Synchronous cultures from the baby machine. A model for animal cells.

A baby-machine system that produces newborn Escherichia coli cells from cultures immobilised on a membrane was developed many years ago in an attempt to attain optimal synchrony with minimal disturbance of steady-state growth, and a model designed to characterise the nature and quality of the synchrony of such cells in a quantitative manner has been published. The baby machine has now been adapted for animal cells, and the present article is an attempt to modify the model to include these cells as well. The model consists of five elements, giving rise to five adjustable parameters (and a proportionality constant): a major, essentially synchronous group of cells with ages distributed normally about zero; a minor, random component from a steady-state population on the membrane that had undergone only very little age selection during the elution process; a fixed background count, to allow for the signals recorded by the electronic particle counter produced by debris and electronic noise; a time-shift, to account for differences between time of cell division and end of sample collection; and the coefficient of variation of the interdivision-time distribution, taken to be reciprocal-normal. It is this last feature, a reciprocal-normal rather than a Pearson type III interdivision-time distribution, that distinguishes this version of the model from its predecessor. The model is fitted by unconstrained non-linear least-squares to data from three different leukemia cell lines. The standard errors of the parameters are quite small in all cases, making their estimates highly significant; the quality of the fit is striking. The five parameters of the model can be divided into two nuisance parameters, two that are associated with the methodology and one that describes an inherent property of the cell itself; it turns out that both methodology parameters are zero in all three data sets studied. We also discuss the partition of the transition-time dispersion between the age distribution of the newborn cells and the age distribution of dividing cells and show that a reliable estimate of the corresponding parameters requires an experiment that extends over at least two and a half doubling times.

P1 and NR1 plasmid replication during the cell cycle of Escherichia coli.

Replication patterns of the miniP1 plasmid pZC176, the miniNR1 plasmid pRR933, and the high-copy miniNR1 derivative pRR942 were examined during the Escherichia coli cell division cycle and compared to the cycle-specific replication pattern of a minichromosome and the cycle nonspecific pattern of pBR322. In E. coli cells growing with doubling times of 40 and 60 min, the miniP1 plasmid was found to replicate with a slight periodicity during the division cycle. The periodicity was not nearly as pronounced as that of the minichromosome, was not affected by the presence of a minichromosome, and was not evident in cells growing more rapidly with a doubling time of 25 min. Both miniNR1 plasmids, pRR933 and pRR942, replicated with patterns indistinguishable from that of pBR322 and clearly different from that of the minichromosome. It is concluded that both P1 and NR1 plasmids can replicate at all stages of the cell cycle but that P1 displays a slight periodicity in replication probability in the cycle of slower growing cells. This periodicity does not appear to be coupled to a specific age in the cycle, but could be associated with the achievement of a specific cell mass per plasmid. During temperature shifts of a dnaC(Ts) mutant, the miniP1 plasmid and pBR322 replicated with similar patterns that differed from that of the minichromosome, but were consistent with a brief eclipse between rounds of replication.

Cell-cycle research with synchronous cultures: an evaluation.

The baby-machine system, which produces new-born Escherichia coli cells from cultures immobilized on a membrane, was developed many years ago in an attempt to attain optimal synchrony with minimal disturbance of steady-state growth. In the present article, we put forward a model to describe the behaviour of cells produced by this method, and provide quantitative evaluation of the parameters involved, at each of four different growth rates. Considering the high level of selection achievable with this technique and the natural dispersion in interdivision times, we believe that the output of the baby machine is probably close to optimal in terms of both quality and persistence of synchrony. We show that considerable information on events in the cell cycle can be obtained from populations with age distributions very much broader than those achieved with the baby machine and differing only modestly from steady state. The data presented here, together with the long and fruitful history of findings employing the baby-machine technique, suggest that minimisation of stress on cells is the single most important factor for successful cell-cycle analysis.

Synchronous cultures from the baby machine: anatomy of a model.

