Regulation of aggregate size and pattern by adenosine and caffeine in cellular slime molds
© Jaiswal et al; licensee BioMed Central Ltd. 2012
Received: 22 July 2011
Accepted: 23 January 2012
Published: 23 January 2012
Skip to main content
© Jaiswal et al; licensee BioMed Central Ltd. 2012
Received: 22 July 2011
Accepted: 23 January 2012
Published: 23 January 2012
Multicellularity in cellular slime molds is achieved by aggregation of several hundreds to thousands of cells. In the model slime mold Dictyostelium discoideum, adenosine is known to increase the aggregate size and its antagonist caffeine reduces the aggregate size. However, it is not clear if the actions of adenosine and caffeine are evolutionarily conserved among other slime molds known to use structurally unrelated chemoattractants. We have examined how the known factors affecting aggregate size are modulated by adenosine and caffeine.
Adenosine and caffeine induced the formation of large and small aggregates respectively, in evolutionarily distinct slime molds known to use diverse chemoattractants for their aggregation. Due to its genetic tractability, we chose D. discoideum to further investigate the factors affecting aggregate size. The changes in aggregate size are caused by the effect of the compounds on several parameters such as cell number and size, cell-cell adhesion, cAMP signal relay and cell counting mechanisms. While some of the effects of these two compounds are opposite to each other, interestingly, both compounds increase the intracellular glucose level and strengthen cell-cell adhesion. These compounds also inhibit the synthesis of cAMP phosphodiesterase (PdsA), weakening the relay of extracellular cAMP signal. Adenosine as well as caffeine rescue mutants impaired in stream formation (pde4 - and pdiA - ) and colony size (smlA - and ctnA - ) and restore their parental aggregate size.
Adenosine increased the cell division timings thereby making large number of cells available for aggregation and also it marginally increased the cell size contributing to large aggregate size. Reduced cell division rates and decreased cell size in the presence of caffeine makes the aggregates smaller than controls. Both the compounds altered the speed of the chemotactic amoebae causing a variation in aggregate size. Our data strongly suggests that cytosolic glucose and extracellular cAMP levels are the other major determinants regulating aggregate size and pattern. Importantly, the aggregation process is conserved among different lineages of cellular slime molds despite using unrelated signalling molecules for aggregation.
During their life cycle, cellular slime molds alternate between unicellular and multicellular forms . The unicellular amoebae feed on bacteria and retain their single cell identity as long as the food is abundant. At the onset of starvation, hundreds to hundreds of thousands of amoebae initiate a chemotactic signal-relay using polyketides, nucleotides or peptides and other unidentified signalling molecules to form a multicellular slug [2–7]. Cells at the anterior of the slug differentiate as a dead stalk while the rest of the cells encapsulate as spores in a fruiting body. Based on the small subunit ribosomal DNA (SSU) rDNA and α-tubulin amino acid sequences, the entire cellular slime mold 'Dictyostelia' are grouped in 4 distinct evolutionary lineages . cAMP is a chemoattractant in all group 4 species including D. discoideum, D. mucoroides and D. giganteum [5, 9] while in other groups at least three different compounds are used for aggregation. Group 3 species like D. lacteum, D. minutum and D. tenue make use of pterin, folic acid, and an unknown compound, respectively [3, 4, 7]. A modified dipeptide, glorin (N-propionyl-Y-L-glutamyl-L-ornithine and lactam ethyl ester) and an unknown compound act as chemoattractants in group 2 species, Polysphondylium pallidum and P. luridum, respectively . It is not clear to what extent the signalling pathways that regulate aggregation are conserved between these different slime mold groups that use structurally unrelated chemoattractants.
The four major determinants known to regulate aggregate size in D. discoideum include the overall cell number and their size within the aggregate, the counting mechanism, cell-cell adhesion and cAMP signal strength [10, 11].
The number and size of the individual cells within an organism determines its overall size or bulkiness [12, 13], and signalling pathways that control cell growth such as the Target of Rapamycin (TOR) kinase pathway  and cell proliferation are important for controlling organ size. The number of cells required to form an aggregate of certain size is regulated by the counting mechanism that precisely counts and foretells when an aggregate of critical size is reached . This is achieved by a set of secreted proteins, the concentration of which determines when an aggregate has to break or continue aggregation to reach certain size. In D. discoideum, the cell number available for aggregation is governed by a secreted factor called conditioned medium factor (CMF; [16, 17]). The threshold concentration of CMF in the medium determines cAMP expression and secretion of countin factors by the amoebae [18, 19]. The counting factor (CF) regulates intracellular glucose levels, cell movement and cell adhesion and maintains the integrity of aggregation streams [10, 15]. CF is comprised of Countin, CF-50, CF-45-1, and CF-60 proteins, and increases cell movement by reducing the cytosolic glucose levels . Indeed, cells starved in the presence of 1 mM glucose exhibit slower movement than untreated controls .
