2025 Volume 62 Article ID: 2025005
Trehalose (Tre) is composed of two molecules of D-glucose joined by an α,α-1,1 glucosidic linkage. Because Tre is utilized by the gut microbiome and enhances gut immunity in chickens, it is used as a feed ingredient. However, taste preference and metabolic dynamics of Tre in chickens are not fully understood. Therefore, in this study, we investigated the taste preference in chickens for Tre and the metabolism of this disaccharide. In a short-term drinking test, chickens preferred low concentrations of Tre solution while avoiding high concentrations. Instead, in a conditioned taste aversion test, chickens did not show taste aversion to Tre, implying that chickens do not have a sufficient taste for Tre. The initial feed intake rate increased when 0.5% Tre was added. Respiratory gas analysis revealed that intragastric administration of 1.0 M Tre weakly increased the respiratory quotient. Furthermore, approximately 50% of Tre was metabolized in chickens. These results suggest that chickens slightly taste the sweetness of Tre. Moreover, adding Tre to feed increases the chickens’ initial appetite, and they use approximately 50% of Tre as an energy source. This information is relevant for using Tre alone or as a supplement in poultry feed.
Trehalose (Tre) is composed of two molecules of D-glucose joined by an α,α-1,1 glucosidic linkage. In broiler chickens, Tre supplementation has been reported to alleviate intestinal inflammation and facilitate animal growth, probably by improving innate immunity, intestinal morphology, and the proliferation of Lactobacilli[1,2,3]. Tre supplementation during the starter/grower period also improves feed conversion in the subsequent finisher period, possibly by decreasing the abundance of Clostridium perfringens in broiler chickens[4]. Tre is the main blood carbohydrate in insects, and its use in poultry feed is expected to increase[5]. However, functional ingredients such as Tre are ineffective if chickens do not actively consume them. Hence, knowledge of the taste preference for Tre, its metabolism, or its avoidance by chickens can improve poultry farming.
In this study, the taste preference for Tre by chickens was examined using a brief access test, whereas a conditioned taste aversion (CTA) test was applied to explore whether chickens experienced the taste of Tre. Moreover, to consider the use of Tre as a poultry feed substrate, we analyzed its intake rate when added or not to feed, partly reflecting taste preference. Finally, we analyzed the metabolic changes before and after Tre administration using respiratory gas analysis and assessed the metabolizability of Tre in chickens. These data can inform breeders on using Tre as an ingredient in poultry feed.
Tre was obtained from Hayashibara Co. Ltd. (Okayama, Japan). LiCl and NaCl were obtained from Wako Pure Chemical Co. (Osaka, Japan).
AnimalsThe use of animals in the study, except for metabolic measurements, was approved by the Animal Research Committee of Hirosaki University (Approval No. A18005, A20001, A20002, and A20003) and followed the Rules for Animal Experimentation of Hirosaki University, Law Concerning the Human Care and Control of Animals (Law No. 105; October 1, 1973), Japanese Government Notification on the Feeding and Safekeeping of Animals (Notification No. 6; March 27, 1980), and Guidelines for Animal Experiments of affiliated institution or Guide for the Care and Use of Agricultural Animals in Research and Teaching (4th edition, 2020). These experiments were conducted from April 1, 2019, to March 31, 2021.
Fertilized eggs of the Rhode Island Red breed were obtained from the National Livestock Breeding Center’s Okazaki station (Okazaki, Japan), and chicks were used for the experiment. The chicks were maintained in a brooder with a heating and humidifying system (Belbird Ltd., Saitama, Japan) at approximately 30 °C under 24-h lighting. They were provided ad libitum access to commercial feed (Power Chick ZK Zenki; JA Zen-Noh Kitanihon Kumiai Feed Co., Ltd., Sendai, Japan) and water.
