The Central Role of cAMP in Regulating Plasmodium falciparum Merozoite Invasion of Human Erythrocytes

All pathogenesis and death associated with Plasmodium falciparum malaria is due to parasite-infected erythrocytes. Invasion of erythrocytes by P. falciparum merozoites requires specific interactions between host receptors and parasite ligands that are localized in apical organelles called micronemes. Here, we identify cAMP as a key regulator that triggers the timely secretion of microneme proteins enabling receptor-engagement and invasion. We demonstrate that exposure of merozoites to a low K+ environment, typical of blood plasma, activates a bicarbonate-sensitive cytoplasmic adenylyl cyclase to raise cytosolic cAMP levels and activate protein kinase A, which regulates microneme secretion. We also show that cAMP regulates merozoite cytosolic Ca2+ levels via induction of an Epac pathway and demonstrate that increases in both cAMP and Ca2+ are essential to trigger microneme secretion. Our identification of the different elements in cAMP-dependent signaling pathways that regulate microneme secretion during invasion provides novel targets to inhibit blood stage parasite growth and prevent malaria.


Introduction
All the clinical symptoms of Plasmodium falciparum malaria are attributed to the blood stage of the parasite life cycle. The intraerythrocytic stage of the life cycle is initiated when liberated P. falciparum merozoites invade and multiply within host red blood cells. Following the development of mature schizonts, next generation merozoites egress from infected erythrocytes and invade uninfected erythrocytes to start a new cycle of infection. Invasion of erythrocytes by P. falciparum merozoites is a complex multi-step process that is mediated by specific molecular interactions between red cell surface receptors and parasite protein ligands [1,2]. A number of parasite ligands that mediate receptor binding during invasion reside in apical membrane-bound organelles known as micronemes and rhoptries [1,2]. Timely secretion of these parasite ligands to the merozoite surface is critical for successful invasion [3,4].
Microneme and rhoptry proteins are secreted from free P. falciparum merozoites in a two-step process [5]. First, exposure of extracellular merozoites to a low [K + ] environment typical of blood plasma leads to a rise in cytosolic Ca 2+ via a phospholipase C (PLC)-dependent pathway, which triggers translocation of microneme proteins such as 175 kD erythrocyte binding antigen (EBA175) and apical merozoite antigen-1 (PfAMA1) to the merozoite surface [5]. Subsequently, binding of EBA175 and its homologs to their erythrocyte receptors triggers secretion of rhoptry proteins such as PfRH2b, Clag3.1 and PfTRAMP [5,6].
The pathways by which exposure of P. falciparum merozoites to a low K + environment triggers a rise in cytosolic Ca 2+ and microneme secretion are not understood. Here, we demonstrate that another ubiquitous second messenger, namely, 3'-5' cyclic adenosine monophosphate (cAMP), plays a central role in regulating cytosolic Ca 2+ levels and microneme secretion during merozoite invasion of red blood cells. We demonstrate that exposure of merozoites to a low K + environment as found in blood plasma activates the bicarbonate-sensitive cytoplasmic adenylyl cyclase b (PfACb) leading to a rise in cytosolic cAMP levels and activation of protein kinase A (PKA), which regulates microneme secretion. In mammalian cells, the cAMP responsive PKA, which regulates diverse cellular processes in response to a rise in cytosolic cAMP levels, is composed of two catalytic subunits and two regulatory subunits [7]. Unlike mammalian cells, P. falciparum has a single inhibitory regulatory subunit (PfPKAr) and a single catalytic subunit (PfPKAc) [8][9][10][11][12]. As the PfPKAr subunit is not predicted to dimerize, the holoenzyme is likely to be composed of a one-to-one ratio of PfPKAr:PfPKAc [12]. The PfPKAr subunit is predicted to have 2 cyclic nucleotide binding domains. When cAMP binds to one or both of these it provokes a conformational change that engenders the dissociation of the PfPKAr:PfPKAc complex and activation of the released PfPKAc subunit that phosphorylates its specific substrates [8][9][10][11][12]. Like Plasmodium, Toxoplasma also encodes cAMP-dependent PKA and its inhibition leads to a block in tachyzoite growth [13]. Increase in cytosolic cAMP levels that would activate Toxoplasma PKA also mediates the tachyzoite to bradyzoite developmental switch [14][15][16]. In addition to activating PKA, we demonstrate that cAMP activates the Epac pathway [17] in P. falciparum merozoites, which triggers a rise in cytosolic Ca 2+ leading to microneme release. cAMP thus plays a central role in regulating microneme secretion during red blood cell invasion by P. falciparum merozoites.

Results
cAMP production by adenylyl cyclase b (PfACb) in P. falciparum merozoites cAMP is produced in eukaryotic cells either by a transmembrane adenylyl cyclase (tmAC), which has a putative K + channel at the N-terminus, or by a cytoplasmic soluble adenylyl cyclase (sAC). Homologs of both the tmAC (PF3D7_1404600) and sAC (PF3D7_0802600) (www.plasmodb.org) are encoded in P. falciparum genome [18,19]. However, the gene encoding the P. berghei transmembrane AC (PbACa) has been knocked out with no deleterious effect on growth of blood stage parasites although sporozoites are impaired in hepatocyte infectivity [20]. Given that PbACa does not appear to have an essential role in blood stage growth we focused our attention on the soluble AC in P. falciparum (PfACb).
Mouse antiserum raised against a synthetic peptide derived from PfACb detects a protein of expected size (.250 kD) in schizont and merozoite lysates by Western blotting (S1 Figure). Detection of PfACb by immunofluorescence assay (IFA) demonstrates that it co-localizes in late stage schizonts and merozoites with the cytoplasmic protein PfNapL [21] (Fig. 1). PfACb is thus expressed in P. falciparum merozoites and is localized in the cytoplasm.
Exposure of P. falciparum merozoites to an ionic environment with low K + as found in blood plasma triggers a rise in cytosolic Ca 2+ and secretion of microneme proteins [5]. To determine if exposure of free merozoites to a low K + environment also leads to a rise in cytosolic cAMP, we measured cAMP levels in purified merozoites suspended in a buffer that mimics the intracellular ionic environment (IC buffer -140 mM KCl, 5 mM NaCl, 1 mM MgCl 2 , 5.6 mM glucose, 25 mM HEPES, pH 7.2) and following transfer to a buffer that mimics extracellular ionic environment with low K + levels (EC buffer -5 mM KCl, 140 mM NaCl, 1 mM MgCl 2 , 2 mM EGTA, 5.6 mM glucose, 25 mM HEPES, pH 7.2). cAMP levels were also measured before and after merozoites were transferred from IC to IC-K low buffer (IC-K low buffer -5 mM NaCl, 5 mM KCl, 135 mM choline-Cl, 1 mM EGTA, 5.6 mM glucose, 25 mM HEPES, pH 7.2) to determine if change in environmental K + level alone can trigger changes in cAMP levels. There is a distinct rise in cAMP when merozoites are transferred from IC to EC buffer, or from IC to IC-K low buffer (Fig. 1B). Importantly, transfer of free merozoites to a low K + buffer in the presence of KH7, which inhibits mammalian sAC [19], blocks the rise in cytosolic cAMP (Fig. 1C).

