INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Myxococcus xanthus fruiting body development is a highly coordinated process. To initiate the developmental program, the cells must sense nutrient limitation at high cell density (20, 28). If both conditions are met, about 100,000 rod-shaped vegetative cells aggregate to form a fruiting body, within which individual cells differentiate into spherical and refractile myxospores. Myxospores remain dormant and environmentally resistant until conditions permit germination (6, 7).

Nutrient limitation is sensed by M. xanthus cells, at least in part, by a rise in intracellular guanosine penta- and tetraphosphate ([p]ppGpp) (16, 43). Cell density appears to be sensed through an extracellular signal, termed A signal (28). The A signal has been identified as a mixture of specific amino acids that appear to be generated by M. xanthus autoproteolysis of surface proteins within the first hours after the onset of starvation (27). When the collective concentration of A signal in a community of starving cells reaches above 10 µM, cells initiate development by aggregating (28).

The expression of certain developmental genes requires independent input from both starvation and A signal (3, 24). The best-characterized gene of this class is designated 4521, in reference to the 4521 Tn5lac insertion. In wild-type strains, 4521 expression begins to increase at 1.5 to 2 h after the onset of development and increases greater than 10-fold by 6 to 8 h (19, 26). The expression of 4521 is cell density dependent (28). Its expression is low in starved cells at densities less than 3 × 10⁸ cells per ml and rises dramatically at densities above this level. The addition of exogenous A signal to low-density starved cells can rescue 4521 expression to near-maximal levels (28). When asg mutants, which are deficient in A-signal generation (19, 26), are starved at high density, 4521 is not expressed. Three unlinked asg genes (asgA, asgB, and asgC) have been identified previously (14, 25, 30). Their DNA sequences indicate that they encode regulatory proteins required for A-signal production (4, 35, 36). The expression of 4521 can be restored to asg mutants by the addition of exogenous A signal (27) or by the presence of asg suppressor mutations, designated sas (for suppressor of asg) (19).

The 4521 promoter has been determined by DNA sequence analysis and mutagenesis to be a member of the sigma-54-dependent family of promoters (23). Related studies have shown that a regulatory region of at most 146 bp upstream of the transcription start site is required for wild-type 4521 expression during growth and development (11). Transcription initiation of the 4521 promoter, similar to that of other sigma-54-dependent promoters, is expected to require an NtrC-like activator to bind upstream of the transcription start site (42).

To study the mechanism by which starvation and A-signal sensing are integrated, six mutants that expressed 4521 at a high level during growth and development were isolated in an asg background after UV mutagenesis (19). One of these mutations, sasB7, did not cluster with the others (sasB5, -14, -15, -16, and -17) (19). The sasB7 mutation is a gain-of-function mutation that maps approximately 4 kb away from the others, within the sasS gene, which encodes a histidine protein kinase component (50) of a putative two-component signal transduction system (49).

We report here the cloning, sequencing, and analysis of the wild-type sasN gene (previously referred to as sasB5 [50]) and the five clustered mutations (sasB5, -14, -15, -16, and -17) that map to it. We have determined that the wild-type sasN gene product is necessary for normal M. xanthus fruiting body formation and sporulation and that SasN functions as a critical regulator that prevents 4521 gene expression during growth.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Plasmids, bacteria, and growth conditions. The plasmids and bacterial strains used in this study are listed in Table 1. Strains such as DK6623 contain three genetic elements. The 4521 Tn5lac (Tc^r) insertion allows the expression of the 4521 reporter gene to be monitored (24, 26). The asgB480 mutation reduces A-signal production and thus blocks 4521 expression. The presence of sasB5, an asgB480 suppressor mutation, restores 4521 expression (19). M. xanthus strains were grown with vigorous shaking at 32°C in CTT liquid medium (1% Casitone [Difco], 10 mM Tris-HCl [pH 7.6], 1 mM K₂HPO₄-KH₂PO₄ [pH 7.7], 8 mM MgSO₄; pH 7.6) or on CTT agar (CTT liquid with 1.5% Bacto Agar). Kanamycin (40 µg/ml), tetracycline (12.5 µg/ml), or trimethoprim (250 µg/ml) was added when appropriate. The growing cells were used in the mid-exponential phase (80 to 160 Klett units, or 4 × 10⁸ to 8 × 10⁸ cells/ml). Escherichia coli was grown in Luria-Bertani (LB) broth (37) or on LB agar (LB broth plus 1.5% agar) containing ampicillin (50 µg/ml), kanamycin (40 µg/ml), or trimethoprim (100 µg/ml) when necessary.

