As opposed to organism-based drug screening approaches, protein-based strategies have the distinct advantage of providing insights into the molecular mechanisms of chemical effectors and thus afford a precise targeting. Capitalising on the increasing number of genome and transcriptome datasets, novel targets in pathogens for therapeutic intervention can be identified in a more rational manner when compared with conventional organism-based methodologies. Trehalose-6-phosphate phosphatases (TPPs) are structurally and functionally conserved enzymes of the trehalose biosynthesis pathway which play a critical role for pathogen survival, in particular, in parasites. The absence of these enzymes and trehalose biosynthesis from mammalian hosts has recently given rise to increasing interest in TPPs as novel therapeutic targets for drugs and vaccines. Here, we summarise some key aspects of the current state of research towards novel therapeutics targeting, in particular, nematode TPPs.
Disease burden caused by parasitic roundworms (nematodes)
Parasitic worms (helminths) of humans, animals and plants cause destructive diseases of major socio-economic importance owing to their impact in agriculture, the environment and human health. Particularly roundworms (nematodes) have a long-term adverse impact on human health with >1 billion infected people  and the resultant morbidity surpassing diabetes and lung cancer in disability-adjusted life years (DALYs) . Parasitic nematodes, such as Onchocerca volvulus (acquired by mosquito bites; causing river blindness) and members of the genus Trichinella (acquired by consumption of raw/undercooked pork; causing trichinosis), can affect humans. Additionally, parasites of companion animals, such as Toxocara canis and Ancylostoma ceylanicum, are an increasing concern, since they can infect other animals and humans . Importantly, A. ceylanicum is one of the commonest hookworms infecting humans [4,5]. T. canis has a worldwide distribution and can cause considerable human infection rates and disease with a substantial human-health impact in some countries . Importantly, toxocariasis has been added to the list of prioritised neglected parasitic infections by the Centers for Disease Control (CDC) [7,8], because it has been reported to be a common parasitic disease in disadvantaged communities in the U.S.A. [9,10], often associated with blindness or increased susceptibility to allergies [11,12]. Furthermore, parasitic nematodes of livestock animals cause substantial suffering, death and/or substantial production losses, and plant–parasitic nematodes give rise to some 80 billion dollars of crop loss in the world every year .
Globally, the control of nematode infections relies heavily on drug treatment, but in many cases also requires improved sanitation and health education. Current anthelmintic drugs are often used across multiple species for the treatment of symptomatic infections and for large-scale prevention of morbidity, particularly of children living in endemic areas. These measures, combined with extensive use in closed livestock populations, have led to the development of increasing resistance to both broad- and narrow-spectrum treatments [14,15]. Therefore, the discovery and design of new chemotherapeutics are required to prepare against the threat from zoonotic diseases as global climate change allows parasitism to spread to new habitats . The dwindling sustainability of periodic deworming and the risk of further drug resistance call for immediate attention and development of novel immunotherapeutics to implement new control measures for neglected nematode diseases of humans and animals . The availability of vaccines for the immunoprotection of animals against parasitic diseases would, thus, constitute a substantial improvement of the current situation .
Drug discovery approaches for anthelmintics
Most of the chemotherapeutics (e.g. mebendazole, thiabendazole, pyrantel and others) currently used to treat parasitic worm infections were discovered more than 40 years ago using in vivo screening in animal models of parasitic diseases. In an effort to overcome the low throughput and concomitantly high costs of this approach, technological advances in the more recent years have led to high-throughput phenotypic screening methods (reviewed in ). Such efforts capitalise on the increasing availability of medium and large compound collections (e.g., see [18,19]), but are inherently limited by the requirement for further experimentation to identify mechanisms of action, particularly if medicinal chemistry approaches are to be rationally applied.
The recent substantial advancement in determination and increasing availability of genomic and transcriptomic datasets enables the deployment of advanced molecular approaches for drug discovery. Through mining of-omic datasets, novel protein targets can be identified whose modulation or inhibition by chemo- or immunotherapeutics may present novel strategies of disease control.
As opposed to organism-based drug screening approaches, protein-based drug discovery methods have the distinct advantage of providing insights into the molecular mechanisms of chemical effectors and thus afford more precise targeting. In particular, the identification of a target protein that is essential and present in multiple pathogenic species, but absent from the host, offers the possibility of developing therapeutics with broad-spectrum activity and minimal side effects.
