Evidence of a M1-muscarinic GPCR homolog in unicellular eukaryotes: featuring Acanthamoeba spp bioinformatics 3D-modelling and experimentations.

Acetylcholine affects the target cellular function via muscarinic and nicotinic cholinergic receptors that are seen to exist in humans. Both the cholinergic receptors are G-protein coupled receptors (GPCRs) that perform cardinal functions in humans. Anti-muscarinic drugs, particularly the ones that target M1 subtype (mAChR1), have consistently shown to kill unicellular pathogenic eukaryotes like Acanthamoeba spp. As the M1 receptor subtype has not been reported to be expressed in the above protist, the presence of an ancient form of the M1 muscarinic receptor was inferred. Bioinformatic tools and experimental assays were performed to establish the presence of a ligand-binding site. A search for sequence homology of amino acids of human M1 receptor failed to uncover an equivalent ligand-binding site on Acanthamoeba, but structural bioinformatics showed a hypothetical protein L8HIA6 to be a receptor homolog of the human mAChR1. Immunostaining with an anti-mAChR1 antibody showed cellular staining. Growth assays showed proliferation and lethal effects of exposure to mAChR1 agonist and antagonist respectively. With the recent authentication of human mAChR1 structure and its addition to the database, it was possible to discover its structural analog in Acanthamoeba; which could explain the effects of anticholinergics observed in the past on Acanthamoeba spp. The discovery of a receptor homolog of human mAChR1 on Acanthamoeba with future studies planned to show its expression and binding to cholinergic agonist and antagonist would help clarify its role in the biology of this protist pathogen.


Introduction
The mammalian cholinergic nervous system has been well established in the context of its ligand, receptor subtypes and their effects on human body functions. The neurotransmitter of the cholinergic nervous system is acetylcholine (ACh), this chemical is released by somatic nerves on skeletal muscles and by autonomic nerves mostly on glands and smooth muscles. ACh is a ligand at nicotinic (nAChR) and muscarinic receptors (mAChR) in humans, both of which are GPCRs (1,2). The nAChR subtypes are nAChR1, nAChR2 and mAChR, which are further sub-classified into five types (M1-M5), namely, mAChR1-5. The tissue distribution of both nicotinic and muscarinic subtypes varies and is mostly related to somatic and autonomic nerve distribution (1,3) though in rare instances they are found in non-neuronal sites as well (3,4). ACh has recently been shown to be present in non-neural sources in human tissues like lungs and immune cells, as well as in malignant tumors like colon cancer and lung cancer (5) by affecting nAChRs and in case of prostate cancer via mAChRs (6,7). The signals originating from mAChRs have been shown to influence cellular functions like smooth muscle contraction via subtype 1 (mAChR1-M1), relaxation (mAChR1-M2) and stimulation of glandular secretion by the M1 and inhibition via M2 (1). During evolution, mAChRs have been shown to remain conserved, all the way from Drosophila to humans (8).
The literature search showed that mAChR can be traced back to insects and small freshwater animals like Hydra (9). There have been reports of the presence ACh in green microalga where they are thought to be important for formation/storage of compounds like lipids and nutritionally essential fatty acids like alpha lenolenic acid (10).
Acanthamoeba has remained a model for studying eukaryotic cellular mechanisms and processes and has shown greater insights into the roles of GPCR in different metabolic and survival processes. The A. castellanii has some downstream cascades synchronized with the GPCRs and hypothetical proteins (11). A comparison of downstream pathways showed that the human mAChR signaling pathways that converge via PI3K, PIP3 to AKT to promote proliferation are also present in Acanthamoeba spp. (12,13). The literature search showed that most of the G-proteins that are known to be present in A. castellanii are hypothetical proteins or belong to an orphan class of GPCR, the ligand of which is yet to be determined. Clues towards the existence of different GPCRs expressed by Acanthamoeba spp. came from few in-vitro studies (14,15) and similar experiments that followed (15,16). The drugs used in most of these studies are known to act as antagonists at human GPCRs like mAChR, 5HT, opioid receptors and mGluR. As amoebicidal and cysticidal effects were observed on Acanthamoeba on the exposure to these drugs, a ligand-binding site was suspected in this protist pathogen. The homology and expression of GPCRs like nAChR or mAChR in Acanthamoeba has yet to be validated. There has been no information about the presence of a cholinergic system and related mAChRs or nAChRs on pathogenic eukaryotes like A. castellanii.
We assumed the presence of a possible homolog of mAChR1 in A. castellanii. This model of ligand binding on Acanthamoeba has to be compared with human cholinergic transmission by this study using bioinformatics, structure activity relationship, 3D modeling, immunostaining and growth assays; which this study aims to do.