The baby-machine system, which produces newborn Escherichia coli cells from cultures immobilized on a membrane, was developed many years ago in an attempt to attain optimal synchrony with minimal disturbance of steady-state growth. In the present article, we describe in some detail a model designed to analyse such cells with a view to characterizing the nature and quality of the synchrony in a quantitative manner; it can also serve to evaluate the methodology itself, its potential and its limitations. The model consists of five elements, giving rise to five adjustable parameters (and a proportionality constant): a major, essentially synchronous group of cells with ages distributed normally about zero; a minor, random component from a steady-state population on the membrane that had undergone only very little age selection during the elution process; a fixed background count, to account for the signals recorded by the electronic particle counter produced by debris and electronic noise; a time-shift, to allow for differences between collection time and sampling time; and the coefficient of variation of the interdivision-time distribution, taken to be a Pearson type III. The model is fitted by nonlinear least-squares to data from cells grown in glucose minimal medium. The standard errors of the parameters are quite small, making their estimates all highly significant; the quality of the fit is striking. We also provide a simple yet rigorous procedure for correcting cell counts obtained in an electronic particle counter for the effect of coincidence. An example using real data produces an excellent fit.

Transcription level of operon ftsYEX and activity of promoter P1 of rpoH during the cell cycle in Escherichia coli.

In Escherichia coli, the genes of the main heat-shock proteins are under the control of the product of gene rpoH, protein sigma 32. The distal promoter P1 of rpoH is located within the terminator of the division operon ftsYEX which could imply some coupling between ftsYEX transcription termination, P1 transcription and cell division. To study the possibility of this coupling, the level of transcription of ftsYEX and the activity of promoter P1 of rpoH were examined in synchronous cultures. Results indicate that transcription of ftsYEX and of rpoH from P1 is continuous, suggesting that ftsYEX transcription termination and P1 activity are not coupled to the cell cycle.

Gravity and the orientation of cell division.

A novel culture system for mammalian cells was used to investigate division orientations in populations of Chinese hamster ovary cells and the influence of gravity on the positioning of division axes. The cells were tethered to adhesive sites, smaller in diameter than a newborn cell, distributed over a nonadhesive substrate positioned vertically. The cells grew and divided while attached to the sites, and the angles and directions of elongation during anaphase, projected in the vertical plane, were found to be random with respect to gravity. However, consecutive divisions of individual cells were generally along the same axis or at 90 degrees to the previous division, with equal probability. Thus, successive divisions were restricted to orthogonal planes, but the choice of plane appeared to be random, unlike the ordered sequence of cleavage orientations seen during early embryo development.

Replication and segregation of a miniF plasmid during the division cycle of Escherichia coli.

Replication of the miniF plasmid pML31 was examined during the division cycle of Escherichia coli growing with doubling times between 40 and 90 min at 37 degrees C and compared to the replication of plasmid pBR322 and the minichromosome pAL70. The replication pattern of pML31 was indistinguishable from that of pBR322 at all growth rates and very different from the cell-cycle-specific replication of the minichromosome. It is concluded that both pML31 and pBR322 plasmids can replicate at all stages of the division cycle, with a probability of replication that increases gradually, but perhaps not exponentially, during the cycle. In contrast, the modes of segregation of pML31 and pBR322 plasmids into daughter cells at division appeared to differ, raising the possibility that pML31 may segregate in a nonrandom fashion similar to that of chromosomes and minichromosomes.

Gene transcription and chromosome replication in Escherichia coli.

Transcript levels of several Escherichia coli genes involved in chromosome replication and cell division were measured in dnaC2(Ts) mutants synchronized for chromosome replication by temperature shifts. Levels of transcripts from four of the genes, dam, nrdA, mukB, and seqA, were reduced at a certain stage during chromosome replication. The magnitudes of the decreases were similar to those reported previously ftsQ and ftsZ (P. Zhou and C. E. Helmstetter, J. Bacteriol. 176:6100-6106, 1994) but considerably less than those seen with dnaA, gidA, and mioC (P. W. Theisen, J. E. Grimwade, A. C. Leonard, J. A. Bogan, and C. E. Helmstetter, Mol. Microbiol. 10:575-584, 1993). The decreases in transcripts appeared to correlate with the estimated time at which the genes replicated. This same conclusion was reached in studies with synchronous cultures obtained with the baby machine in those instances in which periodicities in transcript levels were clearly evident. The transcriptional levels for two genes, minE and tus, did not fluctuate significantly, whereas the transcripts for one gene, iciA, appeared to increase transiently. The results support the idea that cell cycle timing in E. coli is not governed by timed bursts of gene expression, since the overall findings summarized in this report are generally consistent with cell cycle-dependent transient inhibitions of transcription rather than stimulations.

