Habitat-induced changes in forage quality and implications for fitness in Plateau Pika (Ochotona curzoniae)

Abstract

Plateau pikas (Ochotona curzoniae) play a keystone role in the alpine meadow ecosystem on the Qinghai–Tibetan Plateau (QTP). For decades, QTP grasslands have become degraded to some degree, largely as a result of heavy livestock grazing. Concomitantly, the abundance of plateau pikas has increased dramatically as grassland degradation has altered the vegetation community structure and dominant species, shifting the plant nutrient contents toward higher protein conditions that favor them. Considerable research supports the hypothesis that the quantity and quality of food limit herbivore populations. Here, we examined the relationship between the availability of essential amino acids in the diets of plateau pikas and the degree of meadow degradation associated with livestock grazing intensity through a field survey, as well as the fitness of individuals by laboratory feeding experiments with 2 pelleted chows containing 0.26% and 0.45% methionine. Sulfur-containing methionine and cystine were the most limiting amino acids in the diets of pikas. During the pika breeding season, the concentrations of most essential acids, particularly methionine and cystine, were higher in heavily degraded meadows than in lightly and moderately degraded meadows. Individuals fed 0.45% methionine exhibited enhanced cell-mediated immunity, reduced intensity of coccidian infection, and increased concentrations of gonadotropin-releasing hormone, luteinizing hormone, estradiol, progesterone, and testosterone compared to those fed 0.26% methionine. These results showed that heavily degraded meadows provided relatively high-quality food that improved individual fitness, suggesting that the high-methionine food in the heavily degraded meadows may be a key factor in the generation and maintenance of high-density populations of plateau pikas.

摘要

高原鼠兔（Ochotona curzoniae）是青藏高原高寒草甸生态系统的关键物种。然而，近几十年来，青藏高原草地退化程度不断加剧，过度放牧是主要原因。在草地退化过程中，植被群落结构和优势物种发生变化，植物养分含量朝着高蛋白质方向转移，从而导致高原鼠兔数量急剧增加。大量研究表明，食物的数量和质量是限制食草动物数量的重要因素。因此，本研究通过野外调查研究了高原鼠兔食物中必需氨基酸的可用性与牲畜放牧强度相关的草地退化程度之间的关系。此外，我们在实验室使用 0.45%和 0.26%蛋氨酸含量的兔颗粒饲料分组饲喂高原鼠兔，以评估蛋氨酸对个体适合度的影响。结果表明，含硫氨基酸（蛋氨酸 + 胱氨酸）是高原鼠兔食物中最主要的限制性氨基酸。高原鼠兔的繁殖季节，在重度退化草地上，大部分必需氨基酸的浓度，尤其是蛋氨酸和胱氨酸的浓度均高于轻度和中度退化的草地。与饲喂 0.26%蛋氨酸的个体相比，饲喂 0.45%蛋氨酸的个体的细胞介导免疫能力增强，球虫感染强度降低，促性腺激素释放激素、黄体生成素、雌二醇、孕酮和睾酮的浓度增加。这些结果表明，重度退化草地提供了相对优质的食物，提高了个体适合度，揭示重度退化草地中的高蛋氨酸食物可能是产生和维持高原鼠兔高密度种群的关键因素。

Grassland degradation is a global concern, not only affecting pastoralists who rely on healthy grasslands for their survival but also reducing biodiversity and disturbing the ecological environment. Causes of grassland degradation are generally attributed to a combination of overstocking of livestock, unscientific livestock management, global climate change, and excessive herbivory and soil disturbance from small burrowing mammals (Harris 2010). Grassland degradation induces retrograde succession of the plant community, altering structure and function within the community (Wang et al. 2020; Yu et al. 2022). This alteration has greatly affected the geographic distribution and abundance of many small burrowing mammals (Flowerdew and Ellwood 2001; Komonen et al. 2003; Foster et al. 2014). For example, with grassland degradation in recent decades, the population densities of European rabbits (Oryctolagus cuniculus), plateau pikas (Ochotona cruzoniae), and prairie dogs (Cynomys spp.) have considerably increased (Delibes-Mateos et al. 2011). These small burrowing mammals play important roles in shaping grassland ecosystems. However, due to their high density in degraded grasslands, they have been subjected to eradication campaigns because of their putative negative impact on natural habitats and agriculture and their competition with livestock for forage (Miller et al. 2007; Delibes-Mateos et al. 2011). However, poisoning programs affect the food chain and biodiversity. Therefore, understanding how grassland degradation affects the population dynamics of small burrowing mammals is important for maintaining the balance between preserving the important functional role of keystone burrowing mammals and managing livestock production.

