Difference between revisions of "Part:BBa K2500011"

Latest revision as of 23:08, 1 November 2017

AND Gate B: Synthetic Promoter Responsive to LldR and LuxR

The AND gate integrates two signals, namely the presence of high L-lactate concentrations and high bacterial cell density (qorum sensing) and regulates the effector functions of CATE. It allows our engineered bacteria to autonomously decide whether they are currently located in tumor tissue or not.

In the absence of high concentrations of L-lactate, LldR inhibitor proteins bind to the operator sites O1 and O2 surrounding the pLux promoter leading to the formation of a DNA loop. The pLux promoter is sequestered and inaccessible for transcriptional activation by the quorum sensing components.

Sequence and Features

Assembly Compatibility:

10
COMPATIBLE WITH RFC[10]
12
COMPATIBLE WITH RFC[12]
21
COMPATIBLE WITH RFC[21]
23
COMPATIBLE WITH RFC[23]
25
COMPATIBLE WITH RFC[25]
1000
COMPATIBLE WITH RFC[1000]

Usage and Biology

Figure 1:Genetic circuit of CATE. CATE consists of an regulator (upper) and actuator (lower) plasmid. The AND gate integrates the inputs relayed by components encoded on the actuator plasmid and allows our bacteria to autonomously decide whether they are currently located in a tumor environment or not.

Input 1: L-Lactate

Excessive production of lactate even under normoxic conditions¹ is a common feature of cancer cells due to the well-known Warburg effect, which is why we selected this marker as one of the two inputs integrated by our AND gate. LldP is a L-lactate permease transporting L-lactate across the bacterial membrane. At low lactate concentrations, lldR repressor proteins bind the two operator sequences O1 and O2 surrounding the pLux quorum sensing promoter and oligomerize. The oligomerization of bound lldR proteins leads to the formation of a DNA-loop which sequesters the pLux promoter, making it inaccessible for transcription as depicted in Figure 2.

**Figure 2:** In the presence of lower concentrations of L-lactate, lldR repressor proteins bind the operator sequences of the AND gate surrounding the pLux promoter and prevent transcription of genes positioned downstream of this promoter.

Input 2: AHL (quorum sensing)

Quorum sensing allows bacteria to restrict the expression of certain genes to high bacterial cell densities. N-Acyl homoserine lactone (AHL) is a bacterial messenger molecule which is constantly produced at very low levels by LuxI, diffuses across cell membranes and binds transcriptional activator LuxR. The higher the cell density, the higher the concentration of AHL present in the surrounding of each cell. The AHL:LuxR complex together activate transcription of genes positioned downstream of pLux.

CATE is encoded in E. coli Nissle which has the intrinsic ability to survive and home in tumors.² ³ Therefore, an accumulation of bacteria is expected only within tumor cells which is why we chose cell density as the second input. In addition, the integration of quorum sensing also ensures that a sufficient and therefore deadly amount of bacteria carrying the cytotoxic agent have accumulated within the tumor, lowering the incidences of drug resistance due to the incomplete elimination of the tumor.

In conclusion, a combination of high concentrations of lactate and AHL is a unique molecular pattern that should only occur in tumor tissue and therefore represents the only site which can unlock the full cytotoxic capabilities of CATE.

Genetic design

We reasoned that flanking the pLux promoter with O1 and O2 should result in an AND-gate behavior and relied on the previously characterized parts BBa_K1847007 (O1-pconst-O2), BBa_R0062 (plux) and BBa_C0062 (LuxR).

In design B of the AND gate, we adapted the architecture of BBa_K1847007. The Anderson promoter was exchanged with the pLux promoter such that a) the -35 and -10 regions of the new promoter were at the same position as that of the original one and that b) the total distance between O1 and O2 was kept the same as in the original part.

Further, we duplicated the lldR binding sites O1 and O2 in order to achieve a molecular zipper mechanism which would lead to a tighter DNA-loop as more LldR repressors are bound. This design in turn requires even higher, tumor-specific concentrations of L-lactate to completely unzip the loop.

