Characterization of urea transport mechanisms in the intestinal tract of growing pigs

Abstract

Pigs are capable of nitrogen salvage via urea recycling, which involves the movement of urea in the gastrointestinal tract. Aquaporins (AQP) and urea transporter B (UT-B) are involved in urea recycling in ruminants; however, their contribution to urea flux in the intestinal tract of the pig is not known. The objective of this study was to characterize the presence and relative contribution of known urea transporters to urea flux in the growing pig. Intestinal tissue samples (duodenum, jejunum, ileum, cecum, and colon) were obtained from nine barrows (50.8 ± 0.9 kg) and analyzed for mRNA abundance of UT-B and AQP-3, -7, and -10. Immediately after tissue collection, samples from the jejunum and cecum were placed in Ussing chambers for analysis of the serosal-to-mucosal urea flux (J_sm-urea) with no inhibition or when incubated in the presence of phloretin to inhibit UT-B-mediated transport, NiCl₂ to inhibit AQP-mediated transport, or both inhibitors. UT-B expression was greatest (P < 0.05) in the cecum, whereas AQP-3, -7, and -10 expression was greatest (P < 0.05) in the jejunum. The J_sm-urea was greater in the cecum than the jejunum (67.8 . 42.7 ± 5.01 µmol·cm⁻²·h⁻¹; P < 0.05), confirming the capacity for urea recycling in the gut in pigs; however, flux rate was not influenced (P > 0.05) by urea transporter inhibitors. The results of this study suggest that, although known urea transporters are expressed in the gastrointestinal tract of pigs, they may not play a significant functional role in transepithelial urea transport.

NEW & NOTEWORTHY We characterized the location and contribution of known urea transporters to urea flux in the pig. Aquaporins are located throughout the intestinal tract, and urea transporter B is expressed only in the cecum. Urea flux occurred in both the jejunum and cecum. Transporter inhibitors had no affect on urea flux, suggesting that their contribution to urea transport in the intestinal tract is limited. Further work is required to determine which factors contribute to urea flux in swine.

INTRODUCTION

Retention of dietary nitrogen in pigs ranges from 30 to 60% of intake (36), with much of this inefficiency the result of amino acid utilization for body maintenance functions and to support costs associated with catabolism of excess amino acid intake (33). The main waste product of amino acid catabolism is ammonia, which is further detoxified via urea synthesis in the liver. The majority of urea produced in the pig is excreted in urine; however, it has been established in ruminant (51) and nonruminant animals and humans (7, 8, 16, 31, 32) that a proportion of urea enters the gastrointestinal tract (GIT) where resident microbes are capable of using the urea-nitrogen to produce amino acids. These microbially produced amino acids are available to the host animal for body protein synthesis (48, 49). This process of urea recycling represents a potentially important salvage mechanism to improve nitrogen retention during times of protein deficit (11).

The proportion of urea that is recycled in the GIT has been estimated to be between 20 and 50% in pigs, humans, and other nonruminants (11, 31, 32, 43). The regulation of urea transport in the GIT and the factors affecting urea recycling in the gut are poorly understood in the nonruminant and are only recently being elucidated in ruminants (1, 9, 51). Urea is present in salivary, gastric, pancreatic, and biliary secretions (4), and movement of urea from the blood in the GIT is positively correlated with blood urea concentration (19, 46, 51), suggesting substantial movement of urea in the GIT through paracellular routes and with endogenous secretions. In the ruminant, facilitative urea transporters [SLC14A also known as urea transporter B (UT-B)] and aquaporins (AQP), water transporters capable of transporting urea (AQP-3, -7, -8, and -9), have been identified as playing a role in the regulation of urea transport in the rumen (51) in addition to nonspecific paracellular movement. Both UT-B and AQP are expressed in a number of organs and tissues in ruminants (24, 41, 42) and nonruminant animals and humans (6, 14, 15, 23, 25); however, whether these transporters are present in the GIT of the pig and contribute to urea flux is unknown. A greater understanding of the factors regulating urea flux and recycling in the pig could lead to improved protein utilization, aid in efforts to reduce nitrogen excretion, and improve the environmental sustainability of pork production. Therefore, the objective of this study was to characterize the presence and relative contribution of known urea transporters to urea flux in the growing pig.