Brief access testBrief access tests were performed as described previously with slight modifications using the same transparent chambers and monitoring system (MFD-100 combined with ACTIMO-S II; Shinfactory, Fukuoka, Japan)[6]. Briefly, male and female chicks aged between 1–2 weeks at the start of the experiment were monitored over five consecutive days. The chicks were supplied with ad libitum commercial feed (Power Chick ZK Zenki), whereas water intake was restricted to 23 h and 50 min before the tests. In the tests, either the control solution (distilled water) or the test solution was presented to the chicks for 5 min. Then, normal tap water was presented for 5 min to minimize the variance in daily solution intake.
On days 1 and 2, the chicks were presented with the control solution for 5 min for
short-term intake training. On day 3, to get the chicks accustomed to the test solution
and avoid neophobia, they were presented with the test solution for 5 min. The test
solutions used in the present study were 0.05−0.50 M Tre. On days 4 and 5, the chicks were
randomly presented with the test or control solution. The intake of each solution was
measured and compared to evaluate chemosensory responses to the test stimuli. To evaluate
preference for the test stimuli, a preference index for each solution was calculated as
follows:
Chicks were weighed before the test and placed in a chamber where they could drink the solution. The solution was weighed before and after the test, and the difference between the weights was considered the solution intake. To eliminate differences in intake due to chick size, solution intake was divided by the weight of each chick (intake/body weight).
CTA testMale and female chicks aged 1 week at the start of the experiment were used. CTA tests were performed for 8 days following the same methods used in the brief access test, except that they involved conditioning for taste aversion. On days 1–4, the chicks were provided with water for 5 min as a test solution for short-term intake training. On day 5, the chicks were presented with taste solutions (0.25 M Tre or 0.50 M Tre), after which they were intraperitoneally injected with 115 mg/kg body weight of 0.12 M LiCl or the same volume of saline (22.5 mL/kg body weight). On day 6, the chicks were provided with water as a test solution. On days 7 and 8, the chicks were randomly presented again with the taste solutions or water (those presented with the taste solution on day 7 were presented with water on day 8, and vice versa). To analyze the chicks’ behavioral preferences, we compared the intake of taste solutions and water on days 7 and 8. Because in the CTA test, experiences with taste solutions may affect subsequent results, chicks used for this test differed from those employed in the brief access tests and were not subjected to subsequent trials.
One-bowl feed intake testEight male and female chicks aged 4–8 days at the start of the experiment were used. The tests were performed on 3 consecutive days. On day 1, chicks were supplied with ad libitum water and commercial feed (Power Chick ZK Zenki). Water and feed were restricted to 15 h before day 2. On day 2, four chicks were moved to a gray acrylic box (W 600 mm × D 600 mm × H 300 mm) with an open ceiling and divided into four sections by transparent partitions. Each chick was placed in one section (W 300 mm × D 300 mm × H 300 mm) with ad libitum water and experimental feed. The chicks were allowed to observe each other to avoid isolation stress. Feed intake was measured hourly for 3 h. At the end of day 2, the chicks were given ad libitum water and commercial feed, followed by another 15 h of fasting and water deprivation prior to day 3 experiments. The same procedure was performed on day 3, whereby the chicks were fed either a normal commercial feed (Power Chick ZK Zenki) or normal feed with 0.5% (w/w) Tre. The order of experimental feeds on days 2 and 3 was randomized. Two sets of experiments were conducted with four chicks, amounting to data from eight chicks.
Respiratory gas analysisMale and female chicks aged 4–20 days at the start of the experiment were used. The
chicks were individually housed in acrylic chambers (W 120 mm × D 240 mm × H 150 mm) for 6
h before administration to acclimatize them to the experimental environment. During this
time, the chicks were prohibited from accessing feed but had free access to water. Tre
solution (0.2 M and 1.0 M) or distilled water was administered through a gastric tube, and
the expired air was analyzed before and after administration. The volume of the
administered solution was 20 mL/kg. The respiratory quotient (RQ) was calculated based on
oxygen consumption (VO2) and CO2 production (VCO2) per
unit time.
