by Christoph
In his presentation at a meeting in Copenhagen last year, Johan Elf showed several kymograms, which were published afterwards (see frontispiece). No, these are not fancy versions of pie charts that everyone knows, but intuitively accessible visualizations of the localization of macromolecular complexes in living bacterial cells as they change over time.
(click to enlarge)
Figure 1. Spatial organization of the replication machinery in different growth media. Fork plots for strains carrying YPet-DnaN grown in different media. The medium is specified above each fork plot, and the corresponding doubling time is found above each plot. Dashed white lines indicate the average sizes of cells at birth or division. Dashed red lines indicate the average initiation size as determined by tracking replisomes in single cells. Solid white lines indicate average cell pole positions. RDM: rich defined medium. Source. Frontispiece: Johan Elf presenting at 'Major ideas in quantitative microbial physiology', June 2013, Copenhagen DK. Photo by the author
Kymographs (from Greek κῦμα, swell or wave + γραφή, writing) are used in science to measure and record atmospheric pressure, tuning fork vibrations, the functioning of steam engines, animal habits, and the movement of molecules in cells, says Wikipedia. For recording "the movement of molecules in cells", however, you don't need a dedicated machine, a kymograph, but software for image analysis. In a nutshell: microscopic images in a time series of growing cells are each scanned for the distribution of fluorescence, for example of fluorescently labeled proteins. These time-resolved scans are then compiled into a kymogram.
The kymograms in Figure 1 show the positions of the replication forks in E. coli cells growing in different media. Here, the fluorescence-labeled protein in the replication forks is DnaN, the β-clamp subunit of DNA Pol III holoenzyme and an integral part of the replisome. It is known that E. coli DnaN is among the proteins that tolerate an N-terminal, covalent "extension" by a whole yellow fluorescent protein (YPet) without functional impairment. In the false color kymograms, red indicates high fluorescence, which decreases via yellow and green to light blue, while dark blue indicates no fluorescence.
You can see at a glance that "on average, cells grown in acetate initiated replication at one origin; cells grown in lactose –/+ amino acids (AA), succinate+AA, and mannose+AA concurrently initiated replication at two origins; and cells grown in gluconate+AA, glucose+AA, and rich defined medium (RDM) initiated replication at four origins" as the authors state. Note that you "see" the replication forks – two forks, one for each replichore in bidirectional replication – from the initiation of chromosome replication at oriC to its termination, but not when the cells have finished replication but not yet undergone cell division (see the 'acetate' kymogram!).
This was already known from other experimental approaches and served the authors mainly as a basis for investigating the extent to which these cell cycle parameters are altered in mutants. What was not previously known with such detail is that the spatial organization of the replication forks is apparently not dependent on growth conditions or cell division.
Figure 2. See text for details. Source
When Knöppel et al. (2023) succinctly state that "on avarage, cells grown in ace...", this stands for the evaluation of hundreds of images of individual cells for each time point. Without going into the refined and professionally documented statistical analysis here, I would like to give you a brief look at their "raw data" and the first steps of the further analysis in Figure 2:
Legend to Figure 2. Experimental setup and data visualization. (A) Examples of cells tracked over three generations. Detected YPet-DnaN foci are marked with circles, where circles sharing the same color indicate replication started from the same origin, i.e., two replication forks moving bidirectionally on the chromosome, and are often spatially colocalized. The foci were tracked using the u-track algorithm. Initiations (marked with yellow arrows) are defined as the start of u-track tracks, and terminations (marked with blue arrows) are the end of u-track tracks. Cell outlines are displayed with a gray line. (B) Three-generation fork plot. A two-dimensional distribution of the position of YPet-DnaN foci along the long axis of the cells (y axis) and cell area, from here on referred to as cell size (x axis), is plotted. Here, the cell size in generations two and three is defined as if the cells in the previous generation had not divided. We refer to this type of structure as a "supercell." Dashed white lines indicate the average sizes of cells at birth or division. Dashed red lines indicate the average initiation size as determined by tracking replisomes in single cells. Solid white lines indicate average cell pole positions.
Figure 3. The "mother machine" and high-throughput observation of the mother cells. (A) Schematic illustration of the microfluidic mother machine. The old-pole mother cell is trapped at the end of the growth channel. (B) The outer-most branch of the lineage tree represents the old pole mother cell and her progeny. (C) Snapshot of a typical field of view. (D) (top panel) Temporal montage of a single growth channel [within the dotted yellow box in (C)] from the beginning to the end of the experiment. The stable band on the bottom of the montage is the old-pole mother cell, whereas the "feather" of the montage show the growth and escape of her progeny; (mid panel) the average YFP intensity of the mother cell fast fluctuating around the mean without decay with time; (bottom panel) cell length vs. time of the old-pole mother cell, which shows occasional spikes (filamentations). In all three panels, the x-axis represents time in minutes. Source
You may wonder why the photographic images of the E. coli cells in Figure 2A appear so neatly lined up and remain so over the entire observation period. In fact, no manual manipulation was necessary for this (and no software "tricks"). Knöppel et al. (2023) grew their cells in separate channels – lots of channels! – of a microfluidic device equipped with microscope and camera, a "mother machine". Figure 3A schematically shows the experimental setup, and Figure 3CD the evaluation of an exemplary experiment running for ~60 hours. Since E. coli cells grow to different lengths and thicknesses (diameters) under different growth conditions, channels of different dimensions must be used, of course, so that cells line up neatly for photography (Note the rare "hickups" when a cell accidentally skips division).
The "mother machine" that was developed in the lab of Suckjoon Jun, who has previously contributed to STC. The somewhat strange name "mother machine" reflects that it was developed as an alternative to the venerable "baby machine". In a "baby machine", growing E. coli cells were bound to a nitrocellulose filter and unbound new daughter cells were eluted by medium flow. This left the mother cells in complete anonymity, but resulted in an almost synchronous culture of "babies" in the eluate, which were used by Cooper & Helmstetter (1968) in their classic work for the determination of cell cycle parameters as, for example, two time periods: C, the time for a round of chromosome replication, and D, the time between the end of a round of replication and cell division. By contrast, since mother cells and all their subsequent generations of daughter cells grow in the same channel, they do not need to be synchronized at the start of the experiment to assess C- and D-periods.
Do you want to comment on this post? We would be happy about it! Please comment on Mastodon, Bluesky, or on 𝕏 (formerly Twitter).
Comments