Changes in porcine nutrient transport physiology in response to Ascaris suum infection

Background The roundworm Ascaris suum is one of the parasites with the greatest economic impact on pig farming. In this context, lower weight gain is hypothesized to be due to decreased nutrient absorption. This study aims at characterizing the effects of A. suum infection on intestinal nutrient transport processes and potential molecular mechanisms. Methods Three groups of six piglets each were infected orally (10,000 embryonated A. suum eggs) in a single dose (“single infection”). Another three groups were infected orally (1000 embryonated eggs) for 10 consecutive days (“trickle infection”). Animals were necropsied 21, 35 and 49 days post-infection (dpi). Three groups served as respective controls. The Ussing chamber technique was applied for the functional characterization of small intestinal tissues [short-circuit currents (Isc) as induced by glucose, alanine and peptides; 3H-glucose net flux rates; tissue conductance (Gt)]. Transcription and expression levels of relevant cytokines and nutrient transporters were evaluated (qPCR/western blot). Results Peptide- and alanine-induced changes in Isc were significantly decreased in the jejunum and ileum of the trickle-infected group at 49 dpi and in the ileum of the single-infected group at 49 dpi. No significant differences regarding glucose transport were observed between the Ascaris-infected groups and the control group in Ussing chamber experiments. Transcription levels of the glucose and peptide transporters as well as of selected transcription factors (transcription of signal transducer and activator of transcription 6 [STAT6] and hypoxia-inducible factor 1-alpha [Hif-1α]) were significantly increased in response to both infection types after some periods. The transcription of interleukins 4 and 13 varied between decrease and increase regarding the respective time points, as did the protein expression of glucose transporters. The expression of the peptide transporter PepT1 was significantly decreased in the ileal single-infected group at 35 dpi. Hif-1α was significantly increased in the ileal tissue from the single-infected group at 21 dpi and in the trickle-infected group at 35 dpi. The expression levels of Na+/K+-ATPase and ASCT1 remained unaffected. Conclusions In contrast to the current hypothesis, these results indicate that the nutrient deprivation induced by A. suum cannot be explained by transcriptional or expression changes alone and requires further studies. Graphical abstract

Following hepatic migration, they reach the lungs, penetrate into the alveolae, are transported to the upper respiratory tract and swallowed, reaching the intestines again as from 8 dpi [4]. The worms mature in the intestines to reach the adult stage from approximately day 25-29 dpi onwards [5]. Adult A. suum inhabit the jejunal lumen, where each female produces up to 2 million eggs per day [6].
In Germany, up to 43% of pig livers from conventional and up to 64% from organic farming systems are condemned due to "milk spots", i.e. hepatic scarring caused by the migrating larvae [7]. Furthermore, infected pigs show lower weight gain and feed conversion efficiency [8][9][10][11]. Lower weight gain might be explained by a dysfunction of gastrointestinal nutrient absorption induced by alterations of the intestinal ion transport [12], as well as by reduced availability of glucose or cholesterol [13]. A recent experiment confirmed a decreased glucose absorption in A. suum-infected pigs [14]. Chickens infected with Ascaridia galli also showed significantly decreased electrogenic transport of glucose and alanine [15]. These findings were associated with increased transcription of the Th2 cell-associated cytokines interleukin (IL)-4 and IL-13. IL-4 and IL-13, being upregulated during gastrointestinal nematode infections, activate signal transducer and activator of transcription 6 (STAT6), which in turn may lead to decreased sodium-linked glucose absorption [16,17].
IL-4, IL-5 and IL-13 are secreted upon activation of Th2 cells in response to parasitic antigens, which also causes degranulation of immunoglobulin E (IgE)-sensitized mast cells [18]. One leukotriene, which is secreted by mast cells, LTD4, is known to activate the protein kinase C (PKC)-mediated phosphorylation of Raf kinase inhibitor protein, which triggers protein kinase A (PKA) to inhibit the Na-alanine cotransport in intestinal epithelial cells by decreasing the affinity of the cotransporter ASCT1 [19].
The transport of dipeptides in the small intestines as mediated by peptide transporter 1 (PepT1) depends on an inwardly directed H + gradient. This gradient is established by the sodium-hydrogen antiporter 3 (NHE3). The functionality of this antiporter could be modulated by an inhibitory PKA-mediated phosphorylation of serine residues [20]. Therefore, peptide transport may be indirectly affected during A. suum infection.
Furthermore, phosphorylation of the sodium-glucose linked transporter 1 (SGLT1) may affect glucose transport [21][22][23]. Nevertheless, it is not yet clear whether phosphorylation leads to an increase or decrease in SGLT1-mediated transport. Moreover, it was shown that PKA and PKC may also modify the maximal rate of transport depending on the expressed cotransporter [24]. Also, the recruitment of glucose transporter 2 (GLUT2) to the brush border membrane [25] may modify glucose transport. Previous studies also revealed that the transcription of some genes involved in barrier function and glucose metabolism were regulated by hypoxia-inducible factor 1α (Hif-1α) [26,27]. Hif-1α is supposedly involved in the transactivation of the GLUT1 promoter, and therefore, an increase in Hif-1α levels might lead to increased GLUT1 transcription [28].
The present study aimed at characterizing the mechanisms of potential malabsorption in response to A. suum infection in pigs. Animals were infected using a singleinfection and a trickle-infection procedure. In addition to the functional characterization of intestinal nutrient transport using the Ussing chamber technique, potential mechanisms were investigated with regard to mRNA and protein expression levels.

