Traffic on DNA


This Java applet (If you don't see anything above, please install
Java 1.4) models interference between two promoters, through RNA polymerase (RNAP) traffic.

Each polymerase occupies 70bp when sitting on the promoter, and 35 bp when elongating. The interference imply that each of the promoters will be less active due to effect from the opposing promoter. This reduced activity is due to a number of effects associated to collisions between moving RNAP's, occlusion of one promoter by the passage of an RNAP from the opposing promoter, as well as ejecting of RNAP on one promoter due to collision with a moving RNAP. Finally the applet also opens for simulation of road block effect, where a sitting duck also eject a elongating RNAP (with probability set by roadblock parameter).

The interference (I-factor) measure activity of promoter without
opposing promoter, divided with activity in current setting.
The approximate I-factor is based on an analytic estimate, which will be correct when opposing promoter is much stronger, or if both promoters are weak. It basically counts reduction of activity of given promoter, assuming that the opposing promoter is not perturbed.

Realistic strength of promoters covers a wide interval. For the very strong promoter PL in lambda phage 1/K=4.5sec correspond to one firing per 4.5 second. Other typical numbers are 1/K=18sec for PR in phage 186, interfering with the lysogen maintenance promoter promoter  PL in 186 which have 1/K=180sec. The interference increases  when as a promoter is exposed to a stronger promoter (i.e. K(opposing)/K(observed) increases ). The aspect ratio quantifies the ratio between the on rate (kon) and the firing rate (kf) of an undisturbed promoter. For aspect ratio>>1, the on rate is high, and the promoter loads easily, but stays occupied for a long time. For aspect ratio alpha<<1, the promoter loads rarely, but fire instantly when it has loaded. PL in 186 have aspect ratio close to 1. pC from phage P2 on the contrary have aspect ratio of about 0.1. One will see that an aspect ratio of about 1 maximizes the interference (try the applet).

The overall system parameter also include distance N between promoters,  that could be varied fro -120 to 3000, with N<40 corresponding to overlapping or partially overlapping promoters. The velocity is the assumed velocity of elongating  RNAP that experimentally (see paper submitted to JMB) is fitted to be between 40 and 50 bp/sec. 

The applet also opens for toying with the system, as clicking below or above the DNA line in middle of screen, artificially  initiate a RNAP of corresponding type on the DNA. Also one may view interference between distant promoters by adjusting the distance button. For strong promoters with similar strength one then observe a system with possibility for very strong fluctuations.

More applets
Transcriptional Interference Model

K.Sneppen, I.B. Dodd, K.E.  Shearwin, A. Palmer,
R.A. Shubert, B. Callen and J.B. Egan.
(A mathematical model for transcriptional interference by RNA polymerase traffic in Escherichia coli)

The model deals with the interference between the two  promoters pA and pS as shown in the applet. The RNAPs are injected onto the DNA through binding and formation of sitting duck complexes at the respective promoters, followed by subsequent formation of elongating complexes. The traffic is simulated by a stochastic model that is adjusted to take into account all known details of the dynamics. The stochastic dynamics are implemented by updating at any time step [t,t+dt]= [t,t+1bp/v] the presence of any RNAP according to the basic processes (v is the velocity). That is, in time step dt, a promoter forms an open complex with probability  kon  dt unless it is occupied or occluded by other RNAPs. In case the promoter is occupied, it initiates elongation with a probability k_f  dt except when an RNAP from the opposing promoter is positioned such that it will collide with the sitting duck in the time step dt. kon and kf together defines the strength of the promoter, that is firing (going from "sitting duck complex" to elongation) with a total rate K which approximately is equal to    kon *kf/(kon+kf).

Any elongating RNAP is moved vdt step in the direction of elongation, except when it collides with an RNAP moving in opposite direction. When such a collision occurs the RNAP's are removed from the system. The transcription activity of any of the promoters is counted by the number of RNAPs which pass the opposing promoter. The interference is found by comparing this number to the number obtained when the opposing promoter is assumed to be silent.

The model, together with carefull analysis and comparison to experiment is submitted to JMB.

References, experimental:

Measurements of up to 30 fold interference between tandem promoters:
S. Adhya and M. Gottesman (1982), Cell 29, 939-944.

Measurements 2-10 fold interference for various
convergent promoter combinations:
B. P. Callen, I. Dodd, K.E. Shearwin and J.B. Egan (2004), Mol. Cell 14, 647-656.

Valid HTML 4.01!