Introduction
‘The blood is the life!’ Dracula (Stoker, Reference Stoker1897)
Caenorhabditis elegans is a nonparasitic free-living nematode (Sulston, Reference Sulston1976). It has become a powerful tool to model complex biological processes in genetics (Brenner, Reference Brenner1974), neurology (Chalfie et al., Reference Chalfie, Tu, Euskirchen, Ward and Prasher1994) and cell survival (Adams and Cory, Reference Adams and Cory1998), due to its relative ease of growth and maintenance (Stiernagle, Reference Stiernagle2006). More recently, C. elegans has been identified as a suitable model to study parasitic behaviour (Crisford et al., Reference Crisford, Ludlow, Marvin, Kearn, O'Connor, Urwin, Lilley and Holden-Dye2013) and facilitate the discovery of anthelmintic drugs (Burns et al., Reference Burns, Luciani, Musso, Bagg, Yeo, Zhang, Rajendran, Glavin, Hunter, Redman, Stasiuk, Schertzberg, McQuibban, Caffrey, Cutler, Tyers, Giaever, Nislow, Fraser, MacRae, Gilleard and Roy2015).
Parasites infect a quarter of the global population (Bethony et al., Reference Bethony, Brooker, Albonico, Geiger, Loukas, Diemert and Hotez2006). Infections have a negative impact on human health and productivity (Keiser and Utzinger, Reference Keiser and Utzinger2010) as well as economic output, due to the infection of crops (Fuller et al., Reference Fuller, Lilley and Urwin2008) and livestock (Besier, Reference Besier2007). Bearing this in mind international initiatives, such as the Human Hookworm Vaccine Initiative targeted against Necator americanus (Sabin Vaccine Institute, 2014), have been established to develop and test vaccines to prevent infection of humans. The complex life-cycles of parasitic nematodes, which rely on a host for propagation (Chauhan et al., Reference Chauhan, Scurr, Christie, Telford, Aylott and Pritchard2017), may serve as a barrier to the development of therapeutics to prevent and treat infections (Holden-Dye and Walker, Reference Holden-Dye and Walker2007). Therefore, C. elegans, which has an easily maintained lifecycle (Lightfoot et al., Reference Lightfoot, Chauhan, Aylott and Rödelsperger2016) that is independent of host interaction, may provide an alternative model to study parasitic infections.
In the present study, we investigated if C. elegans could ingest and then survive on a diet of human erythrocytes. These experiments were performed as a prelude to nominating a hematophagous C. elegans as a model to further understand haem metabolism in nematodes, coupled with the interrogation of immune responses to vaccines currently under development, and to identify new vaccine candidate molecules involved in the intestinal biochemical pathways of hematophagous nematodes.
Aspartic proteinases (APRs) and glutathione-S-transferase (GST) have assumed prominence in vaccine development due to their ability to digest haemoglobin and neutralize the toxic by-products of haemoglobin digestion, respectively. In this context, the current vaccine under development to combat necatoriasis is bivalent (Hotez et al., Reference Hotez, Bethony, Diemert, Pearson and Loukas2010), comprising of an aspartic haemoglobinase (Na-APR-1) and a GST (Na-GST-1) (Brophy and Pritchard, Reference Brophy and Pritchard1992; Brown et al., Reference Brown, Burleigh, Billett and Pritchard1995; Williamson et al., Reference Williamson, Brindley, Abbenante, Prociv, Berry, Girdwood, Pritchard, Fairlie, Hotez, Dalton and Loukas2002). These enzymes function in tandem in the hookworm gut to process human haemoglobin then detoxify haem. Furthermore, neutralizing antibodies raised to the hookworm enzymes have been linked to the protective capacity of the Na-APR-1 component of the vaccine, indicating that sequence identity around the actives sites and other epitopic regions of the C. elegans and N. americanus enzymes could be of immunological relevance (Pearson et al., Reference Pearson, Pickering, Tribolet, Cooper, Mulvenna, Oliveira, Bethony, Hotez and Loukas2010). Therefore, the demonstration of hematophagy in C. elegans would pave the way for unambiguous experiments to test the modes of action of these neutralizing antibodies and to search for new gastrointestinal tract associated vaccine candidates.
