“…TGGE of transcript PSTVd-Sty 1 under equilibrium conditions+ The transcript in 0+2ϫ TBE was denatured at 70 8C for 15 min and slowly renatured to room temperature with ;0+1 8C/min+ PAGE: 4% PA 30:1, 0+2ϫ TBE, silver-stained+ The marker lanes contain a crude RNA extract with 7: 7 S RNA; P: PSTVd; l: linear PSTVd; c: circular PSTVd+ Circular and linear PSTVd comigrate at low temperature because their shape is nearly identical+ The completely denatured molecules migrate very differently because of the high retardation of the covalently closed circle+ possible through a variety of parameters (temperature, anion concentration, viscosity, specific inhibitors) that, however, also influence RNA structure formation+ Therefore, we have chosen to vary the concentration of NTP, which is known to regulate the transcription rate in a wide range (Masukata & Tomizawa, 1990;Chamberlin & Ring, 1973) without altering the RNA structure+ All other conditions (see Materials & Methods) were kept constant+ For all experiments, the polymerase is used in a 24ϫ molar excess over the template+ Lowering the transcription rate does not lead to an increase in non-full-length transcripts (see Fig+ 3)+ Thus all different bands visible in TGGE will consist of structures with the same sequence and length, differing only in shape+ Of course, prematurely terminated transcripts are also generated, but their size heterogeneity will prohibit their detection in gels+ Furthermore, the absolute yields of transcript are lowered with decreasing concentrations of NTP, but the total amount of NTPs is not limiting for the yields; after 15 min of transcription the NTP concentration is lowered by less than 5% with all NTP concentrations used+ For determination of transcription rates, the amount of transcripts was determined at 15 min (see Fig+ 4); at that time point the rates were influenced neither by a lag phase at short times nor by saturation effects at long times (data not shown)+ At NTP concentrations below 50 mM a sigmoidal dependence of yields is obvious; this is a well known effect (Chamberlin & Ring, 1973;Kadesch & Chamberlin, 1982)+ We have no explanation for the step-like dependence near 300 mM NTP+ From the concentration dependence of the transcript yields, however, values for the transcription rates can be derived+ The value of 230 6 20 nt/s at c NTP ϭ 400 mM and 37 8C is reduced to ;50 nt/s at 20 8C (Chamberlin & Ring, 1973)+ Using that value as a basis of calculation, the rates determined here are roughly between 3 nt/s at c NTP ϭ 30 mM and 20 nt/s at c NTP ϭ 70 mM; that is, we are able to shift the transcription rate of the T7-polymerase into the proper range for polymerase II+ According to these rates of transcription, synthesis of a full-length PSTVd transcript needs between a few seconds and several minutes+ Because folding of RNA structures (for review see Turner et al+, 1990;Riesner & Römer, 1973) requires between microseconds (elongation of helices) and minutes (closure of complex junctions or tertiary foldings), the process of transcription and the process of folding might interfere with each other+…”