Animals
Fifty-four hybrid pigs, ~5 weeks old, were acquired from the Farm for Education and Research in Ruthe, University of Veterinary Medicine Hannover, and divided into three separately housed groups. The animals received a standard pig diet ad libitum (Deuka Ferkelstarter Primo, Deutsche Tiernahrung Cremer, Düsseldorf, Germany), had free access to water and were weighed at arrival and prior to necropsy.

Infection and sampling
Pigs of the single-infected group were each infected orally with 10,000 embryonated A. suum eggs, whereas pigs of the trickle-infected group received 1000 eggs/day for 10 days. Six pigs of each group and of the uninfected control group were necropsied on 21, 35 and 49 dpi. Animals were stunned with a penetrative captive bolt device and exsanguinated. The uninfected and infected pigs were necropsied alternately in the morning and afternoon. Successful infection was verified by checking livers for the presence of milk spots. Serum samples taken in weekly intervals were subjected to antibody detection by enzyme-linked immunosorbent assay (ELISA) (human anti-Ascaris lumbricoides IgG ELISA, Abcam (Netherlands) B.V., Amsterdam, Netherlands). Due to the close genetic relationship of A. lumbricoides and A. suum, as well as the fact that the ELISA kit utilizes protein A as secondary antibody, the test was considered suitable for use in pigs. Furthermore, macroscopically visible worms were collected during the sampling procedure for Ussing chamber experiments and molecular analyses.
To ensure the comparability of sampling, 80 cm of each intestinal segment were removed as follows: Tissue from the duodenum was taken at 80 cm distal of the pylorus, from jejunum at 6.0 m distal of the pylorus, and from ileum at 1.0 m proximal of the ileocaecal valve (the first 30 cm were discarded). Tissues for western blotting and qPCR (first 20 cm of each 80-cm segment) were rinsed with ice-cold 0.9% NaCl, stripped of the muscle layers, frozen in liquid nitrogen and stored at −80 °C until further analyses. Samples for Ussing chamber experiments (last 50 cm of each 80-cm segment) were also rinsed with ice-cold 0.9% NaCl and stored in serosal buffer (Table 1) until use.

