Regulation of multiple tip formation by caffeine in cellular slime molds
© Jaiswal et al.; licensee BioMed Central Ltd. 2012
Received: 27 June 2012
Accepted: 20 August 2012
Published: 28 August 2012
Skip to main content
© Jaiswal et al.; licensee BioMed Central Ltd. 2012
Received: 27 June 2012
Accepted: 20 August 2012
Published: 28 August 2012
The multicellular slug in Dictyostelium has a single tip that acts as an organising centre patterning the rest of the slug. High adenosine levels at the tip are believed to be responsible for this tip dominance and the adenosine antagonist, caffeine overrides this dominance promoting multiple tip formation.
Caffeine induced multiple tip effect is conserved in all the Dictyostelids tested. Two key components of cAMP relay namely, cAMP phosphodiesterase (Pde4) and adenyl cyclase-A (AcaA) levels get reduced during secondary tip formation in Dictyostelium discoideum. Pharmacological inhibition of cAMP phosphodiesterase also resulted in multiple tips. Caffeine reduces cAMP levels by 16.4, 2.34, 4.71 and 6.30 folds, respectively in D. discoideum, D. aureostipes, D. minutum and Polysphondylium pallidum. We propose that altered cAMP levels, perturbed cAMP gradient and impaired signalling may be the critical factors for the origin of multiple tips in other Dictyostelids as well. In the presence of caffeine, slug cell movement gets impaired and restricted. The cell type specific markers, ecmA (prestalk) and pspA (prespore) cells are not equally contributing during additional tip formation. During additional tip emergence, prespore cells transdifferentiate to compensate the loss of prestalk cells.
Caffeine decreases adenyl cyclase–A (AcaA) levels and as a consequence low cAMP is synthesised altering the gradient. Further if cAMP phosphodiesterase (Pde4) levels go down in the presence of caffeine, the cAMP gradient breaks down. When there is no cAMP gradient, directional movement is inhibited and might favour re-differentiation of prespore to prestalk cells.
Cellular slime molds are unicellular, free living soil amoebae alternating its life cycle between growth and multicellular development . As amoebae, they prey on bacteria and multiply until all the food is exhausted. At the onset of starvation, the amoebae secrete chemoattractants to communicate with each other enabling them to form a multicellular aggregate. The aggregates transform to a motile slug which later culminates to a fruiting body with a dead stalk holding a mass of dormant spores. D. discoideum slug consists of two prominent cell types: the anterior prestalk cells and the posterior prespore cells . Four morphogenetic regulators viz., cAMP, adenosine, ammonia (NH3) and differentiation inducing factor (DIF) coordinate and regulate cell fate and cell type proportioning during development in slime molds [3–6].
Cellular slime molds are grouped in 4 distinct evolutionary lineages based on the small subunit ribosomal DNA (SSU) rDNA and α-tubulin amino acid sequences . Group 1 species-D. aureostipes, Group 2 species-Polysphondylium pallidum, Group 3 species-D. minutum and Group 4 species-D. discoideum makes use of, an unknown compound, glorin, folic acid and cAMP respectively, as chemoattractants for their aggregation [8–13]. Caffeine is known to induce multiple tip formation in D. discoideum and it is not clear if multiple tip formation induced by caffeine is common to other cellular slime molds.
Tip dominance is one of the crucial steps in slug volume regulation during morphogenesis in cellular slime molds . The single slug tip, like an organiser of the metazoan embryo regulates multicellular development . The tip of the slug acts as a pacemaker  and secretes cAMP signals periodically with a propagation speed of 200 μm/min . The cell movement within the slug is directed and move with an average period of 3 minutes . The cAMP waves are initiated at the slug tip and propagate towards the back of the slug . Because of the primary tip dominance, additional tip formation is repressed , a phenomenon called tip inhibition and adenosine plays a crucial role in this process by inhibiting new tip formation . The mechanism of multiple tip formation is not well understood but it is believed that cAMP relay might regulate this process . Caffeine is known to inhibit the oscillatory cAMP relay [17, 18] and removes tip inhibition by reducing the amplitude and oscillation frequency of cAMP signals . Caffeine, the antagonist of adenosine favours tip activation inducing multiple tip formation [17, 19].