General molecular biological methods. Cloning and manipulations of plasmid DNA were performed according to standard protocols (37). Plasmid DNA for sequencing was prepared by using a QIAprep spin kit (Qiagen). DNA sequencing was performed at the DNA core facility of the Department of Microbiology and Molecular Genetics, University of Texas-Houston Medical School, with an ABI 373A DNA sequencer (Perkin-Elmer, Applied Biosystems Division).

Methods for genetic and developmental analysis of M. xanthus. The following methods for M. xanthus manipulation used in this study have been previously described elsewhere: electroporation of plasmid DNA into M. xanthus (21), M. xanthus chromosomal DNA preparation (1), M. xanthus development on TPM starvation agar (24) and in submerged culture (25), viable myxospore assay (25), and -galactosidase assays (24). The alterations to the chromosome for each M. xanthus strain constructed in this study were confirmed by Southern blot analysis (37).

In situ cloning of the wild-type sasB locus. A 22-kb fragment of the wild-type sasB locus was cloned from strain DK6662 by the in situ cloning strategy (8). Strain DK6662 contains the 6658 Tn5-132 (Tc^r) insertion linked to the wild-type sasB locus (see Fig. 1A) (19). Briefly, the Kan^r plasmid pHBK429 (12), which contains the IS50 region of Tn5, was electroporated into DK6662 cells. The plasmid integrated into the chromosome by single-crossover homologous recombination with either the IS50R or the IS50L of the 6658 Tn5-132 (Tc^r) insertion to give Kan^r Tc^r recombinants. One Kan^r Tc^r recombinant, HK1406, was determined by Southern analysis to contain pHBK429 oriented in its chromosome such that digestion with EcoRI and recirculation would yield a plasmid containing a 22-kb region of the sasB locus. Such a plasmid, pYC1007, was prepared and used to transform E. coli DH5 cells. The physical distance between the 6658 Tn5-132 insertion and the five sasN mutations was estimated by cotransduction frequency to be about 8 kb (9, 19). Thus, it was possible that the 22-kb region might contain the wild-type alleles of these five mutations.

Genetic complementation. The Kan^r plasmids pYC1007, containing the complete 22-kb region linked to 6658 Tn5-132, and pYC1008, pYC1009, and pYC1010, containing three subclones, were tested for their ability to restore a SasB⁺ phenotype to the sasB5 mutant, DK6623, when integrated by a single-crossover recombination event into the sasB locus. Briefly, these plasmids were electroporated into DK6623 cells, plated onto CTT plates containing kanamycin, and overlaid with 0.04% 5-bromo-4-chloro-3-indolyl--D-galactoside (X-Gal), a chromogenic substrate of -galactosidase. The SasB⁺ Lac Kan^r recombinant colonies, which expressed 4521 Tn5lac at a low wild-type level during growth, were tan and easy to differentiate from the parent-like SasB Lac⁺ Kan^r recombinant colonies, which expressed 4521 Tn5lac at a high level during growth and were blue.

DNA sequence determination and computer analysis. A 4.2-kb MluI fragment, which complemented the sasB5, -14, -15, -16, and -17 mutations, was sequenced after being subcloned into pGEM3 to create plasmid pDX45-2. Nested deletion plasmids were generated with exonuclease III after the digestion of pDX45-2 with KpnI and XbaI by using the Erase-A-Base kit (Promega). Twenty plasmids of the appropriate lengths were sequenced with the T7 primer. To acquire the sequence of the second strand, a set of subclones was constructed by using pDX45-2 and its deletion derivatives and sequenced. The DNA sequence was edited and assembled by using the Genetics Computer Group's sequence software package, version 9.1 (5). The Codonpreference program predicted the open reading frames (ORFs) based on the G-C codon bias of the third position (2) in this high-G+C-content (67.5 mol%) organism (32). Motif and hydrophobicity analyses were performed by the Internet services PrositeScan and TMpred at the Swiss Institute for Experimental Cancer Research (http://ulrec3.unil.ch/software).

