Revision as of 12:49, 17 October 2005

Back to the ETH Zurich main page.

Description of the NOR module

This module behaves like a NOR gate. It has two inputs and one output. The output is high only when neither input 1 nor input 2 is high.

NOR Gate

In the figure below is a parts-view of the NOR gate module with PoPS interfaces (i/o).

              ---------------------------------
              |  ------------\                |
PoPS_in1 ---->|  | Repressor1 | --------      |
              |  ------------/          |     |
              |                         =     | 
              |                       --------|------> PoPS_out
              |                         =     | 
              |  ------------\          |     |
PoPS_in2 ---->|  | Repressor2 | --------      |
              |  ------------/                |
              ---------------------------------

J-K Flip-Flop

The central part of the counter is a J-K flip-flop (a.k.a. interconnected NOR module). When input changes the flip flop starts to express the next repressor. Every Repressor represses all other repressors than the next. For example R1 represses R3 and R4 and in the next cycle the flip flop is forced to express R2.

R1          R3 R2         R4 R1         R3
   \         | |           | |         /        
    \/|___   | |           | |  ____|\/
 ___/  (1) |_=_=____   ____=_=_| (3)  \___
    \  ____|     ^       ^     |____  /
     \|          |       |          |/
                 \       /
                  ---S---
                 /       \
                 |       |
     /|____      |       |      ____|\
 ___/  (2) |_____=__   __=_____| (4)  \___
    \  ____| = =           = = |____  /
    /\|      | |           | |      |/\
  R2         | |           | |         R4
            R4 R3         R1 R2

Design

Repressors

The Interconnected NOR module consists of four repressors. We considered several alternatives but finally decided to put all out bets on a approach based on zinc finger proteins.

RNA polymerase...

Zinc Finger Proteins

A Zinc Finger (ZF) [http://en.wikipedia.org/wiki/Zinc_finger] is a protein domain [http://en.wikipedia.org/wiki/Protein_domain] that binds to three base pairs of double stranded DNA. A Zinc Finger Protein (ZFP) consists of one or several zinc finger domains. Many protein-DNA interaction for ZF domains and triplet of base pairs have been described, therefore making it possible to to construct artificial transcription factors by combining ZF domains in a modular fashion. The idea is to use a ZFP as a repressor by putting a binding site for a ZFP upstream of the coding region and thereby preventing RNA polymerase to transcribe the gene.

We use several different designs of ZFPs.

- Three ZF domains to recognize nine base pairs.

- Two ZF domains and a leucine zipper domain. The leucine zipper will be cause the protein to home-dimerize and hence it will recognize 12 base pairs.

- Six ZF domains to recognize twelve base pairs.

- Three ZF domains fused to negative transcription factors. (e.g. NTD and ERD)

Zinc Finger Protein Binding Sites

Since we are going to synthesize parts of our module, we have the possibility to design our own ZFP binding sites, as well as our ZFPs. We have two alternatives for placing the binding sites.

1. Binding sites in the promoter. This would prevent the polymerase from binding to the promotor. Although this might be most likely to work, we have chosen to not pursue this alternative, while we are quite intrigued by the possibility of a roadblock.

2. Binding sites directly after the start of transcription and before the ribosome binding site. This alternative is attractive, since it would allow for a high degree of modularity. In theory the ZFP would act as an extra "roadblock-operator" and any promotor could be used in front of the protein.

The ZFP roadblock operator regions (from now on refered to as: operators) consists of binding motifs for two repressors. The two bindning motifs are usually spaced with 5 base pairs (gcgcg). Some data of the binding strength were available and we have chosen operators with the affinity estimated to K_d = 3-40 nM.