Regulation of microneme secretion by cAMP
We monitored the secretion of microneme protein PfAMA1 following the transfer of free P. falciparum merozoites from IC to EC buffer with or without prior treatment with KH7. PfAMA1 is proteolytically cleaved into 48 kD and 44 kD fragments that are released into the supernatant following translocation to the merozoite surface [22]. PfAMA1 was detected in merozoite supernatants following transfer from IC to EC buffer by Western blotting using anti-PfAMA1 sera (Fig. 1D). Transfer of extracellular merozoites from IC to EC buffer in the presence of KH7 blocks the secretion of PfAMA1 (Fig. 1D). Treatment of merozoites with KH7 in the presence of 3-isobutyl-1-methylxanthine (IBMX), which inhibits 3',5'-cAMP phosphodiesterase (PDE) and maintains high cAMP levels in cells [23], reverses the block in microneme secretion (Fig. 1D). Secretion of another microneme protein, EBA175, is also inhibited if merozoites are treated with KH7 prior to transfer from IC to EC buffer (S2 Figure).
cAMP production in P. falciparum merozoites upon exposure to a low K + environment Production of cAMP by soluble adenylyl cyclases is regulated by HCO 3 2 ions in diverse organisms [24][25][26]. Incubation of P. falciparum merozoite lysates with ATP and different concentrations of HCO 3 2 ions showed that increasing amounts of cAMP are produced in presence of increasing concentrations of HCO 3 2 ions suggesting the presence of a HCO 3 2 sensitive adenylyl cyclase activity ( Fig. 2A). Moreover, cAMP production is inhibited by KH7 ( Fig. 2A). These observations confirm the presence of a HCO 3 2 -sensitive ACb in free P. falciparum merozoites. Cytosolic pH (pH i ) is controlled in eukaryotic cells by carbonic anhydrase (CA), which catalyzes the reversible hydration of CO 2 to produce H + and HCO 3 2 (CO 2 +H 2 O«HCO 3 2 +H + ) [27]. Acetazolamide (ACTZ), a specific inhibitor of CA, can disrupt intracellular pH homeostasis [28,29]. We tested whether the transfer of free merozoites from IC to EC buffer in presence of ACTZ affects pH i . P. falciparum merozoites collected in IC buffer were loaded with the membrane permeable pH-sensitive fluorescent dye BCECF-AM [30] and transferred to EC buffer in the presence or absence of ACTZ. pH i does not change when free merozoites are transferred from IC to EC buffer. In contrast, there

Author Summary
The blood stage of malaria parasites is responsible for all the morbidity and mortality associated with malaria. During the blood stage, malaria parasites invade and multiply within host erythrocytes. The process of erythrocyte invasion requires specific interactions between host receptors and parasite ligands. Many of the key parasite proteins that bind host receptors are localized in apical organelles called micronemes. Here, we demonstrate that cAMP serves as a key regulator that controls the timely secretion of microneme proteins during invasion. We show that exposure of merozoites to a low K + environment, as found in blood plasma, leads to a rise in cytosolic cAMP levels due to activation of the cytoplasmic, bicarbonatesensitive adenylyl cyclase b (PfACb). A rise in cAMP activates protein kinase A (PKA), which regulates microneme secretion. In addition, cAMP triggers a rise in cytosolic Ca 2+ levels through the Epac pathway. Increases in both cAMP and Ca 2+ levels are essential for triggering microneme secretion. Identification of the different elements in the cAMP-dependent signaling pathways that regulate microneme secretion during invasion provides novel targets to block erythrocyte invasion, inhibit blood stage parasite growth and prevent malaria.
cAMP-Mediated Regulation of Microneme Exocytosis Fig. 1. Production of cAMP in P. falciparum merozoites by adenylyl cyclase b (PfACb) and microneme exocytosis. A) Expression of PfACb in P. falciparum blood stages. Immunofluorescence assays (IFA) were used to detect PfACb (green) in P. falciparum rings (i), trophozoites (ii), schizonts (iii) and merozoites (iv) using mouse antisera against a peptide derived from PfACb. Nuclear DNA was stained with DAPI (blue). Rabbit antiserum against P. falciparum cytoplasmic protein PfNAPL (red) was used for co-localization. Yellow indicates overlap of PfACb and PfNAPL. B) Exposure of P. falciparum merozoites to low K + triggers production of cAMP. P. falciparum merozoites were isolated in buffer mimicking intracellular ionic cAMP-Mediated Regulation of Microneme Exocytosis PLOS Pathogens | www.plospathogens.org is a rapid rise in pH i when free merozoites are transferred from IC to EC buffer in presence of ACTZ (Fig. 2B). These observations indicate that PfCA activity is necessary for maintenance of pH i when merozoites are transferred to a low K + environment.
The observation that pH i rises when merozoites are transferred to EC in presence of ACTZ is consistent with a model in which PfCA maintains pH i by producing H + and HCO 3 2 . To test whether the generation of HCO 3 2 in free merozoites following transfer from IC to EC buffer leads to activation of PfACb, we measured levels of cAMP in merozoites with or without treatment with ACTZ. Treatment of merozoites with ACTZ prior to transfer from IC to EC buffer inhibits rise in cAMP (Fig. 2C) as well as secretion of PfAMA1 (Fig. 2D). Treatment of P. falciparum merozoites with ACTZ in presence of IBMX reverses the block in microneme secretion (Fig. 2D). These results suggest that when free merozoites are exposed to an extracellular-like low K + environment, the production of HCO 3 2 by PfCA to maintain pH i leads to activation of PfACb and generation of cAMP, which triggers microneme secretion. Exposure of merozoites in IC buffer to increasing concentrations of HCO 3 2 also leads to a dosedependent increase in microneme secretion (S3 Figure) confirming that HCO 3 2 plays a key role in regulating microneme secretion.
Potential role of cAMP-responsive protein kinase A (PKA) in regulation of microneme secretion PKA commonly plays a role as a cAMP-responsive effector in signaling pathways [7]. To investigate the potential role of PfPKA in microneme discharge, we measured phosphorylation of kemptide, a specific substrate of PKA, by lysates of P. falciparum merozoites in IC and EC buffers with and without addition of KH7. There is a distinct increase in kemptide phosphorylation with lysates made from merozoites in EC buffer compared to IC buffer (Fig. 3A). Pre-treatment with KH7 inhibits this increase in kemptide phosphorylation by merozoite lysates in EC buffer suggesting that a cAMP-responsive kinase is activated when merozoites are transferred to EC buffer (Fig. 3A).
We used a molecular genetic approach to explore the regulatory role of PfPKA in microneme secretion. The P. falciparum transgenic line PHL dhfr-PfPKAr over-expresses the regulatory subunit (PfPKAr) (11, S4 Figure). Overexpression of PfPKAr results in reduced PKA activity and a parasite growth defect, which is restored by addition of the non-hydrolyzable cAMP analog dibutryl-cAMP (DiB) [11]. In contrast to wild type P. falciparum merozoites (Fig. 3A), lysates of P. falciparum PHL dhfr-PfPKAr merozoites in EC buffer do not phosphorylate kemptide at higher levels compared to lysates made in IC buffer ( Fig. 3B). Treatment of P. falciparum PHL dhfr-PfPKAr merozoite lysates with DiB increases kemptide phosphorylation (Fig. 3B). These observations indicate that kemptide phosphorylation is primarily due to PfPKA and PfPKA activity increases when merozoites are transferred from IC to EC buffer.
