Iberian pig adaptation to acorn consumption: I. Net portal appearance of metabolites

Animals, facilities, and diets analysis and composition

The study protocol was approved by the Bioethical Committee of the Spanish Council for Scientific Research (CSIC, Spain; project reference RECUPERA 2020, FEDER funding).

Six Iberian (Silvela strain; Sánchez Romero Carvajal, Jabugo S.A., Puerto de Santa María, Cádiz, Spain) gilts of similar BW (25 ± 0.4 kg initial BW) were utilized. One week before surgery, gilts were housed in individual pens in a controlled environment room (21 ± 1.5 °C), with ad libitum access to water and a standard barley-soybean meal diet (145 g crude protein (CP)/kg dry matter (DM) and 14.3 MJ ME/kg DM). After surgery, pigs were placed in metabolic cages and fed the standard diet 2.5 × ME for maintenance (422 kJ/kg^0.75 BW/d; Nieto et al., 2012) in two portions, at 09:00 (0.25) and 15:00 h (0.75). The day before the first sampling, the standard diet was substituted for a non-supplemented acorn diet and continued until the end of the experiment. The acorn diet consisted of 2.4 kg of acorns providing 1.425 kg DM and a total CP intake of 74.1 g. The Iberian pig shows the particular feeding behaviour whereby only the inner kernel is consumed and the hull is discarded. Composition and chemical analysis of the pre-experimental diet was performed by standard procedures (AOAC, 2000) and can be found elsewhere (González-Valero et al., 2016b). The nutrient composition of the acorn kernels fed is shown in Table 1. DM (no. 934.01), ether extract (no. 920.39) and total ash (no. 942.05) analysis were performed by standard procedures (AOAC, 2000). Total N was determined according to the Dumas’ method, by total combustion in TruSpec CN equipment (Leco Corporation, St. Joseph, MI, USA) and CP is defined as total N × 6.25. The neutral, acid and lignin detergent fractions (aNDFom (NDF assayed with a heat stable amylase and expressed exclusive of residual ash), ADFom (ADF expressed exclusive of residual ash) and Lignin(sa) (lignin determined by solubilization of cellulose with sulfuric acid), respectively) in kernels were analyzed by the method of Goering & Van Soest (1970). Neutral and acid detergent fibre was determined using an ANKOM²²⁰ Fibre Analyser Unit (ANKOM Technology Corporation, Macedon, NY, USA). Gross energy was measured in an isoperibolic bomb calorimeter (Parr Instrument Co., Moline, IL, USA).

Experimental procedures, schedules and calculations

The day before surgery (28 kg average BW) pigs were fasted and water was removed. Catheters were surgically placed in carotid artery and portal vein for blood sampling, and in ileal vein for infusion of para-aminohippuric acid (PAH; 2% w/v; Sigma-Aldrich Química S.A., Madrid, Spain) which is used as a marker to measure blood flow. Detailed description of design, construction and maintenance of the catheters, surgical procedures and post-surgical care of pigs was previously described (Rodríguez-López et al., 2013).

Pigs were adapted to close contact with the staff involved with sampling procedures to minimize stress. Two sampling periods were established and carried out under identical conditions after pigs completely recovered from surgery. The standard diet was changed to a non-supplemented acorn diet (34 kg average BW) and the first sampling (sampling period 1) was initiated one day after the diet change; the second sampling (sampling period 2) began 7 days later.