Gas analysis was performed using an open-circuit metabolic gas analyzer connected directly to a mass spectrometer (ARCO-2000; Arco System, Inc., Chiba, Japan). Room air was pumped into the chambers at a rate of 0.3 L/min. The expired air was dried in a thin cotton column and directed to an O2/CO2 analyzer for mass spectrometry.
Metabolizable energy and metabolizability of Tre in chickensThe purpose of this experiment was to determine nitrogen-corrected metabolizable energy (MEn) and metabolizability of Tre in chickens according to the digestion test found in ‘Instructions on amino acid, total digestible nutrients or metabolizable energy of feed,’ which is described in ‘Enforcement of Partial Revision of Ministerial Ordinance on the Specifications and Standards of Feeds and Feed Additives’ (Ministry of Agriculture, Forestry and Fisheries (MAFF) Ordinance, 56 Chiku B, No 1594 of July 27, 1981, Japan). This experiment was commissioned by the Japan Scientific Feeds Association (Tokyo, Japan) under code number S-14-P-001. The use of animals throughout this protocol was approved by the Animal Experiment Control Committee of the Japan Scientific Feeds Association (Approval No. 584) and followed the Rules for Animal Experimentation of Japan Scientific Feeds Association, Law Concerning the Human Care and Control of Animals (Law No. 105; October 1, 1973), as well as the Japanese Government Notification on the Feeding and Safekeeping of Animals (Notification No. 6; March 27, 1980). The experiment was conducted from June 12 to June 20, 2014.
Ten male broiler chickens (UK Chunky; Mori Breeding Farm Co., Ltd., Koriyama, Japan) aged ~7 weeks were used. Chickens were randomly allotted to two treatment groups: a group fed a basal diet (Table 1) and a test group fed a basal diet plus Tre (9:1). Chromic oxide was added to all diets at a concentration of 0.1%. The chickens were reared individually in metabolic cages. After acclimatization to the experimental conditions by feeding a basal diet, five chickens in each group were fed ad libitum each diet for 8 days. A mixture of feces and urine was collected for 3 days, starting 6 days from the onset of the experiment, for each bird. The mixture was weighed, dried at 60 °C for 2 days, and then air-dried. Three-day mixtures were combined and ground to obtain analytical samples.
| Ingredients | (%) | Ingredients | (%) |
| Corn | 67.54 | Vitamin ADE 2) | 0.20 |
| Soybean meal | 18.00 | Trace minerals 3) | 0.20 |
| Corn gluten meal | 6.00 | DL-Methionine | 0.13 |
| Fish meal (CP65%) | 5.00 | L-Argininge | 0.11 |
| Dibasic calcium phosphate | 1.22 | L-Lysine-HCl | 0.10 |
| Calcium carbonate | 0.95 | L-Valine | 0.03 |
| NaCl | 0.30 | L-Threonine | 0.01 |
| Vitamin B group 1) | 0.20 | Vitamin K34) | 0.01 |
| Total | 100.00 | ||
1) g/kg: thiamine nitrate 2.0, riboflavin 10.0, pyridoxine-HCl 2.0, nicotinamide 2.0, D- pantothenic acid-Ca 4.35, choline chloride 138.0, folic acid 1.0.
2) /kg: vitamin A oil 10,000 IU, vitamin D3 oil 2,000 IU, dl-α-tocopherol-acetate 20 mg.
3) g/kg: Mn 80, Zn 50, Fe 6, Cu 0.6.
4) 5% preparation.
The nitrogen (N) content of the samples was determined according to Standard Feed Analysis (MAFF, April 1, Shoan 19, No. 14729, 2008). Gross energy (GE) was determined using a bomb calorimeter. For the basal diet, test diet, and fecal-urine mixture sample, N and GE were determined as mentioned above, whereas chromic oxide was determined using a colorimetric method[8].