Next, we followed secretion of PfAMA1 upon transfer of P. falciparum PHL dhfr-PfPKAr merozoites from IC to EC buffer with and without DiB treatment (Fig. 3C). PfAMA1 is not secreted when P. falciparum PHL dhfr-PfPKAr merozoites are transferred from IC to EC buffer (Fig. 3C). However, PfAMA-1 secretion is restored when P. falciparum PHL dhfr-PfPKAr merozoites are treated with DiB in EC buffer presumably due to the activation of PfPKA (Fig. 3C). These observations suggest that PfPKA plays a role in regulation of microneme secretion. Impairment of microneme secretion due to PfPKAr overexpression may be one of the reasons for poor growth of P. falciparum PHL dhfr-PfPKAr parasites.
Crosstalk between cAMP and Ca 2+ signaling in P. falciparum merozoites and regulation of microneme secretion Changes in the levels of cytosolic Ca 2+ and cAMP control a variety of functions in diverse cell types [31]. We examined the crosstalk between these two ubiquitous second messengers in extracellular P. falciparum merozoites. We found that treatment of merozoites with the Ca 2+ chelator, BAPTA-AM, or PLC inhibitor, U73122, does not affect the increase in cAMP levels when merozoites are transferred from IC to EC buffer (Fig. 4A). In contrast, treatment of merozoites with the ACb inhibitor KH7 inhibits the increase in cytosolic Ca 2+ levels when merozoites are transferred from IC to EC buffer ( Fig. 4B) indicating that cAMP plays a key role in regulating cytosolic Ca 2+ levels in P. falciparum merozoites.
Next, we found that Ca 2+ levels increase in transgenic P. falciparum PHL dhfr-PKAr merozoites following their transfer from IC to EC buffer (Fig. 4B). Moreover, treatment of transgenic merozoites with DiB in IC buffer, which activates PfPKA, does not lead to further increase in cytosolic Ca 2+ levels (Fig. 4C). These observations indicate that PfPKA does not play a role in generation of free cytosolic Ca 2+ in P. falciparum merozoites following transfer from IC to EC buffer.
Alternative cAMP-responsive effectors, Epac 1 and Epac 2, have been shown to activate phospholipase C (PLC) in presence of cAMP leading to rise in cytosolic Ca 2+ in mammalian cells [17,32,33]. P. falciparum codes for a single putative parasite PfEpac (PF3D7_1417400). Treatment of merozoites in IC buffer environment (IC buffer -140 mM KCl, 5 mM NaCl, 1 mM MgCl 2 , 5.6 mM glucose, 25 mM HEPES, pH 7.2) and transferred to buffer mimicking extracellular ionic environment (EC buffer -5 mM KCl, 140 mM NaCl, 1 mM MgCl 2 , 2 mM EGTA, 5.6 mM glucose, 25 mM HEPES, pH 7.2) or IC-K low buffer (IC-K low buffer -5 mM NaCl, 5 mM KCl, 135 mM choline-Cl, 1 mM EGTA, 5.6 mM glucose, 25 mM HEPES, pH 7.2). Levels of cytosolic cAMP in merozoite lysates were measured using a colorimetric cAMP Direct Immunoassay Kit (Calbiochem) as described in Materials & Methods. Total protein content in each merozoite sample was determined using Pierce Protein Assay Kit (Pierce). Amount of cAMP per mg of protein was determined for each merozoite sample and used to calculate fold increase compared to cAMP per mg of protein in control merozoites in IC buffer. Graphs represent mean fold change (6 SD) in cAMP levels per mg of protein in merozoites under different conditions with respect to cAMP per mg of protein in merozoites in IC buffer from three independent experiments. C) Mammalian soluble AC inhibitor KH7 blocks rise in cytosolic cAMP levels following transfer of P. falciparum merozoites to low K + buffer. P. falciparum merozoites were isolated in IC buffer and transferred to EC buffer with or without treatment with KH7. Levels of cAMP were measured in merozoite lysates as described above. Graphs represent mean fold change (6 SD) in cAMP levels per mg of protein in merozoites under different conditions with respect to cAMP per mg of protein in merozoites in IC buffer from three independent experiments. Treatment with KH7 inhibits increase in cytosolic cAMP levels when merozoites are transferred from IC to EC buffer. D) Inhibition of microneme secretion by KH7. P. falciparum merozoites were transferred from IC to EC buffer with or without treatment with KH7. Secretion of PfAMA1 in merozoite supernatants (AMA(s)) was detected by Western blotting. Presence of cytoplasmic protein PfNapL was detected in P. falciparum merozoite supernatants (NapL(s)) and lysates of merozoite pellets (NapL(p)) under different conditions by Western blotting to control for merozoite lysis and number of merozoites used in the different conditions, respectively. Treatment of merozoites with KH7 inhibits secretion of microneme protein PfAMA1 following transfer from IC to EC buffer. Treatment of merozoites with KH7+IBMX, KH7+DiB+Epac Agonist and KH7+DiB+ A23187 restores microneme secretion. However, treatment with KH7+DiB does not restore microneme secretion. doi:10.1371/journal.ppat.1004520.g001 cAMP-Mediated Regulation of Microneme Exocytosis PLOS Pathogens | www.plospathogens.org with the Epac agonist, 8-pCPT-2'-O-Me-cAMP [34], leads to a rise in cytosolic Ca 2+ levels (Fig. 4D). This increase in Ca 2+ is blocked if merozoites are treated with ESI-09, which inhibits both mammalian Epac 1 and Epac 2 [35] (Fig. 4D). ESI-05, which only inhibits mammalian Epac 2 [36], does not block increase in cytosolic Ca 2+ following treatment with the Epac agonist (Fig. 4D).