A conventional view of the mechanism by which grassland degradation induces an increase in the population density of small burrowing mammals is that the degraded grasslands provide open habitats with a clear view, which is vital for the survival of small burrowing mammals in grasslands (Shi 1983; Bian and Fan 1997; Badingqiuying et al. 2018). However, habitats with low food quality do not appear to be sufficient for maintaining high-density populations of small mammals (Cole and Batzli 1979; Schetter et al. 1998). Many theories have been advanced to explain the population dynamics of small herbivores; among those, food quality and quantity can be centrally important (Cole and Batzli 1979; Batzli and Lesieutre 1991). Changes in vegetation community structure and dominant species in the process of grassland degradation directly affect animal food resources (Tang et al. 2015). For example, the primary vegetation of alpine meadows on the Qinghai–Tibetan Plateau (QTP) is dominated by grasses and sedges; however, as grassland degradation advances, the proportion of forbs significantly increases, which are rich in proteins (Li et al. 2019; Wu et al. 2023). Wu et al. (2023) found that the increased high density of plateau pikas in heavily degraded meadows (HD) is due to a shift in plant nutrient content toward high protein availability. However, the effect of protein quality on pika fitness, and therefore populations, has not yet been identified.

Protein in the diet supplies essential amino acids that cannot be synthesized in adequate amounts by an organism. The ability of a dietary protein to supply a proper balance of these essential amino acids determines the quality of that protein. Numerous studies have shown that levels of essential amino acids in the diet are seasonally deficient across habitats relative to animal requirements in birds (Sedinger 1984), rodents (Lochmiller et al. 1995; Schetter et al. 1998), and primates (Oftedal 1991). Deficiencies in 1 or more essential amino acids can lead to decreases in immunocompetence, reproductive performance, and survival (Lebas 1988; Wegmann et al. 2015; Langlois and McWilliams 2021), suggesting that essential amino acids may have a prominent role in population limitation (Lochmiller et al. 1995; Schetter et al. 1998; Webb et al. 2005). Therefore, evaluation of protein quality in native food sources can help to better understand the relationship between food availability and population dynamics.

The Plateau Pika (Ochotona curzoniae), a small burrowing lagomorph, endemic to the QTP, is a social animal that lives in family groups (Fan et al. 1999; Smith et al. 2019). They are a keystone species in the alpine meadow ecosystem through foraging and digging activities that increase plant species richness and the nitrogen cycle, serving as prey for a large number of predators, and constructing burrows that are used as shelter and nest sites by other species (Smith and Foggin 1999; Delibes-Mateos et al. 2011). Plateau pikas have coexisted with livestock for about 8,000 years, and their population density increases as grazing intensity and grassland degradation increase (Harris 2010; Wu and Wang 2017). Plateau Pika populations on the QTP fluctuate seasonally, with the lowest levels in early spring and peaks in late July of >350/ha at the end of the breeding season (Wang et al. 1997); multiyear cycles are weak or absent. On heavily degraded grasslands, extreme population density can reach 908/ha (Lai et al. 2006).

Here, we: 1) examined the relationship between availability of essential amino acids in the diets of plateau pikas and habitat changes associated with livestock grazing intensity through a field survey; and 2) examined fitness of individuals through a laboratory feeding experiment with 2 pelleted chows containing either high- or low-limiting essential amino acids. Specifically, we predicted that plateau pikas in HD would consume higher-quality protein foods that have abundant essential amino acids, thereby enhancing their immunocompetence and reproductive performance, leading to increased individual fitness in both males and females.

Materials and methods

Field survey

Study site

We examined variability in essential amino acid levels in the summer and winter diets of pikas at degraded meadows which we inferred were induced by livestock overgrazing at the Haibei Alpine Meadow Ecosystem Research Station (37°33ʹ to 37°38ʹN, 101°16ʹ to 101°20ʹE, elevation approximately 3,200 m; hereafter Haibei) and the Sanjiangyuan Grassland Ecosystem Research Station (34°20ʹ to 34°30ʹN, 100°8ʹ to 100°32ʹE, elevation approximately 3,800 m; hereafter Sanjiangyuan). The 2 study regions are located approximately 155 km north and 458 km southwest of Xining, the capital city of Qinghai Province in the People’s Republic of China (Fig. 1). Both regions are characterized by short, cool summers and long, cold winters. The warm season is from June to September, and the cold season is from October to May, with a mean annual temperature of −3.9 °C to −1.7 °C and most of the precipitation concentrated from June to September (Zhao et al. 2006; Sun et al. 2021). The vegetation is alpine meadow comprising sedges, grasses, and forbs. The community structure of the vegetation is relatively simple, with the dominant plant species being Kobresia pygmaea, K. humilis, Elymus nutans, Stipa purpurea, Poa annua, and Potentilla anserina. Meadows have been grazed year-round by domestic herbivores (yaks, sheep, and horses), and the grazing intensity is mainly controlled by local pastoral practices. Plateau pikas are the most abundant small burrowing mammals in both regions, and their common predators include Tibetan Fox (Vulpes ferrilata), Altai Weasel (Mustela altaica), Saker Falcon (Falco cherrug), and Upland Buzzard (Buteo hemilasius; Smith and Foggin 1999; Lai and Smith 2003; Harris et al. 2014).