**Figure 3:** (Left) AND gate design B. LldR repressor binding sites O1 and O2 were duplicated to generate a molecular zip, which results in a tighter DNA-loop and thus should reduce leakiness of the promoter. (Right) The molecular zip mechanism recruits more LldR repressor proteins, shifting the AND gate activation point to higher lactate levels.

Characterization

We determined the dose-response behavior of our AND gates to the two inducers AHL and L-lactate and in order to assess whether our designs would be capable to distinguish healthy and tumor tissue with respect to lactate and expected AHL concentrations.

Two plasmids, a regulator plasmid and an actuator plasmid encoding AND-gate and sGFP, were transformed into E. coli TOP10 as shown in Figure 4. Exponential-phase cultures were induced in microtiter plates with combinations of 8 different AHL and 8 different L-lactate concentrations and analyzed after 5.5 hours growth in the plate. A detailed protocol is available at Protocols

**Figure 4:** (Top) Circuit used to characterize AND gate behaviors. (Bottom) Dose-response of different AND-gate designs to L-lactate and AHL. Shown are geometric means of two replicates of fold changes over non-induced case. b) Similar experiment, analysis and plot but with an increased range of concentrations of both inducers.

All our synthetic AND gates increase the production of sGFP with increasing inducer concentrations, whereby AND gate B clearly exhibits the lowest level of leakiness as expected due its molecular zipper. Hence, the highest level of activation coincides with the highest amounts of both inducers. No activation is observed at low and intermediary concentrations of inducers.

Modeling

AND gate activation model

Given our experimental results, we created a new model which considers the leakiness of both inputs, L-lactate and AHL, separately instead of combining them into a global leakiness to fit our model better to reality.

where

Fit of the AND gate promoter response

From the experimental data of the AND gate response, we could fit 3 parameters of our model: both leakinesses of our hybrid promoters with regard to LuxR-AHL and lactate, and half-activation concentrations of LuxR-AHL.

**Figure 5:** Fit of the activation of AND gate B. Parameter space fitting the experimental data. Each point represents a parameter vector that significantly fit the experimental data. The blue points fit the data the best (least sum of square) while the yellow ones represent parameters combinations that barely fit the data (but still significant according to the chi2 test of goodness of fit). Fitted parameters are annotated in red.

From our statistically significant fit and under the assumptions made, we were able to determine the following parameters for the AND gate:

Summary

Our AND gate promoters allow CATE to distinguish levels of lactate and AHL in healthy tissue to those in tumor tissue
AND gate B exhibits the lowest leakiness due to its molecular zipper mechanism
We created a new model which realistically describes the behavior of the AND gate

References

1. F. Hirschhaeuser et al. "Lactate: a metabolic key player in cancer." Cancer Research (2011): 6921-6925
2. Y. Zhang et al. "Escherichia coli Nissle 1917 Targets and Restrains Mouse B16 Melanoma and 4T1 Breast Tumors through Expression of Azurin Protein" Applied and Environmental Microbiology (2012): 7603–7610
3. J. Stritzker et al. "Tumor-specific colonization, tissue distribution, and gene induction by probiotic Escherichia coli Nissle 1917 in live mice." International Journal of Medical Microbiology (2007): 151-162