Ussing chamber experiments
The Ussing chamber technique [29] was used to characterize the electrogenic transport of glucose, alanine and the dipeptide glycyl-l-glutamine [gly-gln] in duodenal, jejunal and ileal tissues. The tissues were incubated under voltage-clamped conditions, thus, potential differences and respective electrical currents resulting from active ion transport processes were compensated by inversely directed short-circuit currents (I sc ) as a measure of net electrogenic ion transport. To control epithelial integrity, tissue conductance (G t ) was measured continuously.
Tissues were opened along the mesenteric side, the serosal and the muscle layer were stripped, and the mucosa was mounted in the Ussing chambers (1.0-cm 2 serosal area). Tissues were aerated with carbogen, and temperature was held at 37 °C. The experiment included 24 chambers per pig, of which six were used for peptide and six for alanine transport measurements (buffer I and III, Table 1), whereas 12 chambers were used for glucose transport measurements (buffer I and II, Table 1). After initial equilibration (5-10 min), potential differences were set to 0 mV, and I sc was measured.
For the measurement of electrogenic glucose transport, glucose (5 mM, mucosal) was added after equilibration (20 min) under voltage-clamped conditions, as well as mannitol (5 mM, serosal) to prevent osmotic effects. ΔI sc was calculated by subtracting the basal value before addition from the maximal response after addition.
Additionally, unidirectional glucose flux rates were measured by adding 5 µCi 3 H-d-glucose (185,000 Bq) after 10 min either to the mucosal or the serosal side. Two samples (50 µl each) were taken from the radioactive side at the beginning and after 60 min of incubation. Four samples (500 µl each) were taken from the unlabelled side (15 min intervals) and mixed with 4.3 ml of scintillation liquid Rotiszint ® eco plus LSC-Universal cocktail (Carl Roth GmbH + Co. KG, Karlsruhe, Germany), and decays per minute (dpm) were measured (Packard 2500 TR liquid scintillation analyser, Packard BioScience Company, Meriden, CT, USA). Transepithelial active glucose transport (J net ) was calculated from the difference between both unidirectional flux rates (J ms = flux rate from the mucosal to the serosal side; J sm = flux rate from the serosal to the mucosal side) according to standard equations [30].

Quantitative real-time PCR (qPCR) Design of primer-probe combinations
Nucleotide sequences of selected Sus scrofa genes were derived from the GenBank database, and Clone Manager (Version Professional 9, Sci Ed Software, Westminster, CO, USA) was used to align multiple sequences. Genespecific primers and TaqMan ™ MGB probes (Table 2) were designed with AlleleID software (PREMIER Biosoft, San Francisco, CA, USA). Probes (Life Technologies, Darmstadt, Germany) were labelled at the 3′ end with FAM ™ or VIC ® to enable duplex qPCR, whereas the 5′ end was labelled with a non-fluorescent quencher.
The supernatant was removed, and 500 µl of cold ethanol (70%) was added. After centrifugation at 16,000×g for 10 min, the pellet was dried and resuspended in 21 µl RNAse-free water. The RNA to cDNA EcoDry ™ Premix (OligodT) strips (Takara Bio Europe/Clontech, Saint-Germain-en-Laye, France) were used for cDNA synthesis of 20 μl precipitated mRNA. Quality and quantity were again determined, and cDNA was diluted 1:10 in RNAsefree water for storage at −80 °C until use.

Quantitative real-time PCR
For each duplex qPCR reaction, 12.50 μl ABsolute ™ Blue QPCR Low ROX Mix (Thermo Fisher Scientific, Schwerte, Germany), 0.15 μl of each primer (50 μM) and 0.06 μl of each probe (10 μM) were mixed. Following duplex pairs were used: PPIA/TBP, GLUT1/GLUT2, SGLT1/IL-4, Hif-1α/IL13 and PepT1/STAT6. As template, 2 μl of cDNA and 1 μl of each tenfold-diluted serial plasmid standard were used, respectively. The thermoprofile consisted of 15 min at 95 °C and 40 cycles of 95 °C for 20 s, 52 °C for 20 s and 72 °C for 30 s. The reference genes (PPIA/TBP) were included in each run. All reactions were run in duplicate, and each run was repeated once. The mean gene-specific amplification efficiency, correlation coefficient (R 2 ), slope value and y-intercept were analysed ( Table 4). Normalization of average baseline-corrected reporter signals of duplicates and technical replicates, correction of run-specific amplification efficiencies, inter-run calibration and normalization against reference genes were performed with qBase + software (Version 3.2, Biogazelle, Ghent, Belgium).
The stability of the reference genes was controlled comparing single-infected, trickle-infected and uninfected pigs at 21, 35 and 49 dpi in each intestinal segment for each gene combination. The total average M-value, which represents the stability of PPIA and TBP, was 0.810, with a coefficient of variation (CV) of 0.287. As the average M-value remained < 1.0, the reference genes were considered stable for the normalization of the eight target genes during qPCR analyses.