The cAMP signal strength and its relay are regulated by the activity of adenyl cyclases (AcaA), cAMP phosphodiesterase (PdsA and Pde4) and cAMP phosphodiesterase inhibitor (PDI) [20–24]. cAMP upon binding to its receptors (cAR1) activates adenyl cyclase to catalyze the conversion of ATP into cAMP [23, 24]. The secreted cAMP gets degraded by PdsA into 5’AMP which is negatively regulated by PDI [20, 23]. The intracellular cAMP levels are governed by another cAMP phosphodiesterase, RegA. The proteins kinase-A (PKA), the downstream target of intracellular cAMP, upon binding to its regulatory site (PkaR) activates catalytic domain (PkaC) inducing multicellular development . PkaC is known to regulate cAMP relay and genetic lesions in this gene result in defective aggregation . The genes associated with cyclic nucleotide signaling are well conserved across different slime mold species .
During secondary tip formation, cells within the slug could possibly sort out or transdifferentiate. Cell sorting is chemotactic to cAMP; prestalk cells sort out by moving towards cAMP source . Cell sorting in Dictyostelium is the result of coordinated chemotactic cell movement and cAMP relay kinetics between both the cell types, prestalk and prespore . During tip emergence, cells that move fast and respond strongly to cAMP signalling, collect on the mound tops .
Caffeine is an antagonist of adenosine and consist of a purine ring and three methyl groups at 1, 3, 7th position of the ring, which is commonly named as 1, 3, 7 trimethyl xanthine. Adenosine, a hydrolysed derivative of cAMP, is synthesised within the slug tip . cAMP levels are regulated by secreted cAMP phosphodiesterase (Pde4) known to hydrolyze cAMP into 5’AMP . AMP further gets degraded into adenosine by 5’ nucleotidase .
Here, we show that the multiple tip formation is conserved in all 4 slime mold groups and this effect is not observed when treated with caffeine analogs. The cAMP relay during multiple tip formation was indirectly monitored by checking the expression levels of adenyl cyclase-A (AcaA) and extracellular cAMP phosphodiesterase (PdsA and Pde4). We quantified cAMP levels in slugs with secondary tips based on an enzyme linked immune sorbent assay (ELISA). During caffeine induced multiple tip formation, there is impaired cell movement in slugs leading to spontaneous transdifferentiation. Cell movement within the slugs was monitored by tracking a small fraction of fluorescently labelled cells. Regeneration of ablated prestalk ecmA region in the slug during multiple tip formation suggests transdifferentiation of prespore to prestalk cells.
Polysphondylium pallidum PN500 cells were grown on GYP media  in the presence of E. coli B/r - at 22°C with 70% relative humidity [29, 30]. All Dictyostelium strains except AX2 were grown on SM/5 agar plates with K. aerogens at 22°C. AX2 cells were grown in axenic HL5 media (28.6 g bacteriological peptone (Oxoid), 15.3 g yeast extract (Oxoid), 18.0 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 (150 RPM). When there was visible clearing of the bacterial lawns, the plates were washed thrice with ice-cold phosphate buffer (pH 6.4) and cells were harvested at 400 g with 4 minutes of spinning. The cells were spread on non-nutrient agar plates at density of 2 X 106 cells/cm2 and incubated in high humid conditions.
Amobae stained with 0.06% neutral red was incubated at 22°C for 10 minutes and washed twice with KK2 buffer. Neutral red stained amoebae were spread at a density of 2 X 106 cells/cm2 on non-nutrient agar plates and allowed to develop as slugs. Fully developed slugs were transferred using a fine needle onto a buffered agar plate having 5 mM caffeine and observed for multiple tip formation.
To check the expression levels of PdsA, slugs with multiple tips 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 minutes . 20 μl of the cell lysate was electrophoresed in 10% polyacrylamide gels. Equal loading of the protein lysate was checked by Commassie-blue staining running a parallel gel. Anti-PdsA (1:1000-a kind gift from Carole A. Parent, NIH, USA) polyclonal antibody was incubated over night at 4°C. Then, the secondary HRP conjugated antibodies were incubated at room temperature for one hour.
List of primers used in Real-Time PCR
Slugs were fixed in 3.7% formaldehyde solution in Z-buffer (60 mM Na2HPO4, 40 mM NaH2PO4, 10 mM KCl, 1 mM MgSO4 and 2 mM MgCl2) for 15 minutes. After decanting the fixative, 0.1% NP-40 solution in Z-buffer was added for 15 minutes. Subsequently the plates were washed with Z-buffer and the fixed samples were submerged in freshly prepared staining solution (20 μl of 1 mM X-gal solution in equal volume of 5 mM K3 [Fe (CN6)], and 5 mM K4 [Fe (CN6)] solution and incubated at 37°C for 45 minutes before observation.