DNA sequestration and transcription in the oriC region of Escherichia coli.

In Escherichia coli, gidA and dnaA transcription are shut off transiently after initiation of chromosome replication, while mioC transcription is shut off before initiation. The possible involvements of seqA and dam in these transcriptional periodicities were evaluated by examining transcription of the genes in seqA and dam mutants of E. coli PC2 dnaC2(ts) aligned for initiation of chromosome replication. In both seqA- and dam- cells, gidA and dnaA continued to be transcribed after initiation, whereas the inhibition of mioC transcription before initiation was unaltered. After initiation, transcripts from mioC that traversed oriC reappeared more slowly in seqA+ dam+ cells than in seqA- or dam- cells, but before the release of oriC from sequestration. Apparently, initiation of transcription at a promoter can be completely prevented by sequestration, but established transcription forks can traverse sequestered DNA. These findings, plus analyses of transcript levels in steady-state cultures, support the idea that initiation capacity in seqA mutants is elevated because of the combined influences of increased durations of expression of both gidA and dnaA during the division cycle and defective sequestration at oriC. Accordingly, a proposal for the sequence of events during the interinitiation interval in E. coli is presented based on the evident coupling of transcription to replication.

mioC transcription, initiation of replication, and the eclipse in Escherichia coli.

The potential role of mioC transcription as a negative regulator of initiation of chromosome replication in Escherichia coli was evaluated. When initiation was aligned by a shift of dnaC2(Ts) mutants to nonpermissive temperature (40 degrees C), mioC transcript levels measured at the 5' end or reading through oriC disappeared within one mass doubling. Upon return to permissive temperature (30 degrees C), the transcripts reappeared coordinately about 15 min after the first synchronized initiation and then declined sharply again 10 min later, just before the second initiation. Although these observations were consistent with the idea that mioC transcription might have to be terminated prior to initiation, it was found that the interval between initiations at permissive temperature, i.e., the eclipse period, was not influenced by the time required to shut down mioC transcription, since the eclipse was the same for chromosomes and minichromosomes which lacked mioC transcription. This finding did not, in itself, rule out the possibility that mioC transcription must be terminated prior to initiation of replication, since it might normally be shut off before initiation, and never be limiting, even during the eclipse. Therefore, experiments were performed to determine whether the continued presence of mioC transcription during the process of initiation altered the timing of initiation. It was found that minichromosomes possessing a deletion in the DnaA box upstream of the promoter transcribed mioC continuously and replicated with the same timing as those that either shut down expression prior to initiation or lacked expression entirely. It was further shown that mioC transcription was present throughout the induction of initiation by addition of chloramphenicol to a dnaA5(Ts) mutant growing at a semipermissive temperature. Thus, transcription through oriC emanating from the mioC gene promoter is normally inhibited prior to initiation of replication by the binding of DnaA protein, but replication can initiate with the proper timing even when transcription is not shut down; i.e., mioC does not serve as a negative regulator of initiation. It is proposed, however, that the reappearance and subsequent disappearance of mioC transcription during a 10-min interval at the end of the eclipse serves as an index of the minimum time required for the establishment of active protein-DNA complexes at the DnaA boxes in the fully methylated origin region of the chromosome. On this basis, the eclipse constitutes the time for methylation of the newly formed DNA strands (15 to 20 min at 30 degrees C) followed by the time for DnaA protein to bind and activate oriC for replication (10 min).

Preferential release of one daughter cell during division of immobilized chinese hamster ovary cells.

Chinese hamster ovary (CHO) cells were attached to tiny adhesive sites in poly-2-hydroxyethyl methacrylate(polyHEMA-) coated glass, and their divison properties were examined. The adhesive sites were produced by placing a metal mask, containing 8-mum-diameter holes arranged in a regular pattern, on top of the coated glass and exposing the sandwich to glow discharge treatment. This treatment produced an ordered array of circular cavities in the polyHEMA down to the glass. These adhesive sites were smaller in diameter than a newborn CHO cell, so that, upon division, there would theoretically be room for only one of the two new daughter cells to remain attached. It was found that individual CHO cells attached to, and grew upon, the sites, and that division normally resulted in the releas of one of the two new daughters. It is concluded that this culture technique has applications in research on the mammalian cell cycle, cell partitioning, and cellular senescence. (c) 1995 John Wiley & Sons, Inc.