Cell-cell adhesion, another important determinant of aggregate size, is established through cell surface glycoproteins like Cad-1 (gp24) and CsaA (gp80) . Cad-1 is a cadherin-dependent EDTA-sensitive glycoprotein, expressed in early stages of development and CsaA is an EDTA-resistant glycoprotein, expressed during later stages of development  the activation of which is cAMP dependent. The fourth factor known to affect aggregate size is the cAMP signal strength, which depends on cAMP-dependent adenyl cyclase activation, cAMP phosphodiesterase activity (PdsA and Pde4) and the concentration of the cAMP phosphodiesterase inhibitor (PDI) [21–24]. cAMP upon binding to its surface receptors (cAR1) activates adenyl cyclase to catalyze the conversion of ATP into cAMP . The secreted cAMP further gets converted to 5'AMP by the action of PdsA, which is negatively regulated by a phospho diesterase inhibitor PDI . Genetic lesions in the genes encoding for counting factor (Countin, CF-50, CF-45-1, CF-60), cell-cell adhesion (Cad-1, CsaA), or cAMP signal relay (Adenyl cyclase A (ACA), PdsA, cAR-1) and cAMP phosphodiesterase inhibitor (PDI), respectively, affect the aggregate size in Dictyostelium [23, 25–29].
Adenosine, one of the morphogens identified in D. discoideum increases the aggregate size and influences cell fate during development . It consists of adenine attached to a ribose sugar via a β-N9-glycosidic bond. It is a hydrolysed derivative of cAMP, synthesised within the slug tip, which represses competing tip initiation . Pde4, an extracellular cAMP phosphodiesterase regulates cAMP levels in Dictyostelium slugs by catalysing the conversion of cAMP into 5'AMP , which further gets converted to adenosine by 5' nucleotidase . The adenosine antagonist, caffeine represses signals that prevent tip formation, thereby inducing additional tips in D. discoideum slugs [32, 33]. Caffeine has three methyl groups in a purine ring and is commonly named as 1, 3, 7 tri-methyl xanthine. Caffeine is known to increase intracellular Ca++ by promoting the discharge of Ca++ sequestered in mitochondria and smooth endoplasmic reticulum [34, 35] which in turn is known to block pinocytosis in slime molds . Adenosine is a non-competitive inhibitor of cAMP receptors [36–38] and caffeine reversibly inhibits the cAMP-dependent activation of the adenylate cyclase [34, 39].
Here, we investigated the effect of adenosine and caffeine on the aggregate size of several Dictyostelia species across all 4 slime mold groups. We present evidence that these compounds change the aggregate size by modulating cell number and size, countin expression, cytosolic glucose levels, cell movement, and cell-cell adhesion.
All wild type strains of Dictyostelium were cultured on SM/5 agar plates in association with K. aerogenes at room temperature (22°C) except AX2 which was grown in HL5 media. The Dictyostelium mutant strains were grown in axenic HL5 medium (28.6 g bacteriological peptone (Oxoid), 15.3 g yeast extract (Oxoid), 18 g Maltose (Sigma), 0.641 g Na2HPO4 (Merck) and 0.49 g KH2PO4 (Fluka) per litre, pH-6.4) containing antibiotics (200 units/ml penicillin and 200 μg/ml streptomycin sulphate) at 22°C with constant shaking (180 RPM). Polysphondylium pallidum PN500 (a kind gift from Dr. Edward Cox, Princeton University) was grown on GYP agar plates (1 g glucose, 2 g bacteriological peptone (Oxoid), 0.2 g yeast extract (Oxoid), 4.2 g KH2PO4 (Fluka) 2.7 g Na2HPO4 (Merck) and 15 g agar per litre, pH-6.4) in association with E. coli B/r - at 22°C with 70% relative humidity. When there was visible clearing of the bacterial lawns, the cells were harvested by washing the plates with ice-cold KK2 buffer (2.25 g KH2PO4 and 0.67 g K2HPO4 per litre H2O, pH 6.4). Thereafter, the amoebae were plated at a density of 1 × 106 cells/cm2 on non-nutrient agar plates (KK2 buffer containing 15 g agar per litre, pH 6.4) containing the indicated concentration of adenosine or caffeine and we scored for changes in the aggregation pattern under a microscope (Nikon SMZ1000 and Nikon eclipse 80i).
