Reduced neophobia: A potential mechanism explaining the emergence of self-medicative behavior in sheep

Physiology & Behavior Volume 135, August 2014, Pages 189-197 Physiology & Behavior Author links open overlay panelA. VaninaEgeaaJeffery O.HallbJamesMillercCaseySpackmanbJuan J.Villalbad a Instituto Argentino de Investigaciones de las Zonas Áridas (IADIZA), Mendoza, Argentina b Department of Animal, Dairy, and Veterinary Sciences, Utah State University, Logan, UT, USA c Department of Pathobiological Sciences, Louisiana State University, Baton Rouge, LA, USA d Department of Wildland Resources, Utah State University, Logan, UT, USA Received 13 May 2014, Revised 6 June 2014, Accepted 13 June 2014, Available online 20 June 2014. crossmark-logo https://doi.org/10.1016/j.physbeh.2014.06.019 Get rights and content Highlights • We explored the selection of novel foods by parasitized and healthy lambs. • Parasitized lambs modified their feeding behavior relative to non-parasitized lambs. • Ingestive responses were influenced by the type of novel food and flavor on offer. • Diverse diets and risk proneness may increase the likelihood of finding medicinal plants. Abstract Gastrointestinal helminths challenge ruminants in ways that reduce their fitness. In turn, ruminants have evolved physiological and behavioral adaptations that counteract this challenge. For instance, emerging behavioral evidence suggests that ruminants self-select medicinal compounds and foods that reduce parasitic burdens. However, the mechanism/s leading to self-medicative behaviors in sick animals is still unknown. We hypothesized that when homeostasis is disturbed by a parasitic infection, consumers should respond by increasing the acceptability of novel foods relative to healthy individuals. Three groups of lambs (N = 10) were dosed with 0 (Control—C), 5000 (Medium—M) and 15 000 (High—H) L3 stage larvae of Haemonchus contortus. When parasites had reached the adult stage, all animals were offered novel foods and flavors in pens and then novel forages at pasture. Ingestive responses by parasitized lambs were different from non-parasitized Control animals and they varied with the type of food and flavor on offer. Parasitized lambs consumed initially more novel beet pulp and less novel beet pulp mixed with tannins than Control lambs, but the pattern reversed after 9 d of exposure to these foods. Parasitized lambs ingested more novel umami-flavored food and less novel bitter-flavored food than Control lambs. When offered choices of novel unflavored and bitter-flavored foods or different forage species to graze, parasitized lambs selected a more diverse array of foods than Control lambs. Reductions in food neophobia or selection of a more diverse diet may enhance the likelihood of sick herbivores encountering novel medicinal plants and nutritious forages that contribute to restore health. Previous article Next article Keywords Tannins Umami Bitter Grazing Ovis aries Food selection Feeding 1. Introduction Free-ranging herbivores face many challenges. In addition to avoiding predators, they must cope with disease and acquire a balanced diet from foods that differ greatly in nutritional and toxicological composition [1,2]. Many plant tissues contain plant secondary metabolites (PSMs) which have long been recognized as defensive chemicals that deter herbivory through their aversive orosensory and toxic effects [3]. In contrast to these negative actions, recent evidence suggests that some PSMs can contribute to increase animal fitness if they are ingested at appropriate doses [4,5]. A good example of such duality is represented by condensed tannins, a group of PSM which can have both antinutritional and medicinal effects on consumers [6,7]. Herbivores have evolved mechanisms to maintain homeostasis and contemporarily learn to avoid or prefer certain foods because they lower [8] or increase [9] their fitness, respectively. Medicines like some PSMs raise fitness by reducing disease and thus are expected to be preferred during some instances of the lifetime of the individual [10,5]. In a seminal study, Huffman and Seifu [11] observed that wild chimpanzees suffering from parasite-related diseases consumed the bitter pith of the plant Vernonia amygdalina, which contains sesquiterpene lactones and steroid glucosides with antiparasitic activity at the doses consumed by the animals [12]. Since these pioneering results, additional evidence pointing to the use of PSMs to recover or to maintain health has been reported for ruminants [13,14], insects [15] and birds [16]. In addition to PSMs, mammalian herbivores self-medicate with substances such as polymers and clays to attenuate the negative impacts of toxicosis and ruminal acidosis [17]. The study of self-medication in animals led to the emergence of the word “zoopharmacognosy,” meaning the process by which animals use PSMs or other non-nutritional substances to treat and prevent disease [18]. While past research has uncovered many examples of self-medicative behavior in different animal species, there is a lack of basic understanding of how the behavior is triggered. Learning self-medicative behavior involves consuming a medicine with potential toxic [3] effects to consumers which may also make the food unpalatable [19]. What are the mechanisms by which sick mammalian herbivores would be willing to incorporate potentially toxic and unpalatable PSMs into their diets at the expense of – or in addition to – consuming their required amounts of nutrients? When physiological requirements are met and the animal is in a homeostatic state – such as when balanced rations are available ad libitum – ruminants eat only small amounts of novel foods, i.e., they are neophobic [20]. This neophobic behavior likely evolved to decrease the likelihood of consuming harmful foods which reduce animal fitness. However, departure from homeostasis (e.g., when there is an inadequate or unbalanced supply of nutrients) may reduce neophobia [20]. Birds on a positive energy budget are averse to risk, but prone to it when experiencing a shortage in the supply of energy to their bodies [21]. This is because individuals which have acquired abundant reserves of energy have more potential fitness to lose from taking risks (e. g., selecting novel foods, exploring new places) than individuals which have not [22]. We hypothesized that when homeostasis is disturbed by a parasitic infection (which imposes metabolic constraints), consumers should increase the acceptability of novel foods relative to healthy individuals. Sick individuals may have less potential fitness to lose from taking risks (e.g., selecting novel foods) than healthy ones. Additionally, reduced neophobia represents the potential benefit for the consumer of increasing the likelihood of encountering medicinal compounds and nutrients that restore health. Once the individual experiences the benefit of medicines and nutrients, associative learning (i.e., associations between taste and relief experienced after ingesting a medicinal food) will maintain and/or reinforce self-medicative behaviors. To test this hypothesis we induced different magnitudes of parasitic infections (Haemonchous contortus) in sheep (Ovis aries) and then determined the animals' feeding responses to novel flavors, feeds, and forages. 2. Methods We conducted the study at the Green Canyon Ecology Center (pen study), and at the Pasture Research Facility (grazing study) at Utah State University, Logan, Utah, U.S.A. Research protocols for the study were approved by the Utah State University Institutional Animal Care and Use Committee (IACUC No. 2227). 