On the day of sampling, an initial 300 mg pulse dose of PAH was administered into the ileal vein 45 min prior to blood collection, and followed by a continuous infusion of 16 mg/min according to Yen & Killefer (1987), using a syringe pump (Model 33, Harvard Apparatus Inc., Holliston, MA, USA). Apyrogenic filters (MILLEX GP, Syringe Driven Filters Unit, 0.22 µm; Millipore, Carringtwohill, Ireland) were fitted to infusion syringes. Blood samples collected in 4.5 mL (Monovette VetMed; Sarstedt, Nümbrecht, Germany) heparinized tubes were taken simultaneously from carotid artery and portal vein −5 min, 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4, 5 and 6 h after feeding 0.25 of total daily acorn ration. Haematocrit was determined using a microcentrifuge (11,500 × g for 5 min; Biocen, Orto-Alresa, Ajalvir, Madrid, Spain). Plasma was obtained by centrifugation (4 °C, 1,820 × g for 30 min; Eppendorf 5810 R, Hamburg, Germany) and stored in aliquots at −20 °C for analysis of metabolites (ammonia, urea, glucose, lactate, albumin, creatinine, triglycerides and cholesterol) and PAH. After sampling, pigs were fed the remainder of the daily acorn ration and received 2.4 kg of acorn diet per day for one week after which the second sampling period (period 2) was carried out following identical protocol.

Portal blood flow (PBF) and portal plasma flow (PPF) were determined by the indicator dilution method using portal vein haematocrit and plasma PAH concentrations (Katz & Bergman, 1969). The PBF and NPA of metabolites were calculated according to the Fick principle of arterio-venous concentration difference and flow rate (Zierler, 1961). The PPF was calculated as infusion rate of PAH/(portal plasma PAH concentration − arterial plasma PAH concentration); PBF was calculated as PPF/[1 − (haematocrit/100)]. The NPA of a nutrient or metabolite was calculated as PPF × (portal concentration − arterial concentration). Positive NPA indicates production or release, whereas negative values indicate uptake or transfer of the substance.

Statistical analyses

The experimental unit was the pig. Measurements were made sequentially over time in the same experimental unit. Postprandial data were subjected to multivariate ANOVA analysis using the MIXED procedure of the Statistical Analysis Systems Institute (SAS, 2002) for repeated measures. The main effects in the model were sampling period (1 vs. 2), sampling time (from 0–6 h) and their interaction. Differences were considered significant when P < 0.05.

Ammonia and urea

Portal was greater than arterial ammonia concentration as ammonia is converted into urea by the liver preventing hyperammonemia which would disrupt the central nervous system (Huizenga, Gips & Tangerman, 1996). There was evidence that NPA of ammonia was reduced after one week of adaptation to acorns in spite of similar portal concentration of ammonia. The NPA of ammonia results from enterocyte amino acid catabolism, namely aspartate, glutamine and glutamate (Windmueller & Spaeth, 1975; Windmueller & Spaeth, 1980; Remesy & Demigne, 1989) and microbial hydrolysis of circulating urea taken up by the intestine, as large intestine microflora is able to produce ammonia from dietary or endogenous nitrogenous compounds (Huizenga, Gips & Tangerman, 1996). Nevertheless, we were unable to detect a significant difference in NPA of urea (see below) between sampling periods, which might indicate that bacterial breakdown of urea into ammonia was similar or that ammonia from urea hydrolysis was used for de novo synthesis of bacterial protein (Mosenthin et al., 1992). In the present study, a decrease in NPA of ammonia must be explained by a decreased amino acid catabolism in first pass by the intestinal mucosa. Indeed, details of a decreased NPA of amino acids will be presented in the subsequent publication. When expressed relative to digestible protein intake, NPA of ammonia after one week of acorn consumption was half that of sampling period 1 (0.27 vs. 0.14). Reduced NPA of ammonia would save metabolic energy, as less ammonia would be available for conversion to urea, a process that requires ATP. We found no effect of sampling time on ammonia concentration or NPA, contrary to other authors that offered a much greater CP (160 g/kg) amount to the pigs (Yen & Pond, 1990).