MEn was calculated as reported previously[9], and ME and metabolizability of the sample were determined using the following equations:
Paired t-test, unpaired t-test, and two-way repeated ANOVA were used for statistical analyses performed in Excel 2011 (Microsoft Corp., Redmond, WA, USA) or SPSS Statistics (version 27.0.1.0; IBM Corp., Chicago, IL, USA). Differences were considered statistically significant at P < 0.05.
The brief access test showed no difference between the intake of low-dose Tre solution (0.05–0.10 M) or water. However, chickens drank significantly more 0.25 M Tre solution than water (Fig. 1A) but less of the 0.50 M Tre solution (Fig. 1A). The preference index, which was calculated based on the observed intakes, confirmed the intake values, showing a significant preference for 0.25 M Tre and dislike for 0.50 M Tre compared to water (Fig. 1B).
Behavioral responses to trehalose solutions in chickens. (A) Water intake/body weight (BW) (white bar) and test solution intake/BW (gray bar) were compared. We used 0.05 M, 0.10 M, 0.25 M, and 0.50 M trehalose as test solutions. (B) Preference indices for the corresponding concentrations of trehalose are shown. Values denote the mean ± SD (n = 6–13). Open circles represent individual chicks. *P < 0.05, and **P < 0.01 by paired t-test, compared to water intake.
Next, we examined whether 0.25 M and 0.50 M Tre solutions induced CTA. When conditioned by 0.25 M Tre, no significant differences in water or 0.25 M Tre intake were observed between the saline and LiCl groups (Fig. 2A, B). Similarly, conditioning by 0.50 M Tre revealed no significant differences in water or 0.50 M Tre intake between saline and LiCl groups (Fig. 2C, D). Accordingly, 0.25 M and 0.50 M Tre solutions did not induce CTA in chickens under the present experimental conditions.
Water or trehalose solution intake/body weight (BW) of chickens conditioned to avoid the trehalose solution (conditioned stimulus; CS) by LiCl or given saline (control). (A) Water intake/BW and (B) 0.25 M trehalose solution intake/BW of chickens injected with saline or LiCl. (C) Water intake/BW and (D) 0.50 M trehalose solution intake/BW of chickens injected with saline or LiCl. Values denote the mean ± SD (n = 5–6). Open circles represent individual chicks.
The results of the one-bowl feed intake test are shown in Fig. 3. After 15 h of water and feed deprivation, the chickens were fed a normal diet or the normal diet with 0.5% Tre for 3 h with water. The cumulative intake of the normal diet increased over time (open circles in Fig. 3). In contrast, the intake of the normal diet with 0.5% Tre was significantly higher already 1 h after the start (gray circles in Fig. 3), the cumulative intake at 2 and 3 h did not differ between the two groups. These results suggest that adding 0.5% Tre to the diet increased only the initial feed intake.
Cumulative intake of a normal diet or normal diet with 0.5% trehalose for 3 h under one-bowl diet conditions. Values denote the mean cumulative diet intake ± SE (n = 8). Two-way ANOVA (group, P = 0.016; time, P < 0.001; group × time, P = 0.277). *P < 0.05 by paired t-test.
Respiratory gas analysis showed that intragastric administration of 0.2 M Tre did not change the RQ (Fig. 4A). Administration of 1.0 M Tre significantly increased the RQ compared with water administration (Fig. 4B). This suggests that high doses of Tre enhance carbohydrate metabolism in chickens.
Changes in the respiratory exchange ratio (RQ) after intragastric (IG) administration of trehalose in chickens. (A) Changes in the RQ after IG administration of 0.2 M trehalose or water. Two-way ANOVA (group, P = 0.944; time, P < 0.001; group × time, P = 0.964). (B) Changes in the RQ after IG administration of 1.0 M trehalose or water. Two-way ANOVA (group, P = 0.19; time, P < 0.001; group × time, P < 0.001). *P < 0.05 by unpaired t-test. Values denote the mean ± SE (n = 5–7).