In mammalian cells, Epac activated by cAMP catalyzes the transfer of GTP to Rap1 [17]. Rap1-GTP activates PLC and triggers a rise in cytosolic Ca 2+ through the PLC pathway [17]. Stimulation of merozoites with the Epac agonist in presence of GGTI298 [37,38], which disrupts Rap1 activity, or the PLC inhibitor, U73122 [5,39], inhibits rise in cytosolic Ca 2+ (Fig. 4E). 3 2 sensitive PfACb and regulation of microneme secretion. A) HCO 2 3 sensitive adenylyl cyclase activity in P. falciparum merozoite lysates. P. falciparum merozoite lysates were incubated with increasing concentrations of NaHCO 3 (10 mM, 20 mM and 40 mM) and ATP for 30 min at 30uC and production of cAMP was measured. Levels of cAMP are reported (mean + SD from three independent experiments) as femtomoles per mg of merozoite protein. Merozoite lysates without exogenous ATP were used as negative controls. Production of cAMP increased with increasing concentrations of NaHCO 3 . Addition of KH7 in presence of 40 mM NaHCO 3 inhibited cAMP production. B) Intracellular pH (pH i ) of P. falciparum merozoites in different ionic environments with and without treatment with carbonic anhydrase (CA) inhibitor acetazolamide (ACTZ). P. falciparum merozoites loaded with the pH-sensitive fluorescent dye BCECF-AM (inset) were transferred from IC to EC buffer with or without prior treatment with ACTZ. Fluorescence signal from BCECF was measured and used to determine pH i using a standard curve (S6 Figure) as described in Materials and Methods. The pH i of P. falciparum merozoites remains unchanged following transfer from IC to EC buffer. Pretreatment of P. falciparum merozoites with ACTZ results in a rise in pH i following transfer from IC to EC buffer (EC+ACTZ). C) ACTZ blocks rise in intracellular cAMP following transfer of P. falciparum merozoites from IC to EC buffer. P. falciparum merozoites were transferred from IC to EC buffer with or without treatment with ACTZ. Fold change (mean 6 SD) in cAMP levels is reported from 3 independent experiments. Pre-treatment with ACTZ inhibits rise in cAMP in merozoites following transfer from IC to EC buffer. D) ACTZ blocks microneme secretion following transfer of P. falciparum merozoites from IC to EC buffer. P. falciparum merozoites were transferred form IC to EC buffer with or without prior treatment with CA inhibitor, ACTZ. Secretion of PfAMA1 in merozoite supernatants (AMA1(s)) was detected by Western blotting. Presence of cytoplasmic protein PfNapL was detected in P. falciparum merozoite supernatants (NapL(s)) and lysates of merozoite pellets (NapL(p)) under different conditions by Western blotting to control for merozoite lysis and number of merozoites used in the different conditions, respectively. Treatment of merozoites with ACTZ prior to transfer from IC to EC buffer inhibits secretion of microneme protein PfAMA1. Treatment of merozoites with ACTZ+IBMX, ACTZ+DiB+Epac Agonist and ACTZ+DiB+A23187 restores microneme secretion. However, treatment with ACTZ+DiB does not restore microneme secretion. doi:10.1371/journal.ppat.1004520.g002 cAMP-Mediated Regulation of Microneme Exocytosis PLOS Pathogens | www.plospathogens.org Moreover, treatment of merozoites with either ESI-09 or GGTI298 prior to transfer from IC to EC buffer also blocks rise in cytosolic Ca 2+ (Fig. 4F) and inhibits PfAMA1 secretion (Fig. 5A,  5B). In contrast, ESI-05, which does not inhibit increase in cytosolic Ca 2+ levels (Fig. 4F), does not block secretion of PfAMA1 (Fig. 5A). Treatment of merozoites with GGT1298 together with the Ca 2+ ionophore A23187 restores microneme secretion (Fig. 5B). Thus, a cAMP-responsive Epac-Rap1 pathway is likely to be involved in regulating Ca 2+ levels in P. falciparum merozoites.

Fig. 2. Production of cAMP by HCO
To confirm the role of cAMP-responsive effectors PfPKA and PfEpac in microneme secretion, we treated merozoites in IC buffer with the Epac agonist, the Ca 2+ ionophore A23187 and DiB, either individually or in combination. Treatment of merozoites in IC buffer with Epac agonist, A23187 or DiB individually does not trigger PfAMA1 secretion (Fig. 5C). However, treatment of merozoites in IC buffer with A23187 and DiB, or Epac-agonist and DiB triggers PfAMA1 secretion (Fig. 5C). The PLC inhibitor U73122 blocks PfAMA1 secretion induced by treatment of merozoites in IC buffer with Epac-agonist and DiB (Fig. 5D). However, the inactive analog, U73343 has no inhibitory effect. These observations suggest that Epac-agonist triggers rise in Ca 2+ through the PLC pathway. Moreover, both Ca 2+ and cAMP surges are required to trigger the secretion of microneme proteins in free merozoites.

Blockade of cAMP-& Ca 2+ -signaling and inhibition of erythrocyte invasion by P. falciparum merozoites
Erythrocyte invasion assays were performed with purified extracellular merozoites in the presence of inhibitors of cAMPand Ca 2+ -mediated signaling pathways. Merozoites were treated with ACb inhibitor KH7, the CA inhibitor ACTZ, or Epac inhibitors ESI-09 and ESI-05, and allowed to invade human erythrocytes. Successful invasion events were scored by flow cytometry using DNA intercalating dye ethidium bromide to identify infected erythrocytes. KH7, ACTZ and ESI-09 inhibit invasion (Fig. 5E) adding to the evidence that both cAMP and Ca 2+ -dependent signaling pathways play critical roles in the process of red cell invasion by P. falciparum merozoites.

Discussion
We have previously demonstrated that exposure of merozoites to a low K + environment, as found in blood plasma, triggers a rise in cytosolic Ca 2+ through a PLC mediated pathway leading to microneme secretion [5]. Although it is clear that a rise in cytosolic Ca 2+ plays a key role in regulating microneme release, the signal transduction mechanisms by which exposure of extra-erythrocytic merozoites to a low K + environment leads to a rise in cytosolic Ca 2+ were not known.
Here, we demonstrate the role of cAMP in regulating levels of cytosolic Ca 2+ and microneme secretion in response to exposure of merozoites to a low K + environment. The transmembrane channel linked expressed in Paramecium, which contains a putative K + channel-like transmembrane domain fused to the adenylyl cyclase catalytic domain, produces cAMP when the organism is exposed to a low K + environment [18]. A homolog of tmAC, referred to as ACa is expressed in blood stage Plasmodium parasites. However, deletion of the gene encoding P. berghei ACa does not have any deleterious effect on growth of blood stage parasites [20]. In contrast, deletion of the gene encoding P. berghei ACa impairs hepatocyte invasion by P. berghei sporozoites [20]. These observations suggest that the soluble adenylyl cyclase, ACb may play a more significant role in regulation of cAMP levels in Plasmodium merozoites.
Here, we have used a series of pharmacological inhibitors to explore the role of PfACb as well as other players involved in signaling pathways that control microneme secretion. Based on the observations made here, we propose the following model for the regulation of microneme secretion in response to environmental signals (Fig. 6). We propose that when merozoites are exposed to a low K + environment, PfCA produces HCO 3 2 and H + ions to maintain pH i homeostasis (Fig. 6). Given the lack of information about the nature of ion channels present on the surface of P. falciparum merozoites, it is difficult to develop a molecular model to explain the activation of PfCA for maintenance of pH i leading to production of HCO 3 2 when merozoites are exposed to a low K + environment. Production of HCO 3 2 by PfCA activates PfACb leading to a surge in cAMP levels in free merozoites (Fig. 6). Treatment of merozoites with CA inhibitor, ACTZ, or PfACb inhibitor, KH7, blocks production of cAMP and inhibits microneme secretion. cAMP thus serves as a key second messenger that regulates microneme secretion in free merozoites during red cell invasion (Fig. 6).
Next, we demonstrated that production of cAMP in merozoites leads to activation of the cAMP-responsive kinase, PfPKA. We confirmed using a genetic approach that PfPKA plays a role in microneme secretion. Over-expression of the regulatory subunit, PfPKAr, inhibits PfPKA activity and blocks microneme secretion implicating PfPKA in regulation of microneme secretion. PfPKA has been shown to phosphorylate the cytoplasmic domain of PfAMA1 during invasion [40], although the precise functional role in the invasion process is not yet known. A global study of changes in protein phosphorylation in merozoites following their transfer to a low K + environment is needed to identify the signaling pathways involved in microneme secretion. Determination of changes in protein phosphorylation in the presence and absence of PfPKA inhibition will identify substrates of PfPKA that may be involved in microneme secretion and invasion.
