A role for p38 MAPK in the regulation of ciliary motion in a eukaryote
© Ressurreição et al; licensee BioMed Central Ltd. 2011
Received: 4 August 2010
Accepted: 26 January 2011
Published: 26 January 2011
Skip to main content
© Ressurreição et al; licensee BioMed Central Ltd. 2011
Received: 4 August 2010
Accepted: 26 January 2011
Published: 26 January 2011
Motile cilia are essential to the survival and reproduction of many eukaryotes; they are responsible for powering swimming of protists and small multicellular organisms and drive fluids across respiratory and reproductive surfaces in mammals. Although tremendous progress has been made to comprehend the biochemical basis of these complex evolutionarily-conserved organelles, few protein kinases have been reported to co-ordinate ciliary beat. Here we present evidence for p38 mitogen-activated protein kinase (p38 MAPK) playing a role in the ciliary beat of a multicellular eukaryote, the free-living miracidium stage of the platyhelminth parasite Schistosoma mansoni.
Fluorescence confocal microscopy revealed that non-motile miracidia trapped within eggs prior to hatching displayed phosphorylated (activated) p38 MAPK associated with their ciliated surface. In contrast, freshly-hatched, rapidly swimming, miracidia lacked phosphorylated p38 MAPK. Western blotting and immunocytochemistry demonstrated that treatment of miracidia with the p38 MAPK activator anisomycin resulted in a rapid, sustained, activation of p38 MAPK, which was primarily localized to the cilia associated with the ciliated epidermal plates, and the tegument. Freshly-hatched miracidia possessed swim velocities between 2.17 - 2.38 mm/s. Strikingly, anisomycin-mediated p38 MAPK activation rapidly attenuated swimming, reducing swim velocities by 55% after 15 min and 99% after 60 min. In contrast, SB 203580, a p38 MAPK inhibitor, increased swim velocity by up to 15% over this duration. Finally, by inhibiting swimming, p38 MAPK activation resulted in early release of ciliated epidermal plates from the miracidium thus accelerating development to the post-miracidium larval stage.
This study supports a role for p38 MAPK in the regulation of ciliary-beat. Given the evolutionary conservation of signalling processes and cilia structure, we hypothesize that p38 MAPK may regulate ciliary beat and beat-frequency in a variety of eukaryotes.
Motile cilia are microscopic membrane-bound extensions of certain cells that are vital for the survival and reproduction of many eukaryotes. By beating in a regular pattern these evolutionarily-conserved organelles exert mechanical force; they thus play important roles in motility of small organisms and facilitate fluid movement across epithelial surfaces in complex multicellular animals. In addition to their role in fluid movement, motile cilia have recently been found to possess sensory functions in mammals, a feature previously thought to be restricted to non-motile 'primary' cilia [1, 2]. Given that motile cilia are essential to physiology it is not surprising that defects in these organelles cause multiple human disorders [3–5]. Such ciliopathies include primary ciliary dyskinesia resulting in an inability to clear mucous and debris from airways , hydrocephalus caused by abnormal spinal fluid movement in the ventricles of the brain , and situs inversus (inversions of the normal left/right symmetry of organs) a consequence of altered nodal flow during embryogenesis [8, 9]. Motile cilia (and flagella, which are essentially long motile cilia) are also essential for the completion of the life cycles of various parasites of humans and animals. For example, they power locomotion of schistosome larvae (miracidia) enabling host-finding and thus infection of the snail intermediate host , and are required for migration of trypanosomes between gut and salivary glands in the tsetse fly vector , and for viability of the bloodstream trypanosome form .
Motile cilia (and flagella) are composed of nine microtubular doublets and, usually, two central mictrotubular singlets, comprising the axoneme; dynein arms and radial spokes associated with the axoneme generate and control axonemal bending and thus force generation. The regulation of ciliary beating has been the focus of much research , and proteomic studies including those on the model organism Chlamydomonas reinhardtii, have aimed to describe the repertoire of proteins present within eukaryotic cilia and flagella, or their component fractions [14–19]. The proteomic analysis of Chlamydomonas flagella revealed over 90 putative signal transduction proteins including kinases and phosphatases , some of which might be anchored to the axoneme , highlighting the importance of signalling and reversible protein phosphorylation in the function of motile cilia and flagella. This is further supported by the recent identification in this organelle of 32 flagella phosphoproteins with 126 in vivo phosphorylation sites . Despite these advances, our understanding of kinase-mediated cell signalling mechanisms regulating ciliary motion is still rudimentary, being largely restricted to the roles of the cAMP-dependent protein kinase, protein kinase A (PKA) [13, 22–24], and protein kinase C (PKC) [13, 25–27].