METHODS

Ethical approval.

The experimental protocol was reviewed and approved by the University of Saskatchewan’s Animal Research Ethics Board (Protocol No. 20180009) and followed the Canadian Council on Animal Care guidelines (5).

Animals and management.

Nine barrows weighing 50.8 ± 0.92 kg were obtained from the Prairie Swine Centre (Saskatoon, SK) and housed in the Livestock Research Building at the University of Saskatchewan (Saskatoon, SK). Before the study, pigs had ad libitum access to water supplied through nipple drinkers and a commercial growing pig diet.

Flux measurements.

After a minimum 6-day acclimation period, pigs (1/day) were euthanized by captive bolt stunning and exsanguination at 1000. Immediately after confirmation of death, the digestive tract was removed from the abdominal cavity. Tissue samples from the jejunum (midpoint of small intestine), cecum, and colon were collected, washed with physiological buffer (Dulbecco’s Modified Eagle’s Medium, D5523; Sigma-Aldrich, St. Louis, MO) until clean, and placed in physiological buffer warmed to 39°C with an osmolality of 323.5 ± 2.5 mosmol/L and supplemented with 3.7 g/L of sodium bicarbonate to maintain pH of 7.4. Tissues were maintained in buffer with continual gassing with carbogen (95% O₂-5% CO₂) for transport to the laboratory. Upon arrival at the laboratory, the jejunum and cecum were gently stripped of the underlying muscular layer, cut into 3-cm strips, and placed between two halves of a Ussing chamber (Free University of Berlin, Berlin, Germany) with an exposed surface area of 1 cm². Mounted jejunal and cecal epithelia were bathed with 10 mL buffer solution on both the serosal and mucosal sides. The incubation buffer contained 60 mg/L of penicillin G sodium salt, 100 mg/L kanamycin sulfate, and 50 mg/L fluorocystosine to prevent microbial activity according to Doranalli et al. (9). Serosal and mucosal incubation buffers were continually gassed with carbogen and maintained at 38.5°C with water jacket reservoirs. Intestinal epithelia were maintained under short-circuit conditions by a computer-controlled voltage-clamp system (VCC MC6; Physiologic Instruments, San Diego, CA). Short-circuit current (I_sc) and transepithelial conductance (G_t) were measured every 6 s, and mean values were calculated for each flux period and overall. In total, 24 Ussing chambers were used with 12 each for jejunum and cecum.

Once mounted, tissues were allowed to stabilize for 20 min before initiation of flux measurements. To measure the serosal-to-mucosal urea (J_sm-urea) and mannitol flux, a solution containing [¹⁴C]urea (26 kBq; PerkinElmer, Waltham, MA) and [³H]mannitol (37 kBq; PerkinElmer) was added to the serosal side to achieve a final concentration of 1 mM for both urea and mannitol (51). After addition of isotopic tracers, 30 min were allowed for equilibration to ensure steady-state flux conditions before initiation of inhibitor treatments. Individual jejunal and cecal tissues were assigned to one of four treatments (in triplicate), consisting of a control (no inhibitor added) or the addition of phloretin, NiCl₂, or both phloretin and NiCl₂. Phloretin (1) and NiCl₂ (26) were used as inhibitors of UT-B and AQP function, respectively. Inhibitors were dissolved in ethanol, and an equivalent volume was added to both the serosal and mucosal buffers to achieve a final concentration of 1 mM according to Walpole et al. (51). All reagents and inhibitors were purchased from Sigma-Aldrich unless otherwise stated.

