Anthelmintic activity of trans-cinnamaldehyde and A- and B-type proanthocyanidins derived from cinnamon (Cinnamomum verum)

Cinnamon (Cinnamomum verum) has been shown to have anti-inflammatory and antimicrobial properties, but effects on parasitic worms of the intestine have not been investigated. Here, extracts of cinnamon bark were shown to have potent in vitro anthelmintic properties against the swine nematode Ascaris suum. Analysis of the extract revealed high concentrations of proanthocyanidins (PAC) and trans-cinnamaldehyde (CA). The PAC were subjected to thiolysis and HPLC-MS analysis which demonstrated that they were exclusively procyanidins, had a mean degree of polymerization of 5.2 and 21% of their inter-flavan-3-ol links were A-type linkages. Purification of the PAC revealed that whilst they had activity against A. suum, most of the potency of the extract derived from CA. Trichuris suis and Oesophagostomum dentatum larvae were similarly susceptible to CA. To test whether CA could reduce A. suum infection in pigs in vivo, CA was administered daily in the diet or as a targeted, encapsulated dose. However, infection was not significantly reduced. It is proposed that the rapid absorption or metabolism of CA in vivo may prevent it from being present in sufficient concentrations in situ to exert efficacy. Therefore, further work should focus on whether formulation of CA can enhance its activity against internal parasites.

in plants allows the possibility of targeted compound identification for drug discovery, as well as the use of plants as nutraceuticals. In this way, either whole plant material or extracts are included in the diet, representing an alternative or complementary approach to parasite control, particularly in livestock 12 . This approach has many advantages, such as avoiding drug residues in agricultural products, a lower risk of parasites developing resistance, and easy integration into resource-limited communities in the developing world if and where plants are locally available 13 .
The bark of cinnamon (Cinnamomum verum) has been used for millennia as a traditional remedy in herbal medicine 14 . It contains high amounts of the essential oil trans-cinnamaldehyde (CA), as well as being a good source of proanthocyanidins (PAC), a group of plant polyphenols consisting of flavan-3-ol oligomers and polymers 15 . PAC structures vary widely depending on the degree of polymerization and nature of their flavan-3-ol subunits. The four most common flavan-3-ols are catechin and its cis isomer epicatechin (which give rise to procyanidin-type PAC), or gallocatechin and its cis isomer epigallocatechin (which give rise to prodelphinidin-type PAC). The flavan-3-ol units are linked mainly through the C 4 → C 8 inter-flavanol bond (B-type PAC) but flavan-3-ols can also be doubly linked by an additional ether bond between C 2 → O 7 (A-type PAC) 16 (Fig. 1). Cinnamon bark is relatively unusual in that it has been reported to contain PAC with a high number of A-type bonds 17 . A number of studies have investigated the anti-parasitic properties of PAC from a range of different plant sources, with reports of efficacy against both protozoan 18,19 and helminth 20 parasites, however it has not been established whether increased proportions of A-linkages in the PAC molecules can increase the potency of the anthelmintic activity. Moreover, CA has been shown to have anti-parasitic activity against Eimeria infections in poultry 21 , but there have been no reports on activity against helminths. Therefore, the aim of this work was to investigate the anthelmintic properties of cinnamon bark and characterise its active compounds, in order to determine if this may represent a useful natural resource for control of gastrointestinal nematodes.

Cinnamon bark extract has potent activity against Ascaris suum in vitro.
To test whether cinnamon bark contained compounds with anthelmintic activity, we prepared an extract using a procedure that we have previously shown to be effective in extracting polyphenolic compounds from a wide variety of plant sources 22 . The resulting extract was then tested for in vitro activity against larvae of Ascaris suum, an important nematode parasite of swine that also serves as a model for the closely related human parasite A. lumbricoides 23 .