Materials and methods
Anti-human-mAChR1 receptor antibody directed against the mAChR1 receptor was obtained from Millipore-Merck R . Sigma-Aldrich R products like ACh/pilocarpine (agonist), atropine, dicyclomine and pirenzepine (antagonist) were used to target muscarinic receptors and its subtypes. Phyre 2 , a 3D protein structure prediction tool (17) that is licensed under Creative Commons Attribution-NonCommercial-2.0, was used for modeling and 3D structure predictions. SWISS-MODEL database, which is a web-server dedicated to homology modeling of three dimensional (3D) structures) of proteins, was used to build the 3D protein structure (18). This server has a pipeline that uses BLASTp and has free protein function and a protein structure prediction server (Hhblits) to identify templates. National Center for Biotechnology Information (NCBI), UniProtKB and Amoebadb.org were used to fetch the amino acid sequences human and amoebal proteins, alignments, and pairwise alignment homology.

Acanthamoeba genotype and culture methods
An ocular A. castellanii, acquired from a corneal ulcer (Keratitis), of T4 genotype were grown in a nutrient medium (PYG medium), as described earlier (14). Healthy trophozoites that were attached to the bottom of the flasks were selected for all the experiments.

Bioinformatics sequence homology
Available data banks of A. castellanii at NCBI and Amoebadb. org were blasted by blast search types like BLASTn (nucleotide blast) and BLASTp (protein blast) for fetching FASTA sequence of L8HIA6. For the human mAChR1 and L8HIA6 amino acid comparisons, NCBI, European Molecular Biology Labs (EMBL) -Emboss matcher (19), UniProtKB (20) and Wellcome R Trust-Genome online databank were configured for pairwise amino acid alignment and sequence homology determination.

Structural bioinformatics and ligand-binding predictions
The publicly accessible genome and proteomics used in this study were searched and downloaded from databases at the (NCBI), UniProtKB, the Joint Genome Institute and European Molecular Biology Labs (EMBL). Structure activity relationship determination services available at SWISS MODEL and Phyre 2 were used to define the 3D model specifics by submitting the amino acid sequence (FASTA) of L8HIA6. A templatebased model was built to show any homology to known human proteins and GPCRs.
The relationship between the docking of anticholinergic drugs, the 3D model of mAChR1 receptor and Acanthamoebal L8HIA6 was built to understand the effects of ACh, atropine and dicyclomine on A. castellanii. Amino acid FASTA series of Acanthamoeba proteins were fed into the search engines to generate model structures for homology of GPCRs in Acanthamoeba with human proteins. For ligandbinding predictions, models generated by SWISS MODEL and Phyre 2 database were submitted to 3DLigandsite database. The heterogens, ligand, clusters of binding and amino acid residues were compared.

Immunostaining
Trophozoites forms of the Acanthamoeba were grown on cover slips in PYG overnight at 30 C. The cells were fixed with paraformaldehyde (4%) in 0.1M phosphate buffer saline (PBS) for 30 min. Endogenous peroxidase activity was quenched by incubation with 0.2% hydrogen peroxide in 0.1M phosphate buffer saline (PBS) pH 7.3 containing 0.2% Triton X-100 for 25 minutes at room temperature. After three washes with blocking solution (50 mL of phosphate buffered saline 0.02M PBS, pH 7.4/casein 2%), the cells were immersed with primary mAChR1 antibody (Merck R -Millipore) directed against rat and human mAChR1 were incubated for 1 h at 37 C and then refrigerated overnight. Washing was then done thrice with 0.02M of phosphate buffered saline. The reactivity of immune complexes was identified after coincubation with horseradish peroxidase tagged conjugated goat anti-rabbit antibody (Chemicon Catalog # AP132P) for 1 h at room temperature and then refrigerating overnight. The trophozoites were then incubated with a solution of diaminobenzidine (DAB) at the concentration of 0.0125%, containing 0.05% nickel ammonium sulfate for 10 min at room temperature. Cells were then washed with 0.02M PBS, 4x for 10 min and mounted on coated glass slides and dehydrated in an ascending series of ethanol concentrations as per manufacturer's instructions. Inverted microscope (Olympus) was used to obtain images. The slides were observed at 10x and 20x magnifications. For positive controls, neuron and smooth muscle cells were stained and fat cells we used as negative controls.