Construction of sasN-null mutants. The sasN insertion-deletion null mutation introduced into the wild-type strains DK101 and DK1622 was constructed and used to study the involvement of sasN in M. xanthus fruiting body development on starvation agar and in submerged culture, respectively. In each strain, the Tn5 kanamycin resistance gene, aph, replaced a 713-bp SacI-SmaI internal fragment of the sasN gene. The pDX63.4 plasmid generated for this strain construction contains aph flanked on one side by the 1.8-kb region upstream of the sasN SacI site and on the other side by the 0.8-kb region downstream of the sasN SmaI site. The strains were constructed by double-crossover recombination of pDX63.4 into the wild-type sasB locus, resulting in recombinants HK1210 in the DK101 background and HK1211 in the DK1622 background. A trimethoprim resistance gene located in the vector region of pDX63.4 is lost in the double-crossover recombination event; thus, the sasN-null mutants were identified among the Kan^r recombinants by their Tp^s phenotypes.

Construction of the sasN-lacZ fusion. To generate a strain containing a sasN-lacZ transcriptional fusion, plasmid pDX58 was constructed and electroporated into DK101, creating HK1217. Plasmid pDX58 contains M. xanthus DNA from 128 bp upstream of the first predicted sasN translation start site through 780 bp downstream of the site, followed by the promoterless lacZ gene. When introduced by a single-crossover recombination event into the M. xanthus chromosome, the lacZ gene is located at the SasN amino acid residue 260 and the native sasN promoter controls its expression. Vector sequences separate lacZ from the other sasN gene, which is controlled by what is predicted to be a complete promoter region, beginning at the end of the upstream sasR gene.

Cloning and identification of the mutant alleles. The sasB5, -14, -15, -16, and -17 mutant alleles were cloned from their corresponding mutant strains DK6623, DK6631, DK6632, DK6633, and DK6634, respectively, by a modification of the in situ cloning method (8), following integration of plasmid pDX66 into each mutant sasB locus. Briefly, plasmid pDX66, containing the kanamycin resistance gene, aph, and a 1-kb DNA fragment located 720 bp downstream of the sasN termination codon, was electroporated into the five strains. When this plasmid integrated by a single-crossover recombination event into the sasB locus, it formed a tandem duplication of the M. xanthus DNA present on pDX66 and introduced a selectable marker and an origin of replication adjacent to the mutated sasB alleles. The chromosomal DNA from one representative Kan^r Tc^r recombinant derived from each of the five sasB mutants was isolated, digested with MluI, ligated with T4 DNA ligase, and used to transform E. coli DH5 cells. The resulting plasmids were designated pDX102, pDX98, pDX99, pDX100, and pDX101, respectively. Each plasmid contains the pDX66 vector and a different intact sasN mutant allele. Subsequently, the internal 1.5-kb AgeI fragments containing the sasN mutant alleles were subcloned into the SmaI site of pBluescript SK(+) (Stratagene). Both strands of the AgeI fragments were sequenced, and the mutations were identified.

Nucleotide sequence accession number. The nucleotide sequences of sasN and the surrounding region including orf2 and orf3 have been assigned GenBank accession no. AF076221.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Cloning the wild-type region that complements the sasB5, -14, -15, -16, and -17 mutations. Five clustered point mutations (sasB5, -14, -15, -16, and -17) bypass the starvation and high-cell-density requirements for expression of the 4521 developmental reporter gene. As a result, strains containing these mutations express the 4521 gene at high levels during growth and development (19). A region of the wild-type M. xanthus chromosome adjacent to a Tn5 marker linked to these mutations (Fig. 1A) was cloned by the in situ cloning method developed by R. E. Gill et al. (8). This 22-kb fragment of the wild-type sasB region was able to restore the SasB⁺ phenotype to DK6623, the sasB5 mutant, when present as a tandem duplication in the sasB locus (Fig. 1B). Ninety percent of the selected recombinants were SasB⁺ (tan colonies when overlaid with X-Gal), and 10% of the selected recombinants were SasB (blue colonies when overlaid with X-Gal, identical to the DK6623 parent). An internal 4.2-kb MluI subclone of the 22-kb fragment in pYC1009 was also shown to restore the SasB⁺ phenotype to DK6623, at a similar percentage, when integrated at the sasB locus (Fig. 1B). Restoration of the SasB⁺ phenotype at about 90% suggests that these fragments contain a complete wild-type transcription unit. In tandem duplications, restoration of the wild-type phenotype at less than 100% is typical and is most likely a result of apparent gene conversion (45). Gene conversion in M. xanthus is documented to commonly occur at 10 to 25% in these types of crosses (41, 45).