The surface of the schistosome miracidium is almost entirely covered with numerous motile locomotory cilia which, in Schistosoma mansoni, emerge from 21 ciliated epidermal plates arranged in tiers . The S. mansoni miracidium is fully developed in the egg when it is released in the faeces of the infected definitive (human) host; upon water contact the miracidium hatches and swims rapidly to locate a suitable intermediate freshwater snail-host. Miracidia movements respond rapidly to various environmental cues such as the presence of host-snail components  or salinity , suggesting that these motile cilia might possess sensory functions as is the case with Paramecium [reviewed in ]. While studying kinase-mediated cell signalling in S. mansoni during miracidia development we observed an unexpected event, namely that activation of p38 mitogen-activated protein kinase (p38 MAPK) attenuated miracidial swimming. This kinase, an orthologue of the yeast HOG kinase, participates in signalling cascades that regulate transcriptional responses to stress  as well as having other non-transcription factor targets such as cytosolic phospholipase A2 . Here we report findings that support a role for p38 MAPK in the regulation of ciliary motion of the multicellular eukaryote S. mansoni.
A p38 MAPK immunoprecipitation kinase assay kit was next employed to confirm that the S. mansoni protein recognized by the anti-phospho p38 MAPK antibodies possessed p38 MAPK activity. Immunoprecipitates from adult worm homogenates phosphorylated the p38 MAPK substrate activating transcription factor 2 (ATF-2) (Figure 2B). The anti-inflammatory pyridinylimidazole compounds (SmithKline Beecham, SB) inhibit p38 MAPK with both activity of the phosphorylated enzyme and its autophosphorylation affected . These inhibitors compete with ATP at the ATP binding site and exhibit no, or very weak inhibitory activity toward the closely related MAPKs, c-Jun N-terminal kinase (JNK) and extracellular signal-regulated kinase (ERK) . SB 203580 inhibits both p38α and p38β MAPK and the interactions between the inhibitor and human kinases have been mapped using crystallography and amino acid substitution experiments [42, 43]. The crucial residues thought to be involved in this interaction are shown in Figure 1; although falling outside the S. mansoni sequence fragment, these residues are conserved in S. japonicum p38 MAPK. We thus postulated that SB 203580 would inhibit S. mansoni p38 MAPK. Pre-incubation of immunoprecipitates with SB 203580 prior to kinase assay reduced ATF-2 phosphorylation considerably when compared to immunoprecipitates incubated in dimethyl sulfoxide (DMSO) vehicle alone; inhibition appeared dose dependent with 5 μM SB 203580 almost completely attenuating ATF-2 phosphorylation (Figure 2B). Thus the antibody recognizes an S. mansoni protein with p38 MAPK-like activity that can be inhibited by the p38 MAPK inhibitor SB 203580.
Anisomycin is a known potent stimulator of p38 MAPK phosphorylation, and therefore activation, so we considered that this compound might activate p38 MAPK in freshly-hatched swimming miracidia that lacked the phosphorylated (activated) form of the enzyme (Figure 2A). Western blotting revealed that treatment of these miracidia with 20 μM anisomycin for 15 min induced considerable p38 MAPK phosphorylation when compared to controls, which appeared to increase after 30 min and was sustained over 60 min (Figure 2C).
The results of the above biochemical experiments, coupled with the bioinformatic analysis, are commensurate with the S. mansoni immunoreactive protein being a p38 MAPK orthologue.
While observing miracidia by immunofluorescence microscopy a solitary egg was found on the microscope slide, this egg had ruptured to allow hatching but fortunately still contained the miracidium which stained successfully with antibodies. Strikingly, and in contrast to swimming miracidia, the hatching miracidium possessed considerable p38 MAPK activity that via z-section analysis was found to be cilia-associated. Experiments were then designed in an attempt to re-capture this event and on several occasions miracidia were observed within ruptured eggs; such miracidia contained cilia-associated p38 MAPK activity considerably greater than that of freshly-hatched swimming miracidia (Figures 3H-K).