Trehalose-6-phosphate phosphatase (TPP) is part of the biosynthetic pathway producing trehalose, an essential disaccharide for many micro-organisms that is neither required nor synthesised by vertebrate cells . Components of the trehalose biosynthetic pathway have, thus, been identified as targets for chemotherapeutic intervention. Although there are five different trehalose biosynthesis pathways known, and many organisms possess more than one such pathway, only the commonest of those (the so-called OtsAB pathway) is present in invertebrates . This pathway involves the enzyme trehalose-6-phosphate synthase (TPS) which catalyses the formation of trehalose-6-phosphate (T6P) from UDP (uridine diphosphate)-glucose and glucose-6-phosphate. The phospho group is then removed by TPP to yield trehalose [22,23]. In fungi, TPP functions as part of a cooperative multi-enzyme complex with TPS and the two regulatory proteins, TPS3 and TSL1 [24–26]. In contrast, in Escherichia coli — and possibly nematodes — both TPS and TPP are functional as isolated enzymes, although the expression of their corresponding genes, otsA and otsB, is tightly regulated .
Gene knockdown experiments in the free-living nematode Caenorhabditis elegans, as well as the blocking of the gene product in Mycobacterium tuberculosis, showed that accumulation of T6P leads to a lethal phenotype [28,29]. Thus, while disabling TPP in species that have more than one trehalose biosynthetic pathway may not halt trehalose production altogether, the inability to process T6P and subsequent build-up will still be toxic for the pathogen. Since TPP is conserved in many pathogenic species, including helminthic, protozoan and arthropod parasites, but absent from mammalian hosts, the enzyme fulfils all of the above criteria for a worthwhile drug target against multiple eukaryotic pathogens affecting humans and other animals.
Structurally, TPPs belong to the haloacid dehydrogenase (HAD) superfamily the hallmarks of which include a Rossmann α/β-fold, and an inserted cap module that is believed to regulate access to the active site. In a survey of key parasitic and bacterial pathogens, amino acid sequences of putative single-enzyme TPPs from >40 organisms were identified and clustered into three different groups based on obvious differences in structural topologies . Although the three-dimensional crystal structures of several single-enzyme TPPs from Thermoplasma acidophilum , Brugia malayi  and M. tuberculosis  have been determined, the only available experimental structural information about substrate or inhibitor binding is for fungal TPPs (Cryptococcus neoformans, Candida albicans ) with tandem topology where two HAD folds are covalently joined to form a functional trehalose synthesis complex. The characteristic HAD-core/cap structure and catalytic residues of TPP enzymes are largely conserved across a diverse range of species. However, the amino acid sequences that form this folded structure vary greatly between different species groups (Figure 1) with some TPPs having additional domains [30,32,33]. Thus, inhibitor specificity across species may possibly be tuned according to the regions of varying sequence, or — if a broad-spectrum inhibitor is required — be targeted at the conserved catalytic residues (Figure 1).
Information about residues involved in the catalytic mechanism has been obtained with the aid of transition state analogues, and, as discussed previously , these results have often incorrectly been projected to other TPPs, resulting in mis-assignment of catalytic activities to particular residue side chains. Intriguingly, the structure of wild-type TPP of M. tuberculosis (PDB accession number: 5gvx) features a substrate molecule, but since the structure shows an open conformation (i.e. the cap module is not in a close position relative to the core domain), there remain questions as to whether the state (and substrate binding pose) seen in this structure is enzymatically competent. An in-depth appraisal of this structure is impossible, as the atomic co-ordinates of this structure (PDB accession number: 5gvx) have not been released into the public domain, more than one year after publication .
The enzymatic activity of TPPs is highly specific for T6P and requires the testing of potential inhibitors in a competitive enzyme assay using the genuine substrate; only the TPP from T. acidophilum has moderate enzyme activity against the surrogate substrate p-nitrophenyl phosphate reported . Moreover, known phosphatase inhibitors, including EDTA, fluoride and vanadate which act by removing or blocking the active site magnesium ion, show no significant effects on TPP enzyme activity . On the one hand, the different topologies observed for TPPs from nematodes, bacteria and mycobacteria  suggest that molecules acting as competitive TPP inhibitors will likely differ between the different groups; on the other hand, the high degree of conservation of structural elements and amino acid sequence within the groups fuels the hope of broad-spectrum activity of a suitable inhibitor within the different groups.
Three methodologies might be employed for the discovery of potential TPP chemotherapeutics.
