Notice: Wiley Online Library will be unavailable on Saturday 30th July 2016 from 08:00-11:00 BST / 03:00-06:00 EST / 15:00-18:00 SGT for essential maintenance. Apologies for the inconvenience.
Leishmania major aquaglyceroporin (LmjAQP1) adventitiously facilitates the uptake of antimonite [Sb(III)], an active form of Pentostam® or Glucantime®, which are the first line of defence against all forms of leishmaniasis. The present paper shows that LmjAQP1 activity is modulated by the mitogen-activated protein kinase, LmjMPK2. Leishmania parasites coexpressing LmjAQP1 and LmjMPK2 show increased Sb(III) uptake and increased Sb(III) sensitivity. When subjected to a hypo-osmotic stress, these cells show faster volume recovery than cells expressing LmjAQP1 alone. LmjAQP1 is phosphorylated in vivo at Thr-197 and this phosphorylation requires LmjMPK2 activity. Lys-42 of LmjMPK2 is critical for its kinase activity. Cells expressing altered T197A LmjAQP1 or K42A LmjMPK2 showed decreased Sb(III) influx and a slower volume recovery than cells expressing wild-type proteins. Phosphorylation of LmjAQP1 led to a decrease in its turnover rate affecting LmjAQP1 activity. Although LmjAQP1 is localized to the flagellum of promastigotes, upon phosphorylation, it is relocalized to the entire surface of the parasite. Leishmania mexicana promastigotes with an MPK2 deletion showed reduced Sb(III) uptake and slower volume recovery than wild-type cells. This is the first report where a parasite aquaglyceroporin activity is post-translationally modulated by a mitogen-activated protein kinase.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Leishmaniasis is a parasitic disease caused by protozoa of the genus Leishmania. The disease is spread to human and other mammals by the bite of the female phlebotomine sandfly. Leishmaniasis includes two major diseases, cutaneous leishmaniasis and visceral leishmaniasis, caused by more than 20 different Leishmania species. Approximately two million new cases are considered to occur annually, of which 1.5 million are categorized as cutaneous leishmaniasis and 500 000 as visceral leishmaniasis. Leishmania parasites have a digenetic life cycle. The promastigote form resides in the intestinal tract of the sandfly vector and has a slender, spindle-shaped body with a long, anterior flagellum. The amastigote forms of the parasite are small, spherical, aflagellated structures that reside in macrophages and other mononuclear phagocytes of the mammalian host. The first line of treatment against all forms of leishmaniasis are the pentavalent antimony [Sb(V)] containing drugs, sodium stibogluconate (Pentostam®) and meglumine antimonate (Glucantime®) (Frezard et al., 2009).
Leishmania major aquaglyceroporin (LmjAQP1) is adventitiously permeable to antimonite [Sb(III)], an activated form of Pentostam® or Glucantime® (Gourbal et al., 2004). Disruption of one of the two AQP1 alleles in L. major conferred a 10-fold increase in resistance to Sb(III) (Gourbal et al., 2004). AQP1 mRNA levels are significantly lower in either the Sb(III) or arsenite [As(III)] resistant L. major and Leishmania tarentolae cells, indicating that downregulation of AQP1 expression leads to drug resistance (Marquis et al., 2005). These findings have also been corroborated by studies on Leishmania donovani field isolates, and lower levels of AQP1 expression at the transcript level were noted in Sb(V)-resistant strains compared to sensitive strains (Decuypere et al., 2005; Mandal et al., 2010). Besides being an adventitious metalloid transporter, LmjAQP1 is also permeable to glycerol, glyceraldehyde, dihydroxyacetone and sugar alcohols as well as to water; its water conduction capacity is 65% of human AQP1, a classical water channel (Figarella et al., 2007). LmjAQP1 is localized exclusively to the flagellum of promastigotes, whereas in amastigotes, it is present in the flagellar pocket, rudimentary flagellum, and contractile vacuoles (Figarella et al., 2007). LmjAQP1 is involved in several physiologically relevant processes, such as, water and solute transport, volume regulation, and osmotaxis, which help the parasite to face osmotic challenges during its digenetic life cycle (Figarella et al., 2007). However, how LmjAQP1 activity is modulated during these processes is not known.
Thorsen et al. (2006) have shown that in the budding yeast Saccharomyces cerevisiae, the aquaglyceroporin Fps1p-dependent transport of As(III), Sb(III), and glycerol is modulated by the mitogen-activated protein (MAP) kinase Hog1p. MAP kinases are serine/threonine-specific protein kinases that are highly conserved in all eukaryotes. They respond to extracellular stimuli and regulate critical cellular activities, such as apoptosis, cell shape and motility, differentiation, gene expression, mitosis, proliferation, and stress response (Wiese, 2007; Brumlik et al., 2011). A yeast hog1Δ mutant is sensitive to hyperosmotic shock, partially because glycerol transport through Fps1p could not be controlled. The N-terminus of Fps1p is phosphorylated by Hog1p resulting in its inactivation. Thus an N-terminal deletion mutant of Fps1p is constitutively active and the strain becomes sensitive to hyperosmotic stress (Tamas et al., 2003; Thorsen et al., 2006). S. cerevisiae hog1Δ cells were sensitive to As(III) and Sb(III), while cells with elevated Hog1p showed improved metalloid tolerance (Thorsen et al., 2006). Under metalloid stress, Hog1p kinase contributed to tolerance acquisition by phosphorylation of Fps1p on Thr231 critically affecting Fps1p activity (Thorsen et al., 2006). Hence yeast Fps1p is negatively regulated by Hog1p.
Is Leishmania AQP1 activity similarly modulated by an endogenous MAP kinase? Genomic sequencing has identified 15 MAP kinase homologues (designated as LmjMPK1 through LmjMPK15) in L. major (Wiese, 2007). Each of these L. major MAP kinases has also been identified in Leishmania mexicana, Leishmania infantum and Leishmania braziliensis. The LmjMPKs vary considerably in length; LmjMPK1 being the shortest with 358 amino acids while the longest LmjMPK8 contains 1579 amino acids. Pairwise sequence comparison of S. cerevisiae Hog1p (ScHog1p) and each of the LmjMPKs show a sequence identity between 22% and 38% (Fig. S1 and Table S2). The role of the LmjMPKs in modulating L. major AQP1 activity was examined. The experiments described herein are the first report of a parasitic protozoan aquaglyceroporin being positively regulated at the post-translational level by an MAP kinase. This affects the localization of aquaglyceroporin and alters the drug sensitivity and osmoregulatory activity of the parasite.