Although various kinases including PKA, casein kinase, adenylate kinase, cGMP-dependent protein kinase and a putative MAPK have been identified in proteomic screens of cilia/flagella e.g. [14, 15], to our knowledge p38 MAPK has not been detected by such methods. P38 MAPK has however recently been localized to the post-acrosomal region and upper flagellum mid-piece of human sperm by fluorescence microscopy . Although sperm tails are classified as flagella, their regulation differs from motile cilia in some aspects; sperm also possess a number of unique accessory structures  making them somewhat distinct from other motile cilia.
Additional file 1: Supplementary Movie File. Combined example videos of miracidia in spring water (control) or SB 203580 (1 μM in spring water) for 60 min, anisomycin (20 μM in spring water) for 30 min or 60 min, or revived after anisomycin treatment (60 min anisomycin followed by 20 min in SB 203580 (1 μM)). Miracidia swim speed is increased slightly by SB203580 and is attenuated after 30 min anisomycin treatment; swimming stops after 60 min in anisomycin and is revived following subsequent incubation in SB203580. (MPEG 15 MB)
The mechanism by which p38 MAPK controls ciliary beat is not known. Active p38 MAPK could attenuate ciliary beat either by direct or indirect interactions involving phosphorylation of axonemal components or components within a ciliary signal transduction cascade, respectively. Studies with C. reinhardtii have demonstrated that phosphorylation and dephosphorylation control flagella motility  and have highlighted the complex nature of protein phosphorylation in this organelle . Many potential kinase substrates exist that co-ordinate ciliary motility including the dyneins  and central pair kinesin KLP1 , but further characterisation of kinase substrates within cilia is needed.
Only a few other kinases have been reported to regulate cilia beat. PKA has been shown to attach to Paramecium ciliary axonemes, and strong evidence exists for PKA playing a positive role in Paramecium swimming and in controlling ciliary beat frequency of mammalian cilia [13, 22–24]. Interestingly, PKA has also been implicated in regulating the ciliary motion of S. mansoni miracidia . While cGMP-dependent protein kinase (cGMP) has also been shown to positively regulate ciliary beat frequency [13, 50], like for p38 MAPK, PKC has been implicated in slowing ciliary beat [13, 25–27]. Clearly, more research is needed into the influence of kinase activities on ciliary movement. For example, ERK1/2 has recently been found to bind radial spoke protein 3 in mammals and regulate its interaction with PKA ; however, the extent to which ERK actually influences motility via this interaction warrants investigation.
Here biochemical, innunohistochemical and functional data are presented that are consistent with p38 MAPK playing an important part in the regulation of ciliary beat and thus swimming behaviour of the multicellular eukaryote, S. mansoni. The marked difference in p38 MAPK activation between un-hatched or stationary miracidia and actively swimming miracidia is striking. Localization of active p38 MAPK to both the cilium shaft and the tegument of stationary miracidia implies that p38 MAPK might play multiple parts in co-ordinating swim behaviour, including sensory roles as described/hypothesized for motile cilia in other organisms including parasites [1, 2, 53]. Given the conservation of both signalling processes and structure/function of motile cilia, we hypothesize that p38 MAPK might regulate ciliary beat frequency in a variety of metazoans. Thus our findings could have implications for studies into motility of other important multicellular eukaryotes including parasites of humans, and for research into various human ciliopathies.
S. mansoni p38 MAPK gene candidates were identified from version 4.0 of the schistosome genome assembly by searching S. mansoni GeneDB hosted by the Wellcome Trust Sanger Institute, relying on the existing annotation [33, 54]. Although only partial cDNA reads were found, they were further assessed for similarity to p38 MAPKs from other organisms using the NCBI tBLASTx search tool , limited to bilateria (organism) and the nucleotide dataset. Protein sequences of candidates with matches to p38 MAPK genes were aligned to those of other organisms, including that for S. japonicum  using Geneious Pro 4.85 (Biomatters Ltd, Auckland, New Zealand) with Blosum62 cost matrix, threshold = 1 and default parameters. Pair wise identity scores were also obtained using Geneious Pro.
Adult worms were recovered by portal perfusion of patent mice infected with S. mansoni (Belo Horizonte strain) and were immediately snap frozen in liquid nitrogen and stored at -80°C. Livers and spleens were then removed from the infected mice and S. mansoni eggs isolated; miracidia were then hatched from eggs for up to 2 h in natural spring water (Evian) and were collected under a dissecting microscope using a Pasteur pipette . Miracidia were washed three times in spring water in a Stericup filter (0.45 μm PVDF membrane, Millipore, Watford, UK). The same filter was then used to concentrate the miracidia (to achieve approximately 10,000 miracidia/ml); enumeration of larvae was performed in aliquots under an inverted light microscope. Animal use received appropriate local ethical approval.