Serosal aliquots (100 µL) were obtained after the addition of the inhibitors and at the end of the flux period (15 and 155 min after addition of treatments). These aliquots were diluted with a further 400 µl of buffer solution. Mucosal aliquots (500 µL) were obtained at 10 min after addition of treatments and every 30 min thereafter. A scintillation cocktail was added (Ultima Gold; PerkinElmer). The specific activity of aliquots was determined on a liquid scintillation counter (Tri-Carb 2910TR; PerkinElmer).

Tissue sampling and mRNA abundance.

Concurrent to tissue samples for Ussing chamber experiments, tissue samples were collected from the duodenum (30 cm from pyloric sphincter), jejunum (midpoint of the small intestine), ileum (30 cm from the ileocecal junction), cecum, and apex of the colon for gene expression analysis. The intestinal tissues were cut longitudinally and washed in sterile ice-cold phosphate-buffered saline. The epithelium of each region of the small intestine was scraped using a sterile glass slide on a clean surface while chilled on ice. Collected epithelial samples were placed in a 2-mL cryovial, snap-frozen in liquid nitrogen, and stored at −80°C until analysis of mRNA abundance.

Tissue samples were homogenized and RNA extracted using TRIzol (Applied Biosystems, Waltham, MA) according to Méndez et al. (28). The RNA concentration was determined on a NanoDrop (ND-2000; NanoDrop Technologies, Wilmington, DE) by measuring absorbance at 260 and 230 nm. Complementary DNA (cDNA) was synthesized with the High Capacity Reverse Transcription Kit (Applied Biosystems) following the manufacturer’s instruction using a C1000 Touch Bio-Rad thermal cycler. Quantitative PCR of UT-B and AQP- 3, -7, and -10 was completed using SsoFast Evagreen Supermix (Bio-Rad Laboratories, Mississauga, ON, Canada) on a CFX Connect Real-Time System (Bio-Rad). Briefly, a standard curve was made by fivefold serial dilution from a 10 ng/µL dilution, and each reaction consisted of 2 ng of cDNA, 4 µM forward and reverse primers (Table 1), nuclease-free water, and SsoFast Evagreen Supermix. Cycling conditions included 40 cycles of denaturation at 95°C for 5 min, annealing, and extension at primer-specific temperature for 5 min followed by a melt curve. Expression of UT-B and AQP- 3, -7, and -10 was normalized to the geometric mean of glyceraldehyde 3-phosphate dehydrogenase and hypoxanthine phosphoribosyl transferase 1.

Table 1.

Primers used in quantitative PCR analysis

Gene	Primer Sequence	Annealing Temperature, °C	Accession No. or Ref. No.
UT-B	S: TCAATGGCTGTGTGGGAACG	60	XM_021092502.1
	AS: CCAGGGTGGCATTGTAGC
AQP-3	S: GAAGGAGCTGGTGACCCG	60	NM_001110172.1
	AS: GCCACAGCCAAACATCACG
AQP-7	S: GTGGCAGGGAACATCTCGG	57	NM_001113438.1
	AS: CAGGAAGGAGCCCAGGAAC
AQP-10	S: AGTTTCCCATTTACTCCTTGGT	60	NM_001128454.1
	AS: AGTTCTGTAGGGCATCGTAATAG
GAPDH	S: CTTCACGACCATGGAGAAGG	63	3
	AS: CCAAGCAGTTGGTGGTACAG
HPRT1	S: GGACTTGAATCATGTTTGTG	60	37
	AS: CAGATGTTTCCAAACTCAAC

AQP, aquaporin; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; HPRT1, hypoxanthine phosphoribosyl transferase 1; UT-B, urea transporter B.

Digesta sampling and analysis.

Digesta samples from each of the five locations were collected and diluted at 1:1 (wt/wt) with distilled water. The pH was measured using a pH probe (Fisherbrand 13636AP115; Fisher Scientific), and a 10-mL aliquot of mixed digesta was preserved in a vial with 2 mL of 1% sulfuric acid for analysis of ammonia content according to Fawcett and Scott (10).

