Epigenetic and conventional regulation is distributed among activators of FLO11 allowing tuning of population-level heterogeneity in its expression

Abstract

Epigenetic switches encode their state information either locally, often via covalent modification of DNA or histones, or globally, usually in the level of a trans-regulatory factor. Here we examine how the regulation of cis-encoded epigenetic switches controls the extent of heterogeneity in gene expression, which is ultimately tied to phenotypic diversity in a population. We show that two copies of the FLO11 locus in Saccharomyces cerevisiae switch between a silenced and competent promoter state in a random and independent fashion, implying that the molecular event leading to the transition occurs locally at the promoter, in cis. We further quantify the effect of trans regulators both on the slow epigenetic transitions between a silenced and competent promoter state and on the fast promoter transitions associated with conventional regulation of FLO11. We find different classes of regulators affect epigenetic, conventional, or both forms of regulation. Distributing kinetic control of epigenetic silencing and conventional gene activation offers cells flexibility in shaping the distribution of gene expression and phenotype within a population.

Figure 1. Signals from many trans factors converge at the complex FLO11 promoter.

Regulators of FLO11 transcription. Nucleosomal positions are based on a thermodynamic model for nucleosomal occupancy . Binding sites are approximate and based on literature but most sites have not been confirmed directly. The three locations where a tetO sequence was inserted are also shown. See main text, Text S1, and Figure S1 for further details.

Figure 2. Mixed expression states and independent switching at the FLO11 locus.

(A) Dual reporter assay. Each FLO11 allele turns ON and OFF slowly with identical rates λ and γ because the two reporters are equivalent. (B) Mixed expression states. A dual reporter strain grown in rich media (no glucose) supplemented with 1% ethanol and 2% glycerol (false color overlay CFP = red, YFP = Green). All four possible expression states are seen. (C) Transition rates. Equivalence of reporters implies λ ₁ = λ ₂, λ ₃ = λ ₄, γ ₁ = γ ₂, γ ₃ = γ ₄. Independent switching implies λ ₁ = λ ₃ and γ ₁ = γ ₃. (D) OFF to ON transition rates of different expression states: λ ₁ (X), λ ₂ (X), λ ₃ (•), λ ₄ (•). Each marker represents the fraction of cells observed to switch at the particular time, and the pink curve is the same as the fit curve in Figure 3A. (E) As in (D) but for ON to OFF transition rates: γ ₁ (X), γ ₂ (X), γ ₃ (•), γ ₄ (•). The blue curve is the same as the fit curve in Figure 3A. (D) and (E) demonstrate that transition rates at one allele are independent of the state of the other allele. Even the null hypothesis that γ ₂ and γ ₄ are equivalent cannot be rejected at the 5% significance level (two-way T-test, p = 0.28) nor can the null hypothesis that their distributions are identical (two-way KS test, p = 0.47).

Figure 3. A method for inferring kinetics of switching from static steady state distribution.

(A) (Left) Time evolution of the population distribution of YFP expression from a dual-reporter strain growing in YP 1% ethanol, 2% glycerol within a microfluidic chamber over 20 hours. Colorbar indicates fraction of cells (n = 230 over time course). This strain had been growing in identical conditions in liquid culture prior to transfer to the microfluidic chamber. The distribution changes early on because of the small initial sample size (n = 10). (Right) Marginal transition rates between ON and OFF states. Blue/pink dots indicate fraction of cells ON/OFF at birth and observed to switch OFF/ON. Corresponding curves are fits of the model for exponentially distributed switching times from ON to OFF and OFF to ON, with adjusted rates shown next to the plot. Error bars correspond to 3 s.d. from the mean calculated by a bootstrap analysis. Similar results are obtained when focusing on CFP expression (see Figure S5 and Figure S6). (B) Three-state model of FLO11 activation showing separation of timescales between epigenetic (silencing) and conventional regulation. When slow transitions associated with silencing are present, the fast transitions of transcriptional bursting can be lumped into a single rate μ. The model then collapses into the two-state model in (C). (C) (Left) Static distribution of YFP fluorescence of dual-reporter strain grown in identical media conditions as A but in deep well plates rather than the microfluidic device. Transition rates inferred from this snapshot using a stochastic kinetic model (right) agree closely with those obtained by timelapse microscopy. (D) Modulation of switching rates. The stochastic kinetic model's prediction of the fraction of cells in the mixed expression state corresponds to independent switching (given by 2p(1−p), corresponding to the gray line) for a range of conditions. Error bars (x-axis) are from 95% confidence intervals from MLE fit of switching rate to estimate true fraction of ON cells; error bars (y-axis) are due to errors in the estimation of threshold fluorescence value for autofluorescence (see Text S1).

Figure 4. Three different kinetic roles for regulators of FLO11.

