Martin Howard

A Polycomb-based switch underlying quantitative epigenetic memory

The conserved Polycomb repressive complex 2 (PRC2) generates trimethylation of histone 3 lysine 27 (H3K27me3), a modification associated with stable epigenetic silencing. Much is known about PRC2-induced silencing but key questions remain concerning its nucleation and stability. Vernalization, the perception and memory of winter in plants, is a classic epigenetic process that, in Arabidopsis, involves PRC2-based silencing of the floral repressor FLC. The slow dynamics of vernalization, taking place over weeks in the cold, generate a level of stable silencing of FLC in the subsequent warm that depends quantitatively on the length of the prior cold. These features make vernalization an ideal experimental system to investigate both the maintenance of epigenetic states and the switching between them. Here, using mathematical modelling, chromatin immunoprecipitation and an FLC:GUS reporter assay, we show that the quantitative nature of vernalization is generated by H3K27me3-mediated FLC silencing in the warm in a subpopulation of cells whose number depends on the length of the prior cold. During the cold, H3K27me3 levels progressively increase at a tightly localized nucleation region within FLC. At the end of the cold, numerical simulations predict that such a nucleation region is capable of switching the bistable epigenetic state of an individual locus, with the probability of overall FLC coverage by silencing H3K27me3 marks depending on the length of cold exposure. Thus, the model predicts a bistable pattern of FLC gene expression in individual cells, a prediction we verify using the FLC:GUS reporter system. Our proposed switching mechanism, involving the local nucleation of an opposing histone modification, is likely to be widely relevant in epigenetic reprogramming.

Pushing and pulling in prokaryotic DNA segregation

In prokaryotes, DNA can be segregated by three different types of cytoskeletal filaments. The best-understood type of partitioning (par) locus encodes an actin homolog called ParM, which forms dynamically unstable filaments that push plasmids apart in a process reminiscent of mitosis. However, the most common type of par locus, which is present on many plasmids and most bacterial chromosomes, encodes a P loop ATPase (ParA) that distributes plasmids equidistant from one another on the bacterial nucleoid. A third type of par locus encodes a tubulin homolog (TubZ) that forms cytoskeletal filaments that move rapidly with treadmill dynamics.

Dynamic Compartmentalization of Bacteria: Accurate Division in E. Coli

Positioning of the midcell division plane within the bacterium E. coli is controlled by the min system of proteins: MinC, MinD, and MinE. These proteins coherently oscillate from end to end of the bacterium. We present a reaction-diffusion model describing the diffusion of min proteins along the bacterium and their transfer between the cytoplasmic membrane and cytoplasm. Our model spontaneously generates protein oscillations in good agreement with experiments. We explore the oscillation stability, frequency, and wavelength as a function of protein concentration and bacterial length.

Vernalization - a cold-induced epigenetic switch

Summary Growth and development are modulated by environmental signals in many organisms. These signals are often perceived at one stage and ‘remembered’ until later in development. An increasingly well-understood example of this process in plants is provided by vernalization, which refers to the acquisition of the ability to flower after prolonged exposure to cold. In Arabidopsis thaliana, vernalization involves downregulation and epigenetic silencing of the gene encoding the floral repressor FLOWERING LOCUS C (FLC). This epigenetic silencing is quantitative and increases with the duration of exposure to cold. Vernalization involves a Polycomb-based switching mechanism, with localized nucleation of silencing during periods of cold, and spreading of the silencing complex over the whole gene after the exposure to cold. A number of characteristics of vernalization have recently been elaborated on through the use of mathematical modelling. This has revealed the importance of chromatin dynamics for the switching mechanism and has shown that the quantitative nature of the process is due to cell-autonomous switching of an increasing proportion of cells. The principles derived from vernalization are likely to be widely relevant to epigenetic reprogramming in many organisms.

Movement and equipositioning of plasmids by ParA filament disassembly.

Bacterial plasmids encode partitioning (par) loci that confer stable plasmid inheritance. We showed previously that, in the presence of ParB and parC encoded by the par2 locus of plasmid pB171, ParA formed cytoskeletal-like structures that dynamically relocated over the nucleoid. Simultaneously, the par2 locus distributed plasmids regularly over the nucleoid. We show here that the dynamic ParA patterns are not simple oscillations. Rather, ParA nucleates and polymerizes in between plasmids. When a ParA assembly reaches a plasmid, the assembly reaction reverses into disassembly. Strikingly, plasmids consistently migrate behind disassembling ParA cytoskeletal structures, suggesting that ParA filaments pull plasmids by depolymerization. The perpetual cycles of ParA assembly and disassembly result in continuous relocation of plasmids, which, on time averaging, results in equidistribution of the plasmids. Mathematical modeling of ParA and plasmid dynamics support these interpretations. Mutational analysis supports a molecular mechanism in which the ParB/parC complex controls ParA filament depolymerization.

