Protein-DNA Computation by Stochastic Assembly Cascade

Roy Bar-Ziv, Tsvi Tlusty, and Albert Libchaber

PNAS Sep.3, 2002

Presented by Je-Gun Joung

Overview

• The assembly of RecA on single-stranded DAN is measured and interpreted as a stochastic finite-state machine

• This machine is able to discriminate fine differences between sequences

• In vivo stochastic biomolecular computation based on low-specificity protein-DNA binding

Molecular Machine by Stochastic Assembly Cascade

• Hardware– A test tube filled with a solution of single-stranded DN

A molecule, RecA protein, and ATP molecules

– Output device (molecular machine reporters): a change in the rotational motion of the DNA molecule

• Software– Two basic operation: nucleation and disassembly

– Chang the machine’s internal state (the current length of RecA)

Molecular Machine by Stochastic Assembly Cascade

RecA

ATP

ssDNA

Hardware

Output device(fluorescence anisotropy)

QN

QN-1

Q2

Q1

Q0

……

Two basic operation (nucleation and disassembly)

Software

RecA• The protein responsible for recombination in bacteria

5’ssDNA

RecA coats ssDNA

Alignment with homologous DNA sequence

Formation of joint chromosomal DNA

3’

5’5’3’

3’

RecA

RecA

RecA/Rad51 DNA Filaments

RecA Monomer

Biochemistry of RecA Assembly

TAC TAC TAC TAC TAC TAC TAC TAC TAC TAC TAC TAC TAC

5’ 3’

• Nucleation (assembly): extend the filament by long chunks

TAC TAC TAC TAC TAC TAC TAC TAC TAC TAC TAC TAC TAC5’ 3’

• Disassembly: removes monomers one by one

Simulation of a 26-state Stochastic Assembly Cascade

QN

QN-1

Q2

Q1

Q0

……

A Deterministic Reading Head Changes its Internal State

x e8 x d8 d7 x x d6 d5 x e9 d9 x

Q6

QN QN-1 Q2 Q1 Q0………

………

dn : disassembly from state Qn

en : nucleation to state Qn

x : noting happens

Tape

Reading Head

A Simple Set of Rules

Suppose that the machine is at state Qn

• If it read dn, it moves to state Qn+1

• If it read em and m < n the machine moves to state Qm

• In all other cased, if it read x and dm with m n, or em with m n, then it stays at stateoves to state Qn

“Sequence-detector” Machine

• Assembly cascades on single-stranded DNAs on 39 or 78 bases long (13- and 26-stage machines)

• Infer its internal dynamics from observable output• A sufficient set of observations to overcome the n

oisiness of the output 105-106 fluctuating DNA-RacA complexes• A very smooth ensemble-average signal

An Exponential Sensitivity of the Assembly Cascade

• Both curves decay exponentially close to saturation with a higher slope for the longer sequence that has more cascade stages

• A: fluorescence anisotryopy

• S: the normalized anisotropy

• Open circles: (TAC)13

• Triangles: (TAC)26

Sequence-Discrimination Capability

Very similar triplet repeats Localized differences in nonuniform sequences

the difference between the fluorescence anisotropy signal

Discussion

• By choice of other types of sequence base vectors, the stochastic cascade machinery can encode and decode mixtures in terms of other transforms

• Additional operations to manipulate sequences such as recombination

• The question of whether RecA assembly is used for natural computation requires in vivo testing.