Department of Physics and Astronomy

1
Santa Fe 2005
Modeling Single Gene Transcription Biological
Control without Maxwell Demons?
Andrei Ruckenstein

• Department of Physics and Astronomy
• BioMaPS Institute for Quantitative Biology
• Rutgers University, Piscataway NJ 07030

2
Santa Fe 2005
CENTRAL DOGMA
DNA
?
Transcription
RNA
?
Protein
3
Transcription
Santa Fe 2005

• Multi-subunit molecular motor

• Highly conserved across species

Darst et al. Science 289 (2000)
Kornberg et al. Science 292 (2001)

• Promoter binding

• Transcription initiation

• Transcription elongation

• Transcription termination

4
Model of the elongation complex
12-14 bp bubble
8-9 bp hybrid
3-end
exit channel
secondary channel
active center
RNA transcript
5-end of RNA
5
NTP incorporation
NTP (NMP)n (NMP)n1 PPi
i1
6
NTP (NMP)n (NMP)n1 PPi
i
i1
7
NTP (NMP)n (NMP)n1 PPi
i1
8
NTP (NMP)n (NMP)n1 PPi
i1
0
(UNTRANSLOCATED RNA1)
9
(No Transcript)
10
(No Transcript)
11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
Proofreading editing
18
Proofreading editing
(REVERSE TRANSLOCATED-EDITED)
19
Proofreading
20
Outline

• Pausing sequences where RNAP stalls reversibly
  or irreversibly

• Termination with or without memory?

• New model of translocation two (or more)
  coupled ratchets

21
H. Matsuzaki, J. Mol. Biol. (1994), 235, 1173-1192
22
(No Transcript)
23
(No Transcript)
24
TEC Stability Profile

• Average over bubble configurations

• Pauses local minima of G

• Minima are backtracked from position 0!!

(Result of weak DNA-RNA hybrid)
25
Arrest!
26

• Effect of multiple bubbles

Shifts the HIV-LTR pause transcript length 62
to the correct backtracked position ( -1)

• Effect of RNA folding

It can eliminate pauses
27
(No Transcript)
28
(No Transcript)
29
(No Transcript)
30
Pause criterium

• Identify pauses within pause clusters

• Define pause cutoff

• Do not backtrack past equilibrium RNA folds!

Kinetic barriers from co-transcriptional RNA
folding! (D. Luse 1996) Guess separation of
scales between translocation (?sec) RNA unfolding
(msec)

(10 bp GC hairpin unbinding time scale of 100
years! (Herschlag, 1995))
31

• Results
• In 100 experimental pauses
• lt15 false negatives
• ?35 false positives
• Random assignment of pauses on
• the corresponding templates
• ?70 false negatives and
• ?68 false positives
• Best results for multiple bubble
• with folding

32
Intrinsic Termination
Yarnell Roberts Science 1999 Santangelo
Roberts Mol Cell 2004
126 Terminators 96 predictive rate
33
Another mechanism of transcription termination
Gusarov Nudler (1999)
Phage l tR2 terminator
34
(No Transcript)
35
Santa Fe 2005
Gil Bar-Nahum,Vitaly Epshtein, Ruslan Rafiko,
Evgeny Nudler (NYU), and Arkady Mustaev (Public
Health, NJ)
Two mutant E-coli RNAPs
slow
fast

• Slower elongation

• Faster elongation

• Higher fidelity

• Lower fidelity

• More responsive to signals

• Less responsive

• Preferentially back-tracked

• More forward-tracked

• Less bent than wt

• F-bridge preferentially bent

• Incorrect NTPs enhance back translocation w.r.t.
  the
• case of both correct and no NTP
  Proofreading

36
The Double Ratchet Model
37
Santa Fe 2005
Fidelity
F
Misincorporation rate
F
WT
S
38
Santa Fe 2005
Higher Fidelity ?
F
Misincorporation rate

• Arbitrary fidelity can be achieved in thermal
  equilibrium!!

• Editing requires energy!

39
Santa Fe 2005
Time dependence of NTP incorporation
Allosteric binding site for a second NTP in the
active center
Erie et al. Cell 106. 242 (2001)
40
incorporation
30 mM
1mM
100 mM
5m?
500 mM
15m?
41

• Fit to 22 sets of curves
• with five parameters

42
Santa Fe 2005

• The model explains all available biochemical and
  kinetic
• data on elongation by E-coli RNAP (Nahum et
  al., Cell 120,
• 183-193, January 28 2005)

• The interplay between internal and
  translocational ratchets
• play a role in the fidelity, proofreading and
  processivity of
• transcription already before the chemical
  incorporation step!
• No energy required for increasing fidelity
  (multiple check point
• idea) but energy is needed for editing

• Challange
• Model selection optimize fit, minimize
  complexity of the model,
• and minimize variation with additional data
• (maximize predictibility).

43
Santa Fe 2005
MY GRAND CHALLENGES

• CO-TRANSCRIPTIONAL RNA FOLDING
• STRUCTURAL INF/MODELING OF TERMINATION
• PREDICTIONS MODEL SELCTION

44
Santa Fe 2005
IT IS SOMETIMES EASIER TO MAKE A THEORY OF
EVERYTHING THAN A THEORY OF SOMETHING! Aharon
Katchalsky
45
BC 2005
Acknowledgments
Gil Bar-Nahum Vitaly Epshtein Ruslan
Rafikov Evgeny Nudler
(NYU School of Medicine)
(BioMaPS Insitute/Rutgers)
Daibhid OMaoileidigh Vasisht Tadigotla Richard
Ebright Konstantin Severinov Peter von Hippel
(University of Oregon)