We first used a migration inhibition assay to test for activity against third-stage larvae (L3), the stage of the parasite that emerges from embryonated eggs to infect the host 22 . Strikingly, overnight incubation of L3 in the cinnamon extract at concentrations between 125 and 1000 μ g/mL resulted in 100% inhibition of larval migration ( Fig. 2A). In contrast to this, we previously observed that extracts prepared from a number of other bioactive plant sources induced a less potent, dose-dependent inhibition of migration within the tested range of extract concentrations 22 . Subsequent observation of the larvae during incubation in the cinnamon extract revealed that at concentrations ≥ 250 μ g/mL, all larvae died within two-three hours of incubation. We confirmed the potency of the cinnamon extract in a second experiment using fourth-stage larvae (L4) of A. suum recovered from the intestine of pigs after experimental infection, and again observed mortality of larvae within hours of in vitro incubation (Fig. 2B).

Phytochemical analysis of cinnamon bark extract.
To identify the putative anthelmintic compounds, we analysed the extract by LC-MS. The major compound was CA, which accounted for 7.8 g/100 g of the extract (Fig. 3A). In addition, cinnamic acid and several PAC dimers and trimers with A-type linkages were detected by LC-MS (Fig. 3A). To gain further insight into the PAC composition, the extract was subjected to thiolysis with benzyl mercaptan, which breaks PAC polymers with B-type bonds and allows quantification of their constituent flavan-3-ol units 24,25 . This revealed that these cinnamon PAC were procyanidins that consisted of catechin and epicatechin as terminal and extension units ( Fig. 3B; Table 1). The calculated proportions of flavan-3-ol sub-units are shown in Tables S1-2. From this information, both the total amount of PAC and the mean degree of polymerization (mDP -i.e. the average length of the polymers) can be calculated. This revealed that the extract contained 24.2 g PAC/100 g extract, and the mDP was 5.2. Furthermore, analysis of the inter-flavanol links demonstrated that 21% were A-type and 79% were B-type linkages (Table 1).
To determine if the PAC were mainly responsible for the potent activity of the cinnamon extract, two PAC-enriched fractions were isolated from the extract using Sephadex LH-20 chromatography, which separated the extract on the basis of size to yield two PAC-enriched fractions (F1 and F2). The first fraction (F1) contained 52.4 g PAC/100 g of fraction, and these were lower molecular weight PAC (as determined by thiolysis; mDP of 3.7), whilst F2 contained 55.0 g PAC/100 g of fraction consisting of higher molecular weight PAC (mDP of 7; Table 1). The F1 fraction contained only trace amounts of CA, whilst CA was not detectable in the F2 fraction (data not shown).

Activity of cinnamon bark-derived proanthocyanidins against
Ascaris suum. We next tested the F1 and F2 PAC-enriched fractions in migration inhibition assays with A. suum L3. In contrast to the experiments with the extract, here larvae remained alive after overnight incubation although their migratory ability was impaired (Fig. 4A), consistent with our previous study on A. suum L3 using a range of PAC fractions with exclusively B-type linkages 22 . Similarly, exposure of A. suum L4 to the PAC F2 fraction resulted in a dose-dependent decrease in motility (Fig. 4B), but the activity was far less potent than that observed previously for the whole extract (Fig. 2B). Therefore, we hypothesised that the PAC in the extract played a lesser role in the potent anthelmintic properties. To confirm this, we depleted the whole extract of PAC by overnight incubation in polyvinylpolypyrrolidone (PVPP) 26 and repeated the L3 migration inhibition assay. Depletion of PAC did not affect the potency of the extract, whereas we previously observed that depletion of PAC from a range of other plant extracts reduced or abolished the observed anthelmintic effects 22 . Therefore, we concluded that the mixed A-and B-type PAC in the extract have comparable anthelmintic activity to the B-type PAC obtained from other plant sources 22 , and that in the case of cinnamon bark, they do not seem responsible for the potent activity of the extract.

Anthelmintic activity of trans-cinnamaldehyde in vitro.