Western blot analysis Preparation of cytosol, crude and apical membranes
Western blot analysis was performed for ASCT1, pSGLT1, SGLT1 and PepT1, which are located in the apical membrane, GLUT2 and Na + /K + -ATPase, located in the basolateral membrane, and cytosolic Hif-1α. To enrich apical membrane proteins and crude membranes (for the detection of the basolateral proteins), 1.5 g of tissue was thawed in 13. For cytosol preparation, 250 µl phenylmethanesulfonyl fluoride (Sigma-Aldrich Chemie GmbH, Munich, Germany) was added to 50 ml cytosol buffer (150 mM NaCl, 1% Nonidet ™ P 40 [Sigma-Aldrich Chemie GmbH, Munich, Germany], 50 mM TRIS base, 5 mM EDTA). Tissue (0.2 g) was mixed with 1.85 ml of this buffer and thawed on ice. After homogenization with the Potter grinder and centrifugation (17,500×g, 4 °C, 30 min), the aliquoted supernatant was stored at −20 °C. Protein content was measured using the Bradford method (SERVA Electrophoresis GmbH, Heidelberg, Germany).

Western blotting
The samples were dissolved under reducing conditions (  , 60 min, RT) and incubated overnight with the primary antibody (see Table 5) at 4 °C. For Hif-1α, all steps including TBST were performed using PBST.
Membranes were washed (3 × 10 min, TBST) and the secondary antibodies were applied for 90 min at RT. After washing (3 × 10 min, TBST), secondary antibodies were detected by use of SuperSignal ® West Dura Extended Duration Substrate for ASCT1, Na + / K + -ATPase, PepT1 and SGLT1. SuperSignal ® West Femto Maximum Sensitivity Substrate (both Thermo Fisher Scientific) was used for GLUT2, pSGLT1 and Hif-1α. Signals were detected with the ChemiDoc ™ MP    For the detection of pSGLT1, the membranes incubated with the pSGLT1 antibody were stripped after the first antibody detection. For normalization, membranes were stained with Indian ink (royal blue ink combined with 2% glacial acetic acid, 30 min) and destained with distilled water.

Histomorphometric analysis
For histomorphometric analysis of jejunal and ileal samples from 49 dpi, tissue was rinsed with ice-cold 0.9% saline and stored in 4% formalin for 2 days. After transfer to 70% ethanol, samples were cut to ~ 15-20 × 1-2 mm. The tissue was embedded in Surgipath Paraplast (Leica Biosystems Nussloch GmbH, Nussloch, Germany), and 1-2-µm slices were prepared using an HM 325 microtome (Microm GmbH, Neuss, Germany). Haematoxylin and eosin staining was carried out automatically (Leica CV 5030, Leica Biosystems Nussloch GmbH) after deparaffinization using xylene and rehydration in a descending alcohol series with subsequent transfer to tap water. Slides were digitalized with an Aperio CS2 scanner (Leica Biosystems Nussloch GmbH). Aperio Imag-eScope (version 12.4.3.5008, Leica Biosystems, Wetzlar, Germany) was used to evaluate 15 villi per animal. The total length, i.e. from the top of the villus to the bottom Table 7 Summary of statistical comparisons (Mann-Whitney U-tests) regarding mRNA, protein expression and functional data on 21 dpi nd. not determined, Na + /K + -A Na + /K + -ATPase

Data analysis and statistics
Results are presented as means ± SEM. For creation of figures and statistical analysis, GraphPad Prism 8.4.0 was used. Mann-Whitney U-tests were performed, except for weight and weight gain, which were analysed with Student's unpaired t-test. Differences were regarded as significant if P < 0.05.

Results
All single-and trickle-infected pigs showed liver milk spots upon slaughter, indicating successful infection, while no milk spots were noted in the control group.
Regarding antibody detection, all control pigs remained seronegative throughout the study period, while most infected pigs seroconverted between 7 and 21 dpi (22/36 infected pigs). It should be noted that three pigs seroconverted particularly late during the study period, after 35 dpi. Consequently, 4/6 single-infected and 5/6 trickleinfected pigs each were seropositive upon slaughter at 21 and 35 dpi, while all single-infected pigs and 4/6 trickle-infected pigs slaughtered at 49 dpi had seroconverted. Macroscopically visible worms were collected from the intestinal contents of 10 pigs and microscopically verified as A. suum. Based on the Ussing chamber experiments, which showed no significant changes throughout the experiment in either infection type in the duodenum, only the results of the jejunum and ileum are presented.