Survey plots

We identified 3 types of degraded meadows in the 2 regions based on the percentage of flora. In Haibei, lightly degraded meadows (LD) comprised 64% grasses, 9% sedges, and 27% forbs. Correspondingly, moderately degraded meadows (MD) exhibited proportions of 28%, 16%, and 56%; while HD displayed proportions of 7%, 13%, and 80%. Similarly, in Sanjiangyuan, LD consisted of 67% grasses, 5% sedges, and 28% forbs. Correspondingly, MD had proportions of 32%, 7%, and 61%, whereas HD showed proportions of 15%, 10 %, and 75% (Supplementary Data SD1). Five plots in Haibei (June 2018) and 6 plots in Sanjiangyuan (July 2019) were established within each of LD, MD, and HD meadows (thus, 15 and 18 plots total). Each plot was 50 m × 50 m and at least 0.3 km apart. The grazing intensity in LD, MD, and HD was 3.89 to 4.89 sheep/ha, 7.70 to 9.84 sheep/ha, and 13.03 to 17.28 sheep/ha on Haibei; however, grazing intensity data were lacking at Sanjiangyuan. The average number of active burrow entrances of plateau pikas in LD, MD, and HD was 66/ha, 312/ha, and 1,639/ha on Haibei; and 76/ha, 476/ha, and 1,568/ha on Sanjiangyuan. The specific methods, details, and procedures involved in the above experiments were as previously described (Wu et al. 2023).

Protein quality in the stomach of pikas

We livetrapped a total of 168 adult pikas (more than 120 g in body mass; 7 pikas per sex per degraded meadow) using string nooses (Zhou et al. 1987; Wu et al. 2023) during both summer (June–August 2021) and winter (November–December 2021) in the 2 regions. We necropsied them to examine variations in protein quality in forage. The carcasses were placed in ice boxes and returned to the laboratory where the stomachs were excised, and the digesta was cleared of hair fibers and mucosa prior to being dried at 50 °C to a constant mass (±0.1 mg; each digesta sample weighed more than 0.3 g after drying) for the protein quality assay (Lochmiller et al. 1995). The stomach contents of pikas provided a good estimation of the seasonal quality of diets (Johnson 1967; Hou et al. 2019). Concentrations of 9 essential amino acids (threonine, valine, isoleucine, leucine, histidine, lysine, arginine, methionine + cystine, and phenylalanine + tyrosine) in the stomach were measured using postcolumn derivatization with ninhydrin, following guidelines of the National Standard of the People’s Republic of China (GB 5009.124-2016, National Food Safety Standards, Determination of Amino Acids in Food).

To extract amino acids and determine forage quality, we hydrolyzed 40 mg of the sample with 6 N HCl at 110 °C under nitrogen and then removed it for filtration after 22 h and transferred to a 50-mL volumetric flask. Then, 1 mL of the hydrolysate was dried completely in vacuum at 40 °C, redissolved, and dried again. The processed sample was finally dissolved in sodium citrate buffer (pH 2.2) and then analyzed by a Sykam S-433D automatic amino acid analyzer (Sykam Co., Ltd, Germany). Concentrations obtained for methionine and cystine were combined, as were those of phenylalanine and tyrosine, because it is known that cystine and tyrosine share (biochemically substitute) their respective amino acids nutritionally (Lebas 1988). Tryptophan was not measured due to destruction by acid hydrolysis (Gehrke et al. 1985). Hydrolyzable amino acids in the samples were quantified by comparison with amino acid standard solutions. During acid hydrolysis, losses of methionine and cystine were 35.5% and 26.7%, respectively; therefore, methionine + cystine concentrations were adjusted upwards 35.5% and 26.7% to correct for recovery efficiency; recovery of other amino acids was >95%.

Limiting amino acids

The quality of protein in the diet depends on how well it satisfies essential amino acid requirements. Since the amino acid requirements for growth have never been determined specifically for plateau pikas or other pika species, we estimated the requirements by using recommended levels for domestic rabbits (National Research Council 1977). Recommendations for levels of protein in Plateau Pika (14.9%, Ye 1996) diets are similar to those of domestic rabbits (15.3% to 16%; National Research Council 1977; De Blas and Mateos 1998). We therefore used data on the growth requirements of domestic rabbits as a guide to examine whether diets of pikas might be limited in particular essential amino acids by estimating the ratio coefficient of amino acids (RCAA = AAS/mean of AAS, AAS = a certain essential amino acid in the test protein (g/100 g)/a certain essential amino acid requirement pattern (g/100 g) × 100; Li and Li 2021). RCAA < 1 means that the amino acid is below the growth requirement of the Plateau Pika, and the amino acid with the smallest RCAA is the most limiting amino acid.

Feeding experiment

Experimental design

To examine the effect of the most limiting amino acids on body mass growth, sexual development, and immunity development, we captured 120 juvenile plateau pikas (approximately 20 to 30 days of age, 30 to 60 g, weaned) in Haibei in May 2022. We brought them to the laboratory, where they were housed individually in polypropylene cages (45 cm × 32 cm × 20 cm) under natural conditions, with standard pelleted chow and water ad libitum. The purpose of our experiment was to examine the impact of changes in concentrations of the most limiting amino acid on individual fitness. After a 3-day acclimation period, we randomly divided them into 2 groups (30 males and 30 females per group) and fed them with high- and low-methionine pelleted chows for 53 days to ensure development at puberty (Nie et al. 2022). The initial body mass of the pikas did not differ between the 2 groups (t₁₁₈ = −0.35, P = 0.729). Trapping and handling of animals was done in accordance with guidelines of the American Society of Mammalogists (Sikes et al. 2016), and was approved by the Animal Ethics and Welfare Committee of the Northwest Institute of Plateau Biology, Chinese Academy of Science.