To quantify cAMP levels in the slugs, a cAMP XPTM assay kit (catalog no.4339) was procured from Cell Signaling, USA. This kit contains a cAMP XPTM rabbit mAB coated 96 well plate and HRP linked cAMP, substrate (TMB) and other necessary reagents. The slugs were lysed in a 100 μl of 1X lysis buffer containing 1 mM PMSF (phenyl methyl sulfonyl fluoride) and the lysed sample was incubated in ice for 10 minutes. 50 μl of the test sample and 50 μl of the HRP-linked cAMP solution were added on to the assay plate and was incubated at room temperature for 3 hours on a horizontal orbital shaker. The supernatant was discarded and the wells were washed thrice with 200 μl of 1X wash buffer and thereafter 100 μl of TMB substrate was added to the wells. Subsequently, the plate was incubated at room temperature for 10 minutes and the reaction was terminated by adding 100 μl of stop solution. The absorbance was measured at 450 nm. The cAMP standard curved was used to calculate the absolute amount of cAMP in the test samples.
Nikon SMZ-1000 stereo zoom microscope with epi-fluorescence optics was used for monitoring and taking the pictures. The fluorescence images were taken using a Nikon 80i eclispse upright microscope.
Further, we monitored the development of pde4 - cells on non-nutrient agar plates. pde4 - mound breaks and form tips after 9 hours of development (Figure 5B) and from each tip a small slug develops culminating to a fruiting body (Figure 5C). The fruiting body phenotype of Pde4 mutants was identical to the fruiting bodies formed in the presence of caffeine (Figure 5C).
However ectopic tips in the wild type fruiting bodies (AX2) developed in caffeine containing plates were prominent than the ones developed in the presence of IBMX (Figure 5C). Thus our experiment suggests that a stable cAMP gradient in the slug is required for normal development and its disruption leads to secondary tip formation.
Additional file 1: Movie 1A. Cell movement in control slug. (AVI 8 MB)
List of primers used in semi quantitative PCR
In Dictyostelium, the differentiation of cell types and its fate is regulated by morphogens which are DIF (Differentiation Inducing Factor), cAMP, ammonia and adenosine . In D. discoideum, cAMP acts as a chemoattractant as well as regulating cell differentiation . However, in other slime mold species like D. minutum, P. pallidum and D. aureostipes, cAMP is involved in cell differentiation only and not chemotaxis [7, 9, 13]. The slug tip acts as an organiser and regulates the volume and shape . In D. discoideum, adenosine is known to prevent competing tip formation by favouring tip dominance . In the branched slime mold Polysphondylium the primary tip is known to inhibit secondary tip formation  and surgical removal of the apical tip result in secondary tip emergence suggesting the dominance of the apical tip. During its development Polysphondylium goes through a spontaneous spherical to radial symmetry breaking event and during this transition many tip arise around the equator of a spherical mass of cells equidistant from each other suggesting that that one tip inhibits the other . Probably this lateral inhibiton in Polysphondylium is akin to caffeine induced multiple tip effect. Multiple tip phenotype induced by caffeine was observed in all the slime molds species we investigated suggesting a conserved mechanism regulating secondary tip formation in slime molds. The formation of secondary tips is controlled by relayed cAMP signal strength and suppression of tip dominance . In this study, we examined the mechanism that regulates multiple tip formation in the presence of caffeine. Both in the previous report  and in this work, millimolar concentrations of caffeine have been used to generate multiple tip effect. Being soil amoebae, slime molds have to encounter a variety of environments and may have efficient ABC transporters to thrive and hence all the effects can be observed only at millimolar concentrations .
Few known mechanisms of caffeine action include inhibition of cyclic nucleotide phosphodiesterase, competitive inhibition of adenosine receptors, inhibition of ryanodine receptors  and inactivation of Target of Rapamycin complex (TOR complex) . Adenosine receptors are not known to be present in the Dictyostelium genome and hence caffeine could possibly target cAMP phosphodiesterases (PdsA and Pde4) or ryanodine receptor. It is not known if caffeine targets ryanodine receptors in Dictyostelium but caffeine does not affect PdsA levels. It is likely that caffeine by impairing cAMP relay and altering intracellular calcium levels cause pleiotropic effect on signaling, motility and gene expression all leading to multiple tip formation.