Weight gain
There were no significant differences in weight gain between the groups at 21 dpi and 35 dpi. At 49 dpi, there was a significantly increased weight gain in the trickleinfected group (Table 6).

Ussing chamber experiments
No significant differences in electrogenic glucose transport or glucose flux rates were observed between the infected animals and the control group throughout the study (Tables 7, 8, 9; Figs. 1c, f, 2c, f, 3c, f ).
For gly-gln, a continuous increase in the electrogenic response was noticeable along the intestinal axis of the jejunum and ileum, resulting in the highest ΔI sc in the ileum. No significant differences among groups were observed at 21 and 35 dpi (Fig. 4c, f, 5c, f; Tables 7, 8). At 49 dpi, the trickle-infected group showed a significantly decreased ΔI sc in the jejunum, while a significant decrease was observed in both infection groups in the ileum (Fig. 6c, f; Table 9). Regarding alanine transport, the electrogenic response of trickle-infected pigs showed a significant decrease in the jejunum and ileum at 49 dpi (Fig. 6c, f; Table 9).
Basal tissue conductance did not differ significantly between control and infected animals, except for significantly lower conductance at 49 dpi in the jejunum of the trickle-infected group (control group: 29.66 ± 1.57 Table 9 Summary of statistical comparisons (Mann-Whitney U-tests) regarding mRNA, protein expression and functional data on 49 dpi nd not determined, Na + /K + -A Na + /K + -ATPase Glucose transport-associated results of single-and trickle-infected pigs (21 dpi) compared to the control group: relative quantities from jejunal (a) and ileal qPCR analyses (d), relative protein content for immunoblot analyses of jejunal (b) and ileal mucosa (e) and results of Ussing chamber measurements of jejunal (c) and ileal mucosa (f). Results of Ussing chamber experiments show the short-circuit current (∆I sc ) in response to mucosal addition of 5 mM glucose (left y axis) and the calculated 3H-glucose net flux rate after the addition of 5 µCi 3H-d-glucose (J net , right y axis). Data are shown as means ± SEM, and significant differences are indicated with an asterisk (Mann-Whitney U-test, P < 0.05) Fig. 2 Glucose transport-associated results of single-and trickle-infected pigs (35 dpi) compared to the control group: relative quantities from jejunal (a) and ileal qPCR analyses (d), relative protein content for immunoblot analyses of jejunal (b) and ileal mucosa (e) and results of Ussing chamber measurements of jejunal (c) and ileal mucosa (f). Results of Ussing chamber experiments show the short-circuit current (∆I sc ) in response to mucosal addition of 5 mM glucose (left y axis) and the calculated 3H-glucose net flux rate after the addition of 5 µCi 3H-d-glucose (J net , right y axis). Data are shown as means ± SEM, and significant differences are indicated with an asterisk (Mann-Whitney U-test, P < 0.05) mS/cm 2 , trickle-infected: 19.08 ± 1.83 mS/cm 2 , U = 14, P < 0.0001).

qPCR results
At 21 dpi, a significant increase in SGLT1 transcription was observed in the jejunum of both infection groups and in the ileum of the trickle-infected group (Fig. 1a, d; Table 7). GLUT1 transcription was significantly increased in the jejunum of the trickle-infected group and the ileum of both infection groups at 21 dpi as well as in the jejunum of the trickle-infected and in the ileum of the singleinfected group at 35 dpi (Figs. 1a, d, 2a, d, Tables 7, 8). At 49 dpi, the jejunum of the trickle-infected group showed significantly increased GLUT2 transcription (Fig. 3a; Table 9). Only the jejunum of the single-infected group at 21 dpi showed significantly increased transcription of Hif-1α ( Fig. 1a; Table 7).
A significant increase in PepT1 transcription was noted in the jejunum and ileum of the trickle-infected group at 21 dpi (Fig. 4a, d; Table 7).
Transcription of IL-4 in the jejunum was significantly decreased at 21 dpi in the trickle-infected and 35 dpi in the single-infected group (Figs. 1a, 2a; Tables 7, 8), whereas a significant increase was detected in the ileum of the trickle-infected group at 35 and 49 dpi (Figs. 2d, 3d; Tables 8, 9). For IL-13, a significant decrease was observed in the ileum of the trickle-infected group at 21 dpi ( Fig. 1d; Table 7) and in the jejunum of both infection groups as well as in the ileum of the trickle-infection group at 49 dpi (Fig. 3a, d; Table 9). STAT6 transcription was only affected in the trickle-infection group, with a significant increase in the jejunum and ileum at 21 dpi (Fig. 1a, d; Table 7).