The cell division kinetics of AX2 cells was performed in three different conditions: 1. In the presence of adenosine or caffeine: We inoculated 2 × 106 in test tubes having 10 ml of HL5 medium with either adenosine or caffeine (2 mM) and incubated at 22°C with constant shaking (180 RPM). The kinetics was monitored by counting the number of cells with a haemocytometer at regular intervals under a light microscope. 2. Growth kinetics of starved cells in the presence of adenosine/caffeine: We harvested vegetative cells grown in HL5 media (without caffeine/adenosine) and washed twice with ice cold KK2 buffer. We inoculated 1.2 ± 0.12 × 106 cells in 10 ml of Sorensen buffer (2 g KH2PO4 and 0.29 g Na2HPO4 per litre H2O, pH 6.4) containing (3 mM) caffeine/adenosine/10 mM glucose or combinations of these compounds with 10 mM glucose. After 9 hours of incubation at 22°C, we counted the number of cells with a haemocytometer. 3. Growth kinetics during early developmental stages: Cells that were not exposed to the drugs earlier during growth were allowed to develop with caffeine or adenosine and subsequent to aggregate formation, they were dissociated and the cell number was counted. Aggregates were allowed to form in 90 mm Petri dish submerged in Sorensen phosphate buffer containing either caffeine or adenosine (3 mM). The aggregates were dissociated at the indicated time points by incubating them with dissociation buffer (50 mM Tris-HCl (pH-7.5), 5 mM EDTA, 0.2% Pronase-E) and numbers of cells were counted in a haemocytometer.
To measure cell size of starving cells, Sorensen buffer and Sorensen buffer + 120 mM sorbitol containing the indicated concentrations of caffeine and adenosine were used. Sorensen buffer was complemented with sorbitol to maintain the osmolarity  in case caffeine or adenosine perturbs a change in the cell size. 5 × 106 cells/ml were incubated in with constant shaking at 150 RPM on a horizontal shaker for 6 hours. To measure the size of vegetative AX2 cells, we replenished the medium with HL5c (Formedium-HL5 medium with glucose) medium containing 5 mM adenosine or 5 mM caffeine when cells reached a 70% confluence. After 6 hours of incubation, we collected the cells and measured the cell sizes using CasyR TT cell counter machine. The cell volume was measured by the scale given in packed cell volume tubes (PCV).
AX2 cells were plated in non-nutrient agar containing either 2 mM adenosine or 2 mM caffeine and individual cell movement was recorded for 8 hours after starvation . For each cell, the field of view was recorded for 5 minutes and cell movement was then calculated in 1 μm/min intervals . Bulent screen recorder software was used to record the movement of cells and the entire observation was carried out under a Nikon eclipse 80i upright microscope.
2 × 107 cells were suspended in 10 ml of KK2 buffer containing 2 mM adenosine or 2 mM caffeine and incubated at 22°C with constant agitation (180 RPM). Cell-to-cell adhesion assay was carried out by scoring for the presence of solitary cells or clumps of cells with two or more cells adhered to each other [19, 41].
Axenically grown AX2 cells were harvested and re-suspended in HL-5 medium or in Sorensen buffer (starvation) at a density of 8 × 106 cells/ml and incubated for 6 hours at 22°C with constant shaking in the presence of 3 mM adenosine or 3 mM caffeine. The amoebae were harvested at 1200 RPM for 10 minutes at 4°C in PBM buffer (20 mM KH2PO4, 10 μM CaCl2, 1 mM MgCl2, pH 6.1;  and were lysed by freezing them at -80°C for 8 hours. 35 μl from the supernatant was mixed with 200 μl of glucose assay reagent (GAHK20; Sigma-Aldrich, USA) in a 96 well microtiter plate, incubated for 15 minutes and the absorbance was measured at 340 nm. From a similar sample, 2.0 μl of supernatant was mixed with 250 μl of Bio-Rad protein assay reagent, incubated for 20 minutes and absorbance was measured at 595 nm for protein estimation.
For immuno-blotting of Cad-1 and CsaA cell adhesion proteins, extracellular cAMP phosphodiesterase (PdsA) and Countin, we grew AX2 cells in HL5 media and starved in Sorensen buffer at a density of 1 × 107 cells/ml in the presence or absence of the drug at 22°C with continuous shaking (150RPM). 5 × 106 cells were lysed in 200 μl cell lysis buffer (2% SDS, 0.5 M Tris- pH-6.8) containing 1% mercaptoethanol and the mixture was heated at 95°C for 5 min . 20 μl of the cell lysate was electrophoresed either in 13% (Countin and PdsA) or in 10% (Cad-1, CsaA,) polyacrylamide gels. Equal loading of the protein lysates was checked by staining the nitro-cellulose membrane with ponceau-S dye after blotting. The anti-Dd Cad-1 (1:8000) polyclonal antibody, anti-PdsA (1:1000) polyclonal antibody, anti-CsaA (1:10 from the mice supernatant) monoclonal, anti-countin (1:500), polyclonal antibody were incubated over night at 4°C. Subsequently, secondary HRP conjugated antibodies was incubated for one hour at room temperature along with membrane.