In order to observe the haematophagic C. elegans, erythrocytes were harvested from a human blood donor. Erythrocytes were labelled with the fluorophores [caboxyfluorescein succinimidyl ester (FAM-SE), fluorescein isothiocyanate (FITC) and tetramethylrhodamine succinimidyl ester (TAMRA-SE)] to facilitate the visualization of erythrocyte ingestion and digestion. The viability of C. elegans fed on erythrocytes alone, when compared with nematodes fed on E. coli alone and a mixture of erythrocytes and E. coli, was monitored as a function or their motility. Furthermore, databases were screened to identify if C. elegans, like the parasitic N. americanus, translated proteins capable of the enzymatic processing of haemoglobin.
Experimental materials
C. elegans Bristol N2 and E. Coli OP50 were purchased from Caenorhabditis Genetics Center (CGC). Agar, protease peptone, cholesterol, gentamicin sulphate, EDTA, Alsever’s, Dulbecco’s and sodium hypochlorite solutions were obtained from Sigma-Aldrich (Gillingham,UK). Fluorescein isothiocyanate, carboxyfluorescein-SE, TAMRA-SE and BD Vacutainer® blood collection tubes were obtained from Thermo-Fisher- Scientific (Loughborough, UK). Blood was donated by DIP (blood group B Rh negative). Deionized water (18.2 MΩ) was generated by Elga Purelab Ultra (ULXXXGEM2).
Methods
Blood collection
BD Vacutainer® tubes containing ethylenediaminetetraacetic acid (EDTA) (1.8 mg mL−1 of blood) were used to collect blood from a healthy volunteer (4.5 mL). Initially, centrifugation was used to separate erythrocytes from plasma and leucocytes (800 rpm, 8 min). Erythrocytes were resuspended and washed a further 3 times in Alsever's solution using centrifugation (800 rpm, 8 min); discarding the supernatant, containing plasma and EDTA, and any remaining leucocytes and after each wash. After the final wash, the pellet was made up to 10 mL with Alsever's solution (~ 6 × 108 cells mL−1) and stored at 4 °C.
Establishing erythrocyte concentrations
Aliquots of the stock solution (10 µL) were serially diluted with Alsever's solution (90 µL), to create solutions at concentrations 6 × 108, 6 × 107, 6 × 106, 6 × 105, 6 × 104, 6 × 103, 6 × 102 6 × 101 and 6 × 100cells mL−1. Microscopy was used to determine a cell density that would permit free imaging of both erythrocytes and C. elegans.
Fluorescent labelling of erythrocytes
Erythrocytes (500 µL, 3 × 108 cells mL−1) were added to fluorescein isothiocyanate, carboxyfluorescein-SE or TAMRA-SE in Dulbecco's phosphate buffered saline solution (1 mg mL−1, 500 µL, pH 8.0) and delicately inverted [2 h, 1 rpm, HulaMixer™ (Invitrogen), 4 °C]. Erythrocytes were washed with Dulbecco's solution (15 × times, 10 mL−1) and collected with centrifugation (800 rpm, 8 min). Labelling was confirmed with fluorescence microscopy using a Nikon Eclipse TE300 equipped with a Plan Fluor 40 × 0.75 NA objective and CoolLED pE-4000 and pE-100 light source. Labelled erythrocytes were suspended in Alsever's solution and stored in a light protected container at 4 °C.
Signal-to-noise ratio
The signal-to-noise ratio for erythrocytes was determined by drawing a line profile at the centre of erythrocytes labelled with either FAM-SE, FITC and TAMRA-SE. A Nikon Eclipse T1 and QIMAGING optiMOS camera equipped with CoolLED pE-4000 fluorescence illumination and pE-100 bright field illumination and 10 × (0.30 NA) objectives were used to image samples. Fluorescence was captured through excitation at 490 nm and 540 nm collecting at emission between 519 ± 26 nm 595 ± 33 nm (Exposure times 100 µs, LED power 50%). Images were analysed with FIJI open source software.