GE, ME, and metabolizability of Tre in chickens are reported in Table 2. Because the metabolizability of Tre was approximately 50%, half of the Tre intake was available as a source of energy.
| Gross energy (Mcal/kg) |
Metabolizable energy (Mcal/kg) |
Metabolizability (%) |
| 3.51 | 1.70 ± 0.07 a) | 48.4 ± 2.0 a) |
a) Mean ± SD (n = 5)
In this study, we explored the preference of chickens for Tre, a functional ingredient in poultry feed. First, using a brief access test, we showed that chickens preferred a 0.25 M Tre solution and avoided a 0.50 M Tre solution (Fig. 1). Similarly, using the same experimental protocol, Higashida et al. showed that chickens preferred 0.1 M sucrose and avoided 1.0 M sucrose[7]. Thus, it is possible that chickens prefer sweet solutions but avoid highly concentrated ones.
Next, we performed a CTA test to explore whether the chickens could taste Tre. While there was no statistically significant difference between the saline and LiCl groups when conditioned with 0.50 M Tre (Fig. 2D), Tre intake was 0.078 ± 0.022 g/g in the former and 0.065 ± 0.020 g/g in the latter, corresponding to a more than 10% reduction in the LiCl group (Fig. 2D). Yoshida et al. conducted a CTA test with 0.3 M L-Serine solution using chicks and reported that, although the intake of L-Serine was reduced in the LiCl group during the first conditioning, the difference was not statistically significant and aversion learning was established only after the second conditioning[10]. In contrast, with a 0.3 M L-Alanine solution, CTA learning was established after the first conditioning[10]. These results indicate that CTA learning in chickens progresses stepwise for substances with a weak taste. Therefore, a 0.50 M Tre solution can also be considered taste aversive, but not a solution in which CTA learning is immediately established, providing Tre with a slightly taste stimulatory effect.
In a previous report, although a lactose solution was preferred in chickens in a 30-min drinking test[7], CTA for lactose could not be established[10]. Conversely, glucose, galactose, sucrose, and maltose have been used successfully to establish CTA in chickens[10]. Although carbohydrates that induce CTA learning may be clearly perceived by chickens, carbohydrates that do not induce CTA learning may not impact taste. Because the metabolizability of Tre is approximately 50% (Table 2), there may be a relationship between such low metabolizability and the weak physiological basis for tasting Tre in chickens. Tre has approximately 45% of the sweetness associated with sucrose in humans[11]. Accordingly, Tre may only induce a weak, sweet taste in chickens. In cell-based assays, Tre activated mouse Tas1r2/Tas1r3 (sweet taste receptor), mouse Tas1r1/Tas1r3 (umami taste receptor), and mouse Tas1r3[12]. Tas1r1 (T1R1) and Tas1r3 (T1R3) are expressed in chicken taste buds[13], whereas the tas1r2 gene is absent in chicken[14]. Thus, chickens may weakly taste Tre via the T1R3 homodimer or T1R1/T1R3 heterodimer, although the latter is rarely formed[13].
Tre is usually added as a powdered feed, aiming to regulate intestinal flora, improve intestinal morphology, and boost feed conversion ratios[2,4]. In the brief access test with the Tre solution, a range of concentrations was tested, akin to that used to assess other sugars[7]. However, when mixed with feed, Tre crystals enter the mouth directly, and therefore, it is difficult to obtain a clear correspondence between the concentration of Tre in the solution and the amount of Tre added to the feed. Thus, in this study, 0.5% Tre was used in the feed intake test because this concentration significantly improves intestinal morphology, weight gain, and feed conversion in broilers when added to the feed[2]. This choice makes perfect sense from an economic perspective. Here, 0.5% Tre in feed increased the feed intake rate (Fig. 3), suggesting that it might promote initial appetite due to its weak sweetness. However, because this effect disappeared after 2 h (Fig. 3), we speculate that Tre addition may have increased satiety, resulting in negative feedback.