View larger version (23K): [in this window] [in a new window]   FIG. 1.   Genetic analysis of the sasB locus. (A) Physical map of the M. xanthus sasB locus in strain DK6662 containing the linked Tn5 6658. The black line represents the M. xanthus DNA, and the white box represents Tn5 DNA. (B) Restoration of the SasB+ phenotype to the sasB5 mutant DK6623 by subclones of the wild-type sasB locus integrated into the chromosomal sasB locus. The physical map of the sasB DNA cloned in pYC1007 is shown. The bars beneath the physical map indicate the fragments present in the subclones. The percentage of SasB+ recombinants is calculated as the number of SasB+ recombinants divided by the total number of Kanr recombinants tested for the SasB phenotype, multiplied by 100. The number of recombinants tested in each experiment is more than 1,000. (C) Restoration of the SasB+ phenotype to the sasB5 mutant DK6623 by subclones of the wild-type sasB locus integrated into the chromosomal Mx8 phage attachment site. The physical map of the sasB DNA cloned in pYC1012 is shown. Various restriction enzyme recognition sites used in generating subclones are shown. The exonuclease III digestions were generated starting at the orf3 end of this region. The putative ORFs are represented by open boxes with horizontal arrows that indicate the predicted direction of transcription. The numbers of recombinants tested are the same, and the percentages of SasB+ recombinants are calculated, as described for panel B. The asterisks indicate that the plasmids were introduced into the sasB mutant strains, DK6631, DK6632, DK6633, and DK6634, and that the same results were observed.

Sequence analysis of a 4.2-kb region of the sasB locus. The complete DNA sequence of both strands of the 4,205-bp MluI fragment was determined. Three ORFs, designated ORF1, ORF2, and ORF3, were predicted based on the high G+C contents of the third codon positions in orf1, orf2, and orf3 (89.3, 90.6, and 92.3%, respectively), which is indicative of an ORF in high-G+C-content organisms (2) such as M. xanthus (67.5 mol% G+C) (32). The typical codon usage pattern previously observed for M. xanthus (39) was also observed for these genes.

View larger version (54K): [in this window] [in a new window]   FIG. 2.   DNA sequence and deduced amino acid sequence of sasN and its surrounding region. The DNA sequence and the deduced amino acid sequence of sasN are shown. The first predicted SasN start codon is at nucleotide 129. The predicted stop codon at nucleotide 1218 is marked by an asterisk. The hydrophobic region (amino acids 15 to 36) and the leucine zipper motif (amino acids 56 to 77) are underlined and labeled. The mutated nucleotides are shown above their wild-type alleles. The allele designations and the substituted amino acids are in parentheses. The positions of N1 (663 bp) and N2 (1,145 bp) TnT41 insertion mutations (48) are marked by vertical arrowheads.

The sasN gene encodes a negative regulator of 4521 expression. To determine the effect of a sasN-null mutation on 4521 gene expression in an otherwise wild-type background containing Tn5lac 4521, an internal-disruption sasN mutation was constructed and introduced into strain DK6620. The presence of the sasN-null mutation in this strain increases 4521 expression dramatically and dwarfs the parent strain's roughly 10-fold increase in activity during growth and development (Fig. 3). In growing cells (time zero of development), the -galactosidase specific activity of the sasN-null mutant is about 800 U, whereas the activity in the parent strain is about 20 U. This difference in -galactosidase specific activity between growing cells of the sasN-null mutant and the parent strain indicates that SasN functions as a negative regulator of 4521 expression during growth. Furthermore, it indicates that all of the elements necessary for 4521 expression are present in wild-type cells during growth and points to SasN as the critical regulator that prevents 4521 expression. After 10 h of development, the -galactosidase specific activity of the sasN-null mutant reaches about 2,000 U, whereas the activity in the parent strain is about 200 U. These data suggest that SasN may also function as a negative regulator of 4521 expression during development.