As CA was present in relatively high concentrations in the extract, but was removed almost entirely from the PAC-fractions, we hypothesised that CA was the compound mainly responsible for the anthelmintic activity. To test this, we used pure  CA (> 99%) in an efficacy assay against A. suum. We found that pure CA had similar activity to the whole cinnamon extract against both A. suum L3 and L4 (Fig. 5A). Concentrations of around 200 μ M (equivalent to 25.6 μ g CA/mL) resulted in larval death within three hours, which corresponds very well to the amount of CA that would be present in the extract concentrations used (Fig. 2), given that 7.8% of the extract was determined to be CA. CA has been shown to be a potent antimicrobial with activity  against, amongst others, Salmonella sp. and Campylobacter 27,28 , as well as having activity against parasitic plant nematodes 29 . However, this is the first demonstration of anthelmintic activity of CA against a gastrointestinal nematode parasite. To confirm that the activity was not only restricted to A. suum, we tested the activity of pure CA against Oesophagostomum dentatum and Trichuris suis, two other porcine nematodes which fall in different clades to A. suum and are related to human hookworm and whipworm species, respectively 30 . CA also had potent activity against larvae from these other species (Fig. 5B), demonstrating that CA has in vitro activity against a range of gastrointestinal parasites.
Ultrastructural damage in Ascaris suum exposed to trans-cinnamaldehyde. To gain insight into the possible anthelmintic mechanisms of CA, A. suum L4 that had been exposed to a high concentration of CA (236 μ M) were examined by transmission electron microscopy. Examination of the cuticle and hypodermis revealed no major changes in the CA-exposed larvae, however some localised tissue damage and lesions were observed in the muscular layer underlying the hypodermis (Fig. 6A). The most striking damage to the CA-exposed larvae occurred in the digestive tissues (Fig. 6B). Whilst control larvae had a regular, intact gut with undamaged microvilli, the same tissues were completely destroyed in larvae exposed to CA, with the microvilli having lost all integrity, and with massive lesions and vacuoles also present. Thus, it appears that damage to the internal digestive tissues may be at least partly responsible for the anthelmintic activity of CA.

Testing of trans-cinnamaldehyde against Ascaris suum in vivo.
Given that CA appeared to be the most active compound in the cinnamon extract, we proceeded to test whether pure CA could reduce infection with A. suum in pigs in vivo. Two different approaches were taken to administer CA; as a daily dietary supplement or as a targeted therapeutic dose in encapsulated form, as this latter approach may be 1) more suitable to deliver a concentrated dose of the active compound to the site of infection (the small intestine for A. suum), and 2) may be more applicable to human populations where a therapeutic Figure 6. Ultrastructural changes in Ascaris suum exposed to trans-cinnamaldehyde. Transmission electron micrographs of A. suum fourth-stage larvae exposed to either culture media (Control) or 236 μ M trans-cinnamaldehyde (CA) for 12 hours. For all panels scale bar indicates 2 μ m. (a) Cuticle (cu) and underlying Muscular (mu) tissue-note the lesions in the muscle tissue underlying the cuticle and hypodermis in parasites exposed to CA (red arrows). (b) Digestive tissues showing the microvilli (mv) overlying the intestinal lumen-note the destruction of the villi (red circle) and the presence of large vacuoles (red arrow) in parasites exposed to CA.
dose needs to be applied rather than a preventative dietary approach. Thus, three groups of pigs were used. The first group remained untreated as a control. A second group of pigs was fed a daily supplement of 1000 mg CA, as this dose has been shown to be safe and acceptable to pigs, as well as resulting in decreased Escherichia coli excretion 31 . Five days after the supplemental feeding began, all three groups of pigs were infected orally with 5000 embryonated A. suum eggs. The third group was then orally dosed with 1000 mg of CA, placed in acid-resistant capsules, on two occasions, 11 and 13 days post-infection. At 14 days post-infection, all pigs were killed and larval burdens were enumerated. Pigs that received the CA capsules had a mean larval burden of 2386, compared to 3114 and 2991 in those that received the dietary CA or no treatment, respectively. However, there were no significant differences ( Fig. 7A; P = 0.28 by one-way ANOVA). We also assessed the location of larvae within the small intestine, to determine if CA treatment resulted in a more posterior location of larvae. The majority of larvae (~80%) were found in the third quarter of the SI in all pigs. Pigs dosed with CA capsules had more larvae located in the fourth (most posterior) quarter (20%) compared to the control (10%) or those fed CA in the diet (15%), however these differences were also not significant ( Fig. 7B; P = 0.3 by one-way ANOVA). Therefore, despite its potent in vitro activity, CA failed to have an in vivo effect in this model of A. suum infection.