Expression of LmjMPK2 functionally complements the antimonite-sensitive phenotype of S. cerevisiae hog1Δ
The L. major genome encodes for 15 mitogen-activated protein kinases (LmjMPKs) (Wiese, 2007). Whether any of the LmjMPK complements the metalloid-sensitive phenotype of S. cerevisiae HOG1 deletion was examined. Of the 15 LmjMPKs, the closest ScHog1p homologue: LmjMPK1, LmjMPK2, LmjMPK3, LmjMPK4, LmjMPK5, LmjMPK9, LmjMPK10, LmjMPK11, LmjMPK12 and LmjMPK13 were chosen for further analysis. LmjMPK6, LmjMPK7, LmjMPK8, LmjMPK14 and LmjMPK15 are of longer chain-length than ScHog1p and were not considered any further. The chosen LmjMPK genes were PCR cloned individually from the L. major genome and then subcloned into the multiple cloning site of the galactose-inducible yeast expression vector pYES2. The metalloid sensitive S. cerevisiae hog1Δ strain YSH444 (Table S1) was transformed with the various LmjMPK-pYES2 constructs. In the presence of galactose, cells expressing LmjMPK2 partially complemented the As(III)-sensitive phenotype of YSH444, whereas the other LmjMPKs did not provide resistance to As(III) (Fig. S2). The results of As(III) complementation were also confirmed with Sb(III). Cells expressing LmjMPK2 were tested for their ability to grow in the presence of 0.5 mM Sb(III). Figure 1 shows that YSH444 was hypersensitive to Sb(III), but showed Sb(III) resistance upon episomal expression of HOG1. Expression of LmjMPK2 from plasmid LmjMPK2-pYES2 partially complemented the Sb(III)-sensitive phenotype of YSH444 (Fig. 1). The results indicate that LmjMPK2 activity provides metalloid tolerance to S. cerevisiae in vivo in the absence of Hog1p. Whether the lack of complementation by the other LmjMPKs is associated with problems of heterologous protein expression was not investigated any further.
Expression of LmjMPK2 confers antimonite sensitivity in Leishmania
Since expression of LmjMPK2 complemented the Sb(III)-sensitive phenotype of hog1Δ cells, its role in providing similar metalloid sensitivity in Leishmania was examined. LmjMPK2 was cloned into the Leishmania expression vector pSP72αneoα. Similarly, the Fps1p homologue, LmjAQP1 was cloned into the Leishmania expression vector pSP72αhygroα (Figarella et al., 2007). The L. donovani strain LdBob was transfected with these plasmids as described in Experimental procedures. LdBob cells have the advantage that they can be cycled between the promastigote and axenic amastigote forms using an established protocol (Goyard et al., 2003). Although metalloid sensitivity and transport experiments are ideally performed in an AQP1 deletion strain; despite several attempts, it has not been possible to generate a null mutant, which supports the idea of AQP1 being an essential gene for L. major survival and growth (R. Mukhopadhyay and M. Ouellette, unpubl. obs.). An alternate explanation for not obtaining a null mutant for AQP1 is that chromosome 31 (that encodes AQP1) is polyploid in all Leishmania strains studied (Rogers et al., 2011); thereby complicating the generation of a knockout strain.
Sb(III) sensitivity was examined in LdBob promastigotes, while Sb(V) sensitivity was tested in intracellular amastigotes. Sb(V) is taken up by macrophages, and a portion is reduced to Sb(III), which is then transported into the amastigote by the Leishmania aquaglyceroporin LmjAQP1 (Gourbal et al., 2004; Bhattacharjee et al., 2009). The other portion of the Sb(V) is taken into the amastigote and reduced to Sb(III) by LmACR2 (Zhou et al., 2004) and perhaps other enzymes such as TDR1 (Denton et al., 2004). Promastigotes expressing LmjAQP1 were 50-fold more sensitive to Sb(III) than the vector control (Table 1). In contrast, promastigotes coexpressing LmjAQP1 and LmjMPK2 were 30-fold more sensitive to Sb(III) than cells expressing LmjAQP1 alone, and 1400-fold more sensitive than the vector control. Intracellular amastigotes coexpressing LmjAQP1 and LmjMPK2 were sixfold more sensitive to Sb(V) than cells expressing LmjAQP1 alone, and 30-fold more sensitive than the vector control. Compared to the vector control, promastigotes and amastigotes expressing LmjMPK2 alone were 1.3-fold and 1.5-fold more sensitive to Sb(III) and Sb(V) respectively.
Table 1. Metalloid sensitivity of L. donovani expressing wild-type or altered LmjAQP1 and LmjMPK2.
Details of each strain are described in Table S1. EC50 values for Sb(III) for each cell type were determined for the promastigote form of the parasite. EC50 values for Sb(V) were determined for intracellular amastigotes.
138 ± 6
321 ± 21
3 ± 0.3
60 ± 11
105 ± 2
210 ± 11
AQP1 + MPK2
0.1 ± 0.002
10 ± 2
3.5 ± 0.2
53 ± 5
T197A AQP1 + MPK2
2.5 ± 0.02
65 ± 7
103 ± 3
242 ± 6
AQP1 + K42A MPK2
35 ± 2
150 ± 7
The uptake of Sb(III) into either promastigotes or axenic amastigotes was also examined. Consistent with Sb(III) sensitivity, promastigotes coexpressing LmjAQP1 and LmjMPK2 showed a twofold increased influx of Sb(III) than cells expressing LmjAQP1 alone (Fig. 2A). Cells expressing LmjMPK2 alone exhibited similar Sb(III) uptake as the vector control that was fourfold less than cells expressing LmjAQP1 alone (Fig. 2A). Likewise, axenic amastigotes expressing LmjMPK2 alone showed similar Sb(III) permeability as the vector control (Fig. 2B). Amastigotes expressing LmjAQP1 alone exhibited fourfold higher Sb(III) uptake than vector control, whereas cells coexpressing LmjAQP1 and LmjMPK2 showed a twofold increase in Sb(III) influx compared to cells expressing LmjAQP1 alone (Fig. 2B). Thus, coexpression of LmjAQP1 and LmjMPK2 in Leishmania increases Sb(III) sensitivity due to increased metalloid influx into the cells. This is in contrast to the observation in yeast, where elevated Hog1p activity increases metalloid tolerance via Fps1p inactivation (Thorsen et al., 2006).
LmjMPK2 affects LmjAQP1 expression
Whether the increased Sb(III) influx in LdBob cells coexpressing LmjAQP1 and LmjMPK2 was due to an increase in LmjAQP1 expression was examined by immunoblot analysis. A polyclonal antibody (anti-LmxMPK2) against the synthetic peptide CGLESRPVEREAAVRK, corresponding to amino acids 444–458 of LmxMPK2 (cf. Fig. S3) was used. Anti-LmxMPK2 can detect LmjMPK2 as well as L. donovani MPK2 (LdoMPK2). The C-terminal peptide sequence of LmxMPK2 used to raise the polyclonal antibody is conserved in both LmjMPK2 and LdoMPK2 except for one amino acid replacement; Glu451 in LmxMPK2 is replaced by an alanine in the corresponding position in either LmjMPK2 or LdoMPK2 (cf. Fig. S3). Immunoblot analysis showed that LmjMPK2 was expressed in promastigotes transfected with either LmjMPK2 alone or co-transfected with LmjAQP1 and LmjMPK2 (Fig. 3A). Expression of endogenous LdoMPK2 was also detected in cells transfected with either LmjAQP1 alone or in the vector control (Fig. 3A). Cells expressing LmjMPK2 showed a dense band consisting of endogenous LdoMPK2 and a slightly faster moving LmjMPK2, which is 25-residues shorter than LdoMPK2 (cf. Fig. S3).