Characterization of cell-cycle-specific events in synchronous cultures of Escherichia coli: a theoretical evaluation.

Synchronous growth studies are often used to assess the presence, timing and duration of periodic phenomena in the bacterial cell cycle. In an effort to evaluate the quality and quantity of information on cycle-specific events that can reasonably be expected from such inquiries, a model was constructed of a synchronous culture of Escherichia coli cells as would be derived from a growing population immobilized on a surface, and applied to the case of one stable and one unstable cellular component. The results indicated that, while the presence of cycle-specific events may be easily detectable, their timing and duration are very difficult to establish in synchronous growth experiments. Furthermore, differences in timing can be misconstrued as differences in duration, and vice versa, when interpretations are based on the qualitative analysis of the data.

Relationship between ftsZ gene expression and chromosome replication in Escherichia coli.

Transcriptional levels within the ftsQAZ region of the Escherichia coli chromosome were correlated with chromosome replication and the division cycle. The transcripts were measured either in synchronous cultures generated by the baby machine technique or in dnaC2(Ts) mutants that had been aligned for initiation of chromosome replication by temperature shifts. Transcription within the ftsZ reading frame was found to fluctuate during the cell cycle, with maximal levels about midcycle and a minimum level at division, in cells growing with a doubling time of 24 min at 37 degrees C. Examination of transcription in dnaC(Ts) mutants aligned for chromosome replication indicated that the periodicity was due to a reduction in transcripts coincident with replication of the ftsQAZ region. Transcription originating upstream of the ftsA gene exhibited the periodicity and accounted for a significant proportion of the transcripts entering ftsZ. The most obvious interpretation of the data is that replication of the region transiently inhibits transcription, but alternative explanations have not been ruled out. However, no other relationship between transcription and either replication or division was detected.

Dimensional rearrangement of Escherichia coli B/r cells during a nutritional shift-down.

In a search for the mechanism underlying dimensional changes in bacteria, the glucose analogue methyl alpha-D-glucoside was used to effect a rapid reduction in the mass growth rate of Escherichia coli by competitively inhibiting glucose uptake, a so-called nutritional shift-down. The new steady-state cell mass and volume were reached after 1 h, during which the rate of cell division was maintained; rearrangement of the linear dimensions (cell length, diameter), however, required an additional 2 h and caused an undershoot in cell length, consistent with the view that E. coli is slow to modify its diameter. The results are compared with the overshoot in cell length that occurs following nutritional shift-up.

Correlation of gene transcription with the time of initiation of chromosome replication in Escherichia coli.

Transcriptional levels of the Escherichia coli mioC and gidA genes, which flank the chromosomal origin of replication (oriC) and the dnaA gene, were correlated with the time of initiation of chromosome replication. The transcripts were measured either in dnaC2(ts) mutants that had been aligned for initiation of chromosome replication by a temperature shift or in synchronous cultures of cells obtained using the baby machine technique. In both types of experiments, mioC transcription was inhibited prior to initiation of chromosome replication and resumed several minutes after initiation. Conversely, gidA and dnaA transcription were both inhibited after initiation of replication, coincident with the period of hemimethylation of oriC DNA. It is proposed that mioC transcription prevents initiation of chromosome replication, and must terminate before replication can begin. It is further proposed that the eclipse period between rounds of replication, i.e. the minimum interval between successive initiations, encompasses the time required to methylate GATC sequences in newly replicated oriC plus the time required to terminate mioC transcription. Conversely, the active transcription of gidA and dnaA prior to initiation is consistent with their positive effects on initiation, and their shutdown after initiation could serve to limit premature reinitiation.

Rate maintenance of cell division in Escherichia coli B/r: analysis of a simple nutritional shift-down.