The adaptation for one week to acorn feeding resulted in a decrease of plasma urea concentration of 4 mmol/L although no effect was found on the pattern of postprandial changes in plasma urea concentration. Lower plasma urea concentrations have also been reported in pigs when high compared to low fibre diets were fed, reflecting a larger transfer of fermentable carbohydrates to the large intestine (Malmlof, 1987) or when fermentation processes were increased after starch infusion at the distal ileum (Mosenthin et al., 1992). In our study, the diet was identical for both sampling periods and NPA of volatile fatty acids, the main end products of bacterial metabolism, was similar (González-Valero et al., 2016a) indicating comparable fermentation patterns in the hindgut. Luminal urea may be reabsorbed to the portal circulation or hydrolyzed to ammonia by bacterial ureases (Huizenga, Gips & Tangerman, 1996) that is either reabsorbed or used for bacterial protein synthesis. Decreased portal urea may be due to increased urea degradation to ammonia and subsequent use for bacterial protein synthesis; the reduced luminal concentration of urea after hydrolysis would explain a decreased reabsorption and subsequent appearance at the portal vein.

Coma, Carrion & Zimmerman (1995) reported that N retention and urea N concentration were reflective of one another such that urea N is minimized as N retention is maximized. Certainly, increased N retention, e.g., protein deposition, would depress the plasma urea concentration and urea excretion. Although we found lower plasma urea after one week of acorn feeding we did not measure N retention in our study. Amino acids and ammonia stimulate urea production in the liver (González-Valero et al., 2014), so decreased systemic urea would agree with the lower NPA of ammonia after one week of acorn consumption.

NPA of urea was positive in our study indicating a transfer of urea from the gastrointestinal tract to the circulation. In contrast, Iberian pigs fed a barley-soybean diet of greater CP (101 g/kg; Rojas-Cano et al., 2016) showed negative NPA of urea although diet composition in both studies differed in terms of both amino acid and fibre amount and profile. Indeed, it has been established that dietary type of fermentable energy and protein content affect urea concentration and NPA (Rérat & Buraczewska, 1986; Younes et al., 1995a; Younes et al., 1995b).

Glucose and lactate

Portal glucose peaked at 1.5 h compared to maize or pea starch diets in white pigs (maximum glucose portal concentration at 0.5 and 1 h, respectively; Van der Meulen et al., 1997) and barley-soybean diets in Iberian pigs (maximum glucose portal concentration at 0.5 h; Rojas-Cano et al., 2016) indicating a slower digestion of acorn carbohydrates. A portion of acorn starch is resistant to ileal digestion (Morales et al., 2002) and enters the hind gut where fermentative degradation takes place, as we observed previously in Iberian pigs fed whole acorns (García-Valverde et al., 2007). However, calculations on previous work on glucose absorption kinetics in growing pigs showed greater rates of decrease in absorption after the maximum peak (8–10%/h (Rérat, Giusi-Perier & Vaissade, 1993; Van der Meulen et al., 1997)) to the one found in Iberian pigs fed acorns (5.3%/h on average 4.5 h after the glucose peak) in the present study. NPA of glucose did not change in period 2 in spite of decreased NPA of lactate. Whether lactate originated after glucose fermentation in the stomach and small intestine (Argenzio & Southworth, 1975; Cranwell, Noakes & Hill, 1976) or derived from glucose metabolism in the gut wall (Giusiperier, Fiszlewicz & Rérat, 1989) is not known. Although we cannot differentiate between microbial and metabolic lactate, most (0.87–0.93) of the lactic acid absorbed during the first 5 h after the intake of a diet containing different amounts of soluble and insoluble carbohydrates was L-lactic acid (Rérat, 1996). The very early appearance of lactate in the portal blood in our study seems to be in favor of increased glucose oxidation in the gut wall. Furthermore, the correlation between intestinal absorption of glucose and lactate may also indicate that lactate output is likely caused by a conversion of glucose to lactate by intestinal cells. Nevertheless, NPA of glucose was not decreased in sampling period 2, as it would be expected after increased glucose oxidation. It is possible that decreased glucose fermentation by intestinal microflora could have compensated lactate formation in the gut wall. Lactate is a product of postprandial glucose glycolysis (Mithieux, Rajas & Gautier Stein, 2004). Increased NPA of lactate after one week of acorn consumption will provide more precursors for hepatic gluconeogenesis. Indeed gluconeogenesis may be increased after one week of acorn consumption which is in line with the increased systemic glucose in period 2. As gluconeogenesis is an energy consuming process, this would imply augmented postprandial energy expenditure in Iberian pigs fed acorns for one week.

