Team:Kyoto/Project/Goal C

From 2010.igem.org

(Difference between revisions)
(Conclusion)
(Goal C: Characterization of the anti-killer gene)
 
(18 intermediate revisions not shown)
Line 2: Line 2:
==Goal C: Characterization of the anti-killer gene==
==Goal C: Characterization of the anti-killer gene==
===Introduction===
===Introduction===
-
We selected SΔTMD1 as anti-killer gene of Lysis box.  
+
As far as we know, there are two anti-killer genes against the killer gene (lysis cassette), S107 and S<sub>&Delta;TMD1</sub>.
-
Lysis cassette encodes S gene, R gene and so on and the transmembrane domein 1(TMD1) of this S gene is essential for the function of lysis cassette as killer-gene.
+
S107 is expressed by S gene encoded by lysis cassette and S<sub>&Delta;TMD1</sub> is an S allele with the transmembrane domain 1(TMD1) deleted.
-
So S&Delta;TMD1, which is a S mutant with TMD1 deleted, dominant-negatively inhibits lysis cassette.(check [[Team:Kyoto/LearnMore#Lysis Cassette|Learn more]])
+
Because TMD1 of S gene is essential for the function of lysis cassette as killer gene, S<sub>&Delta;TMD1</sub> inhibits lysis cassette stronger than S107.([[Team:Kyoto/LearnMore#Lysis Cassette|Learn more]])
-
That is why we selected S&Delta;TMD1 as anti-killer gene.
+
So, we selected S<sub>&Delta;TMD1</sub> as anti-killer gene.
-
Then, we checked if S&Delta;TMD1 prevent lysis cassette from causing the cell death.  
+
Then, we checked if S<sub>&Delta;TMD1</sub> prevents lysis cassette from causing the cell death.  
-
To research this function of S&Delta;TMD1, we used the E.coli transformed with both these genes( <partinfo>BBa_K358021</partinfo>).
+
To investigate the function of S<sub>&Delta;TMD1</sub>, we used the E.coli transformed with both killer and anti-killer genes ( <partinfo>BBa_K358021</partinfo>).
-
Because E.coli is immediately dead if killer-gene is constitutively expressed, we made lysis cassette regulated by lac promoter. The E.coli were grown in the medium without IPTG, and after there were enough E.coli, they were cultured in the one with IPTG to express lysis cassette. After that, we checked, by measuring the absorbance (A550), if E.coli were alive due to anti-killer gene, S&Delta;TMD1 expressed constitutively.
+
Because E.coli will be killed immediately if the killer-gene is constitutively expressed, we constructed the lysis cassette regulated by lac promoter. At first, the E.coli bacteria were grown in the medium without IPTG. After the population of E.coli become enough, they were cultured in the one with IPTG to express lysis cassette. After that, we checked the growth rate of E.coli by measuring the absorbance (OD 550) of the growth medium. Thus, we investigated the possibility whether E.coli that contains both the killer and anti-killer genes could alive due to the constitutive expression of S<sub>&Delta;TMD1</sub>.  
===Method===
===Method===
====Bacterial strains====
====Bacterial strains====
-
We used four types of E. coli, E. coli KRX transformed with <partinfo>BBa_K358021</partinfo>, KRX transformed with <partinfo>BBa_K358022</partinfo> KRX transformed with <partinfo>BBa_K358024</partinfo> and KPX transfomed with  <partinfo>BBa_K358019</partinfo>.
+
We used four types of E. coli as follows: (1) E. coli KRX transformed with <partinfo>BBa_K358021</partinfo>, (2)KRX transformed with <partinfo>BBa_K358022</partinfo>, (3)KRX transformed with <partinfo>BBa_K358024</partinfo> and (4)KRX transformed with  <partinfo>BBa_K358019</partinfo>.
-
After this, We call each E.coli to abbreviate in order of name, LB2-1, lacLΔ, LB2-2 and lacL.
+
After this, we call each E.coli to abbreviate in order of name, LB2-1, lacL&Delta;, LB2-2 and lacL.
====Construct====
====Construct====
Line 21: Line 21:
{{clear}}
{{clear}}
*<partinfo>BBa_K358021</partinfo>:Anti-killergene is induced constitutively and killergene is regulated by lactose promoter.
*<partinfo>BBa_K358021</partinfo>:Anti-killergene is induced constitutively and killergene is regulated by lactose promoter.
-
It was used in experiment 1
+
It was used in experiment 2.
-
*<partinfo>BBa_K358022</partinfo>:lacP + SΔTMD1RRz, TMD1 is deleted from lysis cassette[SRRz].
+
*<partinfo>BBa_K358022</partinfo>:lacP + S<SUB>&DELTA;TMD1</SUB>RRz, TMD1 is deleted from lysis cassette[SRRz].
-
It was used in experiment 2
+
It was used in experiment 1.
-