Given the essential roles that both Ca 2+ and cAMP play in regulating microneme secretion in free merozoites, we examined the crosstalk between these key second messengers. Inhibiting rise in cytosolic Ca 2+ by treatment of merozoites with BAPTA-AM or PLC inhibitor U73122 does not affect rise in cAMP when merozoites are exposed to a low K + environment. In contrast, blocking a rise in cAMP with KH7 does inhibit increase in cytosolic Ca 2+ levels. Thus, a rise in cAMP levels precedes and is necessary for an increase in cytosolic Ca 2+ when merozoites are exposed to a low K + environment.
We explored the mechanism by which cAMP regulates cytosolic Ca 2+ levels in merozoites and found that PfEpac, rather than PfPKA, is responsible. We present pharmacological evidence to indicate that merozoites possess functional homologs of Epac and Rap1 that activate PLC to trigger a rise in cytosolic Ca 2+ in response to increase in cAMP levels. A potential PfEpac is encoded by PF3D7_1417400 (www.plasmodb.org). A search for Rap1 homologs identifies a number of P. falciparum genes predicted to encode small GTPases. It remains to be seen which one these is the functional homolog of Rap1 that activates PLC in response to rise in cAMP in P. falciparum merozoites.
Previous studies with mature schizonts demonstrated that activation of a cGMP-dependent protein kinase, PfPKG, triggers release of proteins from micronemes and exonemes in merozoites leading to egress [41]. Secretion of exoneme proteins such as the protease PfSUB1 plays a critical role in merozoite egress from infected erythrocytes [42,43]. The external, or internal signals that trigger a rise in cGMP and activate PfPKG leading to egress are cAMP-Mediated Regulation of Microneme Exocytosis not known. A sharp rise in cytosolic Ca 2+ in intra-erythrocytic merozoites is also observed just prior to release of microneme and exoneme proteins and parasite egress [5,43,44]. The pathways for crosstalk between cGMP and Ca 2+ in P. falciparum merozoites during egress have not been defined and whether cGMPdependent PfPKG plays a role in microneme secretion in free merozoites after egress is not known. It will be important to understand the crosstalk between cGMP, cAMP and Ca 2+ in merozoites leading to apical organelle exocytosis both before and after egress.
One possible mechanism for such crosstalk may be through phosphorylation of the PfPKAr subunit by PfPKG that would relieve inhibition of the catalytic PfPKAc subunit. PfPKAr displays characteristics of both mammalian RI and RII subunits [12]. Mammalian RI and RII subunits have distinct inhibitor sites. PfPKAr has a serine in the P-site typical of RII subunits and is phosphorylated by PfPKA in vitro [12]. However, PfPKAr also has a serine in the P+2 site typical of RI subunits. In mammalian cells this site is phosphorylated by PKG [45]. Like the P-site, the P+2 site in PfPKAr is also phosphorylated in vivo in P. falciparum Kemptide phosphorylation by P. falciparum 3D7 merozoite lysates under different ionic environments. P. falciparum 3D7 merozoites were transferred from IC buffer to EC buffer with or without treatment with ACb inhibitor, KH7. Merozoite lysates made under different conditions were incubated with kemptide, a protein kinase A (PKA) substrate, and c-32 P-ATP. Fold change (mean 6 SD) in phosphorylation of kemptide under different ionic conditions is reported relative to phosphorylation in IC buffer from 3 independent experiments. B) Kemptide phosphorylation by lysates of transgenic P. falciparum PHL dhfr-PfPKAr merozoites under different ionic environments. P. falciparum PHL dhfr-PfPKAr merozoites were isolated in IC buffer and transferred to EC buffer with or without treatment with dibutryl-cAMP (DiB). Phosphorylation of kemptide by merozoite lysates made under different conditions was measured in presence of c-32 P-ATP. Fold change in kemptide phosphorylation (mean 6 SD) by merozoite lysates under different conditions is reported relative to that in IC buffer from 3 independent experiments. C) Microneme secretion in P. falciparum PHL dhfr-PfPKAr merozoites. P. falciparum PHL dhfr-PfPKAr merozoites were transferred from IC to EC buffer, IC buffer+DiB, or EC buffer+DiB and secretion of PfAMA1 into merozoite supernatants (AMA1(s)) was detected by Western blotting. Presence of cytoplasmic protein PfNapL was detected in P. falciparum merozoite supernatants (NapL(s)) and lysates of merozoite pellets (NapL(p)) under different conditions by Western blotting to control for merozoite lysis and number of merozoites used in the different conditions, respectively. PfAMA1 was not secreted when P. falciparum PHL dhfr-PfPKAr merozoites were transferred from IC to IC+DiB, or from IC to EC buffer. PfAMA1 was secreted when P. falciparum PHL dhfr-PfPKAr merozoites were transferred from IC to EC+DiB. doi:10.1371/journal.ppat.1004520.g003 cAMP-Mediated Regulation of Microneme Exocytosis Fig. 4. Crosstalk between cAMP and Ca 2+ in P. falciparum merozoites. A) Ca 2+ does not regulate cytosolic cAMP levels. P. falciparum merozoites were transferred from IC to EC buffer with or without treatment with BAPTA-AM or U73122. Levels of cytosolic cAMP were measured in merozoite lysates before and after transfer to EC buffer. Fold changes in cAMP levels per mg of merozoite protein (mean 6 SD from 3 independent experiments) in different conditions relative to cAMP levels in IC buffer (mean 6 SD) from 3 independent experiments are reported. Treatment of merozoites with BAPTA-AM or U73122 does not have any effect on rise in intracellular cAMP levels following transfer from IC to EC buffer. B) Rise in cytosolic Ca 2+ is inhibited by ACb inhibitor KH7. P. falciparum merozoites were loaded with Fluo-4AM and transferred from IC to EC buffer with or without treatment with KH7. Cytosolic Ca 2+ levels in P. falciparum merozoites were measured before and after transfer by flow cytometry. Treatment with KH7 inhibits the rise in cytosolic Ca 2+ following transfer to EC buffer. C) PKA does not regulate cytosolic Ca 2+ . P. falciparum PHL dhfr-PfPKAr merozoites were loaded with Fluo-4AM and transferred from IC to EC buffer or from IC to IC+DiB or from IC to EC+DiB. Cytosolic Ca 2+ levels rise normally following transfer of P. falciparum merozoites from IC to EC buffer. Cytosolic Ca 2+ levels do not rise when P. falciparum PHL dhfr-PfPKAr merozoites in IC buffer are treated with DiB indicating that PKA does not play a role in regulating Ca 2+ levels in merozoites. D) Regulation of cytosolic Ca 2+ levels in P. falciparum 3D7 merozoites by Epac. P falciparum merozoites loaded with Fluo-4AM were transferred from IC to EC buffer or from IC to IC buffer containing Epac agonist 8-Pcpt-2'-O-Me-cAMP (IC+Epac agonist), DiB (IC+DiB), or Epac agonist and Epac inhibitors (IC+Epac agonist+ESI-09 or IC+Epac agonist+ESI-05). Cytosolic Ca 2+ levels rise when merozoites are transferred from IC to EC buffer, or IC buffer to IC+Epac agonist, but not when they are transferred from IC buffer to IC+DiB. EPAC1 antagonist ESI-09 inhibits rise in Ca 2+ stimulated by Epac agonist. E) Regulation of cytosolic cAMP-Mediated Regulation of Microneme Exocytosis PLOS Pathogens | www.plospathogens.org schizonts [46]. Moreover, PfPKG can phosphorylate PfPKAr in vitro (Baker D. and Langsley G., unpublished data). Thus, PfPKG phosphorylation of PfPKAr could provide a mechanism for crosstalk between cGMP-and cAMP-mediated signaling.