Albumin and creatinine

Albumin concentration and NPA in Iberian pigs fed a barley-soybean meal diet of adequate CP (Rojas-Cano et al., 2016) were similar to what has been observed in the present study. Changes in serum albumin levels are relatively unspecific although hypoalbuminemias are observed in animals with malnutrition or deficient protein absorption (Kaneko, Harvey & Bruss, 2008). Although plasma proteins are sensitive to nutritional influences, the changes are often subtle and difficult to detect and interpret. It may be speculated that it takes at least one week of consumption of a very low protein (acorn) diet to develop hypoalbuminemia in the conditions of our study. Albumin is catabolized by all metabolically active tissues providing amino acids for the natural turnover of protein in peripheral tissues (Evans, 2002) and in line with this, negative NPA of albumin in Iberian pigs fed acorns indicates albumin utilization by the visceral tissue, may be as a source of amino acids or as an antioxidant (Anraku et al., 2001) although adaptation to acorn feeding did not alter the NPA. Albumin also functions as a circulating depot and transports molecules for a large number of metabolites including fatty acids (Evans, 2002).

Creatine is released from muscle in amounts proportional to muscle mass (Rassin & Bhatia, 1992) and it is the main source of the creatinine into the portal blood system (Kim et al., 2008). In our study, higher portal and arterial creatinine in pigs consuming acorns for one week was observed, but we do not think that it is a reflection of a greater muscle mass considering that dietary protein was below requirements. Instead, it is possible that substantial amounts of arginine, methionine and glycine were utilized in the liver to synthetize creatine and not used for protein synthesis, as a result of the lysine deficiency of acorn protein. Since dietary creatine is only provided in animal products, virtually all of the creatine loss in pigs must be replaced via endogenous synthesis, which may represent a considerable metabolic burden. Indeed, the methyl group for creatine synthesis is transferred from S-adenosylmethionine (Brosnan, Da Silva & Brosnan, 2011) the universal methyl donor originated from methionine. Nevertheless, there was no change in creatinine metabolism at the PDV level. Similar NPA of creatinine to this study were found in Iberian pigs fed a 101 g CP/kg DM barley-soybean meal diet (Rojas-Cano et al., 2016).

Triglycerides and cholesterol

Contrary to protein and carbohydrates, absorbed fats are transported through the lymphatic system bypassing the portal vein and the liver, following digestion in the gastrointestinal tract. Although the portal system is commonly associated with transport of short and medium chain fatty acids, long chain fatty acids are also capable of being transported by the portal vein bound to albumin, albeit at a lower level (Weidman et al., 1980; Carlier & Bernard, 2000). Iberian pigs fed acorns (65.0 g fat/kg DM; present study) had greater plasma triglyceride concentration than Iberian pigs fed a sunflower oil diet of lower fat content (20 g/kg; Rojas-Cano et al., 2016), but with similar NPA of triglycerides. Diet composition has been reported to modulate plasma triglyceride content, so that it is increased by feeding high-fat diets (Mersmann et al., 1976) while the influence of dietary CP is controversial (Baker, Diller & Jordan, 1968; Madeira et al., 2016).

Although NPA of cholesterol was not altered, portal and arterial concentration slightly increased after one week of acorn consumption. Unfortunately, we have not measured any of the cholesterol fractions (HDL and LDL) that may have shown a different trend. Although it has been established that fat feeding (Allee, Baker & Leveille, 1971; Brooks et al., 1972; Mersmann et al., 1976) increases plasma cholesterol, Iberian pigs fed a 20 g sunflower oil/kg diet (Rojas-Cano et al., 2016) had similar cholesterol concentration and NPA to what was observed in the present study where fat intake was considerably greater.