Structure-guided design of carbohydrate-based substrate mimics
Substrate mimics with high similarity to trehalose should be able to act as effector molecules of any type of TPP (see Figure 2); it has thus been a sound first attempt to test the sulphate analogue of T6P, which indeed showed the anticipated inhibition of TPP from B. malayi , albeit to a limited extent. Clearly, rationally designed inhibitors for TPPs require two main features: (i) a chelating group in place of the phosphate in 6-position of T6P anchoring the ligand to the protein-bound magnesium ion (and surrounding residues); and (ii) in place of the trehalose unit, a scaffold that is complementary to the ligand-binding interface between the HAD core domain and the cap module, decorated with several H-bond donor/acceptor groups for suitable ligand–protein interactions.
Possible binding pose of the substrate T6P in the active site of B. malayi TPP.
Synthetic carbohydrate chemistry is a notoriously difficult area, since the individual carbohydrate building blocks possess up to five different hydroxyl groups that require careful use of protection and de-protection steps during synthetic modification of the scaffold. In an attempt to simplify drug discovery of carbohydrate-based substrate mimics, the carbohydrate unit, believed to be positioned distally to the active site metal ion in the enzyme, has been replaced with an aromatic phenyl unit, giving rise to a set of aryl-D-glucopyranoside-6-sulphate derivatives . This change affords a larger variety of synthetic modifications and, generally, less challenging synthetic strategies, as fewer carbohydrate hydroxyl functions need to be considered. The inhibiting effects of octylphenyl-D-glucopyranoside-6-sulphate compounds  are promising indications that replacement of at least one glucosyl unit of the initial trehalose scaffold is possible.
Compound library screening
The typical approach used for the discovery of novel inhibitors with skeletons unrelated to the substrate (i.e. non-carbohydrate TPP inhibitors) is the screening of compound libraries. Such an approach may either comprise a one-step, high-throughput screening in an enzyme assay probing the inhibition of the catalytic reaction, or employ a two-step strategy in which a generic ligand-binding assay, such as differential scanning fluorimetry (also known as thermal melt assay), is used to identify compounds that interact with the target protein; only the subset of identified compounds is then subjected to an enzyme activity assay . Given the high substrate specificity of TPPs and the resultant requirement for the rather expensive T6P as a substrate in competitive enzyme activity assays, the latter, two-stage approach seems to be more economical for compound library screening in an academic setting. Based on the experiences from successful screening campaigns , compound libraries with a size of at least 10 000–20 000 compounds (and often up to hundreds of thousands of compounds) need to be screened in order to identify a novel lead molecule.
For TPPs, such efforts have not been reported thus far. Currently, known non-carbohydrate inhibitors of TPPs have been discovered by screening of smaller libraries, and include the antibiotic cephalosporin C and the anthelmintic closantel , albeit no TPP inhibitory parameters for the two compounds have yet been published. Additionally, inhibition of two mycobacterial TPPs by antibiotics of the moenzyme group has been reported: moenzyme (flavomycin) and diumycin (macarbomycin), two antibiotic complexes from Streptomyces spp., showed moderate inhibition of TPPs of M. smegmatis and M. tuberculosis , with estimated IC50 values of ∼25 and ∼70 μM, respectively.
Fragment-based drug discovery
Compared with compound libraries used for conventional screening, fragments are organic molecules of much smaller size (typically M < 300 g mol−1, as defined by Congreve's ‘Rule of Three’ , which sets compliance criteria for fragment-based drug discovery). Another major difference between conventional lead compound discovery and fragment-based drug discovery is that the latter approach puts less emphasis on the potency of a hit, but rather aims to identify hit molecules that optimally bind to a portion of the targeted binding site. In the event that fragment hits can be identified for multiple locations within the binding site, larger molecules that represent the covalently linked individual hits within one chemical entity can be synthesised. To the best of our knowledge, fragment-based drug discovery approaches have not yet been applied to TPPs as targets. However, we are of the view that this approach offers the opportunity to block conformational transitions within these enzymes and thus provides a successful route to ‘non-canonical’ TPP inhibitors. In a recent study that probed the enzyme kinetics of several TPPs, the observed kinetic parameters indicated a burst-like mechanism for all TPPs tested . In general, the observation of burst behaviour for TPPs was not entirely unexpected, as the enzyme forms a covalent intermediate with the substrate during the catalytic turnover. However, a detailed analysis showed that the burst amplitudes of the investigated TPPs were non-stoichiometric, which is indicative of multiple global conformational changes of the enzyme, with the most prominent one probably being the transition between the ‘open’ and ‘closed’ conformations of the cap module . We hypothesise that blocking these conformational transitions with a suitable small organic molecule would provide a means of enzyme inhibition outside of the realm of competitive inhibitors (i.e. substrate analogues). The fragment-based approach may be an excellent tool when exploring this avenue.