To examine the expression of LmjAQP1, both flagellar and pellicular membrane fractions were isolated from LdBob promastigotes. LmjAQP1 localizes specifically on the flagellum of Leishmania promastigotes (Figarella et al., 2007). Immunoblot analysis of the flagellar fraction probed with LmjAQP1 anti-peptide antibody showed that, LmjAQP1 was several-fold overexpressed in cells co-transfected with LmjAQP1 and LmjMPK2 compared to those transfected with LmjAQP1 alone (Fig. 3B). Interestingly, in addition to being in the flagellar fraction, LmjAQP1 was also noted in the pellicular fraction of cells coexpressing LmjAQP1 and LmjMPK2 (Fig. 3C). LmjAQP1 was never detected in the cytosolic fraction (data not shown).
Immunofluorescence experiments also supported the immunoblot data. LdBob promastigotes expressing LmjAQP1 alone showed selective staining of the flagella (Fig. 4A). Staining was not observed on the main body of the parasite. In contrast, cells coexpressing LmjAQP1 and LmjMPK2 showed diffuse localization of LmjAQP1 throughout the flagellum and the entire body of the parasite (Fig. 4B).
LmjMPK2 mediates phosphorylation of LmjAQP1 at Thr-197
The Hog1p MAP kinase in S. cerevisiae controls Fps1p activity by phosphorylating Thr-231 within the Fps1p N-terminal tail (Thorsen et al., 2006). Whether LmjMPK2 phosphorylates LmjAQP1 in Leishmania was examined with an anti-phosphothreonine antibody. The blot containing flagellar and pellicular membrane fractions (Fig. 3B and C) was stripped and re-probed with an anti-phosphothreonine antibody (Fig. 3D and E). A strong band of phosphorylated-LmjAQP1 was detected in both flagellar (Fig. 3D) and pellicular (Fig. 3E) membrane fractions of cells co-transfected with LmjAQP1 and LmjMPK2. Phosphorylated LmjAQP1 was not observed in the vector control. Cells overexpressing LmjAQP1 alone also showed some degree of phosphorylation; most likely LmjAQP1 was phosphorylated by endogenous LdoMPK2.
The web-tool KinasePhos (Huang et al., 2005) was used to identify MAP kinase-specific phosphorylation sites in LmjAQP1. The program predicted a single, putative MAP kinase phosphorylation site at threonine 197 (Thr-197). Thr-197 was altered to alanine by site-directed mutagenesis. LdBob promastigotes were transfected with T197A LmjAQP1 either alone or co-transfected with LmjMPK2. Both flagellar and pellicular membrane fractions were isolated, and the blots were probed with anti-LmjAQP1, followed by anti-phosphothreonine antibody. Cells expressing T197A LmjAQP1 either alone or in tandem with LmjMPK2 exhibited significantly reduced expression of altered LmjAQP1 than cells co-transfected with wild-type LmjAQP1 and LmjMPK2 (Fig. 3B and C). T197A LmjAQP1 was observed only in the flagellar membrane (Fig. 3B) and not in the pellicular membrane fraction (Fig. 3C). The phosphorylation status of T197A LmjAQP1 was significantly lower compared to cells expressing wild-type LmjAQP1 (Fig. 3D and E). It is not clear whether the faint bands in the T197A LmjAQP1 fraction (Fig. 3D and E) are due to additional phosphorylation sites in LmjAQP1. Alternately, it might be phosphorylation of endogenous LdoAQP1 by LmjMPK2. LdoAQP1 exhibits 88% sequence identity to LmjAQP1 and has a conserved threonine at position 197 of its primary sequence (Fig. S4). Another explanation might be the limitation of the polyclonal anti-phosphothreonine antibody cross-reacting with phospho-serine-containing sequences.
Immunofluorescence experiments also supported the immunoblot data. Promastigotes expressing T197A LmjAQP1 either alone or together with LmjMPK2 showed selective staining of the flagellum only (Fig. 4C and D). Staining was not observed on the main body of the parasite. These experiments clearly demonstrate that LmjAQP1 is phosphorylated at Thr-197 in vivo.
Lysine 42 of LmjMPK2 is critical for phosphorylation of LmjAQP1
A multiple sequence alignment of L. major MAP kinases with ScHog1p (Fig. S1) indicated that Lys-42 of LmjMPK2 might be critical for kinase activity. This lysine is conserved in each of the L. major MAP kinases. Carrera et al. (1993) have shown that Lys-273 in the corresponding position of protein-tyrosine kinase pp56lck is critical for its phosphate transfer reaction. The equivalent residue in ScHog1p is Lys-52, and alteration of this residue to arginine resulted in loss of kinase activity (Reiser et al., 1999; Thorsen et al., 2006). Lys-42 of LmjMPK2 was altered to alanine by site-directed mutagenesis. LdBob promastigotes transfected with K42A LmjMPK2 showed considerable cytosolic expression of the altered protein (Fig. 3A). Cells co-transfected with wild-type LmjAQP1 and K42A LmjMPK2 showed significant reduction in LmjAQP1 expression, and LmjAQP1 was observed only in the flagellar (Fig. 3B) but not in the pellicular membrane fraction (Fig. 3C). Negligible phosphorylation of wild-type LmjAQP1 was noted, but the intensity of the band was similar to that of cells that did not express the protein, suggesting non-specific background (Fig. 3D and E). Immunofluorescence experiments were also in concordance with the immunoblot data. Cells expressing both wild-type LmjAQP1 and K42A LmjMPK2 showed that wild-type LmjAQP1 is localized only on the flagella (Fig. 4E). These experiments indicate that Lys-42 plays a critical role for LmjMPK2 kinase activity and the resultant phosphorylation of LmjAQP1.
Cells expressing altered LmjAQP1 or LmjMPK2 show reduced metalloid uptake and decreased metalloid sensitivity
Cells expressing T197A LmjAQP1 were examined for their metalloid tolerance. Promastigotes and amastigotes coexpressing T197A LmjAQP1 and wild-type LmjMPK2 were 25-fold and sevenfold more resistant to Sb(III) and Sb(V), respectively, than cells coexpressing wild-type LmjAQP1 and LmjMPK2 (Table 1). In terms of metalloid permeability, promastigotes and amastigotes coexpressing T197A LmjAQP1 and wild-type LmjMPK2 were threefold and twofold less permeable to Sb(III), respectively, than cells coexpressing wild-type LmjAQP1 and LmjMPK2 (Fig. 2). Interestingly, cells expressing T197A LmjAQP1 alone or coexpressing T197A LmjAQP1 and wild-type LmjMPK2 showed similar Sb(III) uptake as cells expressing wild-type LmjAQP1 alone (Fig. 2), and each of these cell types exhibited similar metalloid sensitivity (Table 1). This indicates that the wild-type and the altered LmjAQP1 channel are equally permeable to Sb(III), and that the properties of the channel are not affected by the T197A mutation. This is in contrast to the observation with ScHog1p where phosphorylation of Thr231 diminished transport activity (Thorsen et al., 2006).