Growth and maintenance
Caenorhabditis elegans were maintained on NGM agar and Escherichia coli (OP50) at 20 °C. Synchronized growth cycles of C. elegans were prepared by harvesting eggs from gravid nematodes (Stiernagle, Reference Stiernagle2006). Briefly, high numbers of gravid nematodes were collected by rinsing NGM agar plates with ultra-pure sterile deionized water (4 mL). Sodium hydroxide (5 M, 0.5 mL) and sodium hypochlorite (5%, 1 mL) was added to the nematode suspension and vortexed (10 min) to release eggs and eliminate bacterial traces. The eggs were pelleted using centrifugation (1500 rpm, 1 min) and the supernatant discarded. The eggs were washed with ultra-pure sterile deionized water and collected using centrifugation (1500 rpm, 1 min). The egg suspensions were aspirated to 0.1 mL and C. elegans and were plated on freshly prepared NGM agar seeded with an E. coli lawn and incubated at 20 °C. The generation time of C. elegans under these conditions was 4–5 days.
Dosing C. elegans with erythrocytes
Synchronized nematodes were collected by washing NGM plates with sterile deionized water. C. elegans were washed in deionized water (10 mL) and collected using centrifugation (3 times, 1500 rpm, 1 min). Nematodes were re-suspended in gentamicin (500 µg mL−1, 10 mL, 30 min) to remove traces of E. coli. C. elegans were washed again in sterile deionized water (10 mL) and collected using centrifugation to remove traces of gentamicin (3 times, 1500 rpm, 1 min). Pelleted C. elegans were added to fluorescently labelled erythrocytes (1 × 106 cells mL−1, TAMRA). Observations of erythrocyte ingestion were made using fluorescence microscopy using AMG F1 Microscope equipped with an AMG Plan Fluor 10 × 1.2 NA objective and Epiphan DVI2USB 3.0 (30 fps, 1920 × 1200 pixels) to capture video. Aliquots of nematode and blood suspending media (10 µL) were added to sterile tryptone soya broth media and incubated overnight to check for microbial contamination.
Motility fraction half-life (Mft50) derivation
To help describe the viability of nematode populations, using motility of nematode population as an indicator, the Mft 50, the time required to reduce the motility of population of nematodes by 50%, was derived for C. elegans treated with (1) erythrocytes, (2) E. coli, and (3) erythrocytes and E. coli. For each data point, 100 nematodes were evaluated in triplicate and the standard deviation was calculated. Statistical analyses were conducted using Student's t-test, where P < 0.05 indicated statistical significance.
Protein sequence analysis
The known sequences for N. americanus aspartic haemoglobinase (Necepsin II/Na-APR-1, Uniprot Q9N9H3) and GST (Na-GST-1, Uniprot D3U1A5) were searched on UniProt for sequence identity in C. elegans. Sequences were aligned using the National Institute for Health's National Center for Biotechnology Information blastp tool. The sequences were annotated, where available, with respect to the signal peptide sequence, active site aspartic acids using Clustal Omega.
Results
Fluorescent labelling of erythrocytes
In order to observe the haematophagic C. elegans, erythrocytes were harvested from a human blood donor. Erythrocytes were separated from whole blood by centrifugation, using EDTA as an anticoagulant and stored in Alsever's solution, to enable preservation and long-term storage of erythrocytes (Li et al., Reference Li, Glover, Szalai, Hollingshead and Briles2007).
C. elegans and erythrocytes, when visualized using a brightfield microscope, are optically transparent, such that only refracted light, due the curvature of the nematode anatomy and torus geometry of the red blood cell, permits their visualization. Therefore, to enhance the contrast between C. elegans and erythrocytes and to augment visualization of hematophagy events using fluorescence microscopy, red blood cells were fluorescently labelled with either FAM-SE, FITC or TAMRA-SE. Succinimidyl esters and isothiocyanates readily conjugate to biological protein rich structures that contain amine functional groups, typically found in lysine residues, via stable carboxyamide and thiourea bonds, respectively (Haugland, Reference Haugland2005).