The effect of the p38 MAPK activator anisomycin on p38 MAPK phosphorylation (activation) in S. mansoni was assessed by western blotting using anti-phospho p38 MAPK (Thr180/Tyr182) monoclonal antibodies (Cell Signalling Technology, New England Biolabs, Hitchin, UK) that recognize only the phosphorylated (activated) form of the enzyme. Freshly-hatched miracidia (~900 per treatment) were incubated in anisomycin (20 μM in spring water) or spring water containing vehicle (0.02% (v/v) DMSO) for varying durations (15, 30 or 60 min) and then immediately placed on ice and proteins extracted by adding an appropriate volume of 5x SDS-PAGE sample buffer followed by brief homogenization. Samples were then boiled for 5 min and sonicated briefly. After cooling, protease and phosphatase inhibitors (Sigma, Poole, UK) were added at the manufacturer's recommended concentrations and samples stored at -20°C prior to electrophoresis.
Schistosoma mansoni protein samples were separated on 10% SDS-PAGE gels and were transferred to nitrocellulose using a semi-dry electrotransfer unit (Bio-Rad, Hemel Hempsted, UK). After staining with Ponceau S to confirm homogeneous transfer, membranes were blocked for 1 h in 5% (w/v) non-fat dried milk, and then incubated anti-phospho p38 MAPK monoclonal antibodies (1/1000 in tris-buffered saline/0.1% Tween-20 (TTBS) containing 1% (w/v) BSA) overnight at 4°C. Next, blots were washed in TTBS and incubated for 2 h at room temperature in horse-radish peroxidase-congugated secondary antibodies (Cell signalling Technology; 1/5000 in TTBS) before further washing and incubation in West Pico chemiluminescent substrate (Pierce, Tattenhall UK) for 5 min. Immunoreactive bands were then visualised using cooled CCD GeneGnome chemiluminescence imaging system (Syngene, Cambridge, UK). Equal loading of proteins on blots was checked by stripping blots for 3 h at room temperature with Restore western blot stripping buffer (Pierce), before briefly washing blots in TTBS and incubating blots with anti-actin antibodies (1:3000 in TTBS; Sigma). Human astrocytoma (U251 MG) cell lysates, used as positive control for detection of phosphorylated p38 MAPK, were kindly provided by Suzanne Newton (Kingston University).
To determine p38 MAPK activities of proteins immunoprecipitated using anti-phospho p38 MAPK antibodies, a non-radioactive p38 MAPK activity assay kit was used (Cell Signalling Technology). Ten adult worm pairs were homogenized in cell lysis buffer (250 μl), the lysate cleared by centrifugation at 13,000 rpm in a microfuge for 10 min at 4°C, and the supernatant recovered. Subsequently, 2 μl of immobilized anti-phospho p38 MAPK (Thr180/Tyr182) monoclonal antibodies were added to the lysate and samples were gently mixed overnight at 4°C. Subsequently, the immune complex was washed twice in cell lysis buffer and re-suspended in kinase buffer before 1 μl of ATP and 1 μl of ATF-2 fusion protein were added to start the kinase reaction. After 30 min the reaction was terminated by adding an appropriate volume of 5x SDS-PAGE sample buffer. The samples were then boiled, sonicated and processed for western blotting using anti-phospho ATF-2 primary antibodies. All buffers and reagents used were provided in the p38 MAPK assay kit. In parallel experiments, the p38 MAPK inhibitor SB 203580 (1 μM, 2 μM, or 5 μM) or vehicle (0.02 % (v/v) DMSO) was added to the immunoprecipitates 15 min prior to the start of the kinase assay.