Theophylline (1, 3 dimethyl xanthine), theobromine (3, 7 dimethyl xanthine) and paraxanthine (1, 7 dimethyl xanthine) are caffeine analogs, sharing a common structural xanthine backbone [41, 42]. These compounds are known to have specific selectivity towards different targets like adenosine receptors, and calcium channels [41, 42]. Caffeine being highly lipophilic (trimethyl xanthine) than theobromine or theophylline (dimethyl xanthine) may be why it is most effective, suggesting an intracellular function of caffeine possibly through opening calcium stores. In the presence of caffeine analogs and adenosine, the slugs continued to move which could be tracked by a trail that it leaves behind (Additional file 3: Figure S1). When slugs were transferred to a plate containing 100 μM A23187 + 1 mM CaCl2 it did not result in secondary tip formation (data not shown) the slugs rounded up similar to a mound seen in controls (without drug). Similarly, when slugs were transferred to a plate containing 5 mM EGTA, there was arrested development (data not shown). Studies with caffeine analogs and supplementing calcium with a specific ionophore did not result in multiple tip formation suggesting that an increase of cytosolic calcium alone is not responsible for additional tip effect and a caffeine specific pool of calcium reservoirs is activated during additional tip formation.
During post aggregate stages of development, the tip of the slug continues to be a source of cAMP signalling . Adenyl cyclases catalyse the conversion ATP into cAMP. During multiple tip formation, cAMP levels decrease as a consequence of reduced adenyl cyclase-A expression and its activity. Adenyl cyclase-A is regulated by the activity of cytosolic regulator of adenyl cyclase (CRAC) and Target of Rapamycin complex-2 (TORC2) . In yeast, caffeine acts on TOR1 unit of TORC1 complex and inactivates it . However, in Dictyostelium TOR is a part of both TORC1 and TORC2 so it is likely that caffeine can inactivate both TORC1 and TORC2. TORC2 regulates chemotaxis and multicellular organization by monitoring the expression levels of adenyl cyclase-A (acaA) and cAMP signal relay. The reduced expression of adenyl cyclses and decreased cAMP levels in slugs with additional tips (in the presence of caffeine) can be due to impaired activity of TORC2. However, in the presence of rapamycin, we did not observe multiple tips in the slugs (data not shown). It is not known if rapamycin affects pde4 levels. Also the specific calcium pool mobilized by caffeine may not be responding to rapamycin though both could impair TORC2 activity. Though caffeine and rapamycin are known to induce identical set of genes in yeast , structurally they are different and caffeine effect is pleiotropic. So, caffeine besides affecting TORC2 and decreasing cAMP levels can interact with other proteins such as cAMP phosphodiesterase (PdsA and Pde4) which are involved in maintaining cAMP gradient and that could also significantly contribute to multiple tip formation and all these effects may not happen with rapamycin alone.
In vitro experiments suggest that PdsA levels are not inhibited by caffeine and so it is likely that Pde4 is a key component responsible for maintaining gradients of cAMP as well as adenosine. cAMP phosphodiesterase (Pde4) hydrolyses cAMP into AMP thus generating cAMP gradient from anterior to the posterior region of the slug. Overproduction of the phosphodiesterase Pde4 is known to arrest development at the tight mound stage, prior to tip formation. During multiple tip formation, cAMP phosphodiesterase (Pde4) levels go down and adenosine levels and cAMP gradient also gets perturbed. Inhibition of cAMP phosphodiesterase (Pde4) activity with iso butyl methyl xanthine (IBMX) results in secondary tips in slugs and fruiting bodies. Though application of IBMX, a proven inhibitor of Pde4 gives rise to multiple tips, the phenotype is prominent with caffeine. Further, many pleiotropic pathways converge to give multiple tip effect whereas IBMX effect may be specific to Pde4 alone.
Adenosine is the breakdown product of cyclic AMP, and known to act as an antagonist to cyclic AMP, providing an intrinsic negative feedback loop [14, 47]. Adenosine is known to inhibit competitive tip formation by tip dominance. Multiple tips formed in the presence of caffeine can also be due to the suppression of tip dominance. Tips formed at the anterior of the slug are smaller compared to the ones at the posterior suggesting that anterior end exert maximum inhibition of tip dominance and it gets weaker to the posterior of slug. Secondary tip formation in Polysphondylium in the presence of extracellular cAMP suggests the disruption of cAMP gradient leading to tip formation elsewhere in the slug . PkaC is known to regulate cAMP relay signal and cells lacking it are impaired in aggregation . It is not clear if a 1.6 fold decrease in the expression of PkaC would significantly contribute to multiple tip effect as real time PCR data showed no significant changes in the expression of PkaC between slugs having multiple tips and control (Figure 4). We propose that reduced cAMP levels, suppression of tip dominance and altered cAMP gradient, together cause secondary tip formation in slugs.