(A) The qualitative shape of the Beta distribution for various values of OFF→ON (λ/δ) and ON→OFF (γ/δ) transition rates (normalized with respect to the growth rate δ). When both rates are slower than growth (lower left quadrant) they characterize slow epigenetic transitions between the silenced and competent states. The expression distribution is bimodal, representing stable ON and OFF populations. These rates can be inferred by measuring the expression distribution by fluorescence microscopy and fitting to the Beta distribution. For unimodal ON distributions, epigenetic silencing no longer occurs. If only fast intrinsic fluctuations between the competent and active promoter state were present, the same two-state model would apply, but now predict the fast transition rates and unimodal distributions (upper half of plot). However, because extrinsic fluctuations also matter, direct fitting of measured unimodal distributions does not yield the fast transition rates (see main text and Text S1 for details). (B) Representative fluorescence histograms of the three activator classes. (Top) Tec1p titrated in wildtype background in SD ura-; Flo8p titrated in flo8Δ background in SD ura- +1% ethanol; Msn1p titrated in wildtype background in SD ura-. (Bottom) rtTA titrated in strain with tetO at −350 (nucleosome occluded site), at −1160 (site occludes Sfl1p binding site), and at −1470 (site directly upstream of the −1200 nucleosome free region). Histograms are derived from fluorescence microscopy (cell number >300). The fluorescence distribution of an OFF strain (Y92) used to measure autofluorescence is shown at the top of each plot. (C) Kinetic roles of regulators. Increasing levels of various activators of FLO11 decrease γ, stabilizing the active state without significantly changing λ. Class I activators cannot decrease γ significantly (blue). Class II activators can shift the transition rates into the lower left quadrant which corresponds to partially silenced, bimodal expression (pink). Flo8p has a less stable silenced state compared to the class II activators. It appears that at a critical value of γ the regulators abolish silencing, and the response enters the upper left quadrant. (D) Synthetic activator titration. Titrations of synthetic activators mimic the three classes of activators, depending on the location of the binding site. All titrations (B, C, and D) were in SD ura- except Sfl1p and Flo8p where 1% ethanol was added. Error bars represent 95% confidence intervals (to fits of experimental data to the Beta distribution for a single value of μ—details in Text S1).

Figure 5. Hda1p is necessary for silencing and a heterogeneous response.

(A) Sfl1p titrated in sfl1Δ background in SD ura- +1% ethanol leads to a heterogeneous response. (B) Sfl1p titrated in sfl1Δ background in SD ura- leads to a graded response. (C) Sfl1p titrated in hda1Δ background in SD ura- +0.5% ethanol reverts to a graded response. Expression is lower with no doxycycline because of endogenous Sfl1p expression. (D) ChIP assay probing for acetylated H3 and H4 histones at FLO11 promoter (strains grown in SD complete or SD leu-). Probes amplified the −1.7 to −1.5 kb region of the promoter. Signal (y-axis) represents anti-acetylated histone/anti-histone ratio, or an effective average acetylation per histone in the region. Deletion of both sfl1 and hda1 result in hyperacetylation of the FLO11 promoter which is associated with the abrogation of silencing. Therefore SFL1-dependent silencing at FLO11 requires HDA1. Error bars are standard error of triplicate quantitative PCR samples. (E) When the activator Tec1p is titrated in an hda1Δ background, the response is also graded (SD ura-). (F) Mean levels of expression during Sfl1p and Tec1p titrations in wildtype (blue curve) and hda1Δ (pink curve) backgrounds. Elimination of silencing because of lack of Hda1p lowers the threshold level at which activators function. Furthermore, the population response is graded (C and E, see Figure S8 for other activator titrations). Error bars represent 3 s.d. around mean calculated from bootstrap analysis. Inset: The square of intrinsic noise of Sfl1p (left) and Tec1p (right) titrated in hda1Δ is proportional to the reciprocal of protein abundance (here shown as the mean fluorescence level). This indicates that without silencing, regulators modulate expression by controlling burst frequency (λ').

Figure 6. Ethanol modulates silencing at the promoter via Flo8p.

(A) Wildtype grown in SD complete with ethanol added to final concentration ranging between 0 and 3%. (B) Activators titrated in wildtype background in SD ura- +1% ethanol. All responses are graded, suggesting loss of silencing at the promoter. (C) Synthetic activator (rtTA) titration in SD ura- +1% ethanol. As in (B), responses are also graded. (D) Activators titrated in flo8Δ in SD ura- +1% ethanol. The response is closer to that in Figure 4A rather than Figure 6B, indicating that ethanol abolishes silencing at the promoter through Flo8p. (E) Representative fluorescence histograms of titrations shown in (A, B, C, D). (Top) Wildtype titrated as in panel A, Tec1p titrated as in (B), and as in (D). (Bottom) rtTA titrated in strain with tetO site at −1160 (occluding Sfl1p binding site), and at −350 (nucleosome occluded site) as in panel C; Msn1p titrated as in (D).

Figure 7. Functional roles of regulators of FLO11 promoter activation shape the population response.

Silencing at the promoter is established by binding of Sfl1p and recruitment of Hda1p. Relative activities of Flo8p and Sfl1p determine the chromatin state of the promoter. The underlying promoter state of an OFF population can be revealed by addition of Class I and II activators (bottom), as Class I activators cannot effectively activate transcription at a silenced promoter, whereas Class II activators can activate expression by sufficiently stabilizing the competent state. At very high levels, Class II activators disrupt silencing; at this point, all cells are also expressing highly. In contrast, an intermediate level of Flo8p activity can “open” the promoter converting the silenced state to a stable competent state, while expression remains low. This opening might be related to chromatin modifications. The combination of Flo8p activation and Class I activators allows the decoupling of chromatin state and expression level, whereas activation by Class II activators alone would not.