A Possible Mechanism of Carcinogenesis The Electric Charge Transport Properties of p53

1
A Possible Mechanism of Carcinogenesis - The
Electric Charge Transport Properties of p53
arXivq-bio/0708.3181 0710.1676

• C. T. Shih(???)
• Dept. Phys., Tunghi University
• 2007/10/08
• Dept. Physics, National Tsing-Hua University

• Collaborators
• Rudolf A. Römer, Department of Physics and Center
  for Scientific Computing, University of Warwick,
  United Kingdom
• Stephan Roche, CEA/DSM/DRFMC/SPSMS, Grenoble,
  France

2
Outline

• Mutation of p53 tumor suppressor and cancers
• A possible mechanism of DNA damage/repair
• Scenario how cancerous mutations get rid of the
  repair process
• Model and method
• Results and Discussion

3
Mutation of p53 tumor suppressor and cancers
Guardian of the Genome
4
Mutants of p53 genes
Summary of carcinogens and mutational events that
can alter the p53 genes
5
Functional Significance of p53

• Activate DNA repair proteins when DNA has
  sustained damage
• Hold the cell cycle at the G1/S regulation point
  on DNA damage recognition (if it holds the cell
  here for long enough, the DNA repair proteins
  will have time to fix the damage and the cell
  will be allowed to continue the cell cycle.)
• Initiate apoptosis, the programmed cell death, if
  the DNA damage proves to be irreparable

6
Role of p53 tumor-repressor protein
7
Database IARC (France)http//www-p53.iarc.fr/ind
ex.html
Version R11 (Oct 2006) 23544 records 20366
point mutations
8
DNA A Schematic View
9
(No Transcript)
10
(No Transcript)
11
A possible mechanism of DNA damage/repair

• BER (base excision repair) enzyme with Fe4S42
  cluster robust to oxidation in the absence of
  DNA
• BER binding to DNA oxidation activated
  (Fe4S42 ?Fe4S43) and an electron is
  mediated
• If another BER enzyme is at a distant site, the
  electron will be caught by the BER. The second
  BER will be reduced (Fe4S43 ?Fe4S42) and
  dissociated
• A base lesion can preclude the DNA mediated
  charge transport. BER will remain localized in
  the vicinity of the lesion and diffuse to the
  site to excise the base

E. Yavin et al. (JK Barton group), PNAS 103, 3610
(2006).
12
Electron Paramagnetic Resonance Experiment
13
(No Transcript)
14
Models for Electric Transport of DNA 1D

• 1L 1-leg model
• hopping integral between the i-th and
  (i1)-th nucleotide
• on-site potential of the basepair
• FB (fishbone model) 1L hopping between
  backbone and basepair ( ) on-site potential
  of the backbone ( )
• DNA for 1?i?N and semi-infinite electrodes for
  ilt1 and igtN

• Energy Parameters
• On-site potential 8.24eV,
  8.87eV, 7.75eV and 9.14eV
• 1L The hopping between pairs base is taken to be
  0.4eV
• FB hopping onto the backbone is 0.7eV and the
  backbone onsite energy is taken to be 8.5eV

Using transfer matrix method to calculate
transmission coefficient T(E) for incident energy
E
15
Models for Electric Transport of DNA 2D

• 2L 2-leg model
• hopping integral between the two strands
• LM (ladder model) 2Lhopping between backbone
  and basepair on-site potential of the backbone
• DNA for 1?i?N and semi-infinite electrodes for
  ilt1 and igtN

• Energy Parameters
• 2L The hopping between like base pairs (AT/AT,
  GC/GC, etc.) is chosen as 0.35eV, between unlike
  base pairs it is 0.17eV, 0.1 eV
• LM Intrachain and interchain hopping strengths
  are as in the two-leg model. In addition, the
  backbone is treated as in the fishbone model.

Using transfer matrix method to calculate
transmission coefficient T(E) for incident energy
E
16
Transmission Coefficient Transfer Matrix Method
E Energy of injected carrier T(E) Transmission
coefficent
17
Sequence-Dependent Transport
18
Sequence-Dependent Transport
19
T(E) changed by point mutations
tDNAtm1.0 eV
Black T0(E) (GC)30 Red Tm(E) 30th base C?G

• i beginning site of the subsequence
• w length of the subsequence
• k mutated site
• s mutant base
• (Emin, Emax)(5.75, 9.75) energy range of
  injected carrier
• W4 bandwith in electrodes