*<partinfo>BBa_K358024</partinfo>:This part is the same to BBa_K358021, except for its constitutive promoter[BBa_J23101]. So, this part also induces constitutively SΔTMD1 gene, the anti-killer gene, and lambda lysis cassette[SRRz], the killer gene, is regulated by lacP.
+
*<partinfo>BBa_K358024</partinfo>: This part is similar to BBa_K358021, except for its constitutive promoter [BBa_J23101]. This part also induces constitutively S<sub>&DELTA;TMD1</sub> gene, the anti-killer gene, and &lambda; lysis cassette[SRRz], the killer gene is regulated by lacP.
-
It was used in the additional experiment, which is not finished.
+
It was used in the additional experiment which is not finished.
-
*<partinfo>BBa_K358019</partinfo>:Lysis cassette regulated by lacP, Lysis cassette[SRRz] is regulated by lactose promoter. Activating lactose promoter and expressing SRRz gene, it causes the cell lysis. So, lactose promoter must be repressed when transform this part.
+
*<partinfo>BBa_K358019</partinfo>: Lysis cassette regulated by lacP, Lysis cassette[SRRz] is regulated by lactose promoter.  
-
It was used in experiment 1 and 2
+
Activating lactose promoter and expressing SRRz gene, it causes the cell lysis. Thus, lactose promoter must be repressed when this part is transformed.
 +
It was used in experiment 1 and 2.
-
====Mesurement====
+
====Measurement====
=====Experiment 1=====
=====Experiment 1=====
-
We pick up three colonies from each plate, and cultivate them in the supplemented M9 medium overnight, about 16 hours. The overnight cultures were diluted to 0.1~0.11 in pre-warmed fresh the supplemented M9 medium. we measure A550 of the culture containing various IPTG at about 100 minute interval.
+
We picked up three colonies from each plate, and cultivated them in the supplemented M9 medium for a night, overnight. The overnight cultures were diluted to 0.1~0.11mM in pre-warmed fresh the supplemented M9 medium. 18h later, we measured OD 550 of the culture containing 0mM or 1mM IPTG.
=====Experiment 2=====
=====Experiment 2=====
-
We pick up three colonies from each plate, and cultivate them in the supplemented M9 medium for a night, overnight. The overnight cultures were diluted to 0.1~0.11mM in pre-warmed fresh the supplemented M9 medium. 18h later, we measure A550 of the culture containing 0mM or 1mM IPTG.
+
We picked up three colonies from each plate, and cultivated them in the supplemented M9 medium ~24-30 h (overnight). The overnight cultures were diluted to 0.1~0.11 in pre-warmed fresh the supplemented M9 medium. We measured OD 550 of the culture containing various IPTG at about 100 minute interval.
===Result===
===Result===
-
====Experiment 1: Function check of S&Delta;TMD1====
+
====Experiment 1:Check the role of TMD1 of S gene====
-
We measured the cell number of lacLΔ at several time after induction by adding several concentration IPTG.
+
[[image:GoalC fig3-2-2.png|450px|right|thumb|Fig.1.the comparison between lacL and lacL&Delta;. The black lines on bars' edges are error bar.]]
-
The result is below(Fig.1).
+
Then, in order to check if TMD1 of S gene is essential for this killer gene, we measured the cell number of lacL&Delta;, which does not have TMD1 of S gene and that is, encodes S<sub>&Delta;TMD1</sub> and that of lysis cassette.
 +
This result is below (Fig.1).
 +
From this result, after IPTG was added, the number of lacL&Delta; was not decreasing, while the number of lacL&Delta; was decresing.
 +
It proved that S<sub>&Delta;TMD1</sub> cannot function as killer gene and TMD1 is essential for this killer gene, lysis cassette.
 +
In addition, in the medium without IPTG, the number of lacL&Delta; is larger than that of lacL&Delta;. This supports a report <sup>[[#RefL001|[1]]]</sup>, which says that more short proteins are expressed than the ones that are long.
 +
====Experiment 2: Function check of S<sub>&Delta;TMD1</sub>====
 +
We measured the cell number of lacL&Delta; at several time after induction by adding several concentration IPTG.
 +
The result is below (Fig.2).
This result showed that in the medium with more than 0.03mM IPTG, the number of E. coli is decreasing at a certain point.
This result showed that in the medium with more than 0.03mM IPTG, the number of E. coli is decreasing at a certain point.
-
It demonstrated that In at least more than 0.03mM IPTG, S&Delta;TMD1 can't inhibit lysis cassette completly.
+
It demonstrated that In at least more than 0.03mM IPTG, S<sub>&Delta;TMD1</sub> can't inhibit lysis cassette completely.
-
Then, we measured the cell number of lacL in the same way, and compared the result of lacLΔ with that of lacL to check how S&Delta;TMD1 inhibited lysis cassette(Fig.2).