A recent study describes a different mechanism for elevation of cytosolic Ca 2+ in merozoites leading to microneme release [47]. This study suggests that interaction of the rhoptry neck protein PfRH1 with its erythrocyte receptor can trigger a surge in cytosolic Ca 2+ levels in merozoites leading to release of microneme proteins such as EBA175 [47]. It remains to be seen if cAMP levels also rise and play a role in regulating Ca 2+ levels following receptorengagement by PfRH1. Moreover, the observation that PfRH1 knock-out parasites invade erythrocytes efficiently and display normal blood stage growth [48] suggests that P. falciparum merozoites must have alternative signaling mechanisms to trigger microneme release. Whether other members of the PfRH family such as PfRH2a/b can also mediate rise in cytosolic Ca 2+ and microneme release remains to be determined. Alternatively, as described here, exposure to low K + as found in blood plasma may provide such an alternative mechanism to trigger rise in cytosolic Ca 2+ and microneme release.
Our study demonstrates that cAMP plays a central role in regulating microneme secretion during invasion (Fig. 6). Our data provides evidence to suggest that cAMP not only regulates microneme secretion by activating PfPKA, but it also regulates cytosolic Ca 2+ levels by activation of PfEpac. Our definition of the signal transduction pathways leading to rise in cAMP and Ca 2+ levels in response to external signals and the identification of cAMP and Ca 2+ responsive effectors provides multiple points for therapeutic intervention. For example, inhibition of PfCA, PfACb and PfEpac blocks microneme secretion and red blood cell invasion by P. falciparum merozoites. Small molecule inhibitors that specifically target these key-signaling molecules may provide valuable leads for the development of novel antimalarial drugs that block blood stage growth of malaria parasites.

Generation of PfACb specific antisera
A peptide (1916-1930 aa) derived from PfACb (PlasmoDB ID PF3D7_0802600) was synthesized, conjugated to keyhole limpet hemocyanin (KLH) and used to immunize mice to obtain anti-PfACb specific antibodies.

Generation of PfPKAr and PfPKAc specific antisera
The coding sequence of PfPKAr was amplified by PCR from P. falciparum cDNA with primers containing BamHI and XhoI restriction sites. The fragment was inserted into pCR2.1-TOPO and verified by sequencing. After digestion by XhoI and BamHI the insert was cloned into pGEX6P-1 (GE Healthcare) and transformed into E. coli BL21 that were grown in 50 mL of LB medium containing ampicillin (100 mg/mL, Euromedex). Recombinant protein expression was induced at 30uC for 4 h by addition of 0.2 mM IPTG (Euromedex). Cells were collected by centrifugation and lysed by sonication in lysis buffer (16 PBS-1 mM salt, 1% Triton, EDTA, protease inhibitors). The lysate was centrifuged (15000 rpm, 4uC, 60 min) and the soluble fraction incubated with glutathione sepharose beads (GE Healthcare) with stirring at 4uC for 90 min. Beads were washed with buffer (16 PBS, 1% Triton, 1 mM EDTA), the GST-tagged recombinant protein was eluted using increasing concentrations of glutathione. Purified recombinant antigen was used to immunize two rats to obtain anti-PfPKAr specific antibodies.
PfPKAc was cloned in the same way as PfPKAr. To increase stability of the recombinant catalytic subunit, we co-expressed PfPKAc-His with PfPKAr-GST in E. coli BL21. Transformed bacteria were grown in 50 mL of LB medium containing ampicillin (100 mg/mL, Euromedex). Recombinant protein expression was induced at 37uC for 4 h by addition of 0.5 mM IPTG (Euromedex). The bacterial pellet was washed once with 4 volumes of PBS and lysed using 4 volumes of Tris 20 mM at 4uC. After sonication, the lysate was centrifuged (6000 g, 4uC, 15 min) and the pellet resuspended in 3 volumes of inclusion body wash solution (2 M Urea, 500 mM NaCl, 2% NP40, 20 mM Tris-HCl, pH 8.0). The inclusion bodies were solubilised (5 mM Imidazole, 1 mM b-mercaptoethanol, 6 M Guanidine-HCl 6M, 20 mM Tris-HCl pH 8.0) and recombinant protein purified by metal affinity chromatography. His-tagged PfPKAc was eluted (8 M Urea to 0 M Urea, 20 mM imidazole to 500 mM imidazole, 500 mM NaCl, 1 mM b-mercaptoethanol, 20 mM Tris-Hcl pH 8.0) and used to immunize two rabbits to obtain anti-PfPKAc specific antibodies.
Isolation of P. falciparum merozoites P. falciparum blood stage cultures were synchronized with 5% sorbitol for at least two successive cycles. Synchronized P. falciparum cultures were used to isolate merozoites as described previously [5,43]. Briefly, when majority of infected erythrocytes reached mature schizont stage with prominent segmentation of merozoites, the culture was resuspended in IC buffer (IC buffer -140 mM KCl, 5 mM NaCl, 1 mM MgCl 2 , 5.6 mM glucose, 25 mM HEPES, pH 7.2) and schizonts were allowed to rupture to release merozoites over one hour at 37uC. Unruptured scizonts and uninfected erythrocytes were separated by centrifugation at 500 g for 5 min. The supernatant containing free merozoites was centrifuged at 3300 g for 5 min to collect merozoites. The merozoites were resuspended in IC buffer. The purity of merozoite Ca 2+ levels in P. falciparum 3D7 merozoites by Epac agonist, PLC inhibitor and Rap1 inhibitor. P falciparum merozoites loaded with Fluo-4AM were transferred from IC to EC buffer, or from IC to IC buffer containing Epac agonist (IC+Epac agonist), or IC buffer containing Epac agonist and PLC inhibitor (IC+Epac agonist+U73122), or IC buffer containing Epac agonist and Rap1 inhibitor GGTI298 (IC+Epac agonist+GGTI298). Cytosolic Ca 2+ levels rise when merozoites are transferred from IC to EC buffer, or from IC to IC+Epac agonist. PLC inhibitor U73122 and Rap1 inhibitor GGTI298 inhibit rise in cytosolic Ca 2+ stimulated by Epac agonist. F) Cytosolic Ca 2+ levels in P. falciparum 3D7 merozoites following transfer to EC buffer in presence of Epac and Rap1 inhibitors. P falciparum merozoites loaded with Fluo-4AM were transferred from IC to EC buffer containing Epac inhibitors (EC+ESI-09 or EC+ESI-05), or EC buffer containing Rap1 inhibitor (EC+GGTI298). Cytosolic Ca 2+ rises when merozoites are transferred from IC to EC buffer. Presence of Epac inhibitor ESI-09 and Rap1 inhibitor GGTI298 inhibits rise in cytosolic Ca 2+ . doi:10.1371/journal.ppat.1004520.g004 cAMP-Mediated Regulation of Microneme Exocytosis PLOS Pathogens | www.plospathogens.org Fig. 5. Role of Epac and PKA in regulation of microneme secretion and erythrocyte invasion by P. falciparum merozoites. P. falciparum merozoites were transferred from IC to EC buffer with or without treatment with various inhibitors. Secretion of PfAMA1 into merozoite supernatants (AMA1(s)) was detected by Western blotting. Cytoplasmic protein PfNapL was detected in P. falciparum merozoite supernatants (NapL(s)) and pellets (NapL(p)) by Western blotting under different conditions to control for merozoite lysis and number of merozoites used, respectively. A) Epac inhibitor ESI-09 blocks microneme secretion. Treatment of merozoites with 25 mM ESI-09 inhibits secretion of microneme protein PfAMA1 following transfer from IC to EC buffer. B) Rap1 inhibitor GGTI298 blocks microneme secretion. Treatment of merozoites with GGTI298 inhibits secretion of PfAMA1 following transfer from IC to EC buffer. Treatment of merozoites with GGTI298 in presence of A23187 does not block PfAMA1 secretion. C) Inhibition of microneme secretion by Epac and PKA pathway blockade. P. falciparum merozoites in IC buffer were treated with Ca 2+ ionophore A23187 (IC+ A23187), Epac agonist 8-Pcpt-2'-O-Me-cAMP (IC+Epac Ag) or dibutryl-cAMP (IC+DiB), or with various combinations (IC+A23187+DiB or IC+Epac Ag+ DiB). Treatment of merozoites in IC buffer with Epac Ag and DiB (IC+Epac Ag+DiB) or A23187 and DiB (IC+A23187+DiB) triggers microneme secretion. D) P. falciparum merozoites in IC buffer were treated with dibutryl-cAMP and Epac agonist (IC+Dib+Epac agonist) or DiB+Epac Ag+PLC inhibitor U73122 or its inactive analog U73343. Treatment with U73122 inhibits microneme secretion induced by treatment with Dib and Epac Ag. E) Effect of various inhibitors of cAMP and Ca 2+ signaling pathways on erythrocyte invasion by P. falciparum merozoites. P. falciparum 3D7 merozoites were isolated and allowed to invade erythrocytes in the presence of KH7 (50 mM), ACTZ (100 mM), ESI-09 (25 mM) and ESI-05 (25 mM). Newly invaded trophozoites were stained with ethidium bromide and scored by flow cytometry. Merozoites were allowed to invade erythrocytes in the absence of inhibitors as control. Percent invasion inhibition rates in presence of inhibitors are reported. Data represent mean (6 SD) from three independent experiments with SD. doi:10.1371/journal.ppat.1004520.g005 cAMP-Mediated Regulation of Microneme Exocytosis PLOS Pathogens | www.plospathogens.org preparations was confirmed by Giemsa staining (S5 Figure). Viability of merozoite preparations was estimated by labeling merozoites with dihydroethidine. Fluorescent merozoites were scored by flow cytometry and viability of merozoite preparations was estimated to be ,65%. Merozoites were separated by centrifugation as described above. Merozoite supernatants and merozoite pellets were separated by SDS-PAGE on a 12% gel under reducing conditions. Microneme proteins, PfAMA1 and EBA175, were detected in merozoite supernatants by Western blotting using anti-PfAMA1 and anti-EBA175 rabbit sera, respectively. Rabbit sera against cytoplasmic protein PfNapL were used to detect PfNapL in supernatants and merozoite pellets to estimate cell lysis and confirm that similar amounts of merozoites were used for each condition, respectively. Horseradish peroxidase (HRP)-conjugated anti-rabbit IgG goat antibodies (Sigma, USA) were used as secondary antibodies and detected by enhanced chemi-luminescence using ECL-Prime Western Blotting Detection reagent (GE Healthcare).

Measurement of intracellular cAMP levels
P. falciparum merozoites were lysed with 0.1 N HCl and clarified by centrifugation at 6000 g. Supernatants were used for measuring the cytosolic levels of cAMP using cAMP direct immunoassay kit (Calbiochem, USA) as per manufacturer's protocol. The kit uses a polyclonal antibody to bind cAMP present in the sample in a competitive manner. A standard curve was generated using known amounts of cAMP standards provided in the kit. Quantitative measurement of cAMP was obtained from the equation derived from the standard curve. The total protein content in each merozoite lysate sample was determined using Adenylyl cyclase activity assay P falciparum merozoites were isolated in RPMI 1640 as described, washed twice with phosphate buffered saline (PBS) and resuspended in 50 mM Tris pH 7.5, 150 mM NaCl. Merozoites were lysed by repeated freeze-thaw cycles and the lysate was clarified by centrifugation at 6000 g for 1 min at 4uC. The supernatant was incubated with different concentrations of NaHCO 3 (0 mM, 10 mM, 20 mM and 40 mM) in the presence of 10 mM MgCl 2 , 10 mM CaCl 2 , 5 mM ATP and 500 mM IBMX for 30 min at 30uC. The soluble adenylyl cyclase inhibitor KH7 (50 mM) was used as negative control. The reactions were stopped by the addition of 0.1 N HCl and the level of cAMP produced was measured using the cAMP direct immunoassay kit as described above.
Protein kinase A activity assay P. falciparum merozoites were isolated as described above and incubated with or without KH7 (50 mM) in IC buffer for 15 min prior to transfer to EC buffer. Merozoite pellets were collected by centrifugation at 3000 g and used immediately to measure cAMPdependent protein kinase A (PKA) activity. Merozoites were resuspended in ice-cold TNET buffer (50 mM Tris 7.4, 150 mM NaCl, 1 mM EDTA, and 1% Triton, protease inhibitor cocktail (Pierce, USA), 50 mM Na 3 VO 4 , 50 mM NaF) and lysed by repeated freeze-thaw cycles (4 cycles). The lysates were cleared by centrifugation at 6000 g for 15 min at 4uC. Total amount of protein in the supernatant was determined by (bicinchoninic acid assay (BCA) (Pierce, USA) using known amounts of bovine serum albumin (BSA) as standard. PKA activity was measured by quantitating incorporation of 32 P in kemptide, a known peptide substrate of PKA (L-R-R-A-S-L-G, Kemptide PKA peptide substrate, Promega, USA) as described previously [9].