Diet composition

Based on the above field survey, we found that sulfur-containing methionine + cystine was the most limiting amino acid in the diets of pikas. In summer, the dietary methionine + cystine contents from HD plots increased by 23% and 37% compared to those from LD and MD plots, respectively; however, there was no significant difference between LD and MD plots (for details, see Results). Cystine is not an essential amino acid, but methionine is an indispensable component of the diet that supplies sulfur for cystine synthesis when it is absent or deficient in the diet. If dietary methionine levels are adequate, cystine does not need to be present in the diet (Nelson and Evans 1958). The 2 types of pelleted chows were therefore designed by adjusting the methionine levels to meet the following 3 criteria: (1) high-methionine pelleted chow had 37% more sulfur-containing amino acid content than low-methionine pelleted chow, which reflects the difference in diet between HD and MD plots; (2) high-methionine pelleted chow had 0.65% sulfur-containing amino acid content, which equates to the reproductive requirement for domestic rabbits (De Blas and Mateos 1998); (3) concentrations of other essential amino acids in both pelleted chows were the same as those in the diet derived from the MD plot. According to our design criteria, we ordered the 2 pelleted chows from Jiangsu Xietong Pharmaceutical Bioengineering Co., Ltd (Jiangsu, China). By arraying concentrations of essential amino acids in the ordered pelleted chows, high- and low-methionine pelleted chows contained 0.45% and 0.26% methionine, as well as 0.66% and 0.48% sulfur-containing amino acid, respectively. The 0.66% sulfur-containing amino acid content was slightly higher than the reproductive requirement for domestic rabbits (0.65%). High-methionine pelleted chow had 37.5% higher sulfur-containing amino acids than low-methionine pelleted chow, which approximates the difference (37%) between HD and MD plots. Other nutrient concentrations were the same in both foods (Supplementary Data SD2). Thus, the 2 foods used in the feeding experiment met our design criteria.

Body mass and food intake

We weighed the pikas every 3 days throughout the experimental period and monitored food intake daily for 5 days from the 45th day to the 49th day. Each individual within each group was provided with 40 g of pelleted chow at 9:00 am daily, exceeding the typical daily intake for an individual. The remaining pellet chow was weighed at 9:00 am the following day.

Cellular immunity assays

In the field of ecology, cellular immunity is usually assessed by a phytohemagglutinin (PHA)-delayed hypersensitivity response, in which a subcutaneous injection of PHA can induce local T-cell stimulation and proliferation and result in swelling (Stadecker and Leskowitz 1974). The degree of swelling roughly reflects the strength of the cellular immune response. The most important requirement to measure the PHA response is that the animal has a consistently measurable surface such as a pinna, fin, or foot pad (Demas et al. 2011). On the 20th day of the feeding experiment, we measured cell-mediated immunity by PHA assay. We first measured the left-hind footpad thickness using a micrometer (Tesa Shopcal, Switzerland). Subsequently, 0.33 mg of PHA (Coolaber CP8341) dissolved in 0.1 mL of sterile saline was injected into the middle of the same footpad. After 6 h, the thickness of the footpad was measured again. Each measurement was replicated 3 times on the same individual. The PHA response was calculated as the difference between the preinjection and postinjection measurements divided by the initial footpad thickness [PHA response = (post-PHA − pre-PHA)/pre-PHA] (Chen et al. 2023).

Humoral immunity assays

The humoral immune response is assessed by measuring the production of immunoglobulin G (IgG) in reaction to specific antigens, such as keyhole limpet hemocyanin (KLH; Goüy de Bellocq et al. 2006; Du et al. 2016), which serves as a common and innocuous antigen for inducing antibody responses in mammals (Harris and Markl 1999). On the 24th day of the feeding experiment, pikas received a single subcutaneous injection of 0.15 mg of novel antigen KLH (H7017-20MD) suspended in 0.1 mL of sterile saline. After 10 days of KLH injection, blood samples were collected (70 to 100 μL) from the retro-orbital sinus and centrifuged at 4,000 rpm for 15 min at 4 °C, and serum was collected and stored at −80 °C until analysis. We measured the serum anti-KLH IgG level with an ELISA using a commercially available rabbit anti-KLH IgG ELISA kit (Shanghai Enzyme Link Biotech Co., Ltd, China). All procedures were performed in accordance with the kit instructions.