The effect on metalloid permeability upon alteration of the catalytically important lysine (Lys-42) in LmjMPK2 was also examined. Amastigotes and promastigotes coexpressing wild-type LmjAQP1 and K42A LmjMPK2 were threefold and sixfold less permeable to Sb(III), respectively, than cells coexpressing wild-type LmjAQP1 and LmjMPK2 (Fig. 2). In terms of metalloid tolerance, either promastigotes or amastigotes coexpressing wild-type LmjAQP1 and K42A LmjMPK2 were 350-fold and 15-fold more resistant to Sb(III) and Sb(V), respectively, than cells expressing wild-type LmjAQP1 and LmjMPK2 (Table 1). Promastigotes and amastigotes coexpressing wild-type LmjAQP1 and K42A LmjMPK2 were 12-fold and 2.5-fold more resistant to Sb(III) and Sb(V), respectively, than cells expressing wild-type LmjAQP1 only (Table 1). This is also reflected in Sb(III) uptake; cells coexpressing wild-type LmjAQP1 and K42A LmjMPK2 exhibited lower metalloid uptake than cells expressing LmjAQP1 alone. Thus, alteration of Lys-42 of LmjMPK2 affects the phosphorylation of LmjAQP1, which results in decreased metalloid permeability and sensitivity.
LmjMPK2 stabilizes LmjAQP1 by increasing its half-life
To determine whether phosphorylation modifies the turnover rate of LmjAQP1, cycloheximide blocking experiments were performed to measure protein half-lives. Mid-log phase promastigotes were treated with cycloheximide, followed by isolation of flagella at defined time intervals, and the decay of LmjAQP1 in the flagellar fraction was determined by SDS-PAGE and immunoblot analysis (Fig. 5). The half-life of wild-type LmjAQP1 was estimated to be > 100 h. T197A LmjAQP1 exhibited a half-life of 23 h. LdBob promastigotes coexpressing wild-type LmjAQP1 and LmjMPK2 showed < 20% degradation of LmjAQP1 during the entire experiment, and the half-life of the protein could not be estimated. When T197A LmjAQP1 was coexpressed with wild-type LmjMPK2, the half-life of the altered protein was determined to be 18 h. The half-life of wild-type LmjAQP1 in cells coexpressing wild-type LmjAQP1 and K42A LmjMPK2 was determined to be 30 h. This experiment clearly depicts that phosphorylation at Thr-197 decreases the turnover number of wild-type LmjAQP1.
Phosphorylation of LmjAQP1 affects volume regulation
Upon exposure to hypo-osmotic stress, several mammalian cell types as well as a number of protozoa, swell initially but later return to their normal volume. This phenomenon of regulatory volume decrease is accomplished by the efflux of various osmolytes to the extracellular environment followed by passive water flow (Rohloff et al., 2003). LmjAQP1 plays a major role in volume regulation in both the promastigote and amastigote forms of the parasite (Figarella et al., 2007). The volume regulation ability of L. donovani LdBob promastigotes and amastigotes coexpressing wild-type or altered LmjAQP1 and/or LmjMPK2 was examined. Either promastigotes or amastigotes were subjected to a 50% reduction in osmolarity (from 300 to 150 mOsm). Since cell swelling leads to a decrease in the absorbance, the process of volume recovery was monitored by following the absorbance of the cell suspension at 550 nm.
Subsequent to a hypo-osmotic shock, LdBob promastigotes expressing wild-type LmjAQP1 show a drop in absorbance, indicating cell swelling, followed by a steady rise in absorbance to near original levels, signifying volume recovery. The volume recovery process was essentially complete over a 3 min period, and is much faster than the vector alone control (Fig. 6A). When LdBob promastigotes coexpressing wild-type LmjAQP1 and LmjMPK2 were subjected to a similar treatment, the swelling and recovery process was observed to be much faster than cells overexpressing wild-type LmjAQP1 alone. The magnitude of initial volume recovery for cells overexpressing T197A LmjAQP1 or coexpressing T197A LmjAQP1 and wild-type LmjMPK2 was similar to that of cells expressing wild-type LmjAQP1. Since T197A LmjAQP1 is not phosphorylated by LmjMPK2, the above data indicate that phosphorylated LmjAQP1 provides a faster volume recovery than unphosphorylated protein. Cells transfected with vector alone had the slowest volume recovery, whereas cells coexpressing wild-type LmjAQP1 and K42A LmjMPK2 exhibited a volume recovery that was in-between cells expressing wild-type LmjAQP1 only and vector control. The volume recovery for promastigotes expressing either wild-type or altered LmjMPK2 was similar to that of the vector control.
A similar pattern in volume recovery was observed for axenic amastigotes. Amastigotes coexpressing wild-type LmjAQP1 and LmjMPK2 showed faster volume recovery than cells expressing wild-type LmjAQP1 only (Fig. 6B). Volume recovery was found to be slower-paced in cells expressing T197A LmjAQP1. Similarly, cells expressing various combinations of altered LmjMPK2 also showed that inhibition of the phosphate transfer reaction slows the volume recovery process (Fig. 6B).