Growth and amoebicidal assays
We incubated A. castellanii with mAChR1 agonists and antagonist to determine the proliferative and cytotoxic effects respectively. 1 Â 10 5 trophozoites of A. castellanii were grown in nutrient (PYG) medium, non-nutrient medium (PBS) and PG (without yeast) with 10-15 mM of Pilocarpine and PBS alone in 24 well plates at 37 C for 24 h. A. castellanii were incubated with atropine, dicyclomine and pirenzepine to observe the effects of these drugs on growth and survival. Inverted microscope was used to capture high resolution pictures at variable magnification after inoculating A. castellanii with cholinergic agonist and anticholinergic drugs.

Results
Nucleotides and amino acid sequence homology Acanthamoeba L8HIA6 and human M1 muscarinic receptors have a nucleotide (mRNA) homology A pairwise nucleotide alignment of the mRNA of Human M1 muscarinic (mAChR1) and L8HIA6 was done by downloading the FASTA sequence of the mRNA of both the proteins. Pairwise alignment showed 47% of identities and a similar percentage of similarity (Figure 1), with a total score of 1889. The human nucleotide sequence is in the top row and amoebal nucleotides are in the bottom row ( Figure 1).

Acanthamoeba L8HIA6 and human M1 muscarinic receptors have an amino acid sequence homology
An alignment for sequence homology of amino acids of the human M1 muscarinic receptor (mAChR1) with amoebal L8HIA6 was done by the UniProtKB database. A total of 60 identities was found with 79 similarities between both the proteins (Figure 2(A)). Summary of L8HIA6 protein shows that like M1 muscarinic receptor it has a 7 transmembrane (7tm.) structure. Pfam details of L8HIA6 show that it is a G-protein coupled receptor with downstream signaling properties of GPCRs (Figure 2(B)). The 7tm structure of GPCRs is known in species like humans and ameba like Naegleria gruberi and A. castellanii (Figure 2(C)).
Structural bioinformatics: template-based 3D structure homology Acanthamoeba L8HIA6 has a 3D structure similar to the human M1 muscarinic receptor To see if a template-based model could be generated, the L8HIA6 amino acid sequence was submitted to the SWISS MODEL and Phyr 2 databases. A model 5cxv.

Acanthamoeba L8HIA6 and human M1muscarinic receptors have identical ligand binding attributes
The 3DLigandsite and protein database (PDB) were used to predict the ligand binding potential of L8HIA6 and M1 muscarinic receptor. A remarkable ligand-binding prediction homology was shown for mAChR1 and L8HIA6. Both the GPCRs showed identical heterogens like NAG, 13 clusters and ligand (Figure 4(A) and (B)). The predictions were generated with 100% confidence for both GPCRs. The amoebal L8HIA6 has 3 predicted binding sites ( Figure 4A) for the ligand as compared to a single site on human mAChR1 (Figure 4(B)).

Immunostaining
Acanthamoeba castellanii T4 genotype stained positive with anti-muscarinic M1 antibody Immunostaining with Anti mAChR1 antibody prepared against rat and human mAChR1 was used on Acanthamoeba trophozoites, which showed positive staining at the cell membrane and cytoplasm ( Figure 5(A) and (B)). The anti-mAChR1antibody binds to a portion of the conserved (i3) intramembranous part of the mAChR1. This is a segment of 127 amino acids (Figure 6(A)) against which the antibody is directed. A strong positive reactivity zone was seen at the cell membrane region of the trophozoites ( Figure 5(A)). A sequence alignment of L8HIA6 with the i3 segment amino acid of mAChR1 showed patchy areas of similarities ( Figure 6(B)). Normal smooth muscle cells were stained with anti-human mAChR1 antibody in controls ( Figure 5(C)). Normal fat cells, which are known to be negative for mAChR1, did not show staining with anti-mAChR1 antibody ( Figure 5(D)).