Coccidian infection assays

To further examine the effect of methionine on immunity, a coccidia infection experiment was performed on the 42nd day of the feeding experiment (18 days after KLH injection) to challenge pika defenses. It should be noted that the concentration of anti-KLH IgG typically reaches its peak at day 10 post-antigen injection and then gradually declines over time (Zysling and Demas 2007), so the previous KLH injection did not influence the coccidia infection outcome. Coccidia is the main parasite of the intestinal tract in plateau pikas (Yang et al. 2018). To ensure homogeneity of infection, we removed initial coccidia prior to inoculation with the coccidian oocysts by orally providing a 0.1-mL anthelmintic that contained 3.3 × 10⁻⁵ g of diclazuril (Weierkong, Sichuan), which has been proven to effectively remove coccidia (Yang et al. 2018). All individuals were inoculated orally with 200 coccidias suspended in 0.1 mL of water. Feces were daily collected per pika for 5 consecutive days and stored at 4 °C until tested. We determined fecal oocyst counts by a modified McMaster method using a saturated NaCl solution for flotation (Du et al. 2016; Yang et al. 2018). Prevalence was calculated as the number of infected hosts in the experiment divided by the total number of hosts that were sampled in the experiment. The intensity of coccidian infection was measured as the number of oocysts per gram (OPG) of fresh feces.

Reproductive hormone concentrations

To examine the effect of methionine on reproductive performance, at the end of the experiment we assessed concentrations of gonadotropin-releasing hormone (GnRH), plasma follicle-stimulating hormone (FSH), luteinizing hormone (LH), estradiol (E2), progesterone (PROG), and testosterone (T) secreted by the hypothalamic–pituitary–gonadal axis. GnRH is secreted by the hypothalamus, stimulating the pituitary gland to release FSH and LH. These 2 hormones promote gametogenesis and regulate the production of steroids including E2, PROG, and T (Izvolskaia et al. 2016). All individuals were exposed to Zoletil 50 (LAT-ST) and decapitated, and trunk blood was collected individually. They were centrifuged at 4,000 rpm for 15 min to obtain serum and stored at −80 °C until assayed for sex hormone concentration. The hypothalamus tissues were excised and homogenized in phosphate-buffered saline (PBS) pH 7.4, and the supernatants were obtained following centrifugation at 4,000 rpm for 25 min at 4 °C, and then stored at −80 °C until analysis for GnRH contents. The concentrations of GnRH, FSH, LH, E2, PROG, and T were assayed with an ELISA using commercially available rabbit ELISA kits (Wuhan Colorful Gene Biotech Co., Ltd, China). All experiments were conducted based on the specifications provided in each kit.

Oxidative stress biomarker

Fertility potential was significantly related to increased superoxide dismutase (SOD) levels and decreased contents of malondialdehyde (MDA) of the testis (Mumtaz et al. 2022). At the end of the experiment, we excised testis tissues and homogenized them in PBS (1:10 w/v) for 5 min to produce 10% testis homogenate before being centrifuged for 10 min at 4 °C and 3,000 rpm to collect the supernatant and stored at −80 °C before use. The concentrations of SOD and MDA were detected with a commercially available rabbit ELISA kit (Wuhan Colorful Gene Biotech Co., Ltd, China). The protein content of testis tissue homogenates was determined by the Bradford method (Bradford 1976). Bovine serum albumin (Sigma, St. Louis, Missouri) was used as the calibrating protein.

Evaluation of spermatogenesis

The epididymis immersed in 1 mL PBS was cut with scissors to release sperm. The sperm suspension was diluted using PBS and then sperm density was counted with a hemocytometer under an optical microscope. A 10 μL sperm suspension was transferred on a slide (prewarmed at 37 °C), and a microscope was used to immediately assess sperm motility which is calculated as the percentage of motile sperm in 100 spermatozoa. To determine sperm abnormalities, 50 μL of semen suspension was transferred to a slide. Air-dried slides were fixed with methanol for 5 min and stained with 2% eosin for 2 h (Zhang et al. 2022b). A total of 1,000 sperm were randomly analyzed to detect morphological abnormalities, and morphology was expressed as percentages of normal cells (without any abnormalities, i.e., head, midpiece, or tail abnormalities).

Observation of ovarian sections

Ovaries were fixed in 4% paraformaldehyde for 4 h at 4 °C followed by embedding in paraffin, sectioning (5 mm), and staining with hematoxylin–eosin (H&E). Follicles were categorized as primordial, primary, secondary, and antral (Li et al. 2016). The number of each of these 4 types of follicles per ovary was determined by taking an average of the counts from 10 sections (5 sections apart) cut along the long axis of the entire ovary.

Statistical analyses

The concentration of essential amino acids was analyzed using generalized linear mixed models in IMB SPSS Statistics 21 (IBM Corp, 2012). The factors of meadow type, season, sex, and their interaction were entered in models as fixed factors, and sampling regions (Haibei and Sanjiangyuan) were entered as random factors (Wu et al. 2023). We used repeated-measures ANOVA to examine differences in body mass, food intake, and the intensity of coccidian infection. Cellular and humoral immunity, reproductive hormone content, and spermatogenesis were analyzed using independent samples t-tests. Food treatment was entered in all models as a fixed factor. Data on the intensity of coccidian infection and sperm density were ln(OPG/R + 1) transformed prior to analysis.

