Quantification and Statistical Analysis

The quantitation shown in Figure 1D was performed in R using the tidyverse collection of R packages. The phosphorimage of the deadenylation assay shown in Figure 1C was converted to a data matrix where each entry corresponds to the intensity value of one pixel of the phosphorimage. This matrix was split vertically according to the individual lanes of the gel. For each resulting lane matrix, the average of each row was calculated. This resulted in one vector per lane containing the average values of intensity along the migration direction of the gel of the particular lane (see Figure 1D panel on the left). The peaks of these average intensity distributions for each individual size marker lane were identified. These peak values were used to calibrate migration behavior in the gel with poly(A) tail lengths of the model substrate. Subsequently the maximum of the intensity distributions for each deadenylation time-point was identified. The respective poly(A) tail length was calculated using the calibration curve. The resulting plot is shown in Figure 1D panel on the right.

To determine the half-lives of the intermediates of the deadenylation reaction (Figure S1A and Table S2) decay of each intermediate was quantified in ImageJ. A total of three phosphorimages of deadenylation time courses were quantified (including Figure 1C). The gels were horizontally quantified with the gel quantification tool of ImageJ to follow the decay of the clearly identifiable Model-90A, Model-70A and Model-40A intermediates. This quantification resulted in average intensity distributions for each deadenylation intermediate on each phosphorimage along the time axis of the experiments. The peaks in these intensity distributions correspond to the signals for the respective intermediate at specific time points in the deadenylation reaction. The areas of these peaks for each intermediate and time point were calculated in ImageJ and used for further analysis. These raw data were normalized by the maximal signal in the respective average intensity distribution (to compensate for differences in total signal, exposure etc. between phosphorimages). The mean normalized signal for each time-point and deadenylation intermediate was calculated. An exponential function of the form f(t)=a∗e−St where a is the signal at time point 0, S is the decay rate and t is the time in seconds was subsequently fitted to the data. In the case of the two intermediates the Model-70A and the Model-40A the time point 0 was chosen as the time-point where most of the specific product had accumulated (i.e. where the densitometric signal was the highest). The resulting model allowed the determination of the half-lives and their mean standard deviation intervals for each intermediate (in Figure S1A vertical lines represent the standard deviation interval; the functions fitted to determine the half-lives of the respective model poly(A) RNPs (see Table S2) are represented as lines).

The quantification in Figure 6A was performed in R using the tidyverse collection of R packages. The raw phosphorimage shown in Figure S6C was quantified as described for the quantitation of the in vitro deadenylation assays. To correct for differences in signal strength between individual poly(A) isolations we normalized each individual intensity distributions by the maximum intensity in the respective sample. Two individual lanes of each strain were averaged and the average background signal subtracted. The resulting poly(A) tail distributions was plotted and is shown in Figure 6A.