring the relative fluorescence level exhibited by the kdpFp and its mutants in DH10B cells. From the lowest concentration of potassium ion to 0.025 mM, there was no significant change of GFP synthesis rate. The GFP rate for C and G mutant promoters were always higher than the wild type and A mutant promoters. Starting from the concentration of potassium ion at 0.025 mM, the values of GFP synthesis rate dropped to nearly 0. After the concentration of potassium ion reached 0.1 mM, the GFP synthesis rate became so low and there was no significant differences for all mutants.

The GFP synthesis rate was calculated by measuring the rate of fluorescence intensity level over time. As it can be seen from the GFP synthesis rate graph, there was no significant difference of the GFP synthesis rate at low concentration of potassium ion until 0.025 mM. From the results that we always obtained, the rate for C and G mutant promoters were always greater than the wild-type and A mutant promoters. This explained why the expression level of C and G mutants were higher than the wild type and A mutant.

When the concentration of potassium ion was higher than 0.025 mM of potassium ion in K minimal medium, the values of GFP synthesis rate dropped to nearly 0. This conveyed that the fluorescence intensity started to decrease significantly at 0.025 mM potassium ion concentration onwards.

Finally, at 0.1 mM, the GFP synthesis rate values were so low, the relative fluorescence intensity, therefore, would also be so low.

Considerations for replicating the experiments

a. OD referring to mid log phase of e coli

b. dilution at the beginning

i. We do measurement when the cells are in the mid log phase. As stated from the paper, E. coli enters mid log phase when the value of OD600 is between 0.3 to 0.5. Therefore to have all cultures entering the log phase together by the time we are doing the measurement, we always do dilution to make the OD values around the same before we incubate the cultures inside the 96 deep well plates. We believe that the growth of the cells will be similar if the starting conditions are the same, which we think are the OD values.