@@ Line 6: / Line 6: @@
 <p>In the absence of high concentrations of L-lactate, LldR inhibitor proteins bind to the operator sites O1 and O2 surrounding the pLux promoter leading to the formation of a DNA loop. The pLux promoter is sequestered and inaccessible for transcriptional activation by the quorum sensing components.</p>
-<br>
-<center>
-<figure>
-    <img src="https://static.igem.org/mediawiki/2017/6/6a/T--ETH_Zurich--ANDgateB.png" width="550px">
-</figure>
-<b>Figure 1:</b>x
-</center>
 </html>
 <br>
-<p>All our synthetic AND gates increase the production of sGFP with increasing inducer concentrations. Hence, the highest level of activation coincides with the highest amounts of both inducers. No activation is observed at low and intermediary concentrations of inducers.</p>
 <h3><span class='h3bb'>Sequence and Features</span></h3>
@@ Line 25: / Line 14: @@
 <html>
-<!--
+<br>
-To realize the second level of control, namely avoid activation of our engineered system outside of tumors, we devised a strategy for CATE to be able to differentiate between healthy and tumor tissue. This is achieved by AND-logic integration of two inputs: AHL and Lactate. AHL is produced by the bacteria themselves and reaches high concentrations only in case the population is dense – ergo only in tumors. The second input is lactate, a chemical produced in high amounts by cancer cells as a consequence of defects in cellular respiration and other aberrations, even under normoxic conditions [2]. Hence, a combination of lactate and AHL is a unique molecular pattern that should only occur in tumor tissue.
-In order to achieve such an AND-gate behaviour, we relied on the previously characterized parts https://parts.igem.org/Part:BBa_K1847007 (LldP/LldR, Plldr) and https://parts.igem.org/Part:BBa_R0062 (Plux)/ https://parts.igem.org/Part:BBa_C0062 (LuxR). LldP is a transmembrane protein that imports L-lactate. It is thought that LldR binds to two sites on the DNA, termed O1 and O2, and leads to formation of a DNA-loop that “hides” the region in between O1 and O2 from the transcription complex. Upon binding of L-lactate to LldR it undergoes a conformational change that leads to opening of the loop whereby intervening regions are exposed again [3]. Conversely, LuxR dimerizes upon binding to N-Acyl homoserine lactone (AHL). The resulting complex then binds to the Plux promoter and recruits RNA polymerase [4].
-We reasoned that flanking the Plux promoter with O1 and O2 should lead to AND-gate behaviour. In the absence of Lactate, Plux should be looped and therefore inaccessible to LuxR, independent of AHL being present or not. If AHL is present but lactate is not, Plux is exposed but will not be activated. Only in the case where both inducers are present, no loop is formed and Plux is accessible to dimerized LuxR. See the illustration for a graphical representation of the action of our synthetic AND-gate promoter.-->
 <h3>Usage and Biology</h3>
-<p>text</p>
 <center
 <figure>
      <img src="https://static.igem.org/mediawiki/2017/a/af/T--ETH_Zurich--thecircuitclean.png" width="800px">
+       <figcaption><b>Figure 1:</b>Genetic circuit of CATE. CATE consists of an regulator (upper) and actuator (lower) plasmid. The AND gate integrates the inputs relayed by components encoded on the actuator plasmid and allows our bacteria to autonomously decide whether they are currently located in a tumor environment or not.</figcaption>
 </figure>
 </center>
-<div width="100%">
+<br>
 <b>Input 1: L-Lactate</b><br>
-</p>Excessive production of lactate even under normoxic conditions is a common feature of cancer cells due to the well-known Warburg effect, which is why we selected this marker as one of the two inputs integrated by our AND gate. LldP is a L-lactate permease transporting L-lactate across the bacterial membrane. At low lactate concentrations, lldR repressor proteins bind the two operator sequences O1 and O2 surrounding the pLux quorum sensing promoter and oligomerize. The oligomerization of bound lldR proteins leads to the formation of a DNA-loop which sequesters the pLux promoter, making it inaccessible for transcription as depicted in Figure 3.<p>