Results

Variation in essential amino acids

We found an effect of meadow type and season on the concentration of all essential amino acids, except for the effect of season on histidine concentration in the stomach digesta (Fig. 2). Concentrations of the 9 essential amino acids measured in the stomach digesta collected from the HD plot were higher than those from the MD plot. Eight (threonine, isoleucine, histidine, lysine, arginine, methionine + cysteine, phenylalanine + tyrosine, valine) of the 9 essential amino acids had higher concentrations in HD than in LD (Fig. 2). Average concentrations of all essential amino acids were 12% and 9% greater in the diets of pikas from the HD plot than in those from the MD and LD plots throughout the year. Except for histidine, concentrations of all essential amino acids were greater in summer than in winter (Fig. 2). The largest seasonal difference in concentration was observed for methionine + cysteine, which averaged 35% higher in summer than in winter. We found an effect of sex on the concentration of arginine, threonine, histidine, leucine, lysine, and valine, with males significantly higher than females (Table 1).

Additionally, we observed an interaction effect between season and meadow type on arginine, methionine + cysteine, leucine, isoleucine, and valine, which were higher in HD than in MD and LD, and there was no significant difference between the MD and LD plots during summer. However, there were no significant differences in arginine, methionine + cysteine, leucine, isoleucine, and valine, among the 3 meadow types during winter (Table 1). The interaction effect between season and sex significantly affected concentrations of all essential amino acids in the stomach digesta except for methionine + cysteine. For the concentration of methionine + cysteine, we found interactions of meadow type × sex and meadow type × sex × season (Table 1).

Limiting amino acids

Compared to the growth requirements of domestic rabbits, the diet of plateau pikas from LD had 3 insufficient essential amino acids (methionine + cysteine, phenylalanine + tyrosine, and threonine) during summer (Fig. 3A) and 4 relatively insufficient essential amino acids (methionine + cysteine, phenylalanine + tyrosine, threonine, and arginine) during winter (Fig. 3B). In the MD and HD plots, the diets of pikas had 3 essential amino acid deficiencies (methionine + cysteine, phenylalanine + tyrosine, and threonine) during summer and winter. Among them, regardless of season and meadow type, methionine + cystine occurred at concentrations below the growth requirements of domestic rabbits and became the most limiting of the essential amino acids (Fig. 3). The limitation was more severe in MD and LD than in HD during summer, being 37% and 23% lower, on average, in MD and LD than in HD but being 9% and 1% lower in MD and LD plots than in HD during winter. In addition, except for the above 4 amino acids, other essential amino acids were present in sufficient concentrations in the diets of pikas from the 3 meadows.

Effect of methionine on body mass and food intake

There was no significant difference in body mass between the 2 food groups (F_1,80 = 0.92, P = 0.34; Fig. 4A), but there was a significant effect of methionine on food intake (F_1,44 = 0.92, P = 0.043; Fig. 4B), which was higher in the low-methionine group than in the high-methionine group.

Effect of methionine on immunity response and parasitic infection

Pikas in the high-methionine group had a higher PHA response than those in the low-methionine group (t₉₄ = 3.20, P = 0.02; Fig. 4C), but the anti-KLH IgG level did not differ between the 2 groups (t₇₉ = 0.43, P = 0.095; Fig. 4D). In addition, there was a significant difference in the intensity of coccidian infection between the 2 groups during the experiment (F_1,32 = 8.71, P = 0.006; Fig. 4E). Infection intensity of the low-methionine group was 20.56% higher than that of the high-methionine group. The mean prevalence in coccidia did not differ between the 2 groups (t₈ = −0.78, P = 0.46; Fig. 4F).

Effect of methionine on reproductive hormones

The pikas in the high-methionine group had higher concentrations of GnRH (t₈₄ = 5.30, P < 0.001; Fig. 5A), LH (t₈₄ = 2.08, P = 0.041; Fig. 5C), E2 (t₃₆ = 3.03, P = 0.004; Fig. 5D), PROG (t₃₆ = 2.50, P = 0.017; Fig. 5E), and T (t₄₅ = 3.62, P = 0.001; Fig. 5F) than those in the low-methionine group, but there was no difference in FSH (t₈₂ = 1.50, P = 0.14; Fig. 5B) contents between the 2 groups.

Effect of methionine on oxidative stress biomarker

The pikas in the high-methionine group had higher SOD concentrations (t₄₆ = 0.21, P = 0.04; Fig. 5G) and lower MDA concentrations (t₄₆ = −3.76, P = 0.001; Fig. 5H) than those in the low-methionine group.

Effect of methionine on spermatogenesis and ovarian follicles

Differences in sperm density (t₄₄ = 1.65; P = 0.11; Fig. 5I), sperm malformation rate (t₄₄ = −2.01; P = 0.051; Fig. 5J), and sperm motility (t₄₄ = 1.43; P = 0.16; Fig. 5K) were not found between the 2 groups, but sperm density was on average 30.1% greater in the high-methionine group, and the malformation rate of sperm decreased by 31.2% compared to those in the low-methionine group. There was no difference in the number of primordial follicles (t₂₈ = 0.52; P = 0.61; Fig. 5L), primary follicles (t₂₈ = 1.53; P = 0.14), secondary follicles (t₂₈ = 1.08; P = 0.29), or antral follicles (t₂₈ = 1.885; P = 0.07) between the 2 groups.