For reconstituting acaA - cells with AX2 cells, we grew both the strains in HL5 medium separately. After pelleting, the cells were washed twice with ice cold KK2 buffer, counted and mixed in two different ratios (4:1 and 2:3). Subsequently, the mixed cells were (1 × 107 cells/ml) starved together in Sorensen buffer for 5 hours at the density of the 1 × 107 cells/ml and plated at density of 1 × 106 cells/cm2 on non-nutrient plates containing either adenosine or caffeine.
The statistical analyses were carried out using the Microsoft Office Excel 2003 software. The statistical significance of experiments was confirmed by performing either One-way ANOVA (Analysis of Variance) or Student's t-test (paired).
Since the effects of adenosine and caffeine were similar in all Dictyostelium and Polysphondylium sp. examined, the mechanisms that regulate aggregate size may also be common. We hence chose D. discoideum to know the factors that might contribute to changes in aggregate size, pattern and streaming in the presence of adenosine or caffeine.
Cell division during development in the presence of either caffeine or adenosine
3.0 ± 0.5 × 106
After 9 hours
3.83 ± 0.14 × 106
3 mM adenosine
4.33 ± 0.51 × 106
3 mM caffeine
3.27 ± 0.43 × 106
Number of cells in starvation buffer (One-way ANOVA, p < 0.001, n = 5, it was performed to check if the values obtained in different treatments were significantly different from the control [without compounds]).
1.2 ± 0.12 × 106
After 9 hours
1.43 ± 0.11 × 106
3 mM adenosine
1.58 ± 0.20 × 106
3 mM caffeine
1.23 ± 0.75 × 106
10 mM glucose
2.25 ± 0.39 × 106
3 mM adenosine + 10 mM glucose
2.59 ± 0.33 × 106
3 mM caffeine + 10 mM glucose
1.90 ± 0.33 X106
Effect of caffeine and adenosine on cell size.
Sorensen's buffer +120 mM sorbitol
7.83 ± 0.10
8.87 ± 0.05
9.75 ± 0.24
5 mM adenosine
8.20 ± 0.12 (p < 0.05)
9.25 ± 0.05 (p < 0.003)
9.93 ± 0.03 (p < 0.2)
5 mM caffeine
6.59 ± 0.19 (p < 0.02)
8.39 ± 0.05 (p < 0.004)
9.30 ± 0.13 (p < 0.05)
Effect of caffeine and adenosine on cell volume
Sorensen's buffer +120 mM sorbitol
5 mM adenosine
5 mM caffeine
If both adenosine and caffeine enhance cell adhesion, then one would expect caffeine to induce large aggregate formation which never happened. To gain molecular insights of this differential response with caffeine, western blot analysis was carried out to examine if caffeine and adenosine affect the expression of the cell-cell adhesion glycoproteins Cad-1 and CsaA. During early aggregation stages, Cad-1 expression was not significant in the presence of adenosine or caffeine compared to untreated controls. A reduced Cad-1 expression by 20% and 25% was observed at 2 h and 4 h of development, respectively (Figure 4B). At 6 h of development, Cad-1 expression was almost equal with both caffeine and adenosine and further increased two-fold at 8 h and 10 h of development.
At 6 hours of starvation, i.e. at the late stage of aggregation, the expression of the developmentally regulated cell adhesion protein CsaA could be observed. Since the aggregate formation was delayed by 4 hours in the presence of both these compounds (Additional file 3, Additional file 4, Fig S3), the expression level of CsaA was also not seen. With adenosine, CsaA expression was delayed until after 8 h of development and increased its expression at 12 h and 15 h of development (Figure 4C). With caffeine, CsaA expression got reduced two-fold at 6 h and 8 h of development but increased its expression two- to three-fold higher at 10 h, 12 h, and 15 h of development (Figure 4C). Delayed expression of the cell adhesion proteins in the presence of both the compounds correlates with the development timing of the cells. At late aggregation stages, the cell adhesion proteins (CsaA and Cad-1) are strongly expressed in the presence of adenosine or caffeine. The elevated expression of these proteins might favour strong cell-cell adhesion thereby favouring large aggregate formation in the presence of adenosine. In spite of an increase of the cell adhesion protein expression with caffeine, the aggregates formed were small and it is likely that caffeine may be involved in maintaining the integrity of small but streamless aggregates. Though caffeine favoured small aggregate formation, still there were many aggregation centers each compact on its own and the expression of cell adhesion proteins may be responsible for this small but compact aggregate formation. These data suggest that enhanced expression of cell adhesion proteins results in increased binding of cells with each other resulting in large aggregate formation in the presence of adenosine.