Statistical analysis.

Data were tested for normal distribution using the UNIVARIATE procedure of SAS (SAS 9.4; SAS Institute, Cary, NC) using the fixed effects of segment and inhibitor treatment and pig as a random effect. For data that were normally distributed, data were then subjected to a two-way ANOVA using the MIXED procedure of the SAS using the same model. Nonnormally distributed data (electrophysiology) were further analyzed using the GLIMMIX procedure. Differences between means were determined using the Tukey test with significance between means determined at P ≤ 0.05. Data are presented as LS Means ± SE.

RESULTS

Digesta was characterized throughout the GIT, with pH and ammonia concentrations reported in Table 2. There was no difference in ammonia concentration between the segments; however, pH was highest (P < 0.05) in the ileum, followed by the colon, and lowest in the duodenum.

Table 2.

Digesta characteristics in segments of the gastrointestinal tract of the growing pig

Item	Duodenum	Jejunum	Ileum	Cecum	Colon	SE	P Value
pH	5.54c	6.16b,c	7.10a	6.06b,c	6.19b	0.17	<0.001
Ammonia, mg/dL	3.71	4.24	6.81	5.09	5.27	1.24	0.23

Values are means ± SE.

^a,b,cMeans within a row without a common superscript differ significantly.

Aquaporin-3 and -10 were highly expressed (P < 0.05) in the jejunum, whereas AQP-7 was expressed equally in the duodenum and jejunum (Table 3). UT-B was only expressed in the cecum.

Table 3.

mRNA abundance for UT-B and AQP in gastrointestinal tissues obtained from growing pigs

Transporter	Duodenum	Jejunum	Ileum	Cecum	Colon	SE	P Value
UT-B	0.00b	0.00b	0.02b	8.59a	0.04b	1.99	<0.001
AQP-3	1.69b	4.11a	1.44b	0.25b	0.15b	0.61	<0.001
AQP-7	2.26a	3.19a	1.49a,b	0.06b	0.05b	0.62	<0.005
AQP-10	1.66b	10.54a	0.27b	0.12b	0.02b	2.43	<0.001

Values are means ± SE. UT-B, urea transporter B; AQP, aquaporins.

^a,bMeans within a row without a common superscript differ significantly.

Despite the expression of various transporters throughout the tissues examined in the Ussing chambers, there was no significant inhibition of urea flux or mannitol flux by either NiCl₂, phloretin, or a combination of the two (P > 0.10; Table 4). However, both urea flux and mannitol flux were significantly greater in the cecum than jejunum (P < 0.001). Electrophysiology also showed no significant effect of treatment within each segment of the GIT (Table 4). There were, however, intestinal segment differences, with I_sc being greater in the jejunum and G_t being greater in the cecum (P < 0.001).

Table 4.

J_sm-urea, J_sm-mannitol, G_t, and I_sc in intestinal tissues obtained from growing pigs

	Treatment
Item	CON	N	P	NP	SE	P Value
Jsm-urea, µmol·cm−2·h−1
Jejunum	49.84	32.02	48.93	40.10	7.31	0.58
Cecum	67.30	60.85	69.58	73.49	7.68	0.58
Jsm-mannitol, µmol·cm−2·h−1
Jejunum	49.49	39.54	56.77	51.82	8.35	0.45
Cecum	75.90	68.81	90.24	98.97	8.35	0.45
Isc, mA
Jejunum	0.38	1.01	0.21	0.28	0.20	0.59
Cecum	0.12	0.27	−0.20	−0.22	0.20	0.59
Gt, mS/cm2
Jejunum	30.91	25.20	22.90	22.4	4.41	0.06
Cecum	32.57	33.78	43.49	33.73	4.61	0.06

Values are means ± SE. J_sm-urea, serosal-to-mucosal flux of urea; J_sm-mannitol, serosal-to-mucosal flux of mannitol; G_t, tissue conductance; I_sc, short-circuit current; CON, no inhibitor; N, NiCl₂; P, phloretin; NP, NiCl₂ and phloretin.