A competitive (nonmetabolizable) inhibitor of glucose uptake, alpha-methylglucoside, was used to limit the growth of Escherichia coli. Cell division during such a nutritional shift-down was studied in batch cultures and with the "baby-machine" technique. Following a brief delay, the rate of division was maintained for 60 to 70 min in batch cultures and for an extended period in the baby machine. Decreases in cell size were due, in part, to a possible reduction in the mass per chromosome origin at the time of replication initiation and a shorter time interval between initiation and the subsequent division. These unusual findings suggest that this method for abrupt change in growth rate without modifying repression patterns is useful for studying the control of various aspects of the bacterial cell.

Relationships between chromosome segregation, cell shape and temperature in Escherichia coli.

The partitioning of chromosomes into daughter cells during the division of Escherichia coli is non-random. As a result, the chromosome containing the older template DNA strand has a higher probability of segregating toward the old cell pole than toward the new cell pole. The numerical value of this probability is a function of the incubation temperature. It is shown here that a recent model for explaining the physiological basis for non-random chromosome segregation also explains the temperature dependence of the segregation process.

Improved bacterial baby machine: application to Escherichia coli K-12.

Exponentially growing derivatives of Escherichia coli K-12 were immobilized onto the surfaces of nitrocellulose membrane filters which had been coated with poly-D-lysine. The cells attached firmly to the surfaces, and when flushed with culture medium, the immobilized cells continued to divide and newborn cells were released into the effluent. Cell cycle parameters were examined with the technique, and it was found that K-12 derivatives possessed differing values for interdivision times, C, D, and average cell sizes when grown in the same culture media. It was also found that the cells released from immobilized populations of one culture consisted of two predominant size classes: newborn cells of unit size with single nucleoids and newborn cells of double this unit size. The results demonstrated that K-12 derivatives can be used in the baby machine culture technique to examine all aspects of the cell cycle of this organism. Furthermore, the yield of newborn cells was about fivefold greater than that obtained previously with cultures of strain B/r immobilized onto uncoated membranes.

Description of a baby machine for Saccharomyces cerevisiae.

A method for the continuous withdrawal of newly formed daughter cells from a growing population of Saccharomyces cerevisiae is described. An exponential-phase culture of cells was immobilized onto a surface and then flushed continuously with culture medium. Upon division of a cell in the immobilized population, the mother cell remained attached to the surface and the daughter cell was released. The method can be applied to research on the cell cycle, the segregation of components between cells, and cellular aging.

Involvement of cell shape in the replication and segregation of chromosomes in Escherichia coli.

Chromosome replication appears to initiate in E. coli when the dnaA boxes in oriC become filled with DnaA protein, which could simultaneously mediate both the unwinding of the origin for the start of polymerization and the attachment of oriC to the cell envelope (Bramhill and Kornberg, 1988; Løbner-Olesen et al., 1989; Pierucci et al., 1989). The attachment takes place somewhere within the cell half in which the oriC resides. The boundaries of this attachment/replication zone, which cannot include the polar cap, could be demarcated by the polar and centrally located periseptal annuli (Rothfield, this Forum). Since attachment and polymerization are two aspects of the same process, the attachment probably takes place via the polymerizing strand. Once polymerization begins, the oriC with the older template strand moves away from the younger one, by mechanisms unknown, to eventually take up residence in the equivalent domain of the complementary sister cell. Thus, the template strand that stays within its domain corresponds to the strand that was attached during the previous round of replication, and the template that moves away is the one that was not attached. The driving force for this translocation is not specified by our model, but a number of plausible alternatives have been proposed by others (reviewed in Leonard and Helmstetter, 1990). Throughout the ensuing replication and cell division, the chromosomes are located (or can move freely) within the attachment/replication zone of the developing daughter cell (lateral cylinder and septum). At some time during the course of this process, but before the next initiation event, the replication origins must be released from the attachment sites so that the entire process can be repeated. Thus, the probabilistic non-random chromosome segregation is due to the asymmetry of the attachment/replication zone in the cell, whereas the partitioning system itself must possess a mechanism to discriminate between template strands of different ages. This apparent mechanistic relationship between chromosome replication, chromosome partitioning and the maintenance of cell shape may provide an interesting framework for future experiments.