Glucose levels are known to affect the colony size and we monitored the size of aggregates formed from cells that were grown in the presence of 3 mM these compounds. There was a precocious development (by 2 hours) when amoebae were grown in the presence of 3 mM caffeine. Vegetative cells treated with caffeine and developed in its absence formed large aggregates (Figure 5C) and the size was even larger from those cells that were exposed to adenosine during vegetative growth (not shown). This suggests that the regulation of aggregate size by either caffeine or adenosine acts via the glucose sensing pathway. The caffeine effect appeared to be stage specific and if the treatment were in vegetative stage followed by no caffeine during development, then large aggregates were formed. However, if the cells were grown in the absence of caffeine and developed in its presence later, then small aggregates were formed. An increase in intracellular glucose levels induced by caffeine during growth may be directly responsible for large aggregate formation.
To test whether increasing cytosolic glucose induces an increased expression of cell adhesion protein, the levels of Cad-1 and CsaA and were quantitated by western blot of cells grown with caffeine and developed in the presence or absence of glucose (without caffeine). The delayed development in the presence of 5 mM glucose can also be monitored by checking the expression levels of CsaA, which is a development regulatory protein . In the presence of glucose, there was an increase in Cad-1 expression and a decrease in CsaA expression compared to controls without glucose. With 10 mM glucose, Cad-1 expression increased by 1.5 to 2.5 fold at 1 h, 2 h, 3 h and 4 h of development (Figure 5D). CsaA expression gets repressed in the presence of glucose with barely detectable expression at 6 h of development. After 10 h of development in the presence of glucose, CsaA expression levels were comparable to the untreated controls (Figure 5E). Irrespective of the glucose concentrations tested (10 mM, 20 mM and 40 mM) CsaA expression was abolished at 6 h of development (Figure 5F).
In contrast, the Cad-1 levels from cells grown in the presence of 3 mM caffeine and developed in its absence were decreased by 25% and 50% at 1 h and 3 h of development, respectively (Figure 5G). CsaA expression was seen during early aggregation stages and increased gradually at 1 h, 3 h and 5 h of development (Figure 5H). Although vegetative cells exposed to caffeine mimics the effect of glucose (aggregate size), the developmental timing and cell adhesion protein expressions were different in the presence of these compounds. A reduced Cad-1 expression level in the presence of 3 mM caffeine (which is known to increase the cytosolic glucose levels) and increased expression of CsaA in the presence of extracellular glucose suggests that there is compensatory effect of these cell adhesion proteins. If the expression of one cell adhesion protein is repressed, the other cell adhesion protein expression gets stronger implying that they complement each other in increasing the overall cell adhesion.
If the cAMP relay is perturbed, it affects aggregate size . Chemotaxis, cell motility, and strength of cAMP relay during aggregation are regulated by the adenylyl cyclase AcaA, the cAMP phosphodiesterases Pde4 and PdsA, and the cAMP phosphodiesterase inhibitor PDI [21–24]. Pde4, a membrane bound extracellular cAMP phosphodiesterase regulates cAMP concentration during development . The extracellular cAMP phosphodiesterase (PdsA) is known to regulate group migration during aggregation by degrading the secreted chemoattractant. cAMP phosphodiesterase (pdsA-) mutants have an impaired relay and adenylyl cyclase (acaA - ) null cells don't secrete cAMP and both mutants do not chemotax and stream [22, 29]. The cAMP phosphodiesterase inhibitor (PDI) is a key component of the cAMP relay mechanism that helps in regulating the aggregate size. pde4- cells do not stream and forms small clumps of cells. The pdiA - cells rarely form spirals and aggregates are small .