Generation of L. mexicana LmxMPK2 null mutant (Lmx-ΔMPK2) and add-back (g-LmxMPK2)
Southern blot analysis using a probe hybridizing to the open reading frame of LmxMPK2 suggested that it is a single copy gene in the genome of L. mexicana (Wiese et al., 2003). To generate a null mutant for LmxMPK2, a construct comprising the flanking regions of LmxMPK2 and resistance marker genes conferring resistance to hygromycin B and phleomycin was created (see Experimental procedures). Several attempts to generate homozygous null mutants with consecutive rounds of electroporation always led to promastigotes retaining a copy of LmxMPK2. An extrachromosomal copy of LmxMPK2 was then cloned into pX63polPAC and inserted into wild-type L. mexicana before attempting to delete the two genomic alleles. The deletion strategy was repeated again, which led to a successful replacement of the genomic alleles of LmxMPK2, but retaining the plasmid carrying the gene. These cells were used to infect female BALB/c mice, and foot-pad lesion development was monitored for over a year. Amastigotes were isolated by aspiration from lesion tissue of mice infected either with wild-type L. mexicana or LmxMPK2 mutant, and the disrupted tissues subjected to staining with SYTO 16 and cell sorting into 96-well PCR plates (Wang et al., 2005). Duplex PCR was performed using oligonucleotide primers specific for LmxMPK2 and LmxMPK9 (Bengs et al., 2005). Moreover, lesion-derived amastigotes were differentiated into promastigotes by avoiding the addition of puromycin to select for the presence of the plasmid carrying LmxMPK2. This resulted in 100% of wild-type amastigotes and promastigotes showing the typical bands for LmxMPK2 and LmxMPK9. In all, 78% of amastigotes showed both bands, 19% lacked the band for LmxMPK9 showing LmxMPK2 only and 3% did only show the band for LmxMPK9 but not the band for LmxMPK2. The much smaller number for the latter than the percentage control indicates that all amastigote samples contained cells retaining the plasmid even without any antibiotic selection for over a year in the mouse. LmxMPK2 is therefore essential for the amastigote form in the infected animal. On the other hand, 29% of promastigote reactions showed LmxMPK9 only but not LmxMPK2, 9% showed LmxMPK2 only and 62% showed both. This indicated that promastigotes might be able to lose the plasmid when grown in culture. To test this scenario, clones were generated by limited dilution in the absence of selective antibiotics and analysed by duplex PCR. This resulted in four clones that showed no amplification for LmxMPK2. Total cell lysates of these clones were subjected to immunoblot analysis using anti-LmxMPK2, confirming the absence of the protein in all of them (Fig. 7A). The blot was stripped and re-probed with an antiserum against myo-inositol-1-phosphate synthase (Ilg, 2002) (Fig. 7B). Promastigote lysates of wild-type L. mexicana and LmxMPK2 genomic null mutants carrying LmxMPK2 on a plasmid were used as controls. Two of the null mutant clones were chosen to generate genomic add-backs, in which one of the resistance marker genes was replaced by a construct comprising of the puromycin resistance marker gene, followed by the intergenic region of DHFR-TS and LmxMPK2 all surrounded by the flanking regions of LmxMPK2. Immunoblot analysis confirmed that in all clones obtained for the two complemented null mutants LmxMPK2 was expressed to near wild-type levels (Fig. 7C).
LmxMPK2 null mutant is resistant to Sb(III)
Consequent to the generation of the MPK2 null mutant and its complement in L. mexicana, the following experiments were performed in L. mexicana promastigotes. Immunoblot analyses of total cell lysates from logarithmically growing L. mexicana wild type (Lmx-WT + neo), LmxMPK2 null mutant (Lmx-ΔMPK2 + neo), LmxMPK2 add-back (g-LmxMPK2), Lmx-WT expressing LmjAQP1 (Lmx-WT + AQP1) and LmxMPK2 null mutant expressing LmjAQP1 (Lmx-ΔMPK2 + AQP1) promastigotes showed LmxMPK2 expression in wild-type and Lmx-WT + AQP1 cells, but none in null mutant and Lmx-ΔMPK2 + AQP1 (Fig. 8D, bottom panel). The add-back cells showed slightly lower expression of LmxMPK2 than the wild-type and Lmx-WT + AQP1 cells (Fig. 8D, bottom panel). Flagellar preparations from these strains were probed with anti-LmjAQP1 and anti-phosphothreonine antibody. As expected, LmjAQP1 expression was noted in Lmx-WT + AQP1 and Lmx-ΔMPK2 + AQP1 cells, and none in the wild type and null mutant (Fig. 8D, top panel). When probed with anti-phosphothreonine antibody, phosphorylation of LmjAQP1 was observed only in Lmx-WT + AQP1 cells, but none in Lmx-ΔMPK2 + AQP1 cells (Fig. 8D, middle panel).
The above L. mexicana strains were examined for their Sb(III) sensitivity and Sb(III) uptake. The null mutant was fivefold more resistant to Sb(III) than the wild type (Fig. 8A), and showed a threefold lower Sb(III) uptake than the wild type (Fig. 8B). The add-back strain regained Sb(III) sensitivity, but was still 1.5-fold more resistant to Sb(III) than the wild type (Fig. 8A), and exhibited a twofold increase in Sb(III) uptake compared to the null mutant (Fig. 8B). In contrast, both wild-type and Lmx-ΔMPK2 cells expressing LmjAQP1 were 250 and 145 times more sensitive to Sb(III), respectively, than the wild type (Fig. 8A). In congruence with Sb(III) sensitivity, wild-type L. mexicana expressing LmjAQP1 exhibited 1.5-fold higher Sb(III) accumulation than Lmx-ΔMPK2 + AQP1 cells (Fig. 8B).
When these strains were subjected to hypo-osmotic stress, the magnitude of the initial volume recovery was slower in the null mutant than in the wild type (Fig. 8C). The add-back strain showed a slightly slower volume recovery than the wild type. Expression of LmjAQP1 in wild-type L. mexicana showed significantly less swelling and a near immediate volume recovery. A relatively small but significant decrease in the magnitude of the initial volume recovery was noted in Lmx-ΔMPK2 + AQP1 cells.
The L. major aquaglyceroporin LmjAQP1 has multiple physiological roles. It is involved in solute transport (water, glycerol, methylglyoxal, dihydroxyacetone and sugar alcohols), volume regulation and osmotaxis (Figarella et al., 2007). It is also responsible for the adventitious uptake of Sb(III), which allows antimony containing drugs (Pentostam® or Glucantime®) to be used as the first line of defence against leishmaniasis. However, how the AQP1 channel is regulated in Leishmania is completely unknown. As opposed to higher eukaryotes, Leishmania has minimal control at the level of transcription initiation, and most of the gene regulation in these organisms occurs post-transcriptionally (Haile and Papadopoulou, 2007).
Several reports have indicated the association between MAP kinase signalling and aquaporin expression. The water channel protein AQP5 was shown to be induced by hypertonic stress in mouse lung epithelial cells, and the induction required activation of extracellular signal-regulated kinase (Hoffert et al., 2000). Arima et al. (2003) have shown that hyperosmotic stress induced by mannitol stimulates the expression of AQP4 and AQP9 in cultured rat astrocytes through a p38 MAP kinase-dependent pathway. In mammalian cells, the MAP kinase group of proteins, specifically, extracellular signal-regulated kinase, c-Jun N-terminal kinase and p38 MAP kinase are key players in osmo-sensing signal transduction pathways (Galcheva-Gargova et al., 1994; Han et al., 1994; Matsuda et al., 1995; Burg et al., 1996). The S. cerevisiae Hog1p is homologous to p38 MAP kinase. Thorsen et al. (2006) were the first to demonstrate that cells impaired in Hog1p function were hypersensitive to As(III) or Sb(III), whereas cells with elevated Hog1p activity contributed to metalloid tolerance by affecting the influx of metalloid through Fps1p.