+
Then, we measured the cell number of lacL in the same way, and compared the result of lacL&Delta; with that of lacL to check how S<sub>&Delta;TMD1</sub> inhibited lysis cassette(Fig.2).
-
In Fig.2, LacS is lacL, CTLS is the lacLΔ.
+
From this result, we knew in more than 0.5mM IPTG both lacL and LB2-1 were decreasing in the number at a certain point, or LB2-1 was a little delayed decreasing in the number.  
-
From this result, we knew in more than 0.5mM IPTG both Lacs and CTLA were decreasing in the number at a certain point, or CTLA was a little delayed decreaing in the number.  
+
This demonstrated that S<sub>&Delta;TMD1</sub> can't correctly inhibit lysis cassette.  
-
This demonstrated that S&Delta;TMD1 can't correctly inhibit lysis cassette.  
+
So, we doubted whether S<sub>&Delta;TMD1</sub> really functioned as anti-killer gene.
-
So, we doubted whether S&Delta;TMD1 really functioned as anti-killer gene.
+
Although S<sub>&Delta;TMD1</sub> has no TMD1 of S gene, S<sub>&Delta;TMD1</sub> couldn't show the function as anti-killer gene.
-
Although S&Delta;TMD1 has no TMD1 of S gene, S&Delta;TMD1 couldn't show the function as anti-killer gene.
+
-
[[image:GoalC fig1.png|440px|Fig.1|left|thumb|Fig.1.the result of function check of S&Delta;TMD1.
+
[[image:GoalC fig1.png|440px|Fig.1|left|thumb|Fig.2.the result of function check of S<sub>&Delta;TMD1</sub>.
-
we measure A 550 changing IPTG concentration, 0, 0.02, 0.03, 0.05, 0.3, 0.5, 0.7, 1.0 (mM). ]]
+
We measure OD 550 with various IPTG concentration, 0, 0.02, 0.03, 0.05, 0.3, 0.5, 0.7, 1.0 (mM). ]]
-
[[image:GoalC fig2-2.png|440px|right|thumb|Fig.2.Comparison between lacL and LB2-1.
+
[[image:GoalC fig2-3-2.png|440px|right|thumb|Fig.3.Comparison between lacL and LB2-1.
-
we compared at IPTG concentration 0, 0.02, 0.03, 0.5, 1.0. ]]
+
We compared IPTG concentration 0, 0.02, 0.03, 0.5, 1.0. ]]
-
====Experiment 2:Check the role of TMD1 of S gene====
+
===Discussion===
-
[[image:GoalC fig3-2-2.png|450px|right|thumb|Fig.3.the comparion between lacL and lacL&Delta;.The black lines on bars' edges are error bar.]]
+
====TMD1 of S gene is essential for lysis cassette====
-
Then, in order to check if TMD1 of S gene is essential for this killer gene, we measured the cell number of lacL&Delta;, which does not have TMD1 of S gene and that is, encodes S&Delta;TMD1 and that of lysis cassette.
+
The result of experiment 1 supports [[#RefL001|Ref.1]].
-
This result is below(Fig.3).
+
That is, we were able to make sure that TMD1 of S gene is essential for lysis cassette as killer gene.
-
From this result, after IPTG was added,the number of lacL&Delta; is not decreasing, while the number of lacL&Delta; is decresing.
+
-
It proved that S&Delta;TMD1 cannot function as killer gene and TMD1 is essential for this killer gene, lysis cassette.
+
-
In addition, in the medium without IPTG,  the number of lacL&Delta; is larger than that of lacL&Delta;.This supports a report <sup>[[#RefL001|[1]]]</sup>.  which says that more proteins which is short are expressed than the ones which is long.
+
-
===Discussion===
+
====S<sub>&Delta;TMD1</sub> cannot inhibit lysis cassette completely====
-
====S&Delta;TMD1 cannot inhibit lysis cassette completly====
+
These results showed that S<sub>&Delta;TMD1</sub> cannot inhibit completely lysis cassette.
-
These results showed that S&Delta;TMD1 cannot inhibit completely lysis cassette.
+
This results can't supported a report on lysis <sup>[[#RefL001|[1]]]</sup>, which says that S<sub>&Delta;TMD1</sub> without TMD1 dominant-negatively inhibit lysis cassette.  
-
This results can't supported a report on lysis <sup>[[#RefL001|[1]]]</sup>which says that S&Delta;TMD1 without TMD1 dominant-negatively inhibit lysis cassette.  
+
Why did we get the result against that report?  
Why did we get the result against that report?  
Probably, there were three reasons below, we thought.
Probably, there were three reasons below, we thought.
-
*S&Delta;TMD1 in the plasmid did not worked correctly.
+
*S<sub>&Delta;TMD1</sub> in the plasmid did not worked correctly.
-
*S&DeltaTMD1 is weaker anti-killer gene than we expected.
+
*S<sub>&Delta;TMD1</sub> is weaker anti-killer gene than we expected.
-
*less S&Delta;TMD1 proteins or more lysis cassette proteins was expressed than we expected.
+
*less S<sub>&Delta;TMD1</sub> proteins or more lysis cassette proteins was expressed than we expected.
-
 