2.1. Animals and infection Thirty lambs of both sexes [commercial crossbreds, 2–3 months old; live weight (LW) of 26 ± 0.5 kg (average ± SEM)] had free access to sodium chloride with trace mineral blocks and fresh water throughout the study. One month before the study, lambs were weaned, penned individually and received alfalfa pellets in ad libitum amounts and barley grain (300 g/lamb/day). The dietary experience of the lambs' mothers was limited to a basal diet of alfalfa hay and barley grain. The lambs were kept outdoors under a protective roof in adjacent pens measuring 2.4 × 3.6 m. Lambs were blocked by LW and randomly assigned to three treatment groups (n = 10). Seven weeks before the start of the experiment (June 8, 2013), sheep were drenched with the anthelmintics pyrantel pamoate and albendazole at 25 mg/Kg and 7.5 mg/Kg LW, respectively, delivered in separate doses. Ten days later (June 18, 2013), fecal samples were taken at 0800 from the rectum of each animal, stored in an ice chest and analyzed for fecal egg counts (FEC) during the same day using the MacMaster technique to ensure that animals had very low to nil [< 100 eggs per gram (epg)] parasitic burden before the start of the study. Subsequently (June 19, 2013), lambs of each group were dosed orally with infective third-stage (L3) larvae of H. contortus in the following amounts: 0 (Control group: C), 5000 (Medium group: M) and 15 000 (High group: H) L3/lamb. 2.2. Novel foods and flavors 2.2.1. General protocol At 0800 all lambs were presented with a three-bucket choice of a novel feed containing 3 concentrations of a novel flavor. Refusals were collected every 20 min, weighed and offered again for a period of 2 h to determine their eating pattern. The position of the buckets was distributed at random inside each individual feeder every day. Lambs were then offered 2 kg of alfalfa pellets from 1200 to 1400 daily; refusals were collected and weighed. No other feed was offered until the next day. The order of presentation of novel flavors was: 1—condensed tannins, 2—umami, 3—bitter, and 4—coconut, maple and apple flavors. Since animals were offered choices among three alternatives, the protocol was devised to give animals enough time of exposure (7 to 9 days) to sample all options. In addition, the design allowed for assessing initial responses to the novel flavored foods relative to ingestive responses occurring when animals gained more experience at ingesting these foods. All lambs were weighed on August 1 (38 ± 0.8 kg average LW) and August 30 (44 ± 0.8 kg average LW) and they were fed ad libitum amounts of alfalfa pellets in-between tests. Test duration was variable, as it depended on stable intake values in three consecutive days (average intake values within ± 5%). Blood samples were drawn from all lambs by venipuncture of the jugular vein 20 days before the start of the study, and on days 1, 20 and 40 of the study for analysis of red and white cell parameters using a Hemavet HV950FS (Drew Scientific, Oxford, UK) hematology analyzer. Fecal samples were collected as described before during July 19 and 25, August 1 and 21 and September 16 to determine FEC using MacMaster technique. 2.2.2. Condensed tannins Fecal samples were taken a month after larval dosage (July 19), and a week later (July 25, 2013) and FEC suggested parasites had reached the adult stage (Table 1). Thus, tests with condensed tannins were conducted from July 26 to August 3, 2013. The novel food was beet pulp (ground to 1–2 mm particle size), and it was mixed (0, 5, and 10% as fed) with quebracho tannin extracted from Aspidosperma quebracho (Tannin Corporation, Peabody, MA). Quebracho tannin was in a fine powder form and it was also novel to all the lambs. Table 1. Fecal egg counts (FEC; epg) in 3 groups of lambs (n = 10) exposed to novel foods and flavors (July 26 to August 30, 2013) and novel pastures (August 31 to September 12, 2013). Lambs of each group were dosed orally with infective third-stage (L3) larvae of Haemonchus contortus in the following amounts: 0 (Control group: C), 5000 (Medium group: M) and 15 000 (High group: H) L3/lamb. Group1 June 18 July 19 July 25 August 1 August 21 September 16 C 0 0a 0a 0a 0a 0a M 0 1235b 2350b 3095b 3165b 3906b H 0 2415c 6245c 8165c 7985c 8240c 1 Means in a column with different superscripts differ (group effect; P < 0.10). SEM = 699. 2.2.3. Umami Testing with umami taste was conducted from August 5 to August 12, 2013. The novel food was tall fescue hay (ground to 1–2 mm particle size), and it was flavored (0, 0.5, and 1% as fed) with umami flavor (ref-5435-LUCTA-Montornés del Vallés, Spain), a fine powder which was also novel to all the lambs. 2.2.4. Bitter Testing with bitter taste was conducted from August 14 to August 20, 2013. The novel food was wheat bran (ground to 1–2 mm particle size), and it was flavored (0, 0.5, and 1% as fed) with bitter flavor (Luctarom, a quinine-based flavor ref-5432-LUCTA-Montornés del Vallés, Spain), a fine powder which was also novel to all the lambs. 2.2.5. Flavors Testing with flavors was conducted from August 22 to August 30, 2013. The novel food was grape pomace (ground to 1–2 mm particle size), and it was flavored (1% as fed) with coconut, maple and apple flavors (LUCTA-Montornés del Vallés, Spain), in fine powders which were also novel to all the lambs. 2.3. Novel pastures The day after testing with novel flavors ended, all lambs were transported to the Utah State University pasture research facility at Lewiston, Utah. Because lambs are reluctant to eat in isolation [23], pairs of lambs were formed at random within each treatment group (n = 5 pairs/group). The pair was considered the experimental unit, and once formed and identified by specific numbers and colors, pairs were always tested together. The study was conducted from August 31 to September 12, 2013 on three 0.08-ha (11 × 73.2 m) adjacent (by the longer side of the rectangular area) experimental paddocks. One paddock was seeded with Festuca arundinacea, (Tall fescue [TF]), variety “Kentucky 31” endophyte-infected. The center paddock was seeded with Onobrychis viciifolia Scop. (Sainfoin), variety “Shoshone.” A third paddock was seeded with Medicago sativa (alfalfa [AA]), variety Vernal. Adjacent to the shorter sides of the rectangular area, two 0.13-ha paddocks (40 × 33 m) were seeded with Dactylis glomerata (orchard grass [OG]). Orchard grass also grew around the perimeter of the entire rectangular area. The perimeters of the three experimental paddocks and the two orchard grass plots were fenced with a temporary electric fence. In addition, an electric fence running perpendicular to the longer side of the rectangular area equally divided in half the experimental paddocks, such that two pairs of lambs could graze simultaneously, each pair released from the opposite shorter sides of the paddocks. At 1700 lambs from each group were penned overnight in two holding pens (4.5 × 6.5 m) built under a protective roof, located between the orchard grass and the experimental paddocks, each located in the opposite shorter sides of the paddocks. Two pairs of lambs from each treatment were penned at one side of the paddock and the remaining three pairs were penned at the opposite side of the paddock. Lambs were allowed to graze the orchard grass paddock from August 31 to September 4. The aim of this washout period was to ensure that differences in grazing responses could be attributed to the treatments animals experienced during conditioning and not to the short-term effects of forages consumed prior to