+</p>Excessive production of lactate even under normoxic conditions<sup>1</sup> is a common feature of cancer cells due to the well-known Warburg effect, which is why we selected this marker as one of the two inputs integrated by our AND gate. LldP is a L-lactate permease transporting L-lactate across the bacterial membrane. At low lactate concentrations, lldR repressor proteins bind the two operator sequences O1 and O2 surrounding the pLux quorum sensing promoter and oligomerize. The oligomerization of bound lldR proteins leads to the formation of a DNA-loop which sequesters the pLux promoter, making it inaccessible for transcription as depicted in Figure 2.<p>
+<center>
 <figure>
-     <img src="https://static.igem.org/mediawiki/2017/8/8d/T--ETH_Zurich--lldRmechanism.png" width="200px" align="right">
+     <img src="https://static.igem.org/mediawiki/2017/8/8d/T--ETH_Zurich--lldRmechanism.png" width="200px">
-       <figcaption><b>Figure 3:</b> In the presence of lower concentrations of L-lactate, lldR repressor proteins bind the operator sequences of the AND gate surrounding the pLux promoter and prevent transcription of genes positioned downstream of this promoter.</figcaption>
+       <figcaption><b>Figure 2:</b> In the presence of lower concentrations of L-lactate, lldR repressor proteins bind the operator sequences of the AND gate surrounding the pLux promoter and prevent transcription of genes positioned downstream of this promoter.</figcaption>
 </figure>
-</div>
+</center>
-<div width="100%">
 <br>
@@ Line 57: / Line 42: @@
 <p>Quorum sensing allows bacteria to restrict the expression of certain genes to high bacterial cell densities. N-Acyl homoserine lactone (AHL) is a bacterial messenger molecule which is constantly produced at very low levels by LuxI, diffuses across cell membranes and binds transcriptional activator LuxR. The higher the cell density, the higher the concentration of AHL present in the surrounding of each cell. The AHL:LuxR complex together activate transcription of genes positioned downstream of pLux.</p>
-<p>CATE is encoded in <i>E. coli</i> Nissle which has the intrinsic ability to survive and home in tumors. Therefore, an accumulation of bacteria is expected only within tumor cells which is why we chose cell density as the second input. In addition, the integration of quorum sensing also ensures that a sufficient and therefore deadly amount of bacteria carrying the cytotoxic agent have accumulated within the tumor, lowering the incidences of drug resistance due to the incomplete elimination of the tumor.</p>
+<p>CATE is encoded in <i>E. coli</i> Nissle which has the intrinsic ability to survive and home in tumors.<sup>2</sup> <sup>3</sup> Therefore, an accumulation of bacteria is expected only within tumor cells which is why we chose cell density as the second input. In addition, the integration of quorum sensing also ensures that a sufficient and therefore deadly amount of bacteria carrying the cytotoxic agent have accumulated within the tumor, lowering the incidences of drug resistance due to the incomplete elimination of the tumor.</p>
 <br>
@@ Line 65: / Line 48: @@
 <p>In conclusion, a combination of high concentrations of lactate and AHL is a unique molecular pattern that should only occur in tumor tissue and therefore represents the only site which can unlock the full cytotoxic capabilities of CATE.</p>
-<div>
 <h3>Genetic design</h3>
-<p> We reasoned that flanking the plux promoter with O1 and O2 should result in an AND-gate behavior and relied on the previously characterized parts <a href="https://parts.igem.org/Part:BBa_K1847007">BBa_K1847007</a> (O1-pconst-O2), <a href="https://parts.igem.org/Part:BBa_R0062">BBa_R0062</a> (plux) and <a href="https://parts.igem.org/Part:BBa_C0062">BBa_C0062</a> (LuxR). In design B of the AND gate, we duplicated the lldR binding sites O1 and O2 in order to achieve a molecular zipper mechanism which would lead to a tighter DNA-loop as more LldR repressors are bound. This design in turn requires even higher, tumor-specific concentrations of L-lactate to completely unzip the loop. The space separating the binding sites and pLux was adapted from BBa_K1847007.</p>