The role of an endogenous MAP kinase in Leishmania to modulate LmjAQP1 activity was investigated. Of the ten L. major MAP kinases examined for their ability to complement the Sb(III) sensitivity of hog1ΔS. cerevisiae, only LmjMPK2 was able to provide partial complementation (Figs 1 and S2). When overexpressed in Leishmania, LmjMPK2 affected the osmotic stress response (Fig. 6) and metalloid sensitivity (Fig. 2) of the parasite by regulating LmjAQP1 in a positive manner. Regulation is attained by LmjMPK2 phosphorylating LmjAQP1 at Thr-197. Thr-197 resides in the small cytosolic loop D between transmembrane domains 4 and 5 of LmjAQP1 (cf. Uzcategui et al., 2008). Consequent to overexpression and phosphorylation of wild-type LmjAQP1, a higher intracellular accumulation of Sb(III) (Fig. 2A and B) and hypersensitivity to Sb(III) was noted (Table 1). Although LmjAQP1 resides in the flagellum of promastigotes, phosphorylation led to increased levels and distribution over the entire surface of the parasite (Fig. 4). In the absence of phosphorylation, the altered protein T197A LmjAQP1 remained segregated to the flagellum of the parasite (Fig. 4). Phosphorylation at Thr-197 stabilizes LmjAQP1 by increasing its half-life (Fig. 5). In contrast, T197A LmjAQP1 exhibited a fivefold shorter half-life than the wild type (Fig. 5). Although the T197A mutation makes the protein unstable, why do cells producing this protein do not show a higher resistance to Sb(III)? The answer lies in the fact that the steady state level of the protein is not altered by the mutation (Fig. 3B). Since the altered protein is constantly synthesized in vivo, it is able to confer similar resistance as the wild type. Another interesting observation is that cells expressing T197A LmjAQP1 either alone or in combination with LmjMPK2 showed similar Sb(III) uptake as cells expressing wild-type LmjAQP1 (Fig. 2), and each of these cell types exhibited similar metalloid sensitivity (Table 1). This suggests that Sb(III) permeability of the wild-type or altered LmjAQP1 channel is not affected by the mutation. Or in other words, phosphorylation of LmjAQP1 does not affect its channel activity per se, but helps in its stabilization and redistribution to combat osmotic stress.
Cells coexpressing wild-type LmjAQP1 and K42A LmjMPK2 showed much lower expression of LmjAQP1 than cells expressing wild type alone (Fig. 3B). Although this event is difficult to explain, it is possible that a dominant-negative effect of K42A LmjMPK2 on endogenous LdoMPK2 is affecting the expression of LmjAQP1. The lower expression of wild-type LmjAQP1 in the presence of K42A LmjMPK2, led to reduced Sb(III) influx (Fig. 2), and reduced metalloid sensitivity (Table 1).
The LmxMPK2 null mutant exhibited reduced Sb(III) uptake and increased Sb(III) resistance compared to wild-type L. mexicana (Fig. 8A). The add-back cells showed a similar phenotype as the wild type. The simplest interpretation is that in the absence of phosphorylation, LmxAQP1 has a shorter half-life, which results in decreased Sb(III) uptake. Conversely, phosphorylation of AQP1 in add-back cells stabilizes channel activity. Wild-type L. mexicana overexpressing LmjAQP1 showed a significant increase in Sb(III) sensitivity and uptake than Lmx-ΔMPK2 + AQP1 cells. This suggests that phosphorylation of LmjAQP1 by the endogenous LmxMPK2 stabilizes the activity of the channel.
It is possible that phosphorylation is stage-specific and dependent on environmental stress. Therefore, Leishmania AQP1 may reside in the unphosphorylated form in the flagellum of promastigotes, but is phosphorylated during the metacyclic stage, leading to its redistribution to the pellicular membrane. This redistribution of AQP1 provides stability against an osmotic gradient as the metacyclics begin their migration from the gut towards the proboscis of the fly. The relocalization of AQP1 to the pellicular membrane also allows the parasite to combat hypo-osmotic shock once it is regurgitated into the blood stream of the mammalian host.
The positive regulation of Leishmania AQP1 by an MAP kinase might provide a new approach for drug targeting. Since protozoan MAP kinases are distantly related to human MAP kinases, it might be possible to design compounds that specifically target the protozoan enzymes. One of the approaches might be the development of small molecule inhibitors of Leishmania MPK2 activity. Such compounds would prevent phosphorylation of the protozoan AQP1 channel, resulting in its increased turnover, and consequently, exposing the parasites to osmotic stress. Another approach would be to look for small molecule inducers of MPK2, which will stabilize AQP1, and thereby either resensitize drug resistant isolates to Sb(III) and/or reduce the effective dose, eliminating drug toxicity.
Leishmania promastigotes face acute changes in osmolarity while thriving inside the sandfly vector or as metacyclics swimming towards the proboscis of the sandfly. The amastigote forms also experience osmotic stress inside the human host. Additionally, the parasite faces temperature and pH stress inside the mammalian host. Exposure to antimonials is yet another stress faced by the parasite when the host is treated with either Pentostam® or Glucantime®. The molecular mechanisms by which Leishmania sense, transduce, and adapt to crucial changes in environmental conditions remains a huge gap in our understanding of parasite biology. This is the first report of positive regulation of a parasitic protozoan aquaglyceroporin activity by an MAP kinase, leading to relocalization of the aquaglyceroporin, which affects the metalloid permeability and osmo-regulatory capacity of the parasite.
Strains and media
Leishmania donovani strain LdBob was kindly provided by Professor Stephen M. Beverley, Washington University School of Medicine. Leishmania promastigotes and amastigotes were grown in culture media as described earlier (Goyard et al., 2003). Promastigotes were grown at 25°C while the axenic amastigotes were cultivated at 37°C with 5% CO2. These strains were always maintained as promastigotes. Human leukaemia monocyte cell line THP1 (ATCC) was maintained in RPMI-1640 medium with 10% fetal bovine serum (Invitrogen) and 50 µM β-mercaptoethanol (Sigma, cell culture grade) at 37°C with 5% CO2. Table S1 details the strains and transfectants used in this study.
Cloning of L. major MAP kinases
Leishmania major MAP kinases: LmjMPK1, LmjMPK2, LmjMPK3, LmjMPK4, LmjMPK5, LmjMPK9, LmjMPK10, LmjMPK11, LmjMPK12 and LmjMPK13 were PCR amplified from genomic DNA isolated from promastigotes, using primer pairs as indicated in Table S3. All PCR products were gel-purified, cloned into pGEM-T Easy vector (Promega), and sequenced (CEQ 2000XL, Beckman) to confirm the integrity of the gene. Each of the above LmjMPK/pGEM-T Easy constructs were subcloned into the yeast expression vector pYES2 (Life Technologies) following digestion with HindIII (or BamHI for LmjMPK10, LmjMPK12 and LmjMPK13) and XbaI. Similarly, LmjMPK2/pGEM-T Easy was digested with HindIII and XbaI and cloned into the same sites of pSP72-αneoα giving rise to LmjMPK2-pSP72αneoα. pSP72αneoα and pSP72αhygroα are episomal constructs that use α-tubulin intergenic regions to drive trans-splicing (Papadopoulou et al., 1992; Gourbal et al., 2004).