testing. During daily tests beginning at 0600, a pair of lambs was released from the center of each of the two holding pens where they had free access to the three experimental pastures: fescue, alfalfa, and sainfoin. Two pairs of lambs were tested simultaneously, released from each one of the 2 holding pens, and each pair of lambs was allowed to forage for 20 min. The foraging behavior of both pairs of lambs was recorded by an observer from a 4-m high observation tower adjacent to the north side of the experimental area and equidistant from both holding pens. Scan samples were used [24], taken at 1-min intervals, to assess the incidence of feeding on each of the forage species and bouts of inactivity. Frequency of feeding on each species was calculated as a percentage of the total number of scans in which lambs were feeding. It was also recorded the total number of scans of grazing events and non-grazing events (bouts of inactivity such as not eating, resting, and searching). Pairs normally grazed together (> 95% of the time) on the same plant species. If individuals were performing different behaviors each behavior was recorded for each individual. After the 20 min period, lambs were returned to their respective orchard grass paddocks and a new pair of lambs was released from the holding pens for a new period of 20 min. This procedure was repeated in a sequential order until all pairs of lambs foraged for 20 min. The abundant biomass in the experimental paddocks (Table 2) relative to the number of animals tested and the length of each foraging session prevented potential changes in foraging behavior induced by sward heterogeneity generated by other groups of lambs grazing previously at the same location. Lambs grazed orchard grass until 1700 when they were penned overnight in the two holding pens and no other food was offered until the next day. Observations were carried out from September 5 to September 12, 2013. The order in which pairs were released to graze was changed at random daily using a randomization matrix such that each pair of lambs grazed in a different order every day during the study. The study lasted for 8 days. Pasture biomass was assessed by taking four random readings of a 0.1089 m2 rising-plate meter [25] in each paddock at beginning and end of the study. Table 2. Nutritional composition (on a dry matter basis) of the foods and biomass and nutritional composition of the pastures used during the study. Foods CPa % NDFb % ADFc % Biomass Kg/ha Alfalfa pellets 15.6 ± 1.6 41.1 ± 1.3 30.1 ± 0.8 Beet pulp 10.1 ± 0.2 38.8 ± 0.8 24.5 ± 0.2 Grape pomace 14.3 ± 1.1 57.0 ± 0.5 52.0 ± 0.6 Tall fescue hay 10.1 ± 0.5 61.1 ± 0.4 38.8 ± 0.5 Wheat bran 17.9 ± 0.2 29.7 ± 0.1 10.7 ± 0.2 Pastures Date Alfalfa September 5 22.1 ± 0.2 37.8 ± 2.0 31.3 ± 1.4 8448 ± 457 September 12 21.5 ± 0.6 39.0 ± 0.2 33.4 ± 0.1 7449 ± 207 Orchard grass September 5 10.9 ± 0.7 60.2 ± 1.3 38.3 ± 1.1 7025 ± 162 September 12 10.9 ± 0.5 61.6 ± 0.2 38.4 ± 0.2 6359 ± 510 Sainfoin September 5 20.7 ± 0.3 32.1 ± 1.1 28.0 ± 0.6 7932 ± 216 September 12 20.5 ± 0.7 29.7 ± 1.6 27.9 ± 2.0 8012 ± 599 Tall fescue September 5 9.3 ± 0.2 54.8 ± 1.2 34.0 ± 1.4 7415 ± 324 September 12 8.6 ± 0.1 57.4 ± 0.7 36.9 ± 0.5 7920 ± 162 a Crude protein. b Neutral detergent fiber. c Acid detergent fiber. 2.4. Chemical analyses Representative samples of foods and pastures were placed in paper bags and dried in a forced-air oven at 60 °C for 48 h. They were subsequently ground through a Wiley mill with a 1-mm screen, and analyzed for dry matter (method 930.15) [26], neutral detergent fiber (NDF), acid detergent fiber (ADF) [27], and nitrogen (N) (method 990.03) [26]. 2.5. Statistical analyses Analyses were computed using a mixed model (MIXED procedure; SAS Inst., Inc., Cary, NC; version 9.1 for Windows) with a significance level of α = 0.10. The variance–covariance structure was selected based on the lowest Bayesian information criterion. The model diagnostics included testing for a normal distribution of the error residuals and homogeneity of variance. Means were analyzed using pairwise differences (DIFF) of least squares means (LSMEANS). We computed a diversity index (Shannon's diversity index; [28]) to estimate and compare the degree of dietary diversity selected by lambs when exposed to novel foods and flavors and to novel pastures. The Shannon's index was estimated as = − ∑ i = 1Spi ∗ ln pi, where S is the number of foods and pi represents the proportional contribution of the i food to the total amount eaten. Fecal egg counts were transformed prior to analysis (log (x + 1)) in order to stabilize the variance. For transformed data, backtransformed means are reported in the results. Food intake (Novel Foods and Flavors), expressed as grams of feed consumed/kg of metabolic body weight (BW0.75), percentage of scans recorded on each variety relative to the total number of grazing events (Novel Pastures), Shannon's diversity index, FEC, and blood parameters analyzed as a mixed model with repeated measures (day and time—novel foods and flavors) and lambs (random factor—novel foods and flavors) or pair of lambs (random factor—novel pastures) nested within group (C, M, and H). Group was the between animal factor and day and time (novel foods and flavors) were the repeated measures in the analyses (fixed factors). 3. Results 3.1. Novel foods and flavors 3.1.1. Condensed tannins Lambs in M (July 27 and 28) and H (July 27) consumed more beet pulp than Control lambs, whereas the pattern reversed by the last day of testing (C > H; Group × Day; P = 0.03; Fig. 1). Lambs in M ingested more beet pulp than lambs in H during July 31 (P < 0.10). No differences were detected among groups of lambs in the ingestion of beet pulp during the 20-min feeding periods (Group × Time; P = 0.65; Group × Time × Day; P = 0.91). Download full-size image Fig. 1. Intake of novel foods and flavors by 3 groups of lambs. Lambs were previously dosed orally with infective third-stage (L3) larvae of Haemonchus contortus in the following amounts: 0 (Control group: C), 5000 (Medium group: M) and 15 000 (High group: H) L3/lamb. During testing all lambs received during 2 h/day a simultaneous offer of: Beet Pulp mixed with quebracho tannins in concentrations 0%, 5% and 10% (condensed tannins); tall fescue hay mixed with umami flavor in concentrations of 0%, 0.5% and 1% (umami), and wheat bran with bitter flavor in concentrations 0%, 0.5% and 1% in (Bitter). Values are means for 10 lambs; SE values are represented by vertical bars. Lambs in C consumed more beet pulp + tannin 5% than lambs in H and M (July 30), but the pattern reversed towards the end of testing (August 3) for lambs in H (i.e., H > C) which also showed greater intakes than lambs in M (August 1 and 3; Group × Day; P = 0.08; Fig. 1). No differences were detected among groups of lambs in the ingestion of beet pulp + tannin 5% during the 20-min feeding periods (Group × Time; P = 0.65; Group × Time × Day; P = 0.33). There were no differences detected between groups of lambs regarding consumption of beet pulp + tannin 10% (Group; P = 0.92; Group × Day; P = 0.29; Group × Time; P = 0.21). However, towards the last days of testing (August 1 to 3) and for the first 20 min of the feeding period, intake was greater for H than for C (0.49, 0.38, and 0.43 versus 0.09, 0.09, and 0.14 g/Kg0.75 respectively; SEM = 0.08; Group × Day × Time; P = 0.0003). Dietary diversity scores (Shannon's diversity index) did not differ among treatment groups (Group; P = 0.63; Group × Day; P = 0.62). There were no differences detected between groups of lambs for ingestion of alfalfa pellets (Group; P = 0.87; Group × Date; P = 0.36). Lambs in C, M, and H consumed, respectively, 50, 47, and 49 g/Kg0.75; SEM = 3.9. Likewise, there were no differences detected between groups of lambs for the total ingestion of food (alfalfa pellets + beet pulp + beet pulp + tannins; Group; P = 0.96; Group × Date; P = 0.42). Lambs in C, M, and H consumed, respectively, 83, 83, and 81 g/Kg0.75; SEM = 4.1. 