+<p> We reasoned that flanking the pLux promoter with O1 and O2 should result in an AND-gate behavior and relied on the previously characterized parts <a href="https://parts.igem.org/Part:BBa_K1847007">BBa_K1847007</a> (O1-pconst-O2), <a href="https://parts.igem.org/Part:BBa_R0062">BBa_R0062</a> (plux) and <a href="https://parts.igem.org/Part:BBa_C0062">BBa_C0062</a> (LuxR).</p>
+<p> In design B of the AND gate, we adapted the architecture of BBa_K1847007. The Anderson promoter was exchanged with the pLux promoter such that a) the -35 and -10 regions of the new promoter were at the same position as that of the original one and that b) the total distance between O1 and O2 was kept the same as in the original part.</p>
+<p>Further, we duplicated the lldR binding sites O1 and O2 in order to achieve a molecular zipper mechanism which would lead to a tighter DNA-loop as more LldR repressors are bound. This design in turn requires even higher, tumor-specific concentrations of L-lactate to completely unzip the loop.</p>
 <figure>
-     <img src="https://static.igem.org/mediawiki/2017/3/34/T--ETH_Zurich--lldrmechanismandgateB.png" width="200px" align="right">
+        <img src="https://static.igem.org/mediawiki/2017/6/6a/T--ETH_Zurich--ANDgateB.png" width="65%">
-          <figcaption><b>Figure 4:</b></figcaption>
+        <img src="https://static.igem.org/mediawiki/2017/3/34/T--ETH_Zurich--lldrmechanismandgateB.png" width="25%">
+          <figcaption><b>Figure 3:</b> (Left) AND gate design B. LldR repressor binding sites O1 and O2 were duplicated to generate a molecular zip, which results in a tighter DNA-loop and thus should reduce leakiness of the promoter. (Right) The molecular zip mechanism recruits more LldR repressor proteins, shifting the AND gate activation point to higher lactate levels.</figcaption>
 </figure>
-<div>
 <br>
 <h3>Characterization</h3>
-<p>We determined the dose-response behavior of our AND gates to the two inducers AHL and L-lactate and in order to assess whether our designs would be capable to distinguish healthy and tumor tissue with respect to lactate and expected AHL concentrations.</p> <p>Two plasmids, a regulator plasmid and an actuator plasmid encoding AND-gate and sGFP, were transformed into <i>E. coli</i> TOP10 as shown in Figure 5. Exponential-phase cultures were induced in microtiter plates with combinations of 8 different AHL and 8 different L-lactate concentrations and analyzed after 5.5 hours growth in the plate. A detailed protocol is available at <a href="http://2017.igem.org/Team:ETH_Zurich/Protocols">Protocols</a></p>
+<p>We determined the dose-response behavior of our AND gates to the two inducers AHL and L-lactate and in order to assess whether our designs would be capable to distinguish healthy and tumor tissue with respect to lactate and expected AHL concentrations.</p> <p>Two plasmids, a regulator plasmid and an actuator plasmid encoding AND-gate and sGFP, were transformed into <i>E. coli</i> TOP10 as shown in Figure 4. Exponential-phase cultures were induced in microtiter plates with combinations of 8 different AHL and 8 different L-lactate concentrations and analyzed after 5.5 hours growth in the plate. A detailed protocol is available at <a href="http://2017.igem.org/Team:ETH_Zurich/Protocols">Protocols</a></p>
 <center>
 <figure>
-     <img src="https://static.igem.org/mediawiki/2017/5/56/T--ETH_Zurich--WL_TS_ANDgate_wo_LuxI.png" width="50%"><br>
+     <img src="https://static.igem.org/mediawiki/2017/5/56/T--ETH_Zurich--WL_TS_ANDgate_wo_LuxI.png" width="50%"> <br><br>
      <img src="https://static.igem.org/mediawiki/2017/9/91/T--ETH_Zurich--AND_gate_induction.png" width="100%">