Measurement of intracellular pH in P. falciparum merozoites
Intracellular pH (pH i ) of P. falciparum merozoites was measured using the pH-sensitive fluorescent dye BCECF (29,79bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein) [30,50]. P. falciparum merozoites were isolated in RPMI 1640 and loaded with BCECF-AM (acetoxymethylester of BCECF, Invitrogen, USA) for 20 min at 37uC in IC buffer. The merozoites were washed to remove excess dye and incubated with or without ACTZ in IC buffer at 37uC for 10 min. Merozoites were collected by centrifugation and resuspended in IC or EC buffer in the presence or absence of ACTZ and transferred to a 96 well microtiter plate (200 mL/well). Samples were excited monochromatically with l 1 = 440 nm and l 2 = 492 nm in succession for 0-10 minutes and emission was recorded at 535 nm on a fluorimeter (Perkin-Elmer Victor, USA). A quantitative measure of the pH i is provided by the ratio of the fluorescence measured at 535 nm following excitation at 492 nm and 440 nm. For each experiment, a pH calibration curve was prepared by resuspending merozoites in buffer containing 130 mm KCl, 25 mM HEPES, 20 mM glucose and 1 mM MgCl 2 with a pH range of 6.6 to 8.0 followed by the addition of 10 mm nigericin (Sigma). Nigericin equilibrates the intracellular and extracellular pH. The equation derived from the linear regression curve obtained by plotting the fluorescence ratio at different pH values (S6 Figure) was used to calculate pH i of merozoites in different conditions. Measurement of cytosolic Ca 2+ levels in P. falciparum merozoites P. falciparum merozoites were isolated as described above, loaded with 10 mM Fluo-4-AM for 20 min at 37uC, washed, resuspended in IC buffer and used for experiments within 5 min, as described earlier [5,51]. Fluo-4-AM loaded P. falciparum merozoites were treated with cAMP or Ca 2+ modulating agents and transferred from IC to EC buffer. Fluorescence signal from Fluo-4 in merozoites under different conditions was measured by flow cytometry using FACSCalibur (Becton Dickinson, USA) and analyzed using CellQuest software. Briefly, Fluo-4-AM loaded merozoites were excited at 488 nm and fluorescence signal was detected with a 430 nm/30 nm band pass filter for periods of 2-3 mins. Merozoites were gated on the basis of their forward scatter and side scatter. The mean fluorescence intensity (MFI), which reflects cytosolic Ca 2+ levels in merozoites, was plotted against time using FlowJo software.
Erythrocyte invasion assay by P. falciparum merozoites P. falciparum merozoites isolated as described above were either mock-treated or treated with 50 mm KH7 (Calbiochem, USA), 100 mm ACTZ (Sigma, USA), 25 mM ESI-05 (Biolog, Germany) or 25 mM ESI-09 (Biolog, Germany) for 15 min at 37uC, washed with IC buffer, resuspended in EC buffer and incubated with erythrocytes in EC buffer at 37uC under mixed gas environment for 2 h to allow invasion. EC buffer was then replaced with complete RPMI. After 18-20 h of incubation in complete RPMI under mixed gas environment to allow development of ring stages, the percentage of infected erythrocytes was scored by flow cytometry to determine invasion rates, as described earlier [52].
Supporting Information S1 Figure Detection of P. falciparjum adenylyl cyclase b (PfACb) in P. falciparum merozoite and schizont lysates by western blotting. P falciparum merozoite and schizont lysates were separated by SDS-PAGE and transferred to nitrocellulose. Presence of PfACb was detected by Western blotting using mouse sera raised against a peptide (1916)(1917)(1918)(1919)(1920)(1921)(1922)(1923)(1924)(1925)(1926)(1927)(1928)(1929)(1930) aa) derived from PfACb (PlasmoDB ID PF3D7_0802600) conjugated to keyhole limpet hemocyanin (KLH). Native PfACb was detected at ,270 kDa in P. falciparum merozoite and schizont lysates. Pre-immune mouse serum was used as negative control. Merozoite lysates were also probed for the cytoplasmic protein, P. falciparum nucleosome assembly protein-L (PfNapL) and P. falciparum merozoite surface protein MSP-1 (PfMSP1) using anti-PfNapL rabbit sera and anti-PfMSP1 19 rabbit sera respectively. (TIFF) S2 Figure Inhibition of EBA175 secretion with KH7. P. falciparum merozoites were transferred from IC to EC buffer with or without prior treatment with mammalian ACb inhibitor KH7. Presence of EBA175 was detected in the merozoite supernatant by Western blotting. Mouse antisera were used to detect cytoplasmic protein NAPL in merozoite pellets as loading control and in supernatants to control for cell lysis. EBA175 is secreted when merozoites are transferred from IC to EC buffer. Secretion is blocked by prior treatment of merozoites with KH7. (TIFF) S3 Figure Regulation of microneme secretion by treatment of P. falciparum merozoites with increasing concentrations of NaHCO 3 . P. falciparum merozoites in IC buffer were transferred to IC buffer containing increasing concentrations of NaHCO 3 (10 mM, 20 mM and 40 mM), EC buffer or EC buffer +40 mM NaHCO 3 for 15 min at 37uC. Secretion of PfAMA1 into merozoite supernatants (AMA1(s)) was detected by Western blotting. Cytoplasmic protein PfNapL was detected in P. falciparum merozoite supernatants (NapL(s)) and pellets (NapL(p)) by Western blotting under different conditions to control for merozoite lysis and number of merozoites used, respectively. Treatment of merozoites with NaHCO 3 triggers secretion of microneme protein PfAMA1 in IC buffer in a concentration dependent manner. (TIFF) S4 Figure Expression of PKA regulatory subunit (PKAr) in P. falciparum schizonts. A) Detection of PKAr in schizonts of P. falciparum lines PHL-dhfr-PKAr and P. falciparum PHLdhfr-luciferase by Western blotting. Lysates of P. falciparum PHLdhfr-luciferase and P. falciparum PHL-dhfr-PfPKAr schizonts were separated by SDS-PAGE, transferred to nitrocellulose and probed for presence of PfPKAr by Western blotting using rat anti-PfPKAr serum. Antisera raised against P. falciparum aldolase were used as loading control. The band intensity corresponding to PKAr was higher in PHL-PfPKAr schizont lysates in comparison to PHL-luciferase schizont lysates, whereas the reactivity of anti-Pfaldolase serum was similar confirming equal loading. The specificity of Western blot detection was confirmed by incubation of anti-PfPKAr serum with excess of recombinant PfPKAr prior to detection by Western blotting. B) Detection of PKAr in P. falciparum 3D7 merozoites by immunofluorescence assay (IFA). Mouse sera raised against PfPKAr and rabbit sera raised against PfMSP1 19 were used in IFA to detect expression and localize PfPKAr and PfMSP1 respectively. There is significant overlap between PfPKAr and PfMSP indicating that a significant portion of PfPKAr is associated with the merozoite plasma membrane. (TIFF) S5 Figure Viabilty and purity of P. falciparum merozoites. Viabilty of P. falciparum merozoites was analyzed by staining with dihydroethidine (DHE). P. falciparum merozoites were incubated with DHE (10 mg/ml) for 20 min at 37uC after purification. A) DHE-stained merozoites were further stained with nuclear staining dye DAPI and visualized using a confocal Nikon A1R microscope. B) DHE-stained merozoites were also analyzed by flow cytometry using a FACS Calibur (Becton & Dickinson, USA). 100,000 RBCs were scored per sample for fluorescence staining with DHT to determine percentage of merozoites that are viable. (TIFF) S6 Figure Standard curve for measurement of intracellular pH in P falciparum merozoites. P. falciparum merozoites were loaded with pH sensitive fluorescent dye BCECF-AM, resuspended in buffer at 10 different pH conditions (6.4, 6.6, 6.8, 7.0, 7.2, 7.4, 7.6, 7.8, 8.0) and treated with nigericin for 10 min at 37uC to allow extracellular and intracellular pH to equilibrate. Samples were excited at 440 nm and 492 nm and mean fluorescent intensity (MFI) was measured at 535 nm. Ratio of MFI measured at 535 nm following excitation at 492 nm and 440 nm was plotted against pH to generate a standard curve. (TIFF)