3.1.2. Umami No differences were detected for intake of tall fescue (Group; P = 0.41; Group × Day; P = 0.34; Group × Time; P = 0.81; Group × Time × Day; P = 0.20) or tall fescue + umami 1% (Group; P = 0.58; Group × Day; P = 0.96; Group × Time; P = 0.53; Group × Time × Day; P = 0.51) among groups of lambs (Fig. 1). Averaged across days, lambs in H and M ingested more tall fescue + umami 0.5% during the last 3 days of testing (P = 0.03; Fig. 1). Lambs in H (2.4) selected a more diverse array of novel foods and flavors than lambs in M (1.9; August 7) and C (1.7; August 8), whereas towards the end of testing lambs in C (2.4—August 9 and 1.9—August 11) selected a more diverse arrays of foods than lambs in M (1.7 and 1.3, respectively; SEM = 0.24; Group × Day P = 0.08; Fig. 2). Download full-size image Fig. 2. Dietary diversity scores by 3 groups of lambs exposed to a choice of novel flavored foods and pastures. Tall fescue hay mixed with umami flavor in concentrations of 0%, 0.5% and 1% (Umami). Wheat bran mixed with bitter flavor in concentrations of 0%, 0.5% and 1% (bitter). Sainfoin (Onobrychis viciifolia), alfalfa (Medicago sativa) and endophyte-infected tall fescue (Festuca arundinacea) (pastures). Lambs were previously dosed orally with infective third-stage (L3) larvae of Haemonchus contortus in the following amounts: 0 (Control group: C), 5000 (Medium group: M) and 15 000 (High group: H) L3/lamb. Values are means for 10 lambs; SE values are represented by vertical bars. There were no differences detected between groups of lambs for ingestion of alfalfa pellets (Group; P = 0.22; Group × Date; P = 0.93). Lambs in C, M, and H consumed, respectively, 37.6, 32.7, and 34.1 g/Kg0.75; SEM = 2.0. Lambs in C tended to ingest more total food (alfalfa pellets + tall fescue + tall fescue + umami) than parasitized lambs (Group; P = 0.13; Group × Date; P = 0.56). Lambs in C, M, and H consumed, respectively, 50, 43, and 48 g/Kg0.75; SEM = 1.8. 3.1.3. Bitter No differences were detected for intake of wheat bran (Group; P = 0.26; Group × Day; P = 0.97; Group × Time; P = 0.90; Group × Time × Day; P = 0.64) or wheat bran + bitter 0.5% (Group; P = 0.39; Group × Day; P = 0.22; Group × Time; P = 0.85; Group × Time × Day; P = 0.43) among groups of lambs (Fig. 1). There were no differences detected between groups of lambs regarding consumption of wheat bran + bitter 1% (Group; P = 0.92; Group × Day; P = 0.58; Group × Time; P = 0.16). However, towards the end of testing (August 17–20), and for 40 and 120 min of the feeding period, intake was greater for C (values ranging from 0.3 to 1.7 g/Kg0.75) than for H (values ranging from 0.3 to 0.9 g/Kg0.75) and M (values ranging from 0.02 to 0.8 g/Kg0.75) (SEM = 0.26; Group × Day × Time; P = 0.08). Lambs in H (1.0) selected a more diverse array of novel foods and flavors than lambs in M (0.8) and C (0.8) (SEM = 0.06; Group; P = 0.05; Group × Day P = 0.88; Fig. 2). There were no differences detected between groups of lambs for ingestion of alfalfa pellets (Group; P = 0.86; Group × Date; P = 0.85). Lambs in C, M, and H consumed, respectively, 62.8, 59.3, and 59.4 g/Kg0.75; SEM = 5.1. Likewise, there were no differences detected between groups of lambs for the total ingestion of food (alfalfa pellets + wheat bran + wheat bran + bitter; Group; Group × Date; P = 0.86). Lambs in C, M, and H consumed, respectively, 98, 94, and 96 g/Kg0.75; SEM = 4.6. 3.1.4. Flavors No differences were detected for intake of apple- (Group; P = 0.73; Group × Day; P = 0.54; Group × Time; P = 0.99; Group × Time × Day; P = 0.96), coconut- (Group; P = 0.71; Group × Day; P = 0.99; Group × Time; P = 0.98; Group × Time × Day; P = 0.72), and maple-flavored grape pomace (Group; P = 0.21; Group × Day; P = 0.65; Group × Time; P = 0.21; Group × Time × Day; P = 0.35) among groups of lambs (data not shown). Dietary diversity scores did not differ among treatment groups (Group; P = 0.50; Group × Day; P = 0.64; data not shown). There were no differences detected between groups of lambs for ingestion of alfalfa pellets (Group; P = 0.80; Group × Date; P = 0.88). Lambs in C, M, and H consumed, respectively, 91.8, 88.8, and 89.8 g/Kg0.75; SEM = 3.3. Likewise, there were no differences detected between groups of lambs for the total ingestion of food (alfalfa pellets + beet pulp + beet pulp + tannins; Group; Group × Date; P = 0.86). Lambs in C, M, and H consumed, respectively, 98, 94, and 96 g/Kg0.75; SEM = 4.6. In addition, there were no differences detected between groups of lambs for the total ingestion of food (alfalfa pellets + grape pomace + grape pomace + flavors; Group; P = 0.90; Group × Date; P = 0.78). Lambs in C, M, and H consumed, respectively, 95, 93, and 93 g/Kg0.75; SEM = 3.2. 3.2. Novel pastures Lambs grazed almost all the time during the observation periods (97.9, 98.0, and 97.3% of scans; SEM = 0.5). No differences in the total number of grazing events were observed among treatment groups throughout the study (Group; P = 0.61; Group × Day; P = 0.50). Averaged across days, no differences among groups were detected in the percentage of scans recorded on tall fescue (Group; P = 0.28). No differences among groups were detected in the percentage of scans recorded on alfalfa (Group and Group × Day; P = 0.42) or sainfoin (Group; P = 0.59; Group × Day; P = 0.60; Fig. 3). Download full-size image Fig. 3. Daily grazing events by 3 groups of lambs when they had a choice of novel pastures: alfalfa (Medicago sativa), sainfoin (Onobrychis viciifolia), and endophyte-infected tall fescue (Festuca arundinacea) recorded during scan sampling (represented as a percentage of total grazing scans recorded). Lambs were previously dosed orally with infective third-stage (L3) larvae of Haemonchus contortus in the following amounts: 0 (Control group: C), 5000 (Medium group: M) and 15 000 (High group: H) L3/lamb. Values are means for 5 pairs of lambs; SE values are represented by vertical bars. Averaged across days, dietary diversity scores did not differ among treatment groups (Group; P = 0.34). However, from September 7 to 9, lambs in H (0.65) selected a more diverse array of pastures than lambs in M (0.46) and C (0.40) (P = 0.08; Fig. 2). 3.3. Nutritional composition of foods and forages During testing in pens, wheat bran and alfalfa pellets had greater concentrations of crude protein, whereas grape pomace and tall fescue hay showed greater concentrations of fiber (Table 2). Regarding novel pastures, concentrations of crude protein were greater for legumes and concentration of fiber greater for grasses. Concentrations of crude protein and fiber remained fairly stable from the beginning (September 5) to the end of testing (September 12) (Table 2). 3.4. Fecal egg counts As expected, FEC were nil before the infection and after infection, differences were detected among treatment groups (Group; P < 0.001; Table 1). These differences were consistent with the amounts of L3 larvae of H. contortus inoculated in the different groups of lambs. Fecal egg counts in Control lambs were nil throughout the study. In addition, differences in FEC between lambs in H and M were maintained throughout the study, and FEC in these groups increased steadily until they reached ~ 8000 and 4000 epg in the H and M groups, respectively (Group × Sample date interaction, P < 0.001; Table 1). 