DISCUSSION

The process of urea recycling in the GIT represents a potentially significant nitrogen salvage method and nutrient production mechanisms in both ruminant and nonruminant animals and humans. Urea transported in the GIT is susceptible to microbial degradation, with the liberated urea-nitrogen (i.e., ammonia) available for incorporation in microbial amino acids. It has been shown previously that pigs are capable of absorbing and using these microbial amino acids for body protein deposition (29, 48, 49), although the extent to which such amino acid absorption contributes to overall amino acid supply is not clear.

Whereas the general concept of urea recycling has been extensively studied in nonruminants (11, 31, 32) and ruminants (9, 20, 27), the regulation of urea transport in the GIT and the factors affecting urea recycling in the gut are poorly understood. It has been suggested that the majority of urea hydrolysis, and thus urea recycling in the GIT, occurs in the lower gut (i.e., cecum and colon) of pigs based on the large population of urease-producing microflora in this part of the intestinal tract (50). However, it is more likely that urea recycling occurs throughout the GIT. Indeed, microbial nitrogen, ammonia, and urea contribute >60% of total nitrogen flow at the ileum (30), suggesting that there is a significant level of urea recycling in the small intestine. Previous research has demonstrated that both the small intestine (31, 32) and cecum (43, 53, 54) are major sites for urea recycling in pigs and other nonruminants. Therefore, our aim for this study was to characterize the presence and relative contribution of known urea transporters (UT-B and AQP-3, -7, and -10) to urea flux in the GIT of growing pigs.

Facilitative urea transporters (UT-B) and AQP-3, -7, and -10 have been implicated in urea transport in the GIT of ruminants (51) and nonruminants (2, 41), but their presence in the GIT of growing pigs has not previously been characterized. We determined that AQP-3, -7, and -10 and UT-B are expressed in the GIT of the pig, with the highest expression observed in the jejunum. In contrast, UT-B was only expressed in the cecum. The high expression of AQP throughout the small intestine is consistent with the results of He et al. (12), which showed that the abundance of most AQP was higher in the duodenum and jejunum of piglets, since they are the major sites for the entry of large volumes of water and other endogenous secretions (39). AQP-3, -7, and -10 examined in our study belong to a subgroup of AQP referred to as aquaglyceroporins, which have also been shown to be permeable to urea (2, 34, 40).

The use of NiCl₂ has been shown previously to result in robust inhibition of the AQP (26). In a previous study using Holstein calves (51), the addition of NiCl₂ decreased J_sm-urea by 23% across ruminal epithelial tissue mounted in Ussing chambers, thereby implicating AQP-mediated transepithelial urea transfer. In the current study, the addition of NiCl₂ resulted in a 17.82 and 6.45 µmol·cm⁻²·h⁻¹ decrease in J_sm-urea in the jejunum and cecum, respectively. Although not significant, the numerical reduction in J_sm-urea in the jejunum combined with the higher expression of the AQP mRNA levels seem to suggest the investigated AQP may play some functional role in transepithelial urea transport in the jejunum of pigs.