FAM-SE, FITC and TAMRA-SE were all able to label erythrocytes (Fig. S1). TAMRA-SE demonstrated highly effective labelling of erythrocytes, Fig. 1A. This is because TAMRA-SE labelled erythrocytes, when subjected to the same excitation power and exposure time for imaging, demonstrated, 1.6 × and 6.7 × greater signal to noise ratio, when compared with FAM-SE and FITC labelled erythrocytes, respectively (Fig. 1B). These observations could be attributed to a combination of factors, which include greater labelling efficiency and stability of succinimidyl esters, when compared with isothiocyanates (Banks and Paquette, Reference Banks and Paquette1995), and the superior quantum yield of TAMRA-SE in comparison with FAM-SE and FITC (Haugland, Reference Haugland2005). Furthermore, the utility of TAMRA-SE labelled erythrocytes permits imaging of C. elegans in the absence of unwanted substantial age-related green lipofuscin auto-fluorescence (Forge and Macguidwin, Reference Forge and Macguidwin1986; Pincus et al., Reference Pincus, Mazer and Slack2016) (Fig. 1C) and unwanted blue excitation light dependent phototaxis (Ward et al., Reference Ward, Liu, Feng and Xu2008), thus augmenting imagining capabilities of haematophagic events. This is because lipofuscin auto-fluorescence could generate unwanted imaging artefacts that could be interpreted as internalized erythrocytes. In addition, blue light, which is used to excite FAM-SE and FITC fluorescence emission, initiates heightened light dependent C. elegans motility that renders continual high frame imaging of haematophagic events challenging.
Fig. 1. (A) Brightfield, fluorescent red (TAMRA-SE) and merged images for TAMRA labelled red blood cells. (B) Cross-sectional intensity line plot across the centre of red blood cells labelled with carboxyfluorescein succinimidyl ester (FAM-SE), fluorescein isothiocyanate (FITC) and tetramethylrhodamine succinimidyl ester (TAMRA-SE). (Ci) Brightfield, (Cii) fluorescent red and (Ciii) merged image of C. elegans. Scale bars = 100 µ m.
Observation of haematophagic events
To permit effective visualization of red blood cells and nematodes in a single field of view, erythrocyte counts were performed by serially diluting the stock solution. A concentration of 1 × 106 cells ml−1 was identified as a concentration that would enable observation of hematophagy events using fluorescence microscopy.
The occurrence of hematophagy was confirmed by feeding erythrocytes to axenic C. elegans for up to 24 h. Internalized erythrocytes were visualized in the pharynx and intestinal tract of nematodes after 5 h, Fig. 2. After 24 h, virtually all red blood cells had been consumed. From our observations, all stages of C. elegans were able to consume erythrocytes, but the labelling of the intestinal tract was limited to a select number of nematodes.
Fig. 2. Brightfield, fluorescent (TAMRA-SE) and merged images for C. elegans feeding on fluorescently labelled (TAMRA-SE) red blood cells at 5 h and 24 h after red blood cell introduction. Scale bars = 100 µ m.