View larger version (21K): [in this window] [in a new window]   FIG. 3.   The effect of a sasN-null mutation on the expression of 4521. -Galactosidase specific activities of Tn5lac 4521 in the sasN+ strain DK6620 () and an isogenic sasN null mutant strain, HK1212 (), are shown. The specific -galactosidase activity was determined during growth and at different times during development on starvation agar. Each strain was tested in at least three independent experiments, and the averages of two recombinants are shown. ONP, o-nitrophenol.

sasN is necessary for normal fruiting body formation and sporulation. To determine if a wild-type sasN gene was required for normal fruiting body development, a sasN-null mutation was introduced in two wild-type backgrounds and each strain was tested for the ability to form fruiting bodies and to sporulate. In contrast to the condensed and darkened fruiting bodies formed by wild-type DK101 cells after 2 days, the isogenic sasN-null mutant HK1210 formed loose asymmetric aggregates that did not compact or darken after prolonged incubation (Fig. 4A and B). In submerged culture, wild-type DK1622 cells formed tight and refractile fruiting bodies, whereas the sasN-null mutant HK1211 cells formed many loose, small, and flat aggregates (Fig. 4C and D). In addition, the sporulation efficiency of the sasN-null mutant HK1211 was about 10% that of wild type. A level of 7.5 × 10⁵ to 1.3 × 10⁶ spores/ml was determined for HK1211, compared to 1.0 × 10⁷ spores/ml for DK1622 after 7 days of development. The sasN-null mutants generated either by internal disruption or by transposon insertion showed the same abnormalities and deficiencies in development (data not shown). These data indicate that sasN is required for normal M. xanthus fruiting body formation and sporulation.

View larger version (152K): [in this window] [in a new window]   FIG. 4.   Developmental morphologies of sasN-null mutants. (A and B) Fruiting body development on TPM starvation agar is shown. Concentrated cells at a density of 4 × 109 cells per ml in TPM buffer were inoculated onto TPM agar at 20 µl per spot and incubated at 32°C for 2 days. (A) DK101 wild-type cells; (B) HK1210 sasN internal-disruption null mutant. (C and D) Fruiting body development in submerged culture is shown. Cells (25 µl at a density of 4 × 109 cells per ml) in MC7 buffer were inoculated into 1.7-cm-diameter wells containing 375 µl of MC7 starvation buffer and incubated at 32°C for 7 days. (C) DK1622 wild-type cells; (D) HK1211 sasN internal-disruption null mutant. Bar, 0.1 mm.

sasN is expressed during growth. The high 4521 expression level in the sasN mutant suggests that SasN is a critical negative regulator that prevents 4521 expression during growth. To determine whether the sasN gene was expressed during growth, a strain containing a sasN-lacZ transcriptional fusion was generated. Strain HK1217 contains a tandem duplication of a sasN-lacZ transcriptional fusion and the wild-type sasN gene in an otherwise wild-type DK101 background. During growth, the -galactosidase specific activity was 120 U. This activity can be compared to the basal level of 3 to 20 U observed during growth for most developmental genes. These data indicate that sasN is expressed at a moderately high level during growth and support the idea that the SasN functions to prevent developmental gene expression during growth. The sasN gene is also expressed during development, suggesting a potential role of SasN in development (49).

Expression of 4521 in a sasN sasS mutant is similar to that in the sasS parent strain. SasS is a sensor histidine kinase of a putative two-component signal transduction system (49, 50) and functions as a positive regulator that is absolutely required for 4521 expression. sasS-null mutants express 4521 at a very low basal level (2 U) during growth and development (Fig. 5) (50). SasS has been proposed to sense the accumulation of A signal during early development (50).

View larger version (20K): [in this window] [in a new window]   FIG. 5.   Expression of 4521 in a sasN sasS mutant during growth and early development. -Galactosidase specific activity profiles for strains HK1218 (sasN sasS), HK1503 (sasN+ sasS), and DK6623 (sasN sasS+) are shown. The specific -galactosidase activity was determined during growth and development on starvation agar for each strain in at least three independent experiments, and results of a representative experiment are shown. ONP, o-nitrophenol.