3.5. Blood parameters Several differences between groups were detected for the blood parameters assessed after infecting lambs with different doses of H. contortus (P < 0.10; Table 3). Table 3. Blood parameters in 3 groups of lambs (n = 10) exposed to novel foods and flavors (July 26 to August 30, 2013) and novel pastures (August 31 to September 12, 2013). Lambs of each group were dosed orally with infective third-stage (L3) larvae of Haemonchus contortus in the following amounts: 0 (Control group: C), 5000 (Medium group: M) and 15 000 (High group: H) L3/lamb. Sampling date June 29 July 29 August 21 SEM P Parameter C M H C M H C M H Plasma protein (g/dl) 5.7 6.0 5.9 6.4a 6.3a 5.9b 6.3a 5.9b 5.6b 0.1 0.009 Fibrinogen (mg/dl) 220 210 160 210 230 200 180 230 260 29.1 0.190 Hemoglobin level (g/dl) 12.9 13.2 13.5 12.6a 11.8a 10.8b 11.9a 11.0b 10.2b 0.4 0.001 Hematocrit (%) 36.1 36.9 36.6 36.1a 33.7b 30.7c 34.3a 31.8b 29.1c 1.1 0.007 Red blood cells (106/μl) 12.6 13.0 13.3 12.5a 11.5a 9.9b 11.7a 10.4b 9.0c 0.7 < .0001 Mean corpuscular volume (fl) 28.8 28.5 27.6 29a 29.6a 30.9b 29.1a 30.5a 32.3b 0.7 < .0001 Mean corpuscular hemoglobin concentration (g/dl) 35.9 35.8 37 35.2 34.9 35.3 34.9 34.8 35.1 0.4 0.070 Red cell distribution width 22.2 21.8 22.2 20.2a 21.1a 22.7b 19.7 20.1 19.1 0.6 0.006 White blood cells (× 1000/μl) 13.9 11.9 12.2 11.4a 10.5bc 8.9b 9.3 8.4 8.2 0.8 0.006 Neutrophils (%) 30.1 27.3 26.1 31.2 33.3 28.9 26.1 27.8 23.9 3.1 0.890 Eosinophils (%) 0.5 0.8 1.3 1.6 1.9 2.5 2.3 2.8 3.3 0.5 0.990 Lymphocytes (%) 66.6 69.5 69.6 65.4 64.1 67.9 70.3 67.4 70.9 3.3 0.890 Monocytes (%) 1.4 1.4 1.6 1.8a 0.8b 0.9b 2.1a 3.1b 1.8c 0.3 0.010 Mean platelet volume (fl) 5.7 5.6 5.5 5.9 5.8 6.3 6.5 6.7 6.9 0.3 0.520 Platelets (103/μl) 581 614 606.1 540a 609a 795b 501a 656b 809c 56.6 0.050 Means in a row with different superscripts differ (group effect; P < 0.10). 4. Discussion 4.1. Parasitism and food selection Gastrointestinal helminths challenge ruminants in ways that reduce their fitness [29,30]. In turn, ruminants have evolved physiological and behavioral adaptations that counteract this challenge. Regarding behavioral adaptations, grazing ruminants minimize the chances of infection by avoiding areas where parasite larvae are most concentrated [31,32]. Another behavioral mechanism for minimizing parasite infection is to self-select foods containing compounds that help treat or control parasite infections [33,34]. In support of this, parasitized lambs ate more of a supplement containing an anthelmintic PSM – condensed tannins – than non-parasitized animals, even when the supplement was of very low nutritional value [35]. Lambs with natural gastrointestinal parasitic burdens [14] or artificially infected [36] showed a greater preference for a tannin-containing ration than non-parasitized lambs. Goats from a breed which typically exhibits low propensity to consume a tannin-containing shrub (Pistacia lentiscus) increase preference for this shrub when challenged by a parasitic infection [37]. The fact that sick animals may seek out and ingest PSMs reflects the influence of a multidimensional array of environmental and physiological variables which have not yet been explored. There is currently a lack of knowledge concerning the mechanisms leading to the emergence of self-medication in sick animals. Which are the mechanisms that trigger self-selection of novel potentially medicinal – yet potentially toxic – foods? We reasoned that when physiological conditions are inadequate and homeostasis is challenged, such as during a parasitic infection, animals should respond by increasing their acceptance of novel foods and flavors. Deviations from homeostasis lead to risk proneness in consumers [21] which can even lead to ingestion of toxic preys [38]. This behavior can be explained from an adaptive point of view, since individuals that have acquired abundant reserves of energy have more potential fitness to lose from taking risks than individuals which are in a negative energy budget [22,39]. Mammalian herbivores often show a high degree of neophobia when first encountering unfamiliar foods [40,41]. Wariness of the unfamiliar confers consumers a substantial survival value, such as reducing the likelihood of ingesting harmful plants [20]. Even a familiar food that is normally readily eaten is sampled carefully when its flavor is changed [42,43]. However, sick individuals may have less potential fitness to lose from selecting novel foods than healthy ones. In addition, reduced neophobia may represent a mechanism which increases the likelihood of encountering medicinal compounds and nutrients that restore health. Parasitized lambs in our study displayed some differences in their ingestion of novel foods, flavors and pastures relative to Control animals which gave support to the aforementioned hypothesis. Nevertheless, lambs' responses appeared more complex than just a general reduction in food neophobia. For instance, ingestive responses appeared to vary with the type of food and flavor on offer. Parasitized lambs consumed more beet pulp than Control lambs initially, but the pattern reversed by the last day of testing. The pattern was the opposite for the ingestion of novel beet pulp + tannins. It is likely that parasitized lambs showed initially a greater degree of acceptance of the novel food with a taste salience presumably “more similar” to the familiar basal diet of alfalfa pellets (i.e., beet pulp). For a consumer, taste salience depends on novelty of the taste not on concentration [44], and the taste of tannins was likely more novel than the taste of beet pulp. Studies in rats suggest that the neural response to taste dimensions that result in the acquisition of a taste aversion is strongly modulated by the novelty of the taste [45]. Mammalian herbivores generalize in a qualitative fashion over cues (sensory stimuli) provided by foods (e.g. [46]) and it is likely that in the present study lambs generalized over familiar cues, despite their limited prior experience with the sensorial and postingestive properties of foods. As beet pulp became more familiar and it did not provide a medicinal effect, it is possible that on subsequent days parasitized lambs displayed an increased acceptance of the “more novel” tannin-containing foods (e.g., during day 9 of testing). The medicinal effects of tannins were not evident in this study as FEC and blood indicators of anemia increased with time. A longer time of exposure to tannins might have been required for animals to achieve a greater ingestion of condensed tannins and thus ingest an effective dose against endoparasites [47]. Several behavioral changes caused by endoparasites are rooted in the induction of anorexia [48]. Anorexia is a common symptom of parasitic infectious diseases, but one that has been qualified as paradoxical because parasites typically impose increased metabolic and nutritional demands on the host [48]. However, only H and M lambs during testing with umami tended to display lower food intake than Control lambs. Since parasitized individuals consuming a nutritious food are anorexic for a shorter period of time than individuals exposed to a food of lower nutritional quality [48], it is likely that the high-quality foods ingested by lambs in the present study (as measured by concentration of crude protein and fiber) contributed to attenuate their anorectic responses. Rather than being a paradoxical response, anorexia may in fact represent