Freshly-hatched swimming S. mansoni miracidia were either fixed immediately in absolute acetone or were treated with anisomycin (20 μM), or vehicle (0.02 % (v/v) DMSO), for 30 min prior to fixing. In some experiments, miracidia were left to hatch from eggs for short durations (10 - 20 min) prior to fixing in absolute acetone; this was done in an attempt to recover miracidia in the process of hatching from the egg. All parasites were then stored at 4 °C. For further preparation acetone was removed and samples washed twice with phosphate buffered saline (PBS) before being permeabilized in 0.3% (v/v) Triton X-100 for 1 h and washed with PBS prior to blocking in 10% (v/v) goat serum (Invitrogen, Paisley, UK) for 1 h. After a further wash with PBS, parasites were incubated with anti-phospho p38 MAPK mAb (1:50 in 5% (w/v) BSA) for 3 days on a microfuge tube rotator. The parasites were then washed twice in PBS for 1 h each and incubated in Alexa fluor 488 secondary antibodies (1:500 in 5% (w/v) BSA) for 24 h in the dark, followed by a further wash in PBS for 1 h. To detect cilia, miracidia were also incubated as above in anti-acetylated tubulin mouse monoclonal antibodies (1:100 in 5% (w/v) BSA; Sigma, Poole, UK) and Alexa fluor 594 secondary antibodies. Next, parasites were placed on microscope slides, left to air dry prior to mounting in Vectashield (Vecta Laboratories, Peterborough, UK) anti-bleaching medium, and sealed with transparent nail polish. All incubations were carried out at room temperature and incubations and washes were done in 2 ml screw cap tubes. Miracidia and eggs were then visualized on a Leica TCS SP2 AOBS confocal laser scanning microscope using a 20x dry objective or 40x and 63x oil immersion objectives and images collected with Leica software. Since S. mansoni miracidia autofluoresce, the signal received for the negative controls (i.e. those incubated only with secondary antibodies) was reduced. This was achieved by reducing the power level of the photomultiplier tube, which was then kept constant for all observations.
For conventional scanning electron microscopy, acetone fixed miracidia or eggs were plated on individual coverslips and left to air dry, they were then placed in PBS for 4 h, dehydrated in ethanol, briefly soaked in hexamethyldisilazane and evaporated, and sputter-coated with gold-palladium. Specimens were then visualized on a Zeiss EVO50 scanning electron microscope.
Freshly-hatched miracidia in spring water were divided into 200 μl aliquots and exposed to either SB 203580 (1 μM), anisomycin (20 μM), vehicle (DMSO, 0.02% (v/v)), or were left untreated. Each sample was then immediately placed into a small sterile Petri dish and the 200 μl droplet spread out using a pipette; care was taken to ensure that the size and spread of the droplet was consistent between experiments to minimize artefacts in measurement owing to the miracidia swimming out of the horizontal plane during recordings. Light influences considerably miracidia swimming behaviour, so light intensity and positioning also remained constant for all experiments which were performed at 27°C. Miracidia were videoed over 60 min. There were approximately 10 miracidia in each sample and at least 30 miracidia per treatment were analysed in three independent experiments. Visualization was achieved using an Olympus SZ4045 binocular dissecting microscope and avi-format video recordings were made using a JVC TK-1481 composite colour video camera linked to Studio Launcher Plus for Windows software. Digital videos were subsequently processed using the freely-available analysis software ImageJ  to determine swim path length of individual miracidia in 5s permitting swim velocities (mm/s) to be calculated at various time points after treatment.
Recovered eggs from schistosome-infected mice were hatched in spring water containing penicillin and streptomycin (100 units/ml each). Collected miracidia were then washed, and concentrated using Stericup filters, in sterile Chernin's balanced salt solution, pH 7.2,  containing glucose and trehalose and the same antibiotics (CBSS+). Approximately 1500 miracidia were placed onto individual wells of 6-well cell culture plates (Nunc, Loughborough, UK) and further 2 ml of either CBSS+, or CBSS+ containing DMSO, SB 203580, or anisomycin (0.02% (v/v), 1 μM, and 20 μM final concentrations, respectively) added. The culture plates were then placed in a dark, humidified chamber in an incubator at 26°C. Three independent experiments were performed and media was not changed during larval development. At various time points during development (4h - 55 h), 30 parasites from each sample were randomly selected using an inverted microscope and the percentage of parasites retaining all of their ciliated plates was recorded. Larvae were determined as being alive if they displayed either swimming or contractile movements, or if flame-cell flickering was visible .
Statistical analysis was performed using Minitab 15 Statistical Software; two sample t-tests or analysis of variance (ANOVA) were performed as appropriate.
activating transcription factor 2
Chernin's balanced salt solution
mitogen-activated protein kinase
phosphate buffered saline
protein kinase A
protein kinase C
tween-tris buffered saline.
We are indebted to Mike Anderson and Jayne King of the Natural History Museum (London) for the maintenance and passage of parasites. We would also like to thank Richard Giddens and Laura Grigis, Kingston University, for support with scanning electron microscopy.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.