Discussion
We found that the C. verum extract contained PAC with a mixture of A-and B-type linkages, consistent with a previous report on the composition of PAC in C. zeylanicum bark 17 , however these authors also noted the presence of prodelphinidins, which were not observed in our current study, where PAC were comprised exclusively of procyanidins. A number of groups have investigated the anthelmintic properties of PAC, and indeed we recently reported that PAC derived from a wide variety of plant sources had strong anthelmintic activity against A. suum in vitro 22 . However, these experiments were performed with PAC that were comprised largely of polymers with B-type linkages. As A-type linkages markedly affect the 3-dimensional structure and rigidity of PAC molecules, we were interested to ascertain whether they may enhance anthelmintic activity. Indeed, a recent report has indicated that increased proportions of A-type linkages may enhance the in vitro activity of PAC against E. coli 32 . The potent, lethal effect of the extract tested here against A. suum indicated the presence of other compounds with superior anthelmintic activity to the range of PAC molecules which have been tested previously against a multitude of helminths including A. suum, O. dentatum 33 , Ostertagia ostertagi 34 and Haemonchus contortus 35 . This rapid lethality also contrasted to our previous experiments with A. suum larvae where incubation in other tested plant extracts, or synthetic anthelmintic drugs such as ivermectin, resulted in impaired motility and migratory ability but generally larvae remained alive (data not shown). However, we established that this potency did not derive from the PAC molecules in this cinnamon extract. Thus, we concluded that the proportion of A-and B-type linkages in PAC molecules does not markedly influence anthelmintic activity, at least in our in vitro experiments with A. suum. Instead, our data strongly pointed towards CA as responsible for the potent anthelmintic properties, with lethal in vitro activity towards a range of different gastrointestinal nematodes, and that the mechanism appears to involve the physical destruction of the intestinal tissue of the parasite. This presumably occurs after ingestion of the compound from the surrounding media, as the integrity of the cuticle did not seem to be noticeably compromised. The exact mechanism of the anthelmintic effect can only be speculated upon, but CA being an α , β unsaturated aldehyde can react with various nitrogen groups and interference with key enzymes and subsequent cell lysis has been proposed as an antibacterial mechanism of CA 36,37 . Similar mechanisms may operate against nematode parasites. Despite these promising in vitro data, we did not observe any in vivo activity when CA was administered to pigs, either as a daily dietary supplement or in encapsulated form by gavage. CA has recently been investigated as a daily feed supplement for pigs and poultry, and some anti-bacterial efficacy has been reported with this strategy 31,38 . Several reasons may exist why we did not detect any anthelmintic activity in our in vivo model. First, a higher dose may be required to achieve in vivo efficacy. Whilst dose-response studies were outside of the scope of this study, the dosage used in the dietary supplement (1000 mg per pig per day, equivalent to approximately 35 mg/kg bodyweight) was chosen based on previous reports showing that this dose was well tolerated by pigs and reduced E. coli excretion 31,39 . Moreover, assuming a small intestinal volume of approximately 2L, we assumed that a dose of 1000 mg would result in approximately the same concentrations in the intestine that resulted in efficacy in our in vitro studies, even assuming for a large degree of compound disappearance (up to 90%) during digestion. Toxicity studies for CA do not exist for pigs, but rodent studies indicated that the LD 50 for CA given daily by gavage is in the region of 2500 mg/kg bodyweight/day-some 100-fold higher than we used here 40 . Therefore, we considered our current dosage to be towards the maximum end of what can be safely administered, allowing a wide safety-margin which is necessary for future usage. Second, a previous study has shown a similar discrepancy between the in vitro effect and in vivo responses in using CA to treat Salmonella in pigs 41 . These authors showed that this may be due to the non-specific binding of CA to elements of the pig diet formula reducing in vivo availability. Certainly, CA