Western blot analysis Expression of glucose transporters, Na + /K + -ATPase and Hif-1α
Representative western blots are shown in Fig. 7. At 21 dpi, no significant changes regarding pSGLT1 or SGLT1 expression occurred ( Table 7). The trickle-infected group showed a significant increase in SGLT1 expression in the jejunum at 35 dpi ( Fig. 2b; Table 8). At 49 dpi, pSGLT1 expression was increased in the ileum of the singleinfected group (Fig. 3e; Table 9).
GLUT2 showed a significant increase at 21 dpi in the ileum of the trickle-infected group (Fig. 1e; Table 7) as well as in the jejunum of the single-infected group at 49 dpi ( Fig. 3b; Table 9), but a significant decrease in the jejunum at 35 dpi ( Fig. 2b; Table 8) and in the ileum at 49 dpi ( Fig. 3e; Table 9). No significant differences in the expression of Na + / K + -ATPase were observed. The expression of Hif-1α was unaltered at 49 dpi (Fig. 3b, e; Table 9), but significantly Fig. 3 Glucose transport-associated results of single-and trickle-infected pigs (49 dpi) compared to the control group: relative quantities from jejunal (a) and ileal qPCR analyses (d), relative protein content for immunoblot analyses of jejunal (b) and ileal mucosa (e) and results of Ussing chamber measurements of jejunal (c) and ileal mucosa (f). Results of Ussing chamber experiments show the short-circuit current (∆I sc ) in response to mucosal addition of 5 mM glucose (left y axis) and the calculated 3H-glucose net flux rate after the addition of 5 µCi 3H-d-glucose (J net , right y axis). Data are shown as means ± SEM, and significant differences are indicated with an asterisk (Mann-Whitney U-test, P < 0.05) increased in the ileum of the single-infected group at 21 dpi ( Fig. 1e; Table 7) and both infection groups at 35 dpi ( Fig. 2e; Table 8).

Expression of peptide (PepT1) and amino acid transporters (ASCT1)
PepT1 expression was significantly decreased in the ileum of the single-infected group at 35 dpi ( Fig. 5e; Table 8), but remained unchanged at 21 and 49 dpi, as well as in the trickle-infected group. Regarding ASCT1, no significant changes occurred at 21 dpi or 49 dpi (Figs. 4b, e, 6b, e; Tables 7, 9). At 35 dpi, no obvious upper bands were detected, and expression could not be measured.

Histomorphometric analysis
Villus length, villus width and crypt depth in jejunal and ileal samples from 49 dpi showed no significant changes in response to infection (Table 10).