-        <figcaption><b>Figure 5:</b></figcaption>
+        <figcaption><b>Figure 4:</b> (Top) Circuit used to characterize AND gate behaviors. (Bottom) Dose-response of different AND-gate designs to L-lactate and AHL. Shown are geometric means of two replicates of fold changes over non-induced case.  b) Similar experiment, analysis and plot but with an increased range of concentrations of both inducers.</figcaption>
 </figure>
 </center>
-<p>All our synthetic AND gates increase the production of sGFP with increasing inducer concentrations, whereby AND gate C clearly exhibits the lowest level of leakiness as expected due its molecular zipper. Hence, the highest level of activation coincides with the highest amounts of both inducers. No activation is observed at low and intermediary concentrations of inducers.</p>
+<p>All our synthetic AND gates increase the production of sGFP with increasing inducer concentrations, whereby AND gate B clearly exhibits the lowest level of leakiness as expected due its molecular zipper. Hence, the highest level of activation coincides with the highest amounts of both inducers. No activation is observed at low and intermediary concentrations of inducers.</p>
-<br>
 <h3>Modeling</h3>
-<br>
 <b>AND gate activation model</b><br>
 <p>Given our experimental results, we created a new model which considers the leakiness of both inputs, L-lactate and AHL, separately instead of combining them into a global leakiness to fit our model better to reality.</p>
@@ Line 111: / Line 94: @@
 </center>
-<b>Fit of the AND gate promoter response</b>
+<br>
+<b>Fit of the AND gate promoter response</b><br>
+<p>From the experimental data of the AND gate response, we could fit 3 parameters of our model: both leakinesses of our hybrid promoters with regard to LuxR-AHL and lactate, and half-activation concentrations of LuxR-AHL.</p>
 <center>
 <figure>
-     <img src="https://static.igem.org/mediawiki/2017/f/fd/T--ETH_Zurich--fit_and_gate_b_annotated.png" width="100%">
+     <img src="https://static.igem.org/mediawiki/2017/f/fd/T--ETH_Zurich--fit_and_gate_b_annotated.png" width="70%">
+    <img src="https://static.igem.org/mediawiki/2017/e/ec/T--ETH_Zurich--fit_and_gate_b_plot.png" width="70%">
+<figcaption><b>Figure 5:</b> Fit of the activation of AND gate B. Parameter space fitting the experimental data. Each point represents a parameter vector that significantly fit the experimental data. The blue points fit the data the best (least sum of square) while the yellow ones represent parameters combinations that barely fit the data (but still significant according to the chi2 test of goodness of fit). Fitted parameters are annotated in red.</figcaption>
 </figure>
 </center>
+<p>From our statistically significant fit and under the assumptions made, we were able to determine the following parameters for the AND gate:</P>
+<figure>
+   <img src="https://static.igem.org/mediawiki/2017/3/3d/T--ETH_Zurich--andgatemodelparamters.png" width="100%">
+</figure>
 <h3>Summary</h3>
 <ul>
     <li>Our AND gate promoters allow CATE to distinguish levels of lactate and AHL in healthy tissue to those in tumor tissue</li>
-    <li>AND gate C exhibits the lowest leakiness due to its molecular zipper design</li>
+    <li>AND gate B exhibits the lowest leakiness due to its molecular zipper mechanism</li>
+   <li>We created a new model which realistically describes the behavior of the AND gate</li>
 </ul>
@@ Line 131: / Line 126: @@
 <!-- -->
-===References===
+<br>
+<h3>References</h3>
+. F. Hirschhaeuser et al. "Lactate: a metabolic key player in cancer." Cancer Research (2011): 6921-6925<br>
+. Y. Zhang et al. "Escherichia coli Nissle 1917 Targets and Restrains Mouse B16 Melanoma and 4T1 Breast Tumors through Expression of Azurin Protein" Applied and Environmental Microbiology (2012): 7603–7610<br>
+. J. Stritzker et al. "Tumor-specific colonization, tissue distribution, and gene induction by probiotic Escherichia coli Nissle 1917 in live mice." International Journal of Medical Microbiology (2007): 151-162