+
-
====Reason1:S&Delta;TMD1 in the plasmid did not worked correctly.====
+
-
To make sure the reason, we sequenced BBa_K358021 we used.
+
-
It’s result showed that this construct is inserted into the plasmid  correctly, which demonstrated  that S&Delta;TMD1 worked correctly.
+
-
So, first reason was not right.
+
-
====Reason2:S&DeltaTMD1 is weaker anti-killer gene than we expected.====
+
-
As anti-killer gene against lysis cassette, there are two genes, S107 and S&Delta;TMD1([[Team:Kyoto/LearnMore#Lysis Cassette|Learn more]]).
+
-
S107 is anti-killer gene, but because it has essential TMD1 of S for function of this killer gene, it’s function as anti-killer gene is weak compared with S&Delta;TMD1, which does not have TMD1 of ([[Team:Kyoto/LearnMore#Lysis Cassette|Learn more]]).
+
-
So, we thought that S&Delta;TMD1 is a strong anti-killer gene, but from result of experiment 1 we doubted S&Delta;TMD1 was  a strong anti-killer gene. However, I could not make sure the reason.
+
-
 
+
-
====Reason3:less S&Delta;TMD1 proteins or more lysis cassette proteins was expressed than we expected.====
+
-
in order to make sure this reason, we researched the expression level of killer gene, Lysis cassette and S&Delta;TMD1. 
+
-
As a report <sup>[[#RefL002|[2]]]</sup> says, it is necessary for the inhibition of lysis cassette that there are at least 4.73 times more S107, anti-killer gene than lysis cassette proteins.
+
-
So, Inhibition of this killer gene needs at least 4.73 times more anti-killer gene than killer gene. Then, we calculated the ratio of killer gene and anti-killer gene, assuming that S gene, which lysis cassette includes, produces 2.5:1 S107 and S105([[Team:Kyoto/LearnMore#Lysis Cassette|Learn more]]), strength of S&DeltaTMD1 and S107 is the same. By this calcuration, we knew that in LB-1, the anti-killer gene couldn't inhibit killer gene in the medium with more than 0.1246mM IPTG. So, we insist that it is because S&Delta;TMD1 was not expressed  adequately that S&Delta;TMD1 cannot inhibit lysis cassette completly.
+
-
 