Discussion
This project aimed at characterizing the intestinal electrophysiological response to glucose, peptides and alanine as well as the transcription and expression of cytokines and nutrient transporters following experimental A. suum infection. To gain a comprehensive picture, different small intestinal segments and different time points pi to cover all gut-associated parasite stages, as well as a single infection and a more natural trickle infection were considered. The presence of milk spots indicated successful infection of pigs, while the serological response was somewhat variable, with most infected pigs seroconverting between 7 and 21 dpi, but some becoming seropositive as late as 42 or 49 dpi. As the ELISA test utilized was designed for use in humans, further validation and cut-off adjustment may be necessary regarding A. suum infection in pigs.
The histomorphometric measurements revealed no differences in villus length or crypt depth at 49 dpi. Thus, changes in mucosal surface in response to the infection are unlikely. Regarding glucose transport, no significant effect of A. suum infection was observed in the present study. This is in contrast to previous studies in A. galli-infected chickens as well as A. suum-infected pigs, which showed a decrease in the electrogenic response to glucose as compared to nematode-free animals [14,15].
Nevertheless, western blot analysis showed an upregulation of relevant transporters, e.g. GLUT2 in the ileum of the trickle-infected group at 21 dpi and in the jejunum of the single-infected group at 49 dpi. Furthermore, SGLT1 transcription was increased in the jejunum of both infection groups, while transcription of GLUT1 and Hif-1α was upregulated in the trickle-infected group at 21 dpi. An increase in both SGLT1 and GLUT1 transcription was also observed in the ileum of the trickle-infected group at 21 dpi. The increase in both SGLT1 and GLUT1 transcription seemingly contradicts previous findings, because GLUT1 is supposed to be upregulated in order to compensate for the downregulation of SGLT1 [33]. It has to be considered, however, that the increased SGLT1 transcription observed in the current study does not necessarily indicate increased expression and therefore altered glucose transport. SGLT1 phosphorylation was also examined, as pSGLT1 is a more active transporter [21]. However, the ratio of SGLT1 and pSGLT1 remained unaffected, corroborating the unaltered ΔI sc .
Furthermore, distinct differences were observed between the jejunum and ileum. In the jejunum, GLUT2 showed significantly increased expression in the singleinfected group and significantly increased transcription in the trickle-infected group at 49 dpi, while a significant decrease in GLUT2 expression was noted in the ileum of the trickle-infected group. These differences might be related to the different functions of the intestinal segments. While the proximal parts of the small intestines are exposed to higher glucose levels, the distal parts receive chyme with lower glucose content [21]. Therefore, the reaction and distribution of glucose transporters might vary depending on the segment examined. Earlier studies reported stronger electrogenic responses in the ileum than in the jejunum of pigs [34,35]. Most previous studies on nematode-related alterations of glucose Alanine and gly-gln transport-associated results of single-and trickle-infected pigs (35 dpi) compared to the control group: relative quantities from jejunal (a) and ileal qPCR analyses (d), relative protein content for immunoblot analyses of jejunal (b) and ileal mucosa (e) and results of Ussing chamber measurements of jejunal (c) and ileal mucosa (f). Results of Ussing chamber experiments show the short-circuit current (∆I sc ) in response to mucosal addition of 10 mM alanine or gly-gln. Data are shown as means ± SEM, and significant differences are indicated with an asterisk (Mann-Whitney U-test, P < 0.05) transport, however, were performed only with jejunal tissue [14,17,33], precluding any comparison with previous studies regarding the ileum.
Because (p)SGLT1 is coupled with a sodium gradient, sodium needs to be exported to maintain the driving force for the glucose transporters. The expression of Na + /K + -ATPase was analysed, but no significant changes were observed. Therefore, a changed expression of Na + / K + -ATPase seems unlikely to create an altered gradient. Nevertheless, a change in activity, e.g. by PKA phosphorylation, cannot be excluded [36].
Earlier studies on the effect of IL-4 and IL-13 on sodium-coupled glucose absorption showed decreased glucose transport in wild-type but not in STAT6-deficient mice [17]. Therefore, elevated production of the STAT6 activators IL-4 and IL-13 might affect glucose transport. Interestingly, during the early phase of infection (21 dpi), STAT6 transcription was significantly upregulated in both segments in the trickle-infected group, supporting this hypothesis. Nevertheless, a significant decrease in IL-13 transcription in the jejunum and ileum of trickle-infected pigs occurred at 49 dpi, whereas IL-4 was significantly increased in the ileum, complicating the interpretation of the findings. In a study on A. galli-infected chickens [15], an increase in IL-4 and IL-13 transcription was observed 2 to 3 weeks pi. However, A. galli does not perform a body migration but develops in the intestine only. Considering the body migration of A. suum and its arrival in the small intestine at ~8 dpi, typical reactions may occur around 35 dpi, conceivably explaining the significant increase in IL-4 observed at this time point in the current study.