The cloning of wild-type LmjAQP1 into pGEM-T Easy vector (LmjAQP1/pGEM-T Easy) has been described previously (Gourbal et al., 2004). Mutations in LmjAQP1 and LmjMPK2 (both in pGEM-T Easy vector) were introduced by site-directed mutagenesis using the QuikChange Site-Directed Mutagenesis procedure (Stratagene), as described previously (Mukhopadhyay et al., 2003). The mutagenic oligonucleotides used for both strands and the respective changes introduced (underlined) are as follows: T197A LmjAQP1, 5′-C ACG GAC GCT CGC ATG GCT CCC GCC GTC GAC TAC-3′ (sense) and 5′-GTA GTC GAC GGC GGG AGC CAT GCG AGC GTC CGT G-3′ (antisense); K42A LmjMPK2, 5′-C CGC GTC GTT GCA CTC GCG AAA ATT TAC GAC GCC-3′ (sense) and 5′-GGC GTC GTA AAT TTT CGC GAG TGC AAC GAC GCG G-3′ (antisense). Each mutation was confirmed by sequencing the entire gene. Altered LmjAQP1 and LmjMPK2 were digested with HindIII and XbaI and cloned into the same sites of pSP72αhygroα and pSP72-αneoα respectively.
Generation of an LmxMPK2 null mutant
A NotI fragment from a lambda clone from a genomic DNA library carrying LmxMPK2 was ligated into the NotI site of pBSKII(+) yielding pBN14mapkin22-0305, partially sequenced and the sequence of the LmxMPK2 open reading frame submitted to GenBank DB (CAC07956). This plasmid was used for two PCRs using 5′-GCGCTGCAGCTAGCCGTCACCGCTGCGCCTCTGCTC-3′ and 5′-CCCAAGCTTCCAAACGACGAGTGCAG-3′, and 5′-GCGCTGCAGCCATGGTGGCGTCGTGAATGCAAATAC-3′ and 5′-CCCTCTAGAAAAAGAAGCTGCAAC-3′ to amplify the upstream and downstream regions of LmxMPK2 respectively. The resulting PstI/HindIII fragment and PstI/XbaI fragment were cloned into pBSKII(+), linearized with the respective restriction enzymes (pB1XPmapkin22, pB13HPmapkin22) and sequenced. The PstI/XbaI fragment liberated from pB1XPmapkin22 was ligated into the PstI/XbaI sites of pB13HPmapkin22 resulting in a construct carrying NcoI, PstI and NheI restriction sites instead of the open reading frame of LmxMPK2 and unique XbaI and HindIII restriction sites at the ends of the flanking regions of LmxMPK2. The NcoI and NheI restriction sites were used to slot in the resistance marker genes for hygromycin B phosphotransferase and phleomycin binding protein. Linear fragments for homologous recombination were generated from all three constructs using XbaI and HindIII.
LmxMPK2 DNA constructs
The open reading frame of LmxMPK2 was amplified by PCR using pBN14mapkin22-0305 as a template and the primer pairs 5′-GCCAACATGTCGGCCGAAATCGAGTC-3′ and 5′-GCCAAGCTTCATTTGCGCACCGCTGC-3′ to introduce a BspLU11I site at the start ATG and a HindIII site at the stop codon. The resulting fragment was cloned into the BspLU11I and HindIII sites of pBSKII(+) and sequenced from both ends. The central region confined by AscI and Acc65I restriction sites was replaced by sequenced DNA from pBN14mapkin22-0305 generating pB7BHMPK2. A BspLU11I/HindIII fragment derived from the latter was filled-in, ligated into pX63polPAC (Wiese, 1998), which had been linearized at EcoRV resulting in the Leishmania expression plasmid pX2PACLmxMPK2. An NcoI site was introduced at the start codon of the puromycin acetyl transferase gene by replacement of a 2089 bp EcoRI/NsiI fragment with the corresponding fragment derived from pX14polNcoIPac (Erdmann PhD thesis Hamburg) yielding pX6NcoIPacLmxMPK2. Briefly, two independent PCRs were carried out on pX63polPAC using the oligonucleotides 5′-CGCATACGCGACGAATTC-3′ and 5′-GTACTCGGTCGCCATGGAAGGTCGTCTCCTTG-3′ in one reaction, and 5′-CGACCTTCCATGGCGACCGAGTACAAGCCCACG-3′ and 5′-AGACGCCGACGGTGGCCA-3′ in the other reaction. The amplified fragments were used as overlapping templates in a third PCR with 5′-CGCATACGCGACGAATTC-3′ and 5′-AGACGCCGACGGTGGCCA-3′. The product was ligated into pCR2.1-TOPO, sequenced and used to liberate a 684 bp EcoRI/MscI fragment which was cloned into pX63polPAC linearized using the same restriction enzymes, resulting in pX14polNcoIPAC. A 3417 bp Nco/XbaI fragment was liberated from pX6NcoIPacLmxMPK2 and ligated into the NcoI and NheI restriction sites of the deletion construct described above. From this a 5010 bp HindIII/XbaI fragment comprising the LmxMPK2 upstream region, the puromycin acetyl transferase resistance gene, the DHFR-TS intergenic region, LmxMPK2 and the LmxMPK2 downstream region was purified and used to re-integrate LmxMPK2 into its original gene locus by homologous recombination.
Parasites, transfection, cell sorting and PCR analysis for L. mexicana
Promastigotes of L. mexicana MNYC/BZ/62/M379 wild type and mutants were grown as described previously (Menz et al., 1991). Transfection of L. mexicana promastigotes was performed as described previously (Bengs et al., 2005) and the cells kept under antibiotic selection as non-clonal cell lines. Female BALB/c mice were infected into the left hind footpad at the age of 6–8 weeks with 1 × 107 promastigote parasites in 30 µl of PBS. Amastigotes were isolated from lesions of BALB/c mice as described before (Wiese, 1998). Small-scale amastigote preparation, staining with SYTO16 (Invitrogen) and cell sorting followed by duplex PCR was performed as described in (Wang et al., 2005) using the oligonucleotide primers 5′-GAAGCCAAGCAATCTGCTCG-3′ and 5′-GTTGGCGAACTGCGAGTTGG-3′ to generate a 342 bp LmxMPK2 fragment. For amplification of the 254 bp control DNA fragment derived from LmxMPK9, the primer pairs 5′-GTCAGCGTGCCAATGAAAT-3′ and 5′-CAAGCTCCGGTGCGCGGTA-3′ were used.
Transfection of Leishmania
Transfection of LmjAQP1-pSP72αhygroα and LmjMPK2-pSP72αneoα into Leishmania promastigotes was accomplished as described previously (Ouellette et al., 1990). All transfectants were maintained in the presence of 300 µg ml−1 hygromycin B and 60 µg ml−1 geneticin.