+
-
====Conclusion====
+
-
We insist that the 3rd reason is right. Certainly, we have not made sure that 2nd Reason is not right. However, a report<sup>[[#RefL001|[1]]]</sup> says that S&Delta;TMD1 inhibits lysis cassette. So, It is natural that we think S&Delta;TMD1 have enough activity to inhibit lysis cassette.That is why the 3rd reason is right.
+
-
Then, in order to make sure the 3rd reason, we changed the promoter of S&Delta;TMD1 to stronger one, and now, are checking the function of S&Delta;TMD1. Surely, this result will help us discuss why S&Delta;TMD1 couldn’t function as we expected.
+
=====Reason1:S<sub>&Delta;TMD1</sub> in the plasmid did not work correctly.=====
-
You must expect it!!
+
To investigate this reason, we sequenced BBa_K358021 we used.
 +
The result showed that this construct is inserted into the plasmid correctly, which demonstrated that S<sub>&Delta;TMD1</sub> worked correctly.
 +
So, we assume that the first reason is negative.
 +
=====Reason2: S<sub>&Delta;TMD1</sub> is weaker anti-killer gene than we expected.=====
 +
As anti-killer gene against lysis cassette, there are two genes, S107 and S<sub>&Delta;TMD1</sub> ([[Team:Kyoto/LearnMore#Lysis Cassette|Learn more]]).
 +
S107 is the anti-killer gene. Because it has essential TMD1 of S for function of this killer gene, the anti-killer function of this gene is weak compared with S<sub>&Delta;TMD1</sub>, which does not have TMD1([[Team:Kyoto/LearnMore#Lysis Cassette|Learn more]]).
 +
Thus, we expected that S<sub>&Delta;TMD1</sub> is a strong anti-killer gene. However, from the result of experiment 2, we could not observe the function of S<sub>&Delta;TMD1</sub> as a strong anti-killer gene under the tested condition.
 +
=====Reason3: Expression balance between the anti-killer and killer genes was not optimized properly=====
 +
In order to investigate the possibility of this reason, we analyzed the expression level of the killer gene, lysis cassette and S<sub>&Delta;TMD1</sub>. 
 +
As a report <sup>[[#RefL002|[2]]]</sup> says, it is necessary for the inhibition of lysis cassette that there are at least ~4.7 times more S107, anti-killer gene than lysis cassette proteins.
 +
So, inhibition of this killer gene needs at least ~4.7 times more expression of the anti-killer gene than that of the killer gene. Thus, we calculated the ratio of the expression levels of the killer and anti-killer gene. The calculation was done under the assumptions that S gene, which includes lysis cassette, produces S105 and S107 at a 2.5:1 ratio([[Team:Kyoto/LearnMore#Lysis Cassette|Learn more]]), and that the strength of function of S<sub>&Delta;TMD1</sub> as anti-killer gene is the same as that of S107. By this calculation, we knew that in LB-1, the anti-killer gene couldn't inhibit killer gene in the medium with more than 0.1246 mM IPTG. So, we assume that S<sub>&Delta;TMD1</sub> was not expressed adequately to inhibit lysis cassette completely.
 +
We assume that our experimental results could be due to the 3rd reason (Reason 3) mentioned above, although we cannot rule out the possibility of the second reason. However, it has been reported <sup>[[#RefL001|[1]]]</sup> that S<sub>&Delta;TMD1</sub> inhibits lysis cassette effectively. So, it is reasonable to assume that S<sub>&Delta;TMD1</sub> exhibits enough activity to inhibit lysis cassette. That is why we think that the Reason 3 is the most likely answer. To confirm this possibility, we changed the promoter of S<sub>&Delta;TMD1</sub> to stronger one. The experiment to check the function of S<sub>&Delta;TMD1</sub> using the strong promoter is currently underway. Surely, this result will help us to understand why S<sub>&Delta;TMD1</sub> could not function as we expected and to develop distinct cell-death controllable devices for the further studies.  
 +
===Conclusion===
 +
1)We made and registered S<sub>&Delta;TMD1</sub> by deleting TMD1 of S gene and  lysis cassette&Delta;TMD1(S<sub>&Delta;TMD1</sub>RRz) in BioBrick as new parts.
 +
2)By the comparison of the function as killer gene between lysis cassette and lysis cassette&Delta;TMD1, We made sure that TMD1 is essential for fuction of lysis cassette as killer gene.
 +
3)By co-expressing lysis cassette regulated by a lactose promoter, we characterized the function of S<sub>&Delta;TMD1</sub> as anti-killer gene.
===Reference===
===Reference===
----
----
# <html><a name="RefL001"></a></html>Rebecca White, Tram Anh T. Tran, Chelsey A. Dankenbring, John Deaton, and Ry Young.2010.The N-Terminal Transmembrane Domain of _ S Is Required for Holin but Not Antiholin Function_. J. Bacteriol. 192: 725–733  
# <html><a name="RefL001"></a></html>Rebecca White, Tram Anh T. Tran, Chelsey A. Dankenbring, John Deaton, and Ry Young.2010.The N-Terminal Transmembrane Domain of _ S Is Required for Holin but Not Antiholin Function_. J. Bacteriol. 192: 725–733  
-
# <html><a name="RefL002"></a></html>CHUNG-YU CHANG, KIEBANG NAM, AND RY YOUNG1. 1995.2 S Gene Expression and the Timing of Lysis by Bacteriophage λ. J. Bacteriol. 177: 3283–3294
+
# <html><a name="RefL002"></a></html>CHUNG-YU CHANG, KIEBANG NAM, AND RY YOUNG1. 1995.2 S Gene Expression and the Timing of Lysis by Bacteriophage &lambda;. J. Bacteriol. 177: 3283–3294