Discussion

In the present study, we found that the concentrations of most essential amino acids, as a measure of the quality of Plateau Pika diets, were significantly variable across meadows, being higher in the HD plot than in the LD and MD plots, indicating that they consumed higher-quality protein foods in the HD. The findings agreed with those of Lochmiller et al. (1995) that the diet concentrations of essential amino acids of eastern cottontails (Sylvilagus jioridanus) declined as herbaceous weedy forbs were replaced by grasses during positive vegetation succession. A shift in the vegetation community composition and plant biomass due to grassland degradation can result in a change in the abundance of preferred plant species for small herbivores (Bakker et al. 2009). On the QTP, the proportion of high-protein plant resources preferred by plateau pikas (e.g., legumes, most of which are toxic to livestock but are favored by plateau pikas) increases with grazing intensity and grassland degradation (Li et al. 2019; Wu et al. 2023). The above results indicate that plateau pikas inhabiting the HD consumed abundant high-protein food resources, including higher-quality protein foods. Numerous studies on rabbits (Casady et al. 1961), birds (Wegmann et al. 2015; Langlois and McWilliams 2021), poultry (Bunchasak 2009), and eastern cottontails (Webb et al. 2003) have found that high levels of essential amino acids in the diet promote growth and modulate immunity by reducing the inflammatory response and increasing the delayed hypersensitivity response. Enhanced reproductive success of mice and eastern cottontails due to the improved quality of dietary protein has also been documented (Webb et al. 2005; Cavalcante-Silva et al. 2022). Consequently, the consumption of protein-rich foods of superior quality may enhance individual fitness through the facilitation of growth, increasing immunity, and improving reproductive performance.

Several studies have shown that sulfur-containing methionine + cystine in natural forages of small herbivores is highly variable and frequently deficient relative to their requirements (Sedinger 1984; Lochmiller et al. 1995; Schetter et al. 1998; Webb et al. 2005). In this study, we found that regardless of season and meadow type, the RCAA of 3 essential amino acids (concentrations of methionine + cystine, threonine, and phenylalanine + tyrosine) was <1, indicating that their concentrations were relatively insufficient in the natural diets of plateau pikas. Of these amino acids, methionine + cysteine had the lowest RCAA, and their levels in stomach digestions from the 3 types of alpine meadows were well below those required for growth in domestic rabbits in summer and winter, indicating that methionine + cysteine was the most limiting essential amino acid for plateau pikas in alpine meadows on the QTP. Similar results have been reported for eastern cottontails (Lochmiller et al. 1995), cotton rats (Schetter et al. 1998), and a variety of avian species (Langlois and McWilliams 2021). More importantly, levels of sulfur-containing amino acids were significantly lower in the MD and LD plots compared to the HD plot during summer, which coincides with the breeding season for plateau pikas. However, there was no significant difference in the contents of sulfur-containing amino acids among the 3 types of alpine meadows during winter. This seasonal variation may be attributed to a decline in protein content as plants wither during the winter season (Redfern et al. 2003; Wu et al. 2023). Some early studies showed that if methionine was absent, laboratory rats (Rattus spp.) rapidly lose body mass and eventually die, even if abundant cystine is supplied (Rose 1937). Cystine stimulates growth only when dietary methionine is present (Rose 1937). In contrast, if dietary methionine levels are adequate, cystine does not need to be present in the diet (Nelson and Evans 1958). Schetter et al. (1998) suggested that sulfur-containing cystine + methionine, especially methionine, may be fundamentally involved in limiting wild populations of cotton rats. Webb et al. (2005) found that methionine supplementation enhanced cotton rat population growth. Reproductive performance during the breeding season determines the amplitude of small mammal population fluctuations. Therefore, based on these studies and our finding that plateau pikas consumed higher levels of sulfur amino acids in HD plots in summer, we postulate that the increased concentrations of methionine resulting from HD may positively affect Plateau Pika populations during the breeding season.

Our results from the feeding experiment provided evidence for the inference described above. We found that a high level of methionine enhanced potential reproductive performance by increasing the levels of GnRH, LH, E2, PRO, T, and SOD, and decreasing the level of MDA of plateau pikas. Webb et al. (2005) found that methionine supplementation increased per capita recruitment and the proportion of individuals in reproductive condition, resulting in an earlier and longer reproductive season in Hispid Cotton Rat populations. For nonperiodic fluctuation in the numbers of small mammals, a key question is what determines the peak density or amplitude of seasonal fluctuations. Schetter et al. (1998) suggested that nutrient levels during the breeding season may be most important in determining peak density in a seasonally fluctuating herbivore population, while the most important factor in determining nutrient quality for herbivores is the availability of methionine because it is frequently deficient and is the most limiting essential amino acid in wild herbivores, as shown by our and other studies (Lochmiller et al. 1995; Schetter et al. 1998). Thus, levels of methionine during the breeding season dictate potential peak densities of herbivore populations (Schetter et al. 1998). Taken together with the results of our previous study on the relationship between availability of high-protein resources and the population density of plateau pikas (Wu et al. 2023), these results suggest that although other environmental factors may be involved in driving fluctuations in pika populations, the interaction of abundant high-protein foods and relatively high levels of methionine both of which were produced in heavily grazed meadows or degraded meadows during the pika breeding season may be an important determinant of the peak density of Plateau Pika populations at the end of breeding.