Clearly, progress in the discovery of novel TPP inhibitors will require further efforts using one or all of the three abovementioned approaches. Structure-based drug design, either starting from a substrate/inhibitor-bound or fragment-bound complex, will require efforts and investments into protein crystallography of TPPs in order to provide the crucial information that will guide the design of new lead compounds.
TPP as a possible vaccine candidate
Compared with drug treatment, prophylactic vaccination has distinct advantages for the control of infectious diseases as chemotherapeutics often do not prevent re-infection. In particular, for soil-transmitted helminths, there is an active search for recombinant subunit vaccines and the need for broad-spectrum activity has been recognised . The fact that trehalose biosynthesis is essential for parasite survival and that TPPs are absent from vertebrate host animals make these proteins attractive as vaccine targets. Immune response to foreign pathogens by the host is orchestrated by T lymphocytes which carry antigen-specific receptors on the cell surface and produce cytokines, the key hormonal messengers involved in immunological processes. T lymphocytes are classified into three groups, called CD4+ T-helper type 1 (Th1), type 2 (Th2) and type 17 (Th17). The cytokines produced by the Th1 and Th2 subsets tend to stimulate the production of that individual subset, but inhibit the development of the other (Th1/Th2 dichotomy); additionally, cytokines promoting the differentiation of Th17 suppress the development of Th1 and Th2 . It is currently thought that susceptibility to nematode infection is associated with an inflammatory Th1 immune response , whereas resistance to nematode infection is associated with a strong Th2 response . An optimal immunotherapeutic response would, thus, require balanced activities of Th1 and Th2  and take into account that the parasite may counteract the Th2 response (which diminishes its fitness and survival) through immunomodulation .
In a recent immunisation feasibility study using recombinant TPP from B. malayi, a nematode causing lymphatic filariasis, promising results in the form of a beneficial Th1/Th2 type response and protective immunity were obtained . Additionally, the antibodies generated by inoculation of mice with recombinant Bmal-TPP inhibited the TPP enzyme itself and were capable of binding to and eliminating multiple life-stages of the parasite .
The critical role of TPPs in pathogens and their absence from mammalian hosts makes this group of enzymes highly attractive therapeutic targets in parasitic, fungal and bacterial infectious agents. Because the OtsAB pathway is the only trehalose biosynthesis pathway in nematodes, TPPs of parasites are ideally suited for chemotherapeutic intervention, as the toxic accumulation of T6P as a result of TPP inhibition cannot be cleared by an up-regulation of a redundant pathway. Additionally, the encouraging results from preliminary immunisation studies in a mouse model of filariasis raise hope that TPPs might also serve as novel immunogens for the design of new vaccines.
Owing to increased infection rates and antiparasitic treatment resistance, there is an urgent need for new anthelminthic treatments in human, animal and plant populations.
TPP is essential for parasite survival but not required by vertebrates, making it an ideal therapeutic target.
Traditional drug discovery approaches in the form of large-scale compound and fragment library screening as well as rational inhibitor design can be applied to identify TPP inhibitors. These approaches are challenged by the enzyme's exquisite substrate specificity and the inherent difficulties of carbohydrate synthesis, respectively. Early attempts have, thus far, shown only moderate success.
TPP shows promise as a vaccine target against parasitic nematodes.
In combination with its omnipresence in invertebrate populations and validation as a drug target, this demonstrates the great potential of TPP as a target for the prevention and treatment of helminthic diseases.
Research in the investigators' laboratories is funded by the Australian Research Council, the National Health and Medical Research Council (A.H. and R.B.G.), the Rebecca L. Cooper Medical Research Foundation (A.H.) and Chonnam National University [2015-0597, J.S.K.].
We gratefully acknowledge the Equity Trustees PhD Scholarship and Australian Government Research Training Program Scholarship (M.C.), as well as support by DAAD RISE and PROMOS scholarships (S.B.).
The Authors declare that there are no competing interests associated with the manuscript.