Latest revision as of 03:55, 28 October 2010

Contents

Goal C: Characterization of the anti-killer gene

Introduction

As far as we know, there are two anti-killer genes against the killer gene (lysis cassette), S107 and SΔTMD1. S107 is expressed by S gene encoded by lysis cassette and SΔTMD1 is an S allele with the transmembrane domain 1(TMD1) deleted. Because TMD1 of S gene is essential for the function of lysis cassette as killer gene, SΔTMD1 inhibits lysis cassette stronger than S107.(Learn more) So, we selected SΔTMD1 as anti-killer gene.

Then, we checked if SΔTMD1 prevents lysis cassette from causing the cell death. To investigate the function of SΔTMD1, we used the E.coli transformed with both killer and anti-killer genes ( <partinfo>BBa_K358021</partinfo>). Because E.coli will be killed immediately if the killer-gene is constitutively expressed, we constructed the lysis cassette regulated by lac promoter. At first, the E.coli bacteria were grown in the medium without IPTG. After the population of E.coli become enough, they were cultured in the one with IPTG to express lysis cassette. After that, we checked the growth rate of E.coli by measuring the absorbance (OD 550) of the growth medium. Thus, we investigated the possibility whether E.coli that contains both the killer and anti-killer genes could alive due to the constitutive expression of SΔTMD1.

Method

Bacterial strains

We used four types of E. coli as follows: (1) E. coli KRX transformed with <partinfo>BBa_K358021</partinfo>, (2)KRX transformed with <partinfo>BBa_K358022</partinfo>, (3)KRX transformed with <partinfo>BBa_K358024</partinfo> and (4)KRX transformed with <partinfo>BBa_K358019</partinfo>. After this, we call each E.coli to abbreviate in order of name, LB2-1, lacLΔ, LB2-2 and lacL.