aca- cells have impaired cAMP synthesis and remain single celled. However, if aca- cells are pulsed and developed in the presence of exogenous cAMP, aggregates phenotype can be rescued. If aca- cells were reconstituted with AX2, the secreted cAMP from AX2 might rescue aca- cells there by aggregating together. To test whether caffeine and/or adenosine affect the levels of cAMP secreted, we mixed a large fraction of aca- cells impaired in cAMP synthesis with AX2 in a ratio of 4:1 (aca - : AX2) and 3:2 (aca - : AX2). If there is sufficient cAMP production, this might rescue the aggregation phenotype of aca- cells and if either caffeine or adenosine affects this process, that could be assayed by the presence of aggregate formation. By scoring the aggregate number and size in reconstituted experiments, we could ascertain if there was a change in cAMP levels in presence of adenosine/caffeine. Mixing acaA- cells with AX2 cells in a 3:2 (aca - : AX2) ratio rescued aggregation phenotype and in 4:1 ratio, aggregation centres alone were formed with no spirals (Figure 7B). In aca - and AX2 combinations, the number of small aggregation centers formed in the presence of caffeine was higher in 3:2(aca - : AX2) ratios than 4:1 (aca - : AX2) ratio (Figure 7B). Aggregates formed in the presence of adenosine were larger than the controls in 3:2 (aca - : AX2) ratios (Figure 7B). Based on these experiments, it is likely that both caffeine and adenosine prevent cAMP degradation possibly by affecting pdsA levels and rescue the acaA- phenotype in the presence of wild type cells.
Pde4 regulates cAMP concentration during development . Caffeine might rescue pde4- function by inducing other phosphodiesrerases such as PdsA. To estimate indirectly the level of extracellular cAMP degradation in the presence of caffeine and adenosine, we performed western blots of PdsA expression in AX2 (pulsed with 30 nM cAMP for 5 hours at 6 minutes intervals), pde4- cells (pulsed with cAMP) and AX2 unpulsed. In AX2 cells pulsed with cAMP, PdsA expression at 6 hours of development was higher than from unpulsed AX2 cells developed for 12 hours (Figure 7C). In the presence of 3 mM adenosine or caffeine, PdsA levels were reduced in both pulsed and unpulsed AX2 cells. In pde4- cells, the PdsA levels at 6 h of development, in the presence of caffeine and adenosine increased by 25% and 40% respectively (Figure 7C). In the presence of both the compounds, the PdsA levels decreased in AX2 cells suggesting that either the cAMP relay becomes weaker or pulsing time gets longer than the expected 6 minutes. Caffeine indirectly might rescue the pde4- phenotype by inducing the over expression of other extracellular cAMP phosphodiesterase (PdsA).
smlA - cells overexpress counting factors (CF) and forms small aggregates and mutants with impaired countin expression favours large aggregation territories. The small aggregates of AX2 cells in the presence of caffeine could possibly be due to overexpression of countin factors. We monitored the expression of countin by western blot analysis in vegetative AX2 cells pulsed with cAMP and cells not pulsed with cAMP. In pulsed control cells without adenosine or caffeine, Countin expression at 6 hours of development was barely detectable whereas in unpulsed cells developed for 12 hours, no sign of Countin expression was observed either (Figure 10B). In the presence of 3 mM adenosine or caffeine, pulsed cells at 6 hours showed an increase of Countin expression while both compounds could not induce the expression of countin in unpulsed cells at 12 hours (Figure 10C). Though ccompared to the two mutants, smlA - and ctnA - wild type cells may have enhanced countin expression, it is interesting to note that ctnA - also forms small aggregates in the presence of caffeine. This suggests that other Countin factors such as CF45-1; CF-50; CF-60 may be involved in regulating the aggregate size in this mutant in the presence of caffeine. On the other hand, adenosine overcomes the smlA - defect in forming large aggregates. Though there was increased countin expression in the presence of adenosine, still large aggregates were formed and this may be independent of counting mechanism.
The total cell number and their individual size contribute to the size of a tissue, organ, or an organism . In D. discoideum, four prominent factors are known to affect the aggregate size, namely, cell number and cell volume, cell density sensing countin factors, cell adhesion and cAMP relay. In this study, we examined all the factors that contribute to a change in aggregate size in the presence of adenosine or caffeine as either one of them changes the size of aggregates. Interestingly, few responses to caffeine and adenosine were opposite to each other (cell proliferation rates, cell size, rates of cell movement) while other responses were similar (cAMP signal relay, cell adhesion and countin expression). However, the aggregate sizes were always large with adenosine and small in the presence of caffeine.