A reduction of electrogenic peptide transport was observed at 49 dpi, underlining the hypothesis of nutrient malabsorption due to helminth infection [14,15]. Both infected groups showed a significant reduction in the ileum, and the trickle-infected group additionally in the jejunum. Since PepT1 expression in the ileum remained unchanged at 49 dpi in both infection groups, the question arises as to whether phosphorylation of NHE3 by PKA might impair the maintenance of the H + gradient and thus peptide transport [20]. This Alanine and gly-gln transport-associated results of single-and trickle-infected pigs (49 dpi) compared to the control group: relative quantities from jejunal (a) and ileal qPCR analyses (d), relative protein content for immunoblot analyses of jejunal (b) and ileal mucosa (e) and results of Ussing chamber measurements of jejunal (c) and ileal mucosa (f). Results of Ussing chamber experiments show the short-circuit current (∆I sc ) in response to mucosal addition of 10 mM alanine or gly-gln. Data are shown as means ± SEM, and significant differences are indicated with an asterisk (Mann-Whitney U-test, P < 0.05)  Regarding peptide transport, both segments of trickleinfected pigs showed a significant increase in PepT1 transcription at 21 dpi. It is possible that the nematode has a high demand for peptides at this stage of development, leading to more intense host-parasite competition for these nutrients. Promoting PepT1 and therefore peptide uptake could be a counterregulatory mechanism of the host to compensate for the parasitic infection. On the level of protein expression, however, no significant changes in PepT1 were observed, with the exception of a reduction in the ileum of the single-infected group at 35 dpi.
Similar to the alterations observed regarding peptide transport, alanine transport was also significantly decreased at 49 dpi in the trickle-infected group. However, the decrease cannot be explained by the transcriptional and expression data, since these revealed no significant changes. It could be speculated that the parasites affect the epithelial tissue unevenly. However, since A. suum resides in the gut lumen without being locally attached, it seems reasonable that the intestinal tissue is evenly exposed to the parasites' excretory-secretory antigens. Based on the induced hypercontractility [37] related to the nematodes' presence, this seems probable. Further immunoregulatory processes may play an important role. The Th2-associated release of IL-4 and IL-13 is one example among other possible mechanisms [17,38,39]. Hence, monitoring of respective immune parameters, for example mast cells [33], might be a valuable approach for future studies. It is known that parasitic antigens cause degranulation of IgE-sensitized mast cells [18]. The released cytokine LTD4 supposedly leads to a phosphorylation of ASCT1 mediated by PKA and/ or PKC-α, which results in the downregulation of ASCT1 activity based on decreased affinity [19]. This may be an explanation for the observed reduced Na-alanine cotransport.
Overall, the trickle infection had a greater impact on the results, highlighting that single, high-dose infections may not be ideal for studying host-parasite interactions. This is in accord with previous reports that trickle infections resulted in more successfully manifested A. suum infections than single infections, which induce a strong expulsive reaction of the host [40,41]. However, as discussed below, it was not possible to quantify worm burden in the current study to confirm this hypothesis.
Furthermore, more differences were observed towards the end of the study, indicating that the parasites' impact may increase with the length of infection. Thus, the pigs may have been slaughtered too early for differences in weight gain to become apparent, namely before or just at the beginning of patency, when most worms were still immature. Adult worms may have a different or stronger impact on the transport processes, compared with larval or subadult stages. In line with this theory, the parasite's excretory-secretory antigens are heterogeneous in their molecular weight specific to the developmental stage of the parasite [42].
Moreover, it has to be considered that the results in general, but especially regarding ΔI sc , showed a high variance. One shortcoming of the study is that individual parasite burdens could not be quantitatively assessed, due to the need to rapidly transfer the mucosa to Ussing chambers. Therefore, parasites were not counted during the sampling procedure. Immunologically mediated parasite expulsion between days 14 and 21 pi leads to considerable variation between individuals, and some animals may even completely eliminate the infection [4,43,44], which may have caused the large variance in the results. Therefore, a higher number of experimental animals and thus larger sample size might allow for more resilient conclusions in future studies, also with regard to differing parasite burdens between individual animals.

Conclusions
Considering electrophysiological, transcriptional and expression data as well as weight gain and histomorphometric analysis, the infected pigs did not experience the comprehensive nutrient restriction that was expected based on previous studies [14,15]. Surprisingly, the weight gain was even significantly increased at 49 dpi in the trickle-infected group. Therefore, the common clinical finding of reduced weight gain in A. suum field studies cannot be fully explained by alterations in nutrient transport. Nevertheless, significant differences between the experimental groups were observed, especially regarding electrogenic peptide and alanine transport. This included a tendency towards an upregulation of the PepT1-mediated transport of gly-gln and PepT1 expression at 21 dpi, followed by downregulation at later points of infection. These changes occurred in the jejunum and ileum, whereas no changes were measured in the duodenum. Moreover, the observed functional alterations were not fully explained by transcriptional or expression changes. Thus, further studies are necessary to fully elucidate