Construct

KyotoFigC001.pngKyotoFigC003.pngKyotoFigC002.pngKyotoFigC004.png

  • <partinfo>BBa_K358021</partinfo>:Anti-killergene is induced constitutively and killergene is regulated by lactose promoter.

It was used in experiment 2.

  • <partinfo>BBa_K358022</partinfo>:lacP + S&DELTA;TMD1RRz, TMD1 is deleted from lysis cassette[SRRz].

It was used in experiment 1.

  • <partinfo>BBa_K358024</partinfo>: This part is similar to BBa_K358021, except for its constitutive promoter [BBa_J23101]. This part also induces constitutively S&DELTA;TMD1 gene, the anti-killer gene, and λ lysis cassette[SRRz], the killer gene is regulated by lacP.

It was used in the additional experiment which is not finished.

  • <partinfo>BBa_K358019</partinfo>: Lysis cassette regulated by lacP, Lysis cassette[SRRz] is regulated by lactose promoter.

Activating lactose promoter and expressing SRRz gene, it causes the cell lysis. Thus, lactose promoter must be repressed when this part is transformed. It was used in experiment 1 and 2.

Measurement

Experiment 1

We picked up three colonies from each plate, and cultivated them in the supplemented M9 medium for a night, overnight. The overnight cultures were diluted to 0.1~0.11mM in pre-warmed fresh the supplemented M9 medium. 18h later, we measured OD 550 of the culture containing 0mM or 1mM IPTG.

Experiment 2

We picked up three colonies from each plate, and cultivated them in the supplemented M9 medium ~24-30 h (overnight). The overnight cultures were diluted to 0.1~0.11 in pre-warmed fresh the supplemented M9 medium. We measured OD 550 of the culture containing various IPTG at about 100 minute interval.

Result

Experiment 1:Check the role of TMD1 of S gene

Fig.1.the comparison between lacL and lacLΔ. The black lines on bars' edges are error bar.

Then, in order to check if TMD1 of S gene is essential for this killer gene, we measured the cell number of lacLΔ, which does not have TMD1 of S gene and that is, encodes SΔTMD1 and that of lysis cassette. This result is below (Fig.1). From this result, after IPTG was added, the number of lacLΔ was not decreasing, while the number of lacLΔ was decresing. It proved that SΔTMD1 cannot function as killer gene and TMD1 is essential for this killer gene, lysis cassette. In addition, in the medium without IPTG, the number of lacLΔ is larger than that of lacLΔ. This supports a report [1], which says that more short proteins are expressed than the ones that are long.

Experiment 2: Function check of SΔTMD1

We measured the cell number of lacLΔ at several time after induction by adding several concentration IPTG. The result is below (Fig.2). This result showed that in the medium with more than 0.03mM IPTG, the number of E. coli is decreasing at a certain point. It demonstrated that In at least more than 0.03mM IPTG, SΔTMD1 can't inhibit lysis cassette completely. Then, we measured the cell number of lacL in the same way, and compared the result of lacLΔ with that of lacL to check how SΔTMD1 inhibited lysis cassette(Fig.2). From this result, we knew in more than 0.5mM IPTG both lacL and LB2-1 were decreasing in the number at a certain point, or LB2-1 was a little delayed decreasing in the number. This demonstrated that SΔTMD1 can't correctly inhibit lysis cassette. So, we doubted whether SΔTMD1 really functioned as anti-killer gene. Although SΔTMD1 has no TMD1 of S gene, SΔTMD1 couldn't show the function as anti-killer gene.


Fig.2.the result of function check of SΔTMD1. We measure OD 550 with various IPTG concentration, 0, 0.02, 0.03, 0.05, 0.3, 0.5, 0.7, 1.0 (mM).
Fig.3.Comparison between lacL and LB2-1. We compared IPTG concentration 0, 0.02, 0.03, 0.5, 1.0.