The diameter of an erythrocyte (~10 µ m, Fig. 1B) is approximately twice the diameter of an adult C. elegans mouth opening [~3–4 µ m (Altun and Hall, Reference Altun and Hall2018)] and more than 3 times the size of E.coli [~2–3 µ m (Reshes et al., Reference Reshes, Vanounou, Fishov and Feingold2008)], such that during feeding C. elegans could be unable to ingest whole red blood cells. Therefore, to decipher the mechanism of erythrocyte ingestion nematodes were continually imaged using fluorescence microscopy (see Supporting Movie 1). Time-lapse images (Fig. 3) show the ingestion of erythrocytes by C. elegans that occurs via a 5-step process: (1) C. elegans survey their immediate vicinity for sustenance (<0.00 s). (2) Upon finding an erythrocyte it is captured by the mouth (0.00 s). (3) Pharyngeal pumping draws the erythrocyte into the pharynx causing it to rupture (red flashes) and release its contents. (0.15 s). The contents of the erythrocyte are taken up into the pharynx via peristaltic action, passing the pharyngeal grinder and pharyngeal-intestinal junction into the intestine (0.15–1.35 s). (4) Erythrocyte contents are also dispersed into the immediate vicinity surrounding the mouth, suggesting erythrocyte ingestion is an inefficient process (0.30–0.60 s). (5) Caenorhabditis elegans seek further sustenance and nutrition (>1.35 s). These observations highlight that C. elegans are capable of adapting, by modifying their mechanism of food ingestion, rather than limiting their diet to smaller bacterial organisms (Fang-Yen et al., Reference Fang-Yen, Avery and Samuel2009).
Fig. 3. Time-lapse images (0.00–1.35 s) for C. elegans ingesting fluorescently labelled red blood cells. For full-length video See Supporting Movie 1. Scale Bar = 100 µ m.
The loss of erythrocyte integrity visualized during C. elegans digestion (Fig. 3, Supporting Movie 1) would release haemoglobin, which is toxic to organisms upon the release of haem. Therefore, to cope with potential haem toxicity, nematodes use enzymatic pathways that include APRs and GSTs to neutralize the toxic by-products of haemoglobin digestion (Perally et al., Reference Perally, LaCourse, Campbell and Brophy2008).
Viability of C. elegans during erythrocyte feeding
The motility of C. elegans can be used to predict viability as nematode body movement gradually declines and stops completely with age (Collins et al., Reference Collins, Huang, Hughes and Kornfeld2008). Using C. elegans motility as an absolute parameter, where motile and non-motile nematodes were classified as viable and non-viable the effects of restricting the nematode diet to erythrocytes alone, E. coli alone or a mixture of erythrocytes and E. coli was investigated. The motility fraction (Mf), as an indicator for C. elegans viability (Chauhan et al., Reference Chauhan, Orsi, Brown, Pritchard and Aylott2013), showed the three diets did not affect the overall viability (Fig. 4, P > 0.05) and were comparable with previously reported survivorship data (Wood et al., Reference Wood, Rogina, Lavu, Howitz, Helfand, Tatar and Sinclair2004).
Fig. 4. Motility fraction as an indicator of nematode survival for C. elegans fed on erythrocytes (red), E. coli alone (green) and erythrocytes & E. coli, monitored for 15 days. Each data point evaluates the motility of ~100 nematodes each monitored in triplicate, where the error bars represent the standard deviation between experimental repeats.
To determine the effect of different diets on the viability of nematodes, the Mft 50, the time required to reduce the motility of population of nematodes by 50%, was derived. The Mft 50 for C. elegans fed on erythrocyte alone, E. coli alone and erythrocyte and E. coli were 7.07 (± 0.96 s.d.) days, 8.71 (± 0.13 s.d.) days and 7.68 (± 0.57 s.d.) days, respectively, and were not statistically different (P > 0.05). Therefore, under an erythrocyte diet was not affected when compared with the control groups of E. coli and erythrocytes and E. coli.
Enzyme sequence identities between C. elegans and N. americanus
A high degree of sequence identity was confirmed. In particular, peptide A291Y, an epitope in Na-APR-1 (Necepsin II) recognized by enzyme neutralizing and host-protective antibodies, shares 71% identity and 10/13 active site amino acids with C. elegans Asp-4 (Fig. 5). The sequence identities between the respective GSTs are shown in Fig. S2.
Fig. 5. (A) Search criteria and (B) amino acid overlay for C. elegans (C. ele, ASpartyl Protease 4, Query) and N. americanus (N. ame, Necepsin II, Subject). Highlighted regions indicate leader sequence (grey), Aspartic active sites (pink) and target epitopes [C. elegans (green), Overlay (77% match, yellow), N. americanus (red)].