It has been well established that the major site of urea recycling in the GIT of ruminants for microbial amino acid production is the rumen (20). As such, ruminal expression of UT-B is high (40, 41, 51), and it has been demonstrated that urea transport across the ruminal epithelium is at least partially dependent on UT-B-inhibitable transport (51). In the current study, expression of UT-B expression was confirmed in the cecum of the growing pig, suggesting a role of UT-B in urea flux across the cecal epithelium. However, despite expression of UT-B and high urea flux, we could not ascertain a functional role of UT-B in urea transfer across the cecal epithelium of the pigs used in this study via inhibition of UT-B. Indeed, J_sm-urea increased when we added phloretin to the apical solution. Whereas phloretin has been successfully used to determine urea transport via UT-B in ruminants (1, 9, 51), its effects can be variable (34, 44). For instance, Stumpff et al. (44) had to abandon attempts to use phloretin to inhibit urea flux in the cecum of pigs because of alterations in tissue electrophysiology. Based on increased G_t and reduced I_sc in that study, the authors attributed their observations to nonspecific effects of phloretin (44) such as effects on uptake of glucose (35). In our study, G_t and mannitol flux numerically increased in the cecum upon addition of phloretin treatment. Because G_t is a measure of the electrical gradient created by ion movement across the epithelial tissue (21), an increase in G_t may be indicative of increased epithelial permeability, which is further corroborated by the observed increase in the diffusion of mannitol. Thus, we suspect the inhibitory effects of phloretin on UT-B may have been negated by increased membrane permeability.

Dietary composition and the resultant effects on the gastrointestinal environment, including pH, metabolite concentration, and microbial activity, may also play a role in the extent of urea recycling and nitrogen salvage; however, the impact of these factors is poorly understood. In ruminants, urea recycling in the rumen is enhanced with increased levels of fermentable substrate (e.g., fiber, carbohydrate), and fermentation products (e.g., short-chain fatty acids), while high levels of rumen ammonia have an inhibitory effect (1, 34, 52). Energy availability from fermentable substrates has been shown to be a limiting factor for microbial protein synthesis (38), and ammonia incorporation in microbial protein is increased with the availability of fermentable fiber (18, 22). Whereas high plasma urea-nitrogen concentration leads to some amount of urea transfer in the GIT, it has been suggested that urea recycling may respond to dietary protein intake and nitrogen requirements of the animal (13, 17, 45, 52), increasing with reduced amino acid supply and/or greater amino acid requirement. It has been suggested that plasma urea concentration (46) and dietary protein content (47) can influence urea flux in the pig. Moreover, Stumpff et al. (44) suggested that urea flux may be a mechanism for pH buffering in the cecum of the pig, since increased flux would increase ammonia concentration; however, the influence of specific digesta characteristics on urea recycling has not been examined. In the current study, neither pH nor ammonia concentration differed between jejunum and cecum even though urea flux differed significantly between these locations. Abdoun et al. (1) demonstrated a correlation between urea flux and pH in the sheep ruminal epithelium mounted in Ussing chambers and suggested J_sm-urea peaked at around a pH of 6.4. Our results suggest that in situ digesta has a similar pH to this, particularly in the jejunum, cecum, and colon. However, our buffer for the urea flux measurements was considerably more basic (pH 7.4) and, therefore, may have affected ex vivo measures of urea flux. Future studies should attempt to determine the impact of gut environmental conditions on urea flux and the regulatory mechanisms involved.

Overall, the current study confirms the expression of known urea transporter genes (AQP-3, -7, and -10 and UT-B) and significant urea flux along the GIT of pigs; however, the results also suggest that the functional role of these urea transports in the pig may be limited. Further research is necessary to establish the abundance of urea transporters and the role of urea transporters in epithelial transport of urea across the porcine intestine in the presence of dietary factors that influence urea recycling.

GRANTS

Funding for this project was provided by the Natural Sciences and Engineering Research Council of Canada.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

J.E.K., M.t.B., K.H., and D.A.C. performed experiments; J.E.K., A.K.A., K.H., G.B.P., and D.A.C. analyzed data; J.E.K., A.K.A., K.H., G.B.P., and D.A.C. interpreted results of experiments; J.E.K., A.K.A., and D.A.C. prepared figures; J.E.K., A.K.A., and D.A.C. drafted manuscript; J.E.K., A.K.A., K.H., G.B.P., and D.A.C. edited and revised manuscript; J.E.K., A.K.A., M.t.B., K.H., G.B.P., and D.A.C. approved final version of manuscript; G.B.P. and D.A.C. conceived and designed research.