is a reactive molecule, and is not only specific for pathogens but will combine with many other endogenous molecules such as proteins in mucus or digesta, particularly those with thiol-containing amino acids 42 . Lastly, and what we consider most likely, the rapid absorption or metabolism of CA may prevent it from reaching the intestinal site of the nematodes in sufficient concentrations to exert in vivo efficacy. Michiels et al. have shown that the vast majority of CA is absorbed from the stomach and the proximal duodenum in pigs 39 . Whilst A. suum larvae pass through the stomach to complete their migratory phase after passing through the lungs and liver, it is unlikely that there will be sufficient contact between the CA and the parasite before the majority of the compound disappears through the stomach. To try and circumvent this problem, we included an additional group in the in vivo study that was dosed with CA in acid-resistant capsules that are designed to protect the dose through the stomach and release it in the small intestine. However, given that the A. suum infection was already established within the intestine and rapidly moving to a posterior location (as noted in our post-mortem data where only a tiny proportion of larvae were found in the anterior half of the small intestine at day 14 post-infection), it may be that even this approach is not suitable for delivering a sufficiently high dose to the site of the infection. However, it may be interesting to ascertain whether CA could have in vivo efficacy on A. suum adults, which reside in the anterior half of the small intestine 43 .
Whilst the rapid absorption of CA from the stomach reduces its potential as an anthelmintic agent against intestinal-dwelling parasites, formulation of the compound to increase its stability may present an option to overcome this issue. Notably, Si et al. have shown that emulsification of essential oils (including CA) in a variety of hydrocolloids can restore in vitro antimicrobial activity which is lost in the presence of pig dietary compounds 41 . Therefore, experiments to determine whether emulsification/stabilisation agents can enhance the in vivo anthelmintic activity of CA are warranted, as are studies against parasites residing in the anterior regions of the small intestine. Moreover, optimisation of microencapsulation procedures to protect the compound and facilitate its slow release in the intestine may also be a suitable option.
In conclusion, we have shown for the first time that cinnamon bark has anthelmintic potential in vitro, and this derives both from its proanthocyanidin tannins and most notably from trans-cinnamaldehyde. However, for the potential of trans-cinnamaldehyde to be realised as an anthelmintic against intestinal helminths in vivo, appropriate formulations to stabilise and protect the compound will likely be necessary.

Ethics statement. All animal experimentation was approved by the Experimental Animal
Unit, University of Copenhagen, and carried out according to the guidelines of the Danish Animal Experimentation Inspectorate (Licence number 2010/561-1914).

Experimental Design.
A schematic outline of the experimental design for the anthelmintic testing is presented in Fig. 8. Cinnamon bark extract was first extracted and tested against A. suum in vitro using migration and motility inhibition assays. Two PAC fractions were derived from the extract and also tested against A. suum in vitro. Pure CA was then purchased commercially and tested against A. suum, O. dentatum and T. suis larvae in vitro and A. suum larvae in vivo.
Extraction procedure and purification of proanthocyanidins. Finely ground cinnamon bark (10 g) was extracted with acetone/water (140 mL, 7:3, v/v) for 1 h to yield a crude extract. Acetone was removed under vacuum on a rotary evaporator at 35 °C; the remaining aqueous solution was centrifuged for 3 min at 3000 rpm and freeze-dried to yield 2.25 g. The crude cinnamon extract (940 mg) was dissolved in distilled water (500 mL) and applied to a Sephadex LH-20 column (20 g, GE Healthcare, Little Chalfont, UK), which was conditioned with water. Deionised water (1 L) was added to remove sugars. Then, the first fraction (F1-fraction) of purified tannins was eluted with acetone/water (1 L, 3:7, v/v) and the second fraction (F2-fraction) was eluted with acetone/water (1 L, 1:1, v/v). The acetone was removed on a rotary evaporator at 35 °C and the remaining aqueous solutions were frozen overnight and freeze-dried.