Discussion

TMD1 of S gene is essential for lysis cassette

The result of experiment 1 supports Ref.1. That is, we were able to make sure that TMD1 of S gene is essential for lysis cassette as killer gene.

SΔTMD1 cannot inhibit lysis cassette completely

These results showed that SΔTMD1 cannot inhibit completely lysis cassette. This results can't supported a report on lysis [1], which says that SΔTMD1 without TMD1 dominant-negatively inhibit lysis cassette. Why did we get the result against that report? Probably, there were three reasons below, we thought.

  • SΔTMD1 in the plasmid did not worked correctly.
  • SΔTMD1 is weaker anti-killer gene than we expected.
  • less SΔTMD1 proteins or more lysis cassette proteins was expressed than we expected.
Reason1:SΔTMD1 in the plasmid did not work correctly.

To investigate this reason, we sequenced BBa_K358021 we used. The result showed that this construct is inserted into the plasmid correctly, which demonstrated that SΔTMD1 worked correctly. So, we assume that the first reason is negative.

Reason2: SΔTMD1 is weaker anti-killer gene than we expected.

As anti-killer gene against lysis cassette, there are two genes, S107 and SΔTMD1 (Learn more). S107 is the anti-killer gene. Because it has essential TMD1 of S for function of this killer gene, the anti-killer function of this gene is weak compared with SΔTMD1, which does not have TMD1(Learn more). Thus, we expected that SΔTMD1 is a strong anti-killer gene. However, from the result of experiment 2, we could not observe the function of SΔTMD1 as a strong anti-killer gene under the tested condition.

Reason3: Expression balance between the anti-killer and killer genes was not optimized properly

In order to investigate the possibility of this reason, we analyzed the expression level of the killer gene, lysis cassette and SΔTMD1. As a report [2] says, it is necessary for the inhibition of lysis cassette that there are at least ~4.7 times more S107, anti-killer gene than lysis cassette proteins. So, inhibition of this killer gene needs at least ~4.7 times more expression of the anti-killer gene than that of the killer gene. Thus, we calculated the ratio of the expression levels of the killer and anti-killer gene. The calculation was done under the assumptions that S gene, which includes lysis cassette, produces S105 and S107 at a 2.5:1 ratio(Learn more), and that the strength of function of SΔTMD1 as anti-killer gene is the same as that of S107. By this calculation, we knew that in LB-1, the anti-killer gene couldn't inhibit killer gene in the medium with more than 0.1246 mM IPTG. So, we assume that SΔTMD1 was not expressed adequately to inhibit lysis cassette completely. We assume that our experimental results could be due to the 3rd reason (Reason 3) mentioned above, although we cannot rule out the possibility of the second reason. However, it has been reported [1] that SΔTMD1 inhibits lysis cassette effectively. So, it is reasonable to assume that SΔTMD1 exhibits enough activity to inhibit lysis cassette. That is why we think that the Reason 3 is the most likely answer. To confirm this possibility, we changed the promoter of SΔTMD1 to stronger one. The experiment to check the function of SΔTMD1 using the strong promoter is currently underway. Surely, this result will help us to understand why SΔTMD1 could not function as we expected and to develop distinct cell-death controllable devices for the further studies.

Conclusion

1)We made and registered SΔTMD1 by deleting TMD1 of S gene and lysis cassetteΔTMD1(SΔTMD1RRz) in BioBrick as new parts. 2)By the comparison of the function as killer gene between lysis cassette and lysis cassetteΔTMD1, We made sure that TMD1 is essential for fuction of lysis cassette as killer gene. 3)By co-expressing lysis cassette regulated by a lactose promoter, we characterized the function of SΔTMD1 as anti-killer gene.

Reference


  1. Rebecca White, Tram Anh T. Tran, Chelsey A. Dankenbring, John Deaton, and Ry Young.2010.The N-Terminal Transmembrane Domain of _ S Is Required for Holin but Not Antiholin Function_. J. Bacteriol. 192: 725–733
  2. CHUNG-YU CHANG, KIEBANG NAM, AND RY YOUNG1. 1995.2 S Gene Expression and the Timing of Lysis by Bacteriophage λ. J. Bacteriol. 177: 3283–3294