Ethics

No specific permissions were required for research on the locations described in this article as all activities were carried out according to the Swedish Right of Public Access law. The field studies did not involve endangered or protected species. All animal handling and sampling was carried out under approval of the Swedish Ethical Committee on animal research (C212/9) and was in compliance with Swedish laws and regulations. The appropriate authority and ethical committee was “Djuretiska nämnden, Uppsala, Sweden”.

Bears and Collections of Samples

Samples of blood were taken from 16 free-ranging sub-adult 2- to 3-yr-old Eurasian brown bears [Ursus arctos, 11 females and 5 males equipped with a Global Positioning System (GPS) collar with a median body weight of 53 kg (range 22–77 kg)] in Dalarna and Gävleborgs Counties, Sweden, 2010–2012. Each bear was captured during hibernation (February or March) (Fig. 1) and during the active period in June of the same year [6]. Bears were immobilized by darting with an anesthetic (see below) in the den and again by darting from a helicopter during June (Fig. 2). Bears were weighed on a stretcher suspended beneath a spring scale. In winter the anesthetic consisted of a mixture of tiletamine-zolazepam (1.1 mg/kg), medetomidine (0.03 mg/kg) and ketamine (1.3 mg/kg), and in summer, in a mixture of tiletamine-zolazepam (4.7 mg/kg) and medetomidine (0.09 mg/kg) [6]. In the field, blood samples were taken from the jugular vein within 20 min from darting and collected in tubes containing the additives EDTA, Lithium heparin or clot activator, respectively (Vacuette®; Greiner Bio-one, Frickenhausen, Germany). Approximately 1 hour after sampling blood was centrifuged at 2000xg for 10 min for collection of plasma or serum and immediately frozen on dry ice until storage at −70°C.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Captured brown bear in its den.

Photo: Andrea Friebe, the Scandinavian Brown Bear Research Project.

https://doi.org/10.1371/journal.pone.0072934.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 2. Aerial darting from helicopter of a free-ranging female brown bear during the active summer period.

Photo: Andreas Zedrosser, the Scandinavian Brown Bear Research Project.

https://doi.org/10.1371/journal.pone.0072934.g002

Laboratory Measurements

The following analytes were measured by spectrophotometer with routine methods on a Konelab 20XT centrifugation analyzer (Thermo Fisher Scientific, Vantaa, Finland); albumin with the bromcresol green (BCG) method; glucose using the glucose oxidase/peroxidase Trinder reaction; creatinine with the Jaffe method; cholesterol and triglycerides in coupled enzyme reactions forming quinonimine, respectively; total protein, calcium and phosphorus forming colored complexes with cupric ions, Arsenazo III and ammonium molobdate, respectively; urea using an urease/glutamate dehydrogenase method converting NADH to NAD; and uric acid with a fully automated Trinder (AOX) method using uricase/peroxidase.

Fructose was measured enzymatically with a recently modified sensitive inulin assay [7]. The fructose in the sample is converted to sorbitol using sorbitol dehydrogenase (SDH) and NADH, which is read at 340 nm on the Konelab 20XT pending an incubation for 20 min. The thiol assay measured free sulphydryls in the plasma in a reaction using Ellman's reagent (5,5′-dithio-bis(2-nitrobenzoic acid)) (DTNB). The TNB²⁻ formed was quantified spectrometrically at 412 nm on the Konelab 20XT [8]. Plasma C-reactive protein (CRP) levels were measured by the Immulite Automatic Immunoassay Analyzer (Siemens Medical Solutions Diagnostics, Los Angeles, CA, USA) with an assay manufactured for this analyzer. The assays are not validated for bear plasma. Measurement of pentraxin-3 (PTX3) was accomplished with ELISA's from R&D Systems Inc. (Abingdon, UK). Free plasma amino acids were determined in both human and bear samples by high-performance liquid chromatography with fluorometric detection [9]. Aspartate, cysteine and proline are not possible to analyze for technical reasons. Serum FGF23 was measured using an intact mouse/human FGF23 assay (Kainos, Japan) that only detects the intact, biologically active, protein by using two monoclonal antibodies for capture and detection, respectively. Although this assay has not been developed and/or validated for brown bears, the results yielded physiologically plausible concentrations comparable with those in humans and demonstrated good linearity in diluted samples. In contrast, a C-terminal FGF23 assay and intact FGF23 assay (Immutopics, CA, USA) did not produce detectable FGF23 levels in the current study.

Statistics

All values are expressed as median (range). Differences between winter and summer samples were analyzed with paired Wilcoxon signed rank test. Comparative statistical between healthy humans and summer bears were analyzed by non-parametric Wilcoxon signed rank test. Changes in the level of some variables between the hibernation and active periods were calculated and shown as “Δ”. Spearman's rank correlation was used to determine correlations between two variables. A p-value <0.05 was considered to be statistically significant. The statistical analysis was performed using statistical software SAS version 9.3 (SAS Campus Drive, Cary, NC, USA).

Nitrogen metabolism

This study confirms the remarkable ability of the brown bear to recycle urea nitrogen during the hibernation period. As mammals cannot hydrolyze urea (because they do not possess the enzyme urease) it has to be excreted by the kidney. A previous study in bears has shown that glomerular filtration rate (GFR) is reduced to about only one fourth of its normal value during hibernation (from 122 to 37 ml/min) [10]. As the bladder becomes leaky so that water, electrolytes and nitrogen wastes are returned to the blood, bears are anuric during the entire hibernation [11]. Despite the decrease in GFR and the recycling of urine, serum urea is lower during hibernation than during the active period due to the unique ability of bears to recycle urea nitrogen back to protein [12]. In contrast to the lower urea concentration, creatinine (another endproduct of protein metabolism) was 2.5-fold higher during hibernation than during summer. Creatinine, the endproduct of creatine, which is the high-energy storage molecule in muscle, is not metabolized