The cell-end factor pom1p inhibits mid1p in specification of the cell division plane in fission yeast.

Intrinsic spatial cues ensure the proper placement of the cell division plane. In the fission yeast Schizosaccharomyces pombe, the position of the nucleus helps to direct the medial positioning of contractile-ring assembly and subsequent cell division . An important factor in this process is mid1p (anillin-like protein), which is a peripheral-membrane protein that forms a broad cortical band of dots overlying the nucleus in interphase and recruits myosin in early mitosis . How mid1p localizes to this cortical band and tracks the nucleus is not clear, especially because its localization is independent of the cytoskeleton . Here, we used a combination of experimental and computational approaches to test mid1p localization mechanisms. We provide evidence that pom1p, a DYRK-family protein kinase that forms a concentration gradient emanating from the nongrowing cell end, inhibits mid1p. In pom1 mutants, mid1p is distributed over half of the cell, covering the nongrowing cell end. This abnormal distribution is established in a dynamic manner in interphase and leads to the formation of misplaced or multiple contractile rings. Our computational and experimental results support a model in which both positive cues from the medial nucleus and negative cues from the cell tips specify the position of the division plane.

Pattern formation inside bacteria: Fluctuations due to the low copy number of proteins

We examine fluctuation effects due to the low copy number of proteins involved in pattern-forming dynamics within a bacterium. We focus on a stochastic model of the oscillating MinCDE protein system regulating accurate cell division in E. coli. We find that, for some parameter regions, the protein concentrations are low enough that fluctuations are essential for the generation of patterns. We also examine the role of fluctuations in constraining protein concentration levels.

Fundamental Limits to Position Determination by Concentration Gradients

Position determination in biological systems is often achieved through protein concentration gradients. Measuring the local concentration of such a protein with a spatially varying distribution allows the measurement of position within the system. For these systems to work effectively, position determination must be robust to noise. Here, we calculate fundamental limits to the precision of position determination by concentration gradients due to unavoidable biochemical noise perturbing the gradients. We focus on gradient proteins with first-order reaction kinetics. Systems of this type have been experimentally characterised in both developmental and cell biology settings. For a single gradient we show that, through time-averaging, great precision potentially can be achieved even with very low protein copy numbers. As a second example, we investigate the ability of a system with oppositely directed gradients to find its centre. With this mechanism, positional precision close to the centre improves more slowly with increasing averaging time, and so longer averaging times or higher copy numbers are required for high precision. For both single and double gradients, we demonstrate the existence of optimal length scales for the gradients for which precision is maximized, as well as analyze how precision depends on the size of the concentration-measuring apparatus. These results provide fundamental constraints on the positional precision supplied by concentration gradients in various contexts, including both in developmental biology and also within a single cell.

Cellular organization by self-organization mechanisms and models for Min protein dynamics

We use the oscillating Min proteins of Escherichia coli as a prototype system to illustrate the current state and potential of modeling protein dynamics in space and time. We demonstrate how a theoretical approach has led to striking new insights into the mechanisms of self-organization in bacterial cells and indicate how these ideas may be applicable to more complex structure formation in eukaryotic cells.

Antagonistic Roles for H3K36me3 and H3K27me3 in the Cold-Induced Epigenetic Switch at Arabidopsis FLC

Summary Posttranslational modifications of histone tails are an important factor regulating chromatin structure and gene expression. Epigenetic memory systems have been predicted to involve mutually exclusive histone modifications that, through positive feedback mechanisms, generate bistable states [1, 2]. How the states are interconverted is not understood, and whether the histone modifications are sufficient for epigenetic memory is still greatly debated [3]. We have exploited the process of vernalization, the slow quantitative epigenetic silencing of Arabidopsis FLC induced by cold, to detail with fine temporal and spatial resolution the dynamics of histone modifications during an epigenetic switch. The profiles of H3K36me3, H3K4me3, and H3K4me2 at FLC throughout the vernalization process were compared to H3K27me3, which accumulates at a local nucleation region during the cold and spreads across the locus on return to the warm [2]. We find for many phases of the vernalization process that H3K36me3 and H3K27me3 show opposing profiles in the FLC nucleation region and gene body, that H3K36me3 and H3K27me3 rarely coexist on the same histone tail, and that this antagonism is functionally important. A lack of H3K36me3 results in a fully silenced state at FLC even in the absence of cold. We therefore propose that H3K36me3 is the opposing modification to H3K27me3 in the Polycomb-mediated silencing of FLC. However, the lack of an absolute mirror profile predicted from modeling suggests that their antagonistic roles contribute a necessary, but not sufficient, component of the mechanism enabling switching between and inheritance of epigenetic states.