20
Comparison of the cancerous/noncancerous mutations
P53 seq. Mutation 14585 C?T (found 133 times in
IARC Database) AGGGGAGCCTCACCACGAGCTGCCCCCAGGGAGCA
CTAAGCGAGGTAAGCAAGCAGGACAAGAAGCGGTGGAGGAGACCAA
Energy-dependence of logarithmic transmission
coefficients of the original
sequence (C solid line) and mutated (A dotted, G
dotted-dashed, T dashed) sequences with length L
20 (from 14575th to 14594th nucleotide) of p53.
The left panel shows results for model 1L, the
right two panels denote the two transport windows
for the fishbone model

• C?T is a cancerous mutation (the 9th highest
  frequency found in various types of cancer)
• C?A and C?G are non-cancerous (not found in the
  human cancer cells up to now)

21
Comparison of the cancerous/noncancerous mutations
P53 seq. Mutation 14585 C?T (found 133 times in
IARC Database) AGGGGAGCCTCACCACGAGCTGCCCCCAGGGAGCA
CTAAGCGAGGTAAGCAAGCAGGACAAGAAGCGGTGGAGGAGACCAA
Energy-dependence of logarithmic squared
differences between the
transmission coefficients of the original
sequence and mutated (C ? T solid line, ? A
dotted, ? G dotted-dashed) sequences.
For the all four models, the cancerous mutation
C?T results in the weakest change in T(E)!
22
CT Change for Different Models and Propagation
Lengths and A Scenario for Carcinogenesis
Cancerous Mutation C?T
Lowest CT Change
23
Scenario how cancerous mutations get rid of the
repair process

• The mutations become cancerous they can get rid
  of the repair processes
• One of the repair processes (proposed by J.K.
  Barton) is to detect the lesions by probing the
  DNA mediated charge transport
• The point mutations can cause the electric
  transport properties change of DNA segments
  containing the mutation points
• The transport change of the cancerous mutations
  must be small, or they will be detected by the
  electric probing process of BER enzymes
• This is a necessary condition, but may not
  sufficient because there are other repair
  mechanism

24
Statistical Analysis for All Possible Mutations
Hotspots, Small CT change

• The average effect of a mutation (k, s) of a
  subsequence with length L on the CT of p53 is
  defined as
• For all 203033 60909 possible mutations,
  calculate the G for various L

Scatter plots of G(k, sw) versus occurrence
frequency of all cancerous mutations (k,s) for
(a) L 20 and (b) 80. The sharp peaks at small
G agree with the scenario that the most cancerous
mutations namely those with high frequency
change the CT only slightly and thus have
smaller G.
25
Comparison of the cancerous/noncancerous mutations

• Define the following 3 set of the mutations
• M all 60909 possible mutations
• Mc 1953 mutations found in the IARC database
  (found in cancer cells at least one time)
• Mc,10 366 mutations found more than 10 times in
  the IARC database (most cancerous mutations)
• For given L, sort the CT results for M according
  to G(k, sL) and determine the rank r(k, sL) of
  the CT change for each mutation (k, s)
• A smaller rank means less CT change for the
  mutation
• ?(k, sL) 100 r(k, sL)/60909 is then the
  relative rank in percentage

26
Statistical Analysis for All Possible Mutations
The Histograms of the Distribution of ?(k, sL)

• Histogram of the distribution of ?(k, sL) in Mc
  (light wide bars) and Mc,10 (dark thin bars)
  which changes the kth nucleotide to s for (a) L
  20 and (b) 80.
• For M, all values are equal to 5 (20 intervals)
  as indicated by the horizontal dashed lines.
• (c) shows the percentage of G(k, sL) values
  inMc,10 for small CT change as a function of DNA
  lengths in the range 05 (black), 510 (dark
  grey), 1015 (light grey) and 1520 (white).
• Similarly, (d) indicates large CT change forMc,10
  in the ranges 8085 (black), 8590 (dark grey),
  9095(light grey) and 95100 (white).
• The horizontal dashed lines in (c) and (d)
  indicates the distributions forM.

Many mutations in Mc,10 have smaller CT change
(G) on the average
The tendency is stronger in L80 case
The number of mutations whose g larger than 80
Is less than average for all L
More than 50 of the mutations in Mc,10 has there
g smaller than 20 for L90
27
Summary

• The conductance of hotspots of cancerous
  mutations is smaller than that of other sites
• On average the cancerous mutations of the gene
  yield smaller changes of the CT
• The tendency is stronger in the set of highly
  cancerous mutations with occurence frequency gt 10
• These results suggest a scenario of how cancerous
  mutations might circumvent the DNA damage-repair
  mechanism and survive to yield carcinogenesis
• The results are robust for a wide range of types
  and parameters of models
• Our analysis is only valid in a statistical sense
  
• Non-cancerous mutations with weak change of CT
  are observed
• Other DNA repair processes should exist and we
  therefore do not intend to claim that the
  DNA-damage repair solely uses a CT-based criterion

28
Thank you for your attention!