LC-MS analysis. Thiolysis reactions were performed as previously described 44 , except that 8 mg of extract or fraction was used for the reaction. Flavan-3-ols and their benzyl mercaptan (BM)-adducts were identified by LC-MS analysis on an Agilent 1100 Series HPLC system and an API-ES instrument Hewlett Packard 1100 MSD detector (Agilent Technologies, Waldbronn, Germany). Samples (20 μ L) were injected into the HPLC connected to an ACE C 18 column (3 μ m; 250 × 4.6 mm; Hichrom Ltd; Theale; UK), which was fitted with a corresponding ACE guard column, at room temperature. The HPLC system consisted of a G1379A degasser, G1312A binary pump, a G1313A ALS autoinjector and a G1314A VWD UV detector. Data were acquired with ChemStation software (version A 10.01 Rev. B.01.03). The flow rate was 0.75 mL min −1 using 1% acetic acid in water (solvent A) and HPLC-grade acetonitrile (solvent B). The following gradient program was employed: 0-35 min, 36% B; 35-40 min, 36-50% B; 40-45 min, 50-100% B; 45-55 min, 100-0% B; 55-60 min, 0% B. Eluting compounds were recorded at 280 nm. Mass spectra were recorded in the negative ionization scan mode between m/z 100 and 1000 using the following conditions: capillary voltage, − 3000 V; nebulizer gas pressure, 35 psi; drying gas, 12 mL min −1 ; and dry heater temperature, 350 °C. Flavan-3-ol terminal and extension units were identified by their retention times and their molecular masses and their peak areas at 280 nm were integrated and quantified using response factors relative to taxifolin 25 . A response factor of 0.55 for A-type procyanidin dimer was calculated using authentic procyanidin A2, and in the absence of commercially available trimers the same response factor was used for A-type procyanidin trimers and their benzylmercaptan adducts (Ropiak et al., submitted). This provided information on the PAC composition in terms of % terminal and % extension flavan-3-ol units. It also allowed calculation of % procyanidins and % cis-and trans-flavan-3-ols according to Gea et al. 25 , excluding the cis/trans ratio for A-type PAC. The percentage of A-type linkages (% A-type) and mean degree of polymerization (mDP) were calculated according to Ropiak et al. (submitted). For quantification of CA, pure CA was dissolved in methanol/water (1:1) at a range of concentrations between 0.03 and 0.13 g/L in order to generate a standard curve from the corresponding peak areas. The whole extract was analysed under identical conditions and the concentration of CA in the extract determined by extrapolation from the standard curve. Parasites. A. suum eggs were obtained from gravid adult worms collected from a local slaughterhouse (Danish Crown, Ringsted, Denmark) and embryonated and stored as described 45 . To obtain the L3, eggs were hatched by mechanical stirring with 2 mm glass beads for 30 minutes at 37 °C, and then viable larvae were purified by incorporating the eggs/hatched larvae into 0.5% agar and incubating overnight at 37 °C in RPMI 1640 media containing L-glutamine (2 mM), penicillin (100 U/mL), streptomycin (100 μ g/mL) and amphotericin B (0.5 μ g/mL). Larvae that migrated into the media were collected and used in the subsequent anthelmintic assays. L4 were collected from the small intestine of pigs 12-14 days after experimental infection, purified from the gut content and washed as described previously 22 . O. dentatum eggs were obtained by collecting faeces from mono-infected donor pigs, and L3 were produced by copro-culture and stored in water at 10 °C. T. suis L1 were produced by mechanical hatching of embryonated eggs, followed by collection of viable larvae after migration through a 20 μ m sieve, as previously described 46 . Anthelmintic assays. Migration and motility assays for A. suum were conducted as previously described 22 . Briefly, for the migration inhibition assay, 100 A. suum L3 were added in triplicate to 48-well plates and incubated (37 °C, 5% CO 2 in air) overnight in either cinnamon extract, PAC fractions or CA. Agar was then added to a final concentration of 0.8% and the number of larvae able to migrate was assessed by light microscopy. In addition, the survival of L3 exposed to either cinnamon extract or CA was assessed at hourly intervals after the start of the incubation. Larvae were considered alive if they had a characteristic coiled appearance and were motile, and were considered dead if straight and immobile even after extended observation 47 . For PAC-depletion experiments, the extract (1 mg/mL) was incubated overnight at 4 °C with PVPP at a ratio of 1 mg extract to 50 mg PVPP. The extract was then centrifuged for ten minutes at 3000 g and the supernatant used in the assays. Controls consisted of media alone with PVPP and whole extract with no PVPP that were incubated in an identical fashion.