Discussion
The scientific community is keen to develop vaccines against parasitic nematodes of humans (Noon and Aroian, Reference Noon and Aroian2017) and livestock (Nisbet et al., Reference Nisbet, Meeusen, Gonzalez, Piedrafita, Gasser and Von Samson Himmelstjerna2016). During vaccine development, APRs have assumed prominence given their ability to digest haemoglobin. It is apparent that neutralizing antibodies that interfere with the activity of these enzymes contribute to host protection. High levels of sequence identity, between enzymes involved in hematophagy, have also been identified with Schistosoma mansoni, Onchocerca volvulus, Strongyloides stercoralis, Ancylostoma spp and Haemonchus contortus. Therefore, the development of a high-throughput model, using a nematode species that is simple to manipulate, to investigate parasite hematophagy could become a high priority for the parasitological research community (Buckingham and Sattelle, Reference Buckingham and Sattelle2009).
In the present paper, we have demonstrated that C. elegans were able to ingest and digest fluorescently labelled erythrocytes. Ingestion would appear to begin with erythrocyte rupture at the mouth and could be followed by mechanical degradation by the pharyngeal grinder (Avery and Thomas, Reference Avery and Thomas1997), with cell membrane rupture complemented by potential haemolysins, such as the C. elegans saponins and amoebapores (Banyai and Patthy, Reference Banyai and Patthy1998), which are activated by a low pH microenvironment (McGhee, Reference McGhee2007). Haemoglobin digestion and haem detoxification could be conducted by C. elegans Aspartyl Protease 4 and GST.
At this stage, we feel that sufficient initial evidence has been attained to support experiments to investigate antibodies raised against protein homologues from parasites, to assess their effects on the blood feeding and associated viability and survival of C. elegans. Proof of principle data on the value of the model would pave the way for the exploration of new targets associated with haem metabolism in nematodes (Chen et al., Reference Chen, Samuel, Krause, Dailey and Hamza2012; Sinclair and Hamza, Reference Sinclair and Hamza2015) and identify alternate gastrointestinal associated vaccine candidates.
With respect to candidate selection, gastrointestinal associated molecules with corresponding homologues in parasites would be selected, then cloned and expressed, enabling the production of mono-specific antibodies against the candidate molecule. The ability of this antibody to inhibit hematophagy by C. elegans, and have a negative impact on its survival, would be indicative of the value of the target molecule as a potential vaccine candidate. For example, the intestinally expressed lipases (Behm, Reference Behm and Lee2002) the NUC-1 nuclease (Lyon et al., Reference Lyon, Evans, Bill, Otsuka and Aguilera2000), for processing cell-free DNA from nucleated tissue cells and leucocytes, and calreticulin (CRT-1) could be considered as new vaccine candidates (Park et al., Reference Park, Lee, Yu, Jung, Choi, Lee, Kim, Lee, Kwon, Singson, Song, Eom, Park, Kim, Bandyopadhyay and Ahnn2001; Winter et al., Reference Winter, Davies, Brown, Garnett, Stolnik and Pritchard2005).
In conclusion, Walker stated in 2005 that it was ‘difficult to disagree with the lament that unfortunately the biology of digestion (in C. elegans) represents something of a blind spot in this remarkably well-characterised organism’ (Walker et al., Reference Walker, Houthoofd, Vanfleteren and Gems2005). The conversion of C. elegans to hematophagy will hopefully promote new interest in the functioning of its intestine, where parallels may be drawn with the biochemistry of hematophagic parasites.
Acknowledgements
The authors acknowledge Dr Gary Telford for technical laboratory guidance. VMC gratefully acknowledges Dr Jonathan W Aylott for his continued financial support and academic guidance.
Author contributions
VMC and DIP conceptualized experiments and wrote the manuscript. VMC conducted experiments, prepared figures and supporting information.
Financial support
This work was supported by the School of Pharmacy Research Committee, University of Nottingham (VMC).
Conflict of interest
The authors declare no conflict of interest.
Ethical standards
Not applicable.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0031182018001518