Cloning and identification of the sasB5, -14, -15, -16, and -17 mutant alleles. The five clustered sasB mutant alleles, sasB5, -14, -15, -16, and -17, were cloned, sequenced, and identified. The mutation sasB16 is a C-to-T transition at nucleotide 267, which changes Q47 to an amber termination codon. Mutations sasB5, -14, -15, and -17 are identical C-to-A transversions at nucleotide 966 which change T280 to P. Mutation sasB15 also carries a G-to-A transition at nucleotide 729 which changes D201 to N. It is likely that sasB5, -14, -15, and -17 are siblings, because all of the mutations were isolated in the same screen (19).

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

The M. xanthus sasN gene was identified and determined in this report to encode a negative regulator of the developmental reporter 4521 and to be required for wild-type fruiting body formation and sporulation. In the absence of SasN, as a result of either the original point mutations (19) or the null mutations constructed in this study, 4521 expression is extremely high during both growth and development (Fig. 3). These data indicate that all of the elements necessary for 4521 expression are present during growth and that SasN serves a critical role as a regulator that prevents 4521 expression during growth.

The mechanism of SasN function has yet to be determined. The two most likely options for SasN negative control of the sigma-54-dependent promoter 4521 are that SasN is a direct negative regulator, a repressor of the 4521 promoter, or that it is an indirect negative regulator, an inhibitor of an activator of 4521 expression. Biochemical evidence is necessary to definitively determine how SasN functions. The sasN DNA sequence and its predicted amino acid sequence have provided only limited insight into the mechanism of SasN action. However, the examination of 4521 expression in strains containing mutations in either or both of the genes encoding SasN or the 4521 activator, the SasS sensor kinase, has been more enlightening. The expression of 4521 during growth and development in the sasS mutant, in which the activated cognate response regulator of SasS should not accumulate, remains at a very low basal level (Fig. 5). The expression of 4521 in a mutant containing both sasS and sasN mutations shows the same low basal level (Fig. 5). Data such as these are typical of sigma-54-dependent promoters and have been used to argue that these promoters have no need for repressors. Specifically, in the absence of the activator the sigma-54-RNA polymerase holoenzyme can form a stable closed complex with the promoter but is unable to undergo isomerization to form an open complex that is transcriptionally competent (42). To our knowledge, all previously identified negative regulators of sigma-54-dependent promoters function indirectly by controlling the activity of activators. Thus, if SasN controls 4521 expression as a repressor, it would be the first identified repressor controlling a sigma-54 promoter. In addition, this low basal expression level in the absence of both the positive and the negative regulators suggests that there are no other independent regulators controlling 4521 expression.

The very high 4521 expression level during growth in the sasN mutant implies that there is a high level of the activated positive regulator. If SasN is a repressor, one would predict that the level of the activated positive regulator is normally high during growth and that SasN directly blocks activation of 4521 transcription. This block would have to be relieved soon after starvation to permit 4521 developmental expression. If SasN is an inhibitor of an activator, one would predict that the sasN mutation would cause an increase in the normally low level of the activated positive regulator. The potential targets of SasN inhibition are both elements of the two-component SasS-SasR system, which are the known positive regulators of 4521 expression (49, 50). It is possible that SasN inhibits SasS autohistidine kinase activity or stimulates SasS phosphatase activity. If SasR is the SasN target, SasR~P would be stabilized as a result of the sasN mutation. A SasN-SasS interaction that stimulates dephosphorylation of SasR~P would be similar to the nitrogen assimilation pathway in the enteric bacteria that controls transcription of many promoters (33), including the best-studied sigma-54-dependent promoter, glnA (22, 34).

SasN is an intriguing regulator that controls developmental gene expression during the transition from growth to development. This activity is reminiscent of the Bacillus subtilis transition state regulators AbrB, Hpr, and Sin (46). However, SasN is also required for M. xanthus fruiting body formation and sporulation, a characteristic that differentiates it from these B. subtilis regulators that do not confer a Spo phenotype. It is likely that SasN plays additional roles in M. xanthus development, such as regulating other developmentally expressed genes. Understanding the mechanism of SasN action and the elements that control SasN activity and/or expression should help to clarify how M. xanthus cells limit the critical transition from growth to development to the appropriate conditions by integrating information on their nutrient status and cell density.