Motility of L4 exposed to either cinnamon extract or PAC fractions was assessed at 12-hour intervals for up to 60 hours using a motility scoring system where 5 is fully motile and 0 is no movement 22,48 . Larvae that scored 0 on two successive time-points were considered dead. The motility of L4 exposed to CA was assessed after 6 hours incubation. All assays included media only, as a negative control, and ivermectin (50 μ g/mL) as a positive control. O. dentatum L3 survival was also assessed by an agar-based migration inhibition assay 33 . The number of migrating worms was used as a surrogate measure for the number of live worms, as we have observed with O. dentatum in this assay that close to 100% of viable worms are able to migrate. Survival of T. suis exposed to CA was assessed as described above for survival of A. suum L3, after 2 hours incubation. Electron Microscopy. Transmission electron microscopy was carried out as described previously 22 .
Briefly, worms were washed well in PBS and fixed in 2% glutaraldehyde in 0.05 M phosphate buffer. Samples were post-fixed and dehydrated before sectioning at 70 nm. Imaging was done using a Phillips CM100 microscope. Images were obtained with an Olympus Veleta camera and processed using ITEM software.
In vivo testing. Fifteen helminth-naive pigs (Danish Landrace/Yorkshire/Duroc, mean weight 26 kg) of mixed sex (females and castrated males) were obtained from a specific-pathogen free (SPF) farm with no history of helminth infection. The animals were stratified on the basis of sex and weight to three groups (n = 5) and housed in separate pens with concrete floors. Straw was provided daily. All the pigs were fed restrictively (1 kg per pig/day) with a diet consisting of 75% ground barley and 25% protein/mineral supplement (NAG, Helsinge, Denmark) and had free access to water. Group 1 served as a control. Group 2 was fed a daily supplement of 1000 mg cinnamaldehyde/kg of feed, which was mixed thoroughly into the feed immediately prior to feeding. Five days after the commencement of the supplementary feeding, all 15 pigs were inoculated with 5000 A. suum eggs by stomach tube. On days 11 and 13 post-infection (p.i.), a time-point when the larvae had returned to the small intestine following migration, pigs in group 3 were administered 1000 mg of cinnamaldehyde encapsulated in acid-resistant gelatin capsules (size 2 DR capsules, Capsugel, Bornem, Belgium) directly into the stomach by gavage. Pigs in the control group were similarly administered empty capsules by gavage on days 11 and 13 p.i. On day 14 p.i. all pigs were killed by captive bolt pistol and exsanguation. The entire small intestine was removed, and separated into equal halves. The distal half of the intestine was further separated into two equal segments. Thus, three segments of the small intestine were obtained (first half, third quarter and fourth quarter), and worms were isolated from a 50% subsample of each of these segments separately by the agar-gel method using two hours of incubation 49 . Worms were stored in 70% ethanol and then counted using a dissecting microscope.
Data analysis and statistics. Where mentioned, ANOVA analyses were carried out using GraphPad Prism version 6.0.