a behavioral adaptation [49]. For instance, it has been hypothesized that anorexia allows the host to become more selective in its diet, and thus choose foods that either minimize the risk of infection or contain high concentrations of antiparasitic compounds [50]. In support of this idea, when parasitized herbivores were offered a choice between non-contaminated and feces-contaminated pastures, they avoided the latter pastures [31,51]. Parasitized sheep offered choices among foods with different protein contents, increase their preference for high-protein feeds, likely to meet the increased protein requirements induced by parasitism [52,53]. Protein-deficient lambs also increase preference for umami-flavored feeds, as the umami taste signals protein [54]. Consistent with this, lambs in H and M consumed more tall fescue + umami 0.5% than lambs in C towards the end of testing. Differences in the diversity of novel foods and flavors selected were also observed when lambs were fed the novel umami-flavored food. Parasitized lambs also showed lower concentrations of plasma protein than Control lambs. Thus, it is likely that an enhanced requirement for protein primed lambs to select greater proportions of umami-flavored novel foods. Protein restricted lambs increase preference for umami-flavored foods without prior exposure to the umami taste [54]. Parasitism may influence preference of other taste dimensions that signal palatable foods. Some by-products of infection like lipopolysaccharides seem to mediate the production of proinflammatory cytokines that downregulate sweet taste receptors genes in taste buds of mice [55]. It has been hypothesized that mammals should avoid anything that tastes unpalatable, particularly if it is bitter [56]. This rule is supported by the observations that virtually all naturally occurring poisons taste bitter to humans [19], and most chemicals that taste bitter to humans also elicit an aversive response in other mammals [57]. However, herbivores can substantially benefit by consuming bitter-tasting plant tissues with therapeutical compounds [57]. Repeated sampling of the bitter-tasting anti-malarial agent chloroquine by malaria-infected mice resulted in significant reductions in parasitemia and risk of mortality [58]. In caterpillars, infection by parasites enhances acceptance of the taste of some antiparasitic pyrrolizidine alkaloids, which also has a positive impact on fitness [59]. However, parasitized lambs in the present study did not increase their intake of bitter-flavored feed relative to Control lambs. On the contrary, towards the end of testing and for some feeding intervals, intake of wheat bran + bitter 1% was greater for C than for H and M lambs. If animals generalized bitterness from their prior exposure to condensed tannins, an experience which didn't lead to recovery from parasitism due to the low doses ingested, then subsequent responses to bitter taste could have been attenuated as a consequence of such experience. Alternatively, mammalian herbivores may have not evolved an enhanced acceptance for bitter taste in response to parasitism. Consumers with a high occurrence of bitter in their diet like mammalian herbivores have evolved a high tolerance for bitter [57] since these animals cannot “afford” to reject all foods that taste bitter. In addition, bitter taste thresholds vary independently of toxicity thresholds [57], and they may vary independently of therapeutic thresholds as well since some of the antiparasitic effects of bioactives like tannins rely on their toxic actions [29]. Thus, enhanced preference for a taste dimension typically encountered by mammalian herbivores like bitter may not consistently imply increased medicinal effects. Nevertheless, lambs in H selected a more diverse array of novel foods and flavors than lambs in M and C which suggests that the parasitic infection induced lambs to display a more “generalistic” approach towards wheat bran and wheat bran containing a bitter taste. No differences between treatment groups were detected regarding the ingestion of coconut- apple and maple-flavored grape pomace. The lack of differences in feeding behavior between groups of lambs observed during exposure to these novel food and flavors and the differences in ingestion of other novel foods and flavors observed in previous testing periods suggests that ingestive responses by parasitized lambs involves something more than just the expression of the rule of thumb ‘when sick, reduce food neophobia.’ Such a rule may involve a risk-prone foraging strategy in combination with some intrinsic orosensorial attributes of the specific novel food encountered (e.g., bitter, umami taste, astringency, or generalization based on the presence of familiar orosensorial cues in the novel food). Lambs generalize from familiar to unfamiliar foods based on a common flavor cue [60] and during grazing novel pastures lambs were familiar with alfalfa and tall fescue hay. Sheep generalize aversions from one legume (sainfoin) to another (alfalfa) and to a greater extent than to a grass (tall fescue) [61]. Thus, lambs' prior experiences with grass and legume hay may have enhanced acceptability of all plant species during exposure to novel fresh forages, attenuating potential differences among groups. Nevertheless, from September 7 to 9, lambs in H selected a more diverse array of pastures than lambs in M and C, suggesting that high parasitic loads primed animals to select a more diverse diet during grazing. 4.2. Influence of parasitism on deviations from homeostasis Infected lambs in the present study likely experienced negative internal states caused by H. contortus, as reflected by the high FEC values and deviations of blood parameters from reference values [62] and in relation to Control lambs (Tables 1 and 2). The values obtained for blood parameters and FEC in parasitized lambs were consistent with clinical feature of anemia secondary to Haemonchosis [63]. H. contortus infections induce reductions in hematocrit, hemoglobin and red blood cell counts, size and shape (i.e., red cell distribution width, mean corpuscular volume) attributed to the blood loss caused by the blood sucking activities of the parasite [64]. Red blood cell distribution width was greater for parasitized lambs than unparasitized lambs. This parameter is a calculation of the variation in the size of red blood cells. In some cases of anemia, the amount of variation (anisocytosis) in red blood cell size (along with variation in shape, poikilocytosis) causes an increase in this parameter [65]. Values for mean corpuscular volume were greater in parasitized lambs, suggesting a tendency for a macrocytic anemia. Macrocytic red blood cells are large and tend to have higher mean corpuscular hemoglobin, which may explain the elevated values for this parameter. The compensatory liberation of platelets from megakaryocyte medullary reserves in response to the gastric hemorrhage would explain the higher platelet counts in parasitized animals [66], while the proportional loss of proteins with erythrocytes in hemorrhage could explain the lower concentration of plasma proteins in parasitized lambs [66]. No lethargy or lack of motivation to eat was observed in parasitized animals. Thus, despite clinical signs of infection animals were in an appropriate motivational state to explore and ingest novel foods. It is likely that these were the ideal conditions for parasitized animals to sample novel foods and flavors as their intake capacity was not depressed by parasitic burdens. 