In the present study, we also found that high-methionine foods improved cell-mediated immunity and reduced the intensity of coccidian infection. These 2 results corroborated each other, indicating that high-level methionine foods enhance individual immunocompetence. Webb et al. (2003) found that methionine supplementation enhanced platelet and leukocyte counts, which may confer advantages to the overall health of male cotton rats. Immunity suppression can increase parasite infection, and in turn lead to an increase in the risk of predation (Shang et al. 2019). Thus, the involvement of host immunocompetence may influence survival of individuals within herbivore populations. Additionally, in this study, although no positive effect of high-methionine foods on body weight growth was found, when individuals were fed low-methionine foods, they ate more food to compensate for the lack of methionine contents, which is widely believed to be the universal optimal foraging strategy for animals. In natural populations, an increase in the number of bouts of foraging may increase the risk of exposure to both parasites and predators, inducing stress responses in individuals (Lima and Dill 1990; Winternitz et al. 2012; Shang et al. 2019). Stress responses can induce a vicious cycle interaction between stress, immunity, and parasites (Beldomenico and Begon 2010, 2015; Yang et al. 2018). In other words, individuals in poor physiological condition are more susceptible to infection, and once it occurs further deteriorating their condition, making them even more susceptible to infection and ultimately death. Shang et al. (2019) found that this vicious cycle led to a population decline by reducing the overwinter survival of individuals in root voles (Microtus oeconomus). Taken together, these results indicate that insufficient or low concentrations of methionine reduce immunity, and therefore fitness, and it is reasonable to speculate that food resources in heavily degraded grasslands may be involved in regulating populations of plateau pikas by increasing survival rate.

High densities of plateau pikas indicate a poor condition of grassland and further exacerbate grassland degradation (Fan et al. 1999; Harris 2010). A vicious circle thus ensues: heavy livestock grazing leads to grassland degradation, with dramatic increases in pika populations, resulting in further grassland degradation (Kang et al. 2007). Numerous studies showed that the presence of open habitats is a crucial factor contributing to significant increases in pika populations (Shi 1983; Bian and Fan 1997; Badingqiuying et al. 2018). Therefore, the mechanism underlying this vicious circle can be attributed to the synergistic effects of open habitat availability and food quality in degraded grasslands. We suggest that the best strategy to prevent dramatic increases in population sizes of plateau pikas is to manage livestock numbers and grassland practices to avoid grassland degradation.

In conclusion, plateau pikas in HD associated with livestock grazing intensity consumed relatively high levels of essential amino acids, especially methionine. High-level methionine foods enhance individual fitness by improving immunity and reproductive performance, which supports the proposition of both Chapman et al. (1982) and Lochmiller et al. (1995) that the availability of sulfur-containing amino acids may be a critical determinant of life-history events in populations of small herbivores. Additionally, grassland degradation or livestock grazing created open habitats for pikas to increase their ability to detect approaching predators (Shi 1983; Bian and Fan 1997; Fan et al. 1999). We suggest that methionine, together with open habitats, may be fundamentally involved in the population dynamics of plateau pikas in degraded grassland on the QTP.

Supplementary data

Supplementary data are available at Journal of Mammalogy online.

Supplementary Data SD1. Vegetation cover, height, relative biomass, and vegetation community parameters of 3 meadow types in Haibei and Sanjiangyuan regions.

Supplementary Data SD2. Nutrient of experimental diets.

Supplementary Data SD3. All data used in this study.

Acknowledgments

We thank 2 reviewers, Dr. Rhiannon Jakopak and R. B. Harris, for their thoughtful reviews and comments on earlier versions of this manuscript.

Author contributions

JHB and YW conceived and designed the experiments. HQC, GZS, LZ, XD, and XQW performed the experiments. HQC and JHB analyzed the data. HQC wrote the manuscript. All authors contributed critically to the drafts and gave final approval for publication.

Funding

This work was supported by the Natural Science Foundation of Qinghai Province (grant number: 2021-ZJ-945Q), joint grant from Chinese Academy of Sciences–People’s Government of Qinghai Province on Sanjiangyuan National Park (grant number: LHZX-2021-05 and LHZX-2020-01), and Sanjiangyuan Animal Genome Project and Strategic Priority Research Program of the Chinese Academy of Sciences (grant number: XDA2005010406).

Conflict of interest

None declared.

Data availability

All data used in this study are available in Supplementary Data SD3.

References

Author notes