Intragastric administration of 0.2 M Tre did not change the RQ, whereas 1.0 M Tre increased the RQ from 30 to 100 min after administration. Because an increase in RQ indicates enhanced carbohydrate metabolism, we believe that Tre was metabolized and used as an energy source. Birds do not possess a gene encoding trehalase, which is also absent from the proteome of the avian intestinal brush border membrane[15]. Interestingly, Chotinsky et al. detected trehalase activity in enterocytes obtained from the jejunum and ileum of 18-day embryos and until 7-day post-hatch chicks, with the activity gradually decreasing with age[16]. Brun et al. speculated that such weak avian trehalase activity was due to contamination of the homogenate assay with exogenous trehalase from invertebrate tissue remains or the intestinal microbiome[15]. Tre uptake occurs through simple diffusion in chicken brush-border membrane vesicles lacking disaccharidase activity[17]. Thus, the metabolic changes induced by Tre administration observed in this study may involve simple diffusion and exogenous trehalase from the intestinal microbiome. Even though exogenous trehalase may be functional, total trehalase activity in chickens is so low it may not be sufficient to degrade Tre to glucose in the intestinal tract. Accordingly, Tre might slightly alter the RQ in the case of high-dose administration.
The present study implies that approximately 50% of Tre, which enters the gastrointestinal tract, is metabolized by chickens through two pathways: it is either absorbed from the intestinal tract by simple diffusion or metabolized by the gut bacteria, and the metabolites are absorbed from the intestinal tract. The remaining 50% may be utilized to grow gut bacteria, which are excreted in the feces and urine. These speculations align with previous reports on the role of gut bacteria[1,2,3,4].
Young chicks aged 0–2 weeks were used for most experiments in this study. This age is akin to those previously used to examine taste sensing[6,7,10]. In contrast, the metabolizability experiments used chickens aged ~7 weeks. This is because, in young chicks, many feathers get mixed with feces and urine due to feather change, and the volume of feces and urine is lower than in older chickens. Consequently, caution should be exercised when comparing metabolizability results across different studies due to age discrepancies.
In the present study, the concentration and administration route of Tre differed between experiments. This was because we searched for a concentration range, which allowed observation of the Tre response in each experiment. The amount of Tre entering the stomach varied among individuals in drinking and feeding behavior tests. However, for respiratory gas analysis, we considered it important to administer a defined amount of Tre into the stomach accurately. Therefore, each experiment was conducted within the concentration range, in which changes caused by Tre were observed. Although it would be ideal to match the concentrations and routes of administration in all experiments, this was not feasible in the present study.
In summary, our findings suggest that chickens have a slight taste for Tre, such that its addition to feed stimulates initial appetite, allowing about 50% of Tre to be used as an energy source. Based on these features, Tre is a valuable and functional carbohydrate for the gut microbiome in chickens.
This study was supported by a joint research grant to Fuminori Kawabata by Hayashibara Co., Ltd., Japan.
Conceptualization: Fuminori Kawabata, Kazuhisa Mukai; Methodology: Fuminori Kawabata; Analysis and investigation: Fuminori Kawabata, Misako Sakai, Hiroki Murasawa, Yu Komine, Yuko Kawabata; Writing – original draft preparation: Fuminori Kawabata, Misako Sakai, Hiroki Murasawa, Yu Komine; Writing – review and editing: Fuminori Kawabata, Kazuhisa Mukai, Yuko Kawabata; Writing – revised manuscript: Fuminori Kawabata, Misako Sakai, Hiroki Murasawa, Yu Komine, Kazuhisa Mukai, and Yuko Kawabata.
This study was supported by a joint research grant to Fuminori Kawabata by Hayashibara Co., Ltd., Japan. Kazuhisa Mukai is an employee of Hayashibara Co. Ltd. The company name has been changed to Nagase Viita Co., Ltd.