Metalloid sensitivity assays
Saccharomyces cerevisiae W303-1A and YSH444 bearing plasmids pYES2, pYES2 harboring LmjMPKs or HOG1 (Table S1) were grown overnight at 30°C in a synthetic medium containing 2% galactose and 0.7% yeast nitrogen base (Difco), supplemented with appropriate amino acids. Spots of 3 µl of serial dilutions of cultures were then applied onto 1% agar plates with the above medium in the presence or absence of 0.5 mM potassium antimonyl tartrate. The plates were incubated at 30°C for 2–3 days.
Metalloid sensitivity of the promastigote transfectants was determined as described previously (Gourbal et al., 2004). Briefly, log phase promastigote cultures were diluted to 107 cells ml−1 in a culture medium containing various concentrations of Sb(III) in the form of potassium antimonyl tartrate (Sigma). Following 72 h incubation, cell growth was monitored from the absorbance at 600 nm using a microplate reader (Spectramax 340, Molecular Devices). Percentage survival was plotted against Sb(III) concentrations and EC50 was determined using SigmaPlot 11.0. Each assay was performed at least three times. Error bars were calculated from the mean ± SE.
Metalloid sensitivity of amastigotes inside macrophages was determined after infecting the human leukaemia cell line THP-1 with LdBob promastigotes, as described previously (Zhou et al., 2004), with some minor modifications. THP-1 cells were differentiated with phorbol myristate acetate and infected with LdBob promastigotes at a ratio of 20:1 for 4 h. Non-internalized parasites were washed away and the infected macrophages were treated with varying concentrations of potassium hexahydroxoantimonate [Sb(V)] (Sigma). After 4 days of culture, wells containing adherent differentiated THP-1 cells were washed, and luciferase activity was determined as described (Zhou et al., 2004), using a microtiter plate luminometer (LMaxII, Molecular Devices). EC50 was calculated from the sigmoidal analysis of percentage light emitted (compared to untreated macrophages) versus antimonate concentrations using SigmaPlot 11.0. Each assay was performed at least three times and the results are represented as mean ± SE.
Log phase Leishmania promastigotes or axenic amastigotes were washed twice with PBS, pH 7.4 (Invitrogen) and suspended in PBS at a density of 108 cells ml−1. Either promastigotes or amastigotes were then incubated with 10 µM Sb(III) for 15 min at room temperature. A 200 µl portion was filtered through a 0.22 µm nitrocellulose filter and the filter washed once with 5 ml of ice-cold PBS. The filters were digested with 0.4 ml of concentrated HNO3 (69–70%) (EM Science) for 1 h at 70°C, allowed to cool to room temperature, and diluted with high-pressure liquid chromatography grade water (Sigma) to produce a final concentration of HNO3 of approximately 3%, and then analysed by a PerkinElmer SCIEX ELAN® DRC-e inductively coupled plasma mass spectrometer. Standard solutions were made in the range of 0.5–10 p.p.b. in 3% HNO3 using antimony standards (Ultra Scientific). Each transport experiment was repeated at least three times with duplicate samples. Error bars were calculated from the mean ± SE.
Cell volume measurements
Relative changes in cell volume following the induction of hypo-osmotic shock were measured as described earlier (Rohloff et al., 2003). Briefly, log phase promastigotes or amastigotes were washed twice in PBS and re-suspended at a density of 109 cells ml−1. One hundred microlitre portions of the cell suspension were transferred to a microtiter plate. Hypo-osmotic shock was induced by dilution of the isotonic cell suspension with an equal volume of deionized water and the absorbance at 550 nm was recorded every 15 s for 3 min in a microplate reader (Spectramax 340, Molecular Devices). A decrease in absorbance corresponds to an increase in cell volume. Isosmotic control experiments consisted of dilution of cell suspensions with appropriate volumes of isosmotic buffer. Unless otherwise noted, all hypo-osmotic shock experiments were conducted at a final osmolarity of 150 mOsm (1:1 dilution of isosmotic buffer and water). Each experiment was repeated at least three times. Error bars were calculated from the mean ± SE.
Cell lysates and immunoblots
The flagellar and pellicular plasma membrane fractions from LdBob promastigotes transfected with vector alone or wild-type or altered LmjAQP1 and LmjMPK2 (Table S1) were isolated as described previously (Figarella et al., 2007). The total protein content of either the flagellar or pellicular plasma membrane fractions were determined by a filter assay as described previously (Schaffner and Weissmann, 1973). For immunoblot analysis, thirty micrograms of each protein sample were electrophoresed into a 12% SDS-polyacrylamide gel, and electroblotted onto nitrocellulose membrane (Whatman). The blot was incubated with affinity-purified LmjAQP1 antiserum as described previously (Figarella et al., 2007). The labelling was visualized with horseradish peroxidase-conjugated goat anti-rabbit serum (Sigma) using a Western Lightning Chemiluminescence Reagent Plus system (PerkinElmer). To check for the phosphorylation status, the blot was stripped and re-probed with an anti-phosphothreonine antibody (Fitzgerald, USA), and the labelling was visualized by chemiluminescence.
Rabbit polyclonal anti-peptide antibody (anti-LmxMPK2) were used to detect LmjMPK2 expression in Leishmania. For analysis of L. mexicana lysates, ∼ 2 × 107 cells were boiled in 100 µl of lysis buffer [1× PBS, 0.1% (w/v) SDS, 50 mM DTT, 50 µM leupeptin, 25 µM Nα-p-Tosyl-L-lysine chloromethyl ketone, 1 mM phenylmethanesulfonyl fluoride, 10 mM 1,10-phenanthroline, and 1× Laemmli sample buffer]. Twenty-five microlitre of each sample were subjected to 12% SDS-PAGE and electro-blotted onto nitrocellulose membrane. The blots were incubated with the anti-LmxMPK2 antibody (1:100 dilution) for overnight at 4°C, and the labelling was visualized by chemiluminescence. As a loading control, the blots were also stripped and re-probed with a polyclonal antiserum against myo-inositol-1-phosphate synthase (1:500 dilution) (Ilg, 2002).
Immunofluorescence analyses of LdBob promastigotes transfected with wild-type or altered LmjAQP1 and LmjMPK2 were performed as described previously (Figarella et al., 2007). The slides were examined by Olympus BX61 Photomicroscope with a 100× oil immersion objective lens.
Mid-log phase LdBob promastigotes were treated with 70 µg ml−1 cycloheximide and cells were harvested at timed intervals. Flagella were isolated from cells collected at each time-point. Densitometric analyses of expression of LmjAQP1 were performed from immunoblots using the ImageJ software (http://imagej.nih.gov/ij/). The expression ratio was determined after normalizing LmjAQP1 band intensity at each time-point against that at zero hour.
This work was supported by National Institutes of Health Grants GM55425 and AI58170. M. J. T. acknowledges the support of Olle Engkvist Byggmästare Foundation. M. W. was funded by the Deutsche Forschungsgemeinschaft DFG (WI 2044/5-1). We thank Dr Helen G. Tempest for the use of her photomicroscope.