It is known that a positional signal in isolated prestalk or prespore part induces localized transdifferentiation of the appropriate cell type to form the prestalk and prespore pattern. During secondary tip formation, we observed the prestalk marker EcmA-GFP expression in the prespore region indicating a position–dependent mechanism of regulation. Cells in the prespore region of the slug move in periodic fashion with a speed of 20 μm / minute . If the slug tip is removed it impairs cell movement in the prespore region and all the prespore cells form a mound. ALCs scattered along the slug length and prespore cells re-differentiate to prestalk cells until a proportioned slug is formed . Cells within the caffeine treated slugs do not migrate far and undergo transdifferentiation, forming local aggregation centres resulting in secondary tips. ALCs scattered in the prespore region are likely to initiate the first organising center and additional prespore cells are recruited along with ALCs to form multiple tips . Prestalk EcmA marker expression analysis during secondary tip formation suggests cell movement is severely impaired during ectopic tip formation. If there is no cell movement, then transdifferentiation could be the reason for multiple tip effect, which is what we observed. Semi-quatititative analysis of ecmA and pspA expression showed that prestalk and prespore cells are disproportionate during secondary tip emergence. The redifferentiation of prespore and prestalk cells is controlled by cAMP and DIF-1 (Diffrentiation inducing factor) in a combinatorial manner. In the presence of DIF-1, spore differentiation is inhibited and the cells become stalk instead  so the stalk cell diffrentiation in prespore part can be modulated by DIF-1. cAMP is known to induce prespore cell differentiation  and thus the expression of prespore marker in prestalk part might be carried out by cAMP. The surface sheath of pseudoplasmodium also plays an important role regulating spore differentiation . The low molecular weight, diffusible effectors mainly NH3 and cAMP determines the thickness of the sheath . The sheath covering the posterior prespore region is thicker compared to the slug anterior and this may also play a critical role in establishing gradients of cAMP . During secondary tip formation, the levels of these effector molecules in the posterior region of the slug may decrease and it is not sure if the thickness of the sheath also reduces. The thinner sheath of the pseudoplasmodium does not favour spore cell differentiation and hence cells sort out to the prestalk and form secondary tips in varied prestalk/prespore cell proportions.
Soon after transferring the slugs to caffeine containing plates, the cell migration stops within the slug and impaired cell movement alone may not be responsible for giving the multi-tipped slug. The multiple tip effect gets severe with time suggesting that the phenotype that we observe depends on changes in gene expression also. Prolonged observation with intense beam of light for microscopic observations prevented multiple tip formation in the presence of caffeine (data not shown).
Few examples of mutants with additional tips include the ones carrying genetic lesions in 5’ nucleotidase and TipA and TipB mutants [50, 51]. The latter two have defects in sorting of prestalk cells which also show a multiple tip phenotype in the mound stage c. Treating these Tip mutants with caffeine failed to show any change in their phenotype and adenosine failed to rescue these defects either. (Additional file 4: Figure S2). Caffeine besides inducing multiple tips in P. pallidum slugs, affects fruiting body branching with tertiary branches coming out of the secondary stalks (Additional file 5: Figure S3).
The novel finding of this work includes 1. Conserved action of caffeine in inducing multiple tips in different slime mold species. 2. In all species examined, caffeine perturbs cAMP signaling, its levels and gradient within the slug possibly favoring multiple tip formation. 3. The origin of multiple organizing centers is a caffeine specific effect and its analogs fail to give ectopic tips. 4. During ectopic tip formation, there is a highly restricted cell movement and trans-differentiation of prespore to prestalk cell types takes place within the slug.
The effect of caffeine was monitored on different slime mold species for multiple tip formation and in this study, we show that certain mechanisms of multiple tip formation seems to be conserved among distantly related slime molds and factors/mechanism regulating tips formation is also similar in other slime molds. The multiple tip formation is specific to the presence of caffeine. Caffeine reduces cAMP levels in slugs altering its gradient and relay thereby inducing tips elsewhere in the slug. Our work also suggests that cAMP playing a critical role in other slime molds during later stages of development.
This work was supported by Department of Biotechnology (DBT), Council of Scientific and Industrial Research (CSIR), New Delhi, Government of India.
We thank Dr. Carole A. Parent, for providing PdsA antibodies, Dr. Leandro Sastre for providing AX2-AcaA-lacZ strain for this work. We thank Dr. Suresh Rayala for allowing us to use the Real Time-PCR machine. We thank Bill Loomis and Ted Cox for their critical review of the earlier version of this manuscript. All the authors gratefully acknowledge the help of the Dictyostelium stock center for sending out various strains used in this study. SPS thanks University Grants Commission for a scholarship.
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.