5. Conclusion Our findings suggest that negative internal states induced by endoparasites have the potential to modify the selection of novel foods and flavors by lambs. Reduced neophobia may enhance the likelihood of herbivores ingesting therapeutic doses of novel medicinal plants or nutritious forages that enhance their nutrition to better combat disease. This risk-prone strategy may help herbivores “cross the rejection threshold” imposed by novelty which prevents animals from experiencing the medicinal effects of a novel feed, and thereby prevents the emergence of self-medicative behavior. If by reduced neophobia sick animals “cross the rejection threshold” and experience the benefits of consuming beneficial plants, they may continue ingesting these plants through taste-postingestive feedback associations that lead to learned food preferences [43]. Finally, our results show that parasitized lambs did not treat all novel foods as equal; they displayed different and changing degrees of acceptance, which could be explained by generalizations from their limited prior experiences with the salience/taste of familiar feeds, and by the intrinsic characteristics of the novel taste dimensions (i.e., bitter, umami) and nutritional composition of the foods encountered. A better understanding of the emergence of therapeutic self-medication in herbivores will contribute to the development of innovative and sustainable parasite control strategies that enhance animal health and well-being. Acknowledgments This research was supported by grants from the Utah Agricultural Experiment Station (Grant UTA01068). This paper is published with the approval of the Director, Utah Agricultural Experiment Station, and Utah State University, as journal paper number 8667. We acknowledge R. Stott for veterinary services and Romain Cabassu and Travis Lisonbee for technical support. Financial support for A.V.E. from Consejo Nacional de Investigaciones Científicas y Técnicas de la República Argentina (CONICET) is acknowledged. References [1] W.J. Freeland, D.H. Janzen Strategies in herbivory by mammals: the role of plant secondary compounds Am Nat, 108 (1974), pp. 269-289 CrossRef [2] M. Westoby What are the biological bases of varied diets? Am Nat, 112 (1978), pp. 627-631 CrossRefView Record in Scopus [3] P.R. Cheeke Natural toxicants in feeds, forages, and poisonous plants (2nd ed.), Interstate Publisher, Danville, IL (1998) [4] A. Cozier, M.N. Clifford, H. Ashihara Plant secondary metabolites: occurrence, structure and role in the human diet Blackwell Publishing, Oxford, UK (2006) [5] G.A. Lozano Parasitic stress and self-medication in wild animals A.P. Moler, M. Milinski, P J B (Eds.), Slater advances in the study of behavior, Elsevier Science, London, UK (1998), pp. 291-317 ArticleDownload PDFView Record in Scopus [6] J.P. Muir The multi-faceted role of condensed tannins in the goat ecosystem Small Rumin Res, 98 (2011), pp. 115-120 ArticleDownload PDFView Record in Scopus [7] J.J. Villalba, F.D. Provenza, A.K. Clemensen, R. Larsen, J. Juhnke Preference for diverse pastures by sheep in response to intraruminal administrations of tannins, saponins and alkaloids Grass Forage Sci, 66 (2011), pp. 224-236 CrossRefView Record in Scopus [8] F.D. Provenza, J.A. Pfister, C.D. Cheney Mechanisms of learning in diet selection with reference to phytotoxicosis in herbivores J Range Manag, 45 (1992), pp. 36-45 CrossRefView Record in Scopus [9] F.D. Provenza, J.J. Villalba Foraging in domestic herbivores: linking the internal and external milieu V.L. Bels (Ed.), Feeding in domestic vertebrates: from structure to function, CABI Publ, Oxfordshire (2006), pp. 210-240 CrossRefView Record in Scopus [10] J. Janzen Complications in interpreting the chemical defenses of trees against tropical arboreal plant-eating vertebrates G. Montgomery (Ed.), The ecology of arboreal folivores, Smithsonian Institution Press, Washington, DC, USA (1978), pp. 73-84 View Record in Scopus [11] M.A. Huffman, M. Seifu Observations on the illness and consumption of a possibly medicinal plant, Vernonia amygdalina (Del.), by a wild chimpanzee in the Mahale Mountains National Park, Tanzania Primates, 30 (1989), pp. 51-63 CrossRefView Record in Scopus [12] K. Koshimizu, H. Ohigashi, M.A. Huffman Use of Vernonia amygdalina by wild chimpanzee: possible roles of its bitter and related constituents Physiology and Behaviour, 56 (1994), pp. 1209-1216 ArticleDownload PDFView Record in Scopus [13] J.T. Gradé, J.R.S. Tabuti, P. Van Damme Four footed pharmacists: indications of self-medicating livestock in Karamoja, Uganda Econ Bot, 63 (2009), pp. 29-42 CrossRefView Record in Scopus [14] J.J. Villalba, F.D. Provenza, J.O. Hall, L.D. Lisonbee Selection of tannins by sheep in response to gastro-intestinal nematode infection J Anim Sci, 88 (2010), pp. 2189-2198 CrossRefView Record in Scopus [15] M.S. Singer, K.C. Mace, E.A. Bernays Self-medication as adaptive plasticity: increased ingestion of plant toxins by parasitized caterpillars PLoS ONE, 4 (3) (2009), p. e4796, 10.1371/journal.pone.0004796 CrossRef [16] C. Catoni, A. Peters, H.M. Schaefer Life history trade-offs are influenced by the diversity, availability and interactions of dietary antioxidants Anim Behav, 76 (4) (2008), pp. 1107-1119 ArticleDownload PDFView Record in Scopus [17] J.J. Villalba, F.D. Provenza, R. Shaw Sheep self-medicate when challenged with illness-inducing foods Anim Behav, 71 (2006), pp. 1131-1139 ArticleDownload PDFView Record in Scopus [18] E. Rodriguez, R. Wrangham Zoopharmacognosy: the use of medicinal plants by animals K.R. Downum, J.T. Romeo, H.A. Stafford (Eds.), Recent advances in phytochemistry, Phystochemical Potential of Tropical Plants, vol. 27, Plenum Press, New York, NY, USA (1993), pp. 89-105 CrossRefView Record in Scopus [19] J.I. Glendinning How do predators cope with chemically defended foods? Biol Bull, 213 (2007), pp. 252-266 CrossRefView Record in Scopus [20] F.D. Provenza, J.J. Villalba, C.D. Cheney, S.J. Werner Self-organization of foraging behavior: from simplicity to complexity without goals Nutr Res Rev, 11 (1998), pp. 199-222 CrossRefView Record in Scopus [21] T. Caraco, W.U. Blanckenhorn, G.M. Gregory, J.A. Newman, G.M. Recer, S.M. Zwicker Risk-sensitivity: ambient temperature affects foraging choice Anim Behav, 39 (1990), pp. 338-414 View Record in Scopus [22] J.S. Brown Patch use as an indicator of habitat preference, predation risk, and competition Behav Ecol Sociobiol, 22 (1988), pp. 37-47 CrossRefView Record in Scopus [23] A.M. Sibbald, R.J. Hooper Sociability and the willingness of sheep to move away from their companions in order to graze Appl Anim Behav Sci, 86 (2004), pp. 51-62 ArticleDownload PDFView Record in Scopus [24] J. Altman Observational study of behavior: sampling methods Behavior, 49 (1974), pp. 227-265 [25] J.H. Scrivner, D.M. Center, M.B. Jones A rising plate meter for estimating production and utilization J Range Manag, 39 (1986), pp. 475-477 CrossRefView Record in Scopus [26] AOAC Official methods of analysis (17th ed.), Association of Official Analytical Chemists, Arlington, VA (2000) [27] P.J. Van Soest, J.B. Robertson, B. Lewis Methods for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition J Dairy Sci, 74 (1991), pp. 3583-3597 ArticleDownload PDFCrossRefView Record in Scopus [28] K. Hutcheson A test for comparing diversities based on the Shannon formula J Theor Biol, 29 (1970), pp. 151-154 ArticleDownload PDFView Record in Scopus [29] F. Jackson, J. Miller Alternative approaches to control quo vadit? Vet Parasitol, 139 (2006), pp. 371-384 ArticleDownload PDFView Record in Scopus [30] F. Jackson, M. Varady, D.J. Bartley