There was an opposing effect of adenosine and caffeine on cell division rates and cell size. The amoebae became smaller with caffeine and the cell doubling time got longer than the controls. On the other hand, adenosine increased the growth rates thereby increasing the cell number and it also increased the cell size. Caffeine is an inhibitor of Target of Rapamycin (TORC1), and yeast cells lacking Tor1p, a component of TORC1, are resistant to caffeine . The TORC1 complex is a major pathway that determines cell growth and cell size . In nutrient poor conditions or in glucose depleted conditions, the TORC1 complex is inactivated , leading among other downstream effects to autophagy induction . The TORC2 complex maintains cell integrity, cell morphology, chemotaxis and cell polarity [50, 51]. Based on the observations in yeast and other organisms, it is likely that caffeine induced reduction in cell number and cell size may be mediated by down regulating TORC1 activity in Dictyostelium also . Adenosine might be enhancing the TORC1 activity favouring an increase in cell number and cell size. In yeast, caffeine is also known to affect DNA repair mechanisms and inhibits cell cycle progression . Both the compounds may be targeting the genes involved in regulating cell proliferation thus altering the cell number in Dictyostelium also. Small aggregates formed in the presence of caffeine may be due to fewer cells available for aggregation and the cell sizes were also smaller compared to controls. Contrastingly, if there is an increase in cell number and an increase in cell size that would significantly contribute to enlarged aggregate size as with adenosine. Adenosine is likely to contribute additional cell divisions beyond starvation thereby making more cells available in a defined area for aggregation compared to controls and caffeine treated cells. Caffeine's effect seems to be pleiotropic and may depend on cell cycle stages. Caffeine might affect a fraction of the cells to have delayed cell division while in others it may completely stop cell division and relay, and this would result in a number of single cells without participating in aggregation as observed in caffeine containing plates.
During development, pulses of extracellular cAMP attract the amoebae located farther apart towards the aggregation centres or the territories. We monitored aggregate pattern and size of the cells with mutations in the key enzymes in cAMP relay that regulate synthesis (ACA), and degradation of extracellular cAMP. acaA- cells do not make cAMP, so cells remain unicellular without aggregating during starvation [54–56]. In acaA- cells, neither caffeine nor adenosine induced aggregation suggesting that both compounds do not activate other adenyl cyclases (ACB and ACG) for rescuing the defective aggregation phenotype. In wild type cells, caffeine might enhance PDI activity while down regulating PdsA levels and if that happens, then it would favour small aggregate formation. However, caffeine induced the formation of large aggregate territories in pdiA - and pde4 - mutants. PdiA- mutants form aggregates with no streams but the streaming defect could be rescued in the presence of caffeine or adenosine. This suggests that the cAMP phosphodiestrease levels might have increased in the presence of these two compounds resulting in decreased cAMP levels in the extracellular environment that is responsible for large aggregation territory sizes. pde4- cells do not stream and forms small clumps of cells. By activating other phosphodiesterases or proteins similar pdiA, caffeine might possibly rescue the function of these mutants that have defective aggregation. Unlike wild type cells, in pde4 - cells there was enhanced expression of PdsA in the presence of caffeine and adenosine. The large aggregation stream and increased expression of PdsA in pde4 - cells in the presence of caffeine suggests reduced cAMP accumulation in extracellular media. These results with adenosine and caffeine on mutants impaired in cAMP relay suggest that cAMP level is a major factor contributing to changes in aggregate sizes. Reconstitution of wild type cells with acaA- cells rescue aggregation defect in controls while in the presence of caffeine or adenosine, there were large number of aggregates implying that cAMP degradation is prevented by both these compounds. This was indirectly corroborated by western blotting showing that expression of extracellular cAMP phosphodiesterases is reduced significantly in wild type cells in the presence of caffeine or adenosine. Adding extracellular cAMP resulted in small aggregate formation  and decreasing the levels of cAMP by ammonia increased the aggregate size (Figure 8) suggesting, extracellular cAMP levels play a crucial role in determining the streaming, patterning and territory size. It is also likely that the ratio of cAMP to adenosine determining the aggregate size in Dictyostelium.
During development, caffeine is known to inhibit adenyl cyclase-A (ACA) activation  thereby delaying the cAMP relay and weakening the signal strength. These all contribute to affect the timing of aggregate formation. It is known that in the constant presence of caffeine, cAMP relay gets affected  which might be responsible for small aggregate formation.
Though the strength of cAMP relay gets weak with both the compounds, the differential directional motility of cells could still be a factor determining the aggregate size . It is known that in conditions of enhanced cell movement small sized aggregates were formed . Thus it is likely that a weak cAMP relay and increased cell motility in the presence of caffeine favouring the formation of many small aggregates. Although the strength of cAMP is weak in the presence of adenosine, the amoebae move far slowly which might influence the amoebae to join the existing aggregates rather than to form a new one thereby forming large aggregation territories.