Part	Binding Site Seq	# exact matches	Hamming Dist 1	Hamming Dist 2	Hamming Dist 3
BBa_J05100	ggaggggac	1	132	3125	32788
BBa_J05102	ggaggcggg	5	396	6293	47897
BBa_J05101	gggggcgag	8	303	4955	41302
BBa_J05103	ggggccgga	23	433	5606	45492
BBa_J05108	gtcccctccggaggggac	0	0	0	0
BBa_J05109	ctcgcccccgggggcgag	0	0	0	2
BBa_J05110	cccgcctccggaggcggg	0	0	0	3
BBa_J05111	tccggccccggggccgga	0	0	0	3
BBa_J05115	gggggcgaggggggcgag	0	0	0	1

We need to have 4 operators = 4 permutations of ZFP. The operator regions (i.e. the "roadblocks" that will prevent of the RNApolymerase to transcribe the gene) form a BioBrick that should be inserted between the promoter region and the ribosome binding site in order to keep the design modular.

Zinc Finger Tester

To test whether our assumptions about using multiple zinc finger proteins (ZFP) as repressors (i.e. roadblocks) will actually work, we will build a tester/debugging device in parallel with the counter.

Interconnected NOR System

The NOR system consists of 4 proteins with 4 operator regions. It has an interface boundary with input module (Pr and Prm). Repressor R3 is connected to a reporter to be able to count to modulo two.

Repression

Note that in both cases, it is still under discussion as whether the repression domain should really be included or not. If it turns out that repression has been used in all the literature we (well, actually Hervé) can find, then we would be better off to play safe and include them:

Beerli PNAS 1998: they fused ZF to KRAB repressor and it has a stronger effect compare to ZF alone (see p14632 graph A)
Beerli Nat Biotech review feb2002 : if you read the complete paragraph on gene repression (p 132), it gives strong evidence that we should fuse the ZF to a repressor (apparently at the N-Term of the ZF) (either KRAB or SID repressor). It is stated that polymerase blockade through ZF only is not very efficient.

However, the discussion is ongoing: As Alex pointed out we can not be sure that the 45 amino acid long KRAB domain will work, if it doesn't exist in prokaryotes. KRAB domains have a very specific interaction with a co-repressor molecule. It might be better to find something already known to work in bacteria or skip the repressor approach altogether. An alternative could be a negative transcription domain fused to the ZFP.

Degradation

We considered adding a variant with degradation tag, since the system might be slow otherwise. However, due to the fact that we have limited resources - in terms of money, manpower, and time - and the fact that the degradation is decoupled and it will be clearly visible if the system turns out to be too slow, we decided against it for now.

Additional Comments

We changed our strategy a couple of times in heated debates. There is quite obviously a trade-off between probability of success through redundancy, i.e. trying multiple approaches in parallel, and limited resources, i.e. financial, manpower, and time. Thus we decided against certain additional variants and cloning steps since we simply won't have the time for it. Another important aspect is the overall goal of keeping the design modular - one of the most important aspects of the this contest. The modification of the promoter regions of the lambda-system, i.e. Pr and Prm, although a valid approach will probably be dropped, since it is too much work and cost and no longer modular (as opposed to having the operators after the promoter).

There has been some discussion about having the operator around 0 (-9/+1)

Synthesis and Assembly

Ordered Sequences
Nickname	Cost	SacI	KpnI	NheI	EcoRI	XbaI	SpeI	PstI
BBa_J05215+BBa_J05108	$874.35	1	91	598	7, 97	22, 112	71, 578	85, 592
BBa_J05216+BBa_J05109	$843.9	1	91	577	7, 97	22, 112	71, 557	85, 571
BBa_J05217+BBa_J05110	$883.05	1	91	604	7, 97	22, 112	71, 584	85, 598
BBa_J05218+BBa_J05111	$870	1	91	595	7, 97	22, 112	71, 575	85, 589
BBa_J05221+BBa_J05101	$722.1	1	112	493	7, 118	22, 133	92, 473	106, 487
BBa_J05114	$1238.3	1	849	-	7	22	829	843
BBa_J05112	$759.8	1	519	-	7	22	499	513
BBa_J05222+BBa_J05115	$980.2	1	87	671	7, 93	22, 108	67, 651	81, 665

ZFP Tester

Cut each system into BioBricks Pieces using different restriction enzymes.
Assemble in parallel for each genes 1-4 in two variants: promoter+operator, rbs+gene
Assemble the two constructs
...