Managing anthelmintic resistance in goats—can we learn lessons from sheep? Small Rumin Res, 103 (2012), pp. 3-9 ArticleDownload PDFView Record in Scopus [31] M.R. Hutchings, I.J. Gordon, I. Kyriazakis, E. Robertson, F. Jackson Grazing in heterogeneous environments: infra- and supra- parasite distributions determine herbivore grazing decisions Oecologia, 132 (2002), pp. 453-460 CrossRefView Record in Scopus [32] M.R. Hutchings, S. Athanasiadou, I. Kyriazakis, I. Gordon Can animals use foraging behaviour to combat parasites? Proc Nutr Soc, 62 (2003), pp. 361-370 CrossRefView Record in Scopus [33] M.A. Huffman Current evidence for self-medication in primates: a multidisciplinary perspective Yearb Phys Anthropol, 40 (1997), pp. 171-200 CrossRefView Record in Scopus [34] M.A. Huffman Self-medicative behavior in the African great apes: an evolutionary perspective into the origins of human traditional medicine Journal of BioScience, 51 (2001), pp. 651-661 CrossRefView Record in Scopus [35] L.D. Lisonbee, J.J. Villalba, F.D. Provenza, J.O. Hall Tannins and self-medication: implications for sustainable parasite control in herbivores Behav Process, 82 (2009), pp. 184-189 ArticleDownload PDFView Record in Scopus [36] J. Juhnke, J. Miller, J.O. Hall, F.D. Provenza, J.J. Villalba Preference for condensed tannins by sheep in response to challenge infection with Haemonchus contortus Vet Parasitol, 188 (2012), pp. 104-114 ArticleDownload PDFView Record in Scopus [37] S.Y. Landau, H. Azaizeh, H. Muklada, T. Glasser, E.D. Ungar, H. Baram, et al. Anthelmintic activity of Pistacia lentiscus foliage in two Middle Eastern breeds of goats differing in their propensity to consume tannin-rich browse Vet Parasitol, 173 (2010), pp. 280-286 ArticleDownload PDFView Record in Scopus [38] M. Chatelain, C.G. Halpin, C. Rowe Ambient temperature influences birds' decisions to eat toxic prey Anim Behav, 86 (4) (2013), pp. 733-740 ArticleDownload PDFView Record in Scopus [39] C.A. Barnett, M. Bateson, C. Rowe Educated predators make strategic decisions to eat defended prey according to their toxin content Behav Ecol, 23 (2012), pp. 418-424 CrossRefView Record in Scopus [40] F.D. Provenza, J.J. Lynch, C.D. Cheney Effects of a flavor and food restriction on the intake of novel foods by sheep Appl Anim Behav Sci, 43 (1995), pp. 83-93 ArticleDownload PDFView Record in Scopus [41] F.D. Provenza Acquired aversions as the basis for varied diets of ruminants foraging on rangelands J Anim Sci, 74 (1996), pp. 2010-2020 CrossRefView Record in Scopus [42] F.D. Provenza, J.J. Lynch, J.V. Nolan The relative importance of mother and toxicosis in the selection of foods by lambs J Chem Ecol, 19 (1993), pp. 313-323 CrossRefView Record in Scopus [43] F.D. Provenza Postingestive feedback as an elementary determinant of food preference and intake in ruminants J Range Manag, 48 (1) (1995), pp. 2-17 CrossRefView Record in Scopus [44] J.W. Kalat Taste salience depends on novelty, not concentration, in taste-aversion learning in the rat Journal of Comparative and Physiological Psychology, 86 (1974), pp. 47-50 CrossRefView Record in Scopus [45] I.L. Bernstein, M.T. Koh Molecular signaling during taste aversion learning Chem Senses, 32 (2007), pp. 99-103 CrossRefView Record in Scopus [46] K.L. Launchbaugh, F.D. Provenza The effects of flavor concentration and toxin dose on the formation and generalization of flavor aversions in lambs J Anim Sci, 72 (1994), pp. 10-13 CrossRefView Record in Scopus [47] S. Athanasiadou, G. Arsenos, I. Kyriazakis Animal health and welfare issues arising in organic ruminant production systems I. Kyriazakis, G. Zervas (Eds.), Organic meat and milk from ruminants, Academic Publishers, Wageningen (2002), pp. 39-56 View Record in Scopus [48] I. Kyriazakis Is anorexia during infection in animals affected by food composition? Anim Feed Sci Technol, 156 (2010), pp. 1-9 ArticleDownload PDFView Record in Scopus [49] W.J. Bell Searching behaviour. The behavioral ecology of finding resources Cambridge University Press, Cambridge (1991) [50] I. Kyriazakis, B.J. Tolkamp, M.R. Hutchings Towards a functional explanation for the occurrence of anorexia during parasitic infections Anim Behav, 56 (1998), pp. 265-274 ArticleDownload PDFView Record in Scopus [51] J. Cooper, I.J. Gordon, A.W. Pike Strategies for the avoidance of faeces by grazing sheep Appl Anim Behav Sci, 69 (2000), pp. 15-33 ArticleDownload PDFView Record in Scopus [52] I. Kyriazakis, J.D. Oldham, R.L. Coop, F. Jackson The effect of subclinical intestinal nematode infection on the diet selection of growing sheep Br J Nutr, 72 (1994), pp. 665-677 CrossRefView Record in Scopus [53] I. Kyriazakis, D.H. Anderson, J.D. Oldham, R.L. Coop, F. Jackson Long-term subclinical infection with Trichostrongylus colubriformis: effects on food intake, diet selection and performance of growing lambs Vet Parasitol, 61 (1996), pp. 297-313 ArticleDownload PDFView Record in Scopus [54] A. Bach, J.J. Villalba, I.R. Ipharraguerre Interactions between mild nutrient imbalance and taste preferences in young ruminants J Anim Sci, 90 (2012), pp. 1015-1025 CrossRefView Record in Scopus [55] X. Zhu, L. He, L.P. McCluskey Ingestion of bacterial lipopolysaccharide inhibits peripheral taste responses to sucrose in mice Neuroscience, 258 (2014), pp. 47-61 ArticleDownload PDFView Record in Scopus [56] J. Garcia, W.G. Hankins The evolution of bitter and the acquisition of toxiphobia D.A. Denton, J.P. Coghlan (Eds.), Olfaction and taste, V Proceedings of the 5th International Symposium in Melbourne, Australia, Academic Press, New York, USA (1975), pp. 39-45 ArticleDownload PDFView Record in Scopus [57] J.I. Glendinning Is the bitter rejection response always adaptive? Physiol Behav, 56 (1994), pp. 1217-1227 ArticleDownload PDFView Record in Scopus [58] S.K. Vitazkova, E. Long, A. Paul, J.I. Glendinning Mice suppress malaria infection by sampling a “bitter” chemotherapy agent Anim Behav, 61 (2001), pp. 887-894 ArticleDownload PDFView Record in Scopus [59] E.A. Bernays, M.S. Singer Taste alteration and endoparasites Nature, 436 (2005), p. 476 CrossRefView Record in Scopus [60] J.J. Villalba, F.D. Provenza Postingestive feedback from starch influences the ingestive behaviour of sheep consuming wheat straw Appl Anim Behav Sci, 66 (2000), pp. 49-63 ArticleDownload PDFView Record in Scopus [61] C. Ginane, B. Dumont Do sheep (Ovis aries) categorize plant species according to botanical family? Anim Cogn, 14 (2011), pp. 369-376 CrossRefView Record in Scopus [62] D.E. Ullrey, E.R. Miller, C.H. Long, B.H. Vincent Sheep hematology from birth to maturity I. erythrocyte population, size and hemoglobin concentration J Anim Sci, 24 (1965), pp. 135-140 CrossRefView Record in Scopus [63] A. Rowe, K. McMaster, D. Emery, N. Sangster Haemonchus contortus infection in sheep: parasite fecundity correlates with worm size and host lymphocyte counts Vet Parasitol, 153 (3–4) (2008), pp. 285-293 ArticleDownload PDFView Record in Scopus [64] R.A. Mir, M.Z. Chishti, M.A. Zarger, H. Tak, S.A. Ganie Clinicopathogical changes in sheep experimentally infected with Haemonchus contortus World Journal Agriculture Science, 3 (2007), pp. 562-566 View Record in Scopus [65] H.K. Walker, W.D. Hall, J.W. Hurst Clinical methods: the history Physical and laboratory examinations (3rd ed), Butterworths, Boston, USA (1990) [66] M.G. Weiser Response and disorders of the erythrocyte S.J. Ettinger, E.C. Feldman (Eds.), Textbook of veterinary internal medicine (4th ed.), Editorial Inter-Médica, Buenos Aires, Argentina (1997), pp. 2248-2388 View Record in Scopus