The integrity of the aggregates is mainly maintained by the cell adhesion protein Cad-1 and CsaA and high intracellular glucose concentrations increase cell adhesion in D. discoideum [10, 26]. Glucose 6-phosphatase and glucokinase are the key enzymes regulating internal glucose levels by reversibly converting glucose-6-phosphate into glucose and vice versa . Supplementing glucose increases cytosolic glucose levels and cell-adhesion protein expression thereby favouring large aggregate size. Both adenosine and caffeine increased cytosolic glucose levels and cell adhesion protein expression causing a strong cell-cell adhesion. However, the altered aggregate sizes that we observed in the presence of these two compounds may not be due to cell adhesion protein expression alone as caffeine, similar to adenosine increased cell adhesion via increased expression of cell adhesion protein. The small but tight aggregate formed in the presence of caffeine may require strong adhesion so that the integrity is not lost and this is supported by enhanced CsaA expression and not Cad-1 (Figure 5B). Maintaining the large aggregate size may also require high expression levels of both CsaA and Cad-1. Adenosine strongly induced the expression of both the cell adhesion proteins. Aggregates formed in the presence of adenosine formed large streams which may require the both CsaA and Cad-1 expression while compact aggregates formed with caffeine had little streaming and this might have caused the expression of CsaA alone. High expression levels of the cell adhesion proteins CsaA and Cad-1is likely to contribute to changes in aggregate sizes. Adenosine may be particularly exerting its effect on cell adhesion during stream formation.
In 1986, Hangmaan  had shown that cells grown in medium containing caffeine develop more rapidly (2 hours earlier) when developed in a non-nutrient medium lacking caffeine, which is likely by inducing overexpression of adenyl cyclase-A. Amoebae grown with caffeine and developed in its absence formed large aggregates with normal streaming, and in these conditions, adenyl cylacse-A activity may be high. However, caffeine during development may be inhibiting adenyl cyclase activity favouring small aggregate formation with little streaming. When adenyl cyclase-A expression is high large aggregates were formed and when the expression is low small aggregates were formed. Caffeine, increased intracellular glucose levels which might have an influence on the aggregate size as high glucose levels are known to increase the aggregate size. However, glucose levels were still high when cells were developed in the presence of caffeine and aggregates were still small. During starvation in the presence of caffeine, factors other than glucose may be deciding the aggregate size.
Similarly, we observed precocious development and increased expression of early differentiation markers in cells pulsed with cAMP in the presence of caffeine, which may be due activation of adenyl cyclase-A (acaA) and cAMP receptors (cAR1) (Additional file 3, Additional file 8, Fig S7, Additional file 9).
The four prominent factors regulating the aggregate size were mainly examined in D.discoideum. The conserved effect of adenosine and caffeine on the aggregates size of other slime molds suggests that other species may also be having a similar mechanism in aggregate size regulation. Polysphondylium also strikingly forms large and small aggregates with adenosine and caffeine an effect which was quantitated. Though extracellular chemotactic signals may be different, the intracellular pathways regulating aggregate size may be common in all these slime molds. It is interesting that a chemotactic signal molecule used for cell-cell communication is also used for regulating the multicellular tissue size. It is likely that in other slime molds using chemoattractants other than cAMP, both adenosine and caffeine may cause a defective relay as they also form large and small aggregates, respectively.
In this study, we show that certain mechanisms such as the regulation of aggregate size are conserved among distantly related slime molds. One conserved phenotype is that for all studied organisms, the aggregates become larger in the presence of adenosine and smaller in the presence of caffeine suggesting that the factors/mechanism regulating aggregation sizes might also be similar in other slime molds. Adenosine increases cell size as well as cell number contributing to increased aggregate size while caffeine, reduced the cell size and number, favouring small aggregate formation. Both the compounds affect the speed of the chemotactic amoebae during aggregation causing a variation in the size of aggregates. Though, certain molecular processes induced by adenosine and caffeine are similar the aggregation phenotype was always different between these two compounds. The identical effects induced by both the compounds include increased cell adhesion protein (Cad-1) expression, high glucose levels and altered cAMP relay. The large aggregate with many streams formed in the presence of adenosine may be due to enhanced expression of the cell adhesion protein CsaA and Cad-1. The compactness of the small streamless aggregates formed in the presence of caffeine may be due increased expression of the cell adhesion protein CsaA and not Cad-1 (Figure 11). The synthesis, secretion of other chemoattractants unrelated to cAMP could possibly be also affected by these compounds in a similar way they affect cAMP synthesis, secretion and relay (Figure 11).
RB wishes to thank John Bonner, Edward Cox and Pauline Schaap for their helpful comments to an earlier version of this manuscript. We thank Navin Gopaldas and Sujata R. for helping us in various stages of the work. We also would like to thank Richard Gomer, Carole A. Parent, Chi-Sui and Ludwig Eichinger for providing antibodies for this work. We thank IC&SR, IIT Madras for their partial financial assistance for this project. We thank Rupert Mutzel's enthusiastic support for PJ's stay in his lab. We gratefully acknowledge the help of the Indo Swiss Joint Research Programme for their financial support. All the authors gratefully acknowledge the help of the Dictyostelium stock center for sending out various strains used in this study.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.