Growth assays
Effects of muscarinic agonist and antagonist drugs:

Pilocarpine causes proliferation of Acanthamoeba spp
The cholinergic agonist Pilocarpine showed growth promoting and proliferative influence on Acanthamoeba. A. castellanii trophozoites exhibited proliferation even in the absence of yeast as nutrient in proteose peptone and glucose medium (PG þ Pilocarpine) due to survival stimulus of the muscarinic agonist Pilocarpine (Supplementary File-S1).

Antimuscarinic drugs inhibit Acanthamoeba proliferation
Anticholinergic drugs showed differential effects on A. castellanii. When compared to the counts in PYG medium (Figure  (Figure 7(D)), the nonselective muscarinic receptor antagonist atropine (125 lg/ml) did not show any noteworthy effect on trophozoite survival (Figure 7(C)). Dicyclomine, a comparatively more specific mAChR1 antagonist revealed inhibition of proliferation at doses of 125 lg/ ml (Figure 7(B)).

Highly specific M1-receptor antagonist pirenzepine showed substantial inhibition of Acanthamoeba proliferation
The most specific M1 receptor antagonist Pirenzepine was tested on Acanthamoeba trophozoites for 12 to 24 h at a dose of 150-200 lg/ml (Figure 8). Compared to PYG (Figure 8(A)) and solvent controls (Figure 8(F)) there was a remarkable reduction in the trophozoite count at the 24th hour (Figure 8(C-E)). The trophozoites were stained with Trypan Blue (images not shown) reflecting the lethal effects of Pirenzepine.

Discussion
In this study, we investigated A. castellanii to provide the evidence of the presence of an ACh binding site, that could be a possible target of anticholinergic drugs used in this study and in the past (14,15). With the evidence of ACh in very primitive species like Micrasterias denticulata and Laurencia obtuse, where they are needed for cell growth and proliferation (10), this study was aimed to establish the presence of an ACh binding receptor on Acanthamoeba spp. Based on the findings listed below, we show the existence of a ligand binding site in A castellanii. Our cardinal findings include: i. Identification of a protein L8HIA6 which is a structural homolog of mAChR1. ii. Identical ligand-binding clusters and heterogens of L8HIA6 and mAChR1. iii. Positive immunostaining of Acanthamoeba with antihuman M1 receptor antibody. iv. A proliferative response in Acanthamoeba to M1 agonist in the growth assays. v. An anti-proliferative effect of M1 antagonist drugs in Acanthamoeba spp.
The evidence of the presence of all of the components of the cholinergic transmission cascade like synthetic and degradation enzymes in A. castellanii implored us to search for a muscarinic receptor expressed on this unicellular eukaryote for the ligand acetylcholine. Additionally, this discovery was expected to provide an explanation for the inhibitory effects of the anticholinergic drugs (particularly M1 specific ones) on A. castellanii which we observed in the past (14). At the beginning, we searched for a protein at Acanthamoeba databases that had a sequence homology with the amino acids of human -M1 muscarinic receptor and its i3 intramembranous part, but we were not able to find a match at NCBI database (Supplementary file S4). Gene regulation and expressions are occasionally alike between different species (21), but our exhaustive genomic and protein searches failed to find any proteins with significant identity rates, which understandably appears to be due to the vast evolutionary distance between both the species. In our attempt to search for a primitive muscarinic receptor on Acanthamoeba we compared many GPCRs and nearly 126 hypothetical proteins (data not shown), but none of them had a substantial homology to the human cholinergic mAChR receptors. The proliferation on exposure to agonists and growth inhibition on exposure to M1 receptor antagonist kept baffling us, but despite failed attempts to find a protein similar to muscarinic receptor on Acanthamoeba spp, we attempted to seek an alternative logical explanation for the effects of the agonist and antagonist mentioned above. The fact that only limited amino acids are required to be placed in a constellation for ligand interaction at GCPRs and that muscarinic receptors have also been shown to form homodimers structural complexes for ligand binding (22), inspired us to use structural bioinformatics to solve the paradox. We got back to look for the known G-Protein receptors and an array of hypothetical proteins in Acanthamoeba spp. to scan them for a constellation of amino acids needed for ligand binding and structural homology with human muscarinic receptors. Additionally, this time our research was also focused only on limited but structurally related hypothetical  proteins in Acanthamoeba spp., which had similar evolutionary origins with human muscarinic receptors.
We were able to uncover a hypothetical protein in Acanthamoeba spp., L8HIA6 coded by gene ACA1_153000, which showed an identical constellation of amino acids needed for cholinergic ligand binding ( Figure 4) and a substantial structural homology (Figure 3) with the human mAChR1. Important findings of our study show that the gene which codes for the G-Protein L8HIA6, like the human mAChR1, belongs to the GPCR_A superfamily ( Figure 2). The nucleotide (mRNA) sequence homology ( Figure 1) and a 7tm structure with G-Protein coupled receptor activity originating from cell surface signaling (Figure 2(B)) are its similarities with human mAChR1. For structural bioinformatics 3D model building, we submitted the amino acid FASTA sequence of L8HIA6 to the databases that generated a template-based model, which is nearly identical to human mAChR1 ( Figure  3). The model template 5cxv.1.A showed a pairwise alignment with human M1 muscarinic receptor ( Figure 3) and a ligand docking site with conserved amino acid residues ( Figure 3). These residues that are unique to the M1 muscarinic receptor, were also found to be present in amoebal L8HIA6 template 5cxv.1.A. The zoomed 3D structure of the model (Figure 3 black circles) revealed the above conserved constellation of amino acids, which are known binding sites of ligand such as Tiotropium and N-Methyl-Scopolamine (23). Extensive exploration (24) for additional ligand and heterogens for human M1 muscarinic receptor and L8HIA6 showed identical heterogen NAG, 13 ligand and source structures (Figure 4).  The last challenge for us was to show that in addition to structural homology and ligand-binding similarity, and whether there is was sufficient evolutionary homology of L8HIA6 with the conserved i3 segment of the human mAChR1. Immunohistochemistry done with anti-mAChR1 antibody showed reactivity over the cell surface and cytoplasmic regions of Acanthamoeba spp ( Figure 5). The M1 muscarinic receptors are located over the cell membrane in eukaryotic mammals, but they have also been reported to be present in the cytosol (25). A beta-arresting assisted autophagic vacuole could have been the reason for the cytoplasmic staining in our study. We compared the L8HIA6 with the i3 conserved region (target of the antibody used) for homology to find patchy regions of pairing ( Figure 6) in the sequence that could be a plausible explanation of its reactivity with the anti mAChR1 antibody used for immunostaining.
In growth assays, we showed that Pilocarpine caused proliferation of Acanthamoeba even in the non-nutrient medium (Supplementary File-S1) that has shown to prevent the process of encystation. Growth assays showed an inhibitory effect of the nonspecific and specific mAChR antagonist (Figure 7) on Acanthamoeba spp. New in this study was the use of Pirenzepine on Acanthamoeba. Unlike other antagonists, Pirenzepine is a specific antagonist of the human M1 receptor (26). In 24 h, 150-200 lg/ml of this drug tested on a 1 Â 10 6 trophozoites of Acanthamoeba was able to substantially kill ($92%) them ( Figure 8).

Conclusion
Evolution of GPCRs, like muscarinic receptors, from primitive eukaryotes to human beings must have gone through complex changes in the number of amino acids and their structure. Keeping in mind the unchanged chemistry of their ligand, the amino acid residues that engage the ACh must have retained a consistent constellation over billions of years of the evolutionary period. We showed a primitive Acanthamoebal Muscarine Binding (AMB) M1 receptor homolog in Acanthamoeba spp. and future studies are planned towards cloning the L8HIA6 gene, showing its expression and showing reaction with the specific antibody using fluorophores. We also seek to establish the nature of proteins coupled with G subunits and the use of gene knockout of ACA1_153000 to show the role of L8HIA6 amoebal receptor in the biology of Acanthamoeba spp. Bioinformatic tools have real strength for uncovering microbial GPCRs and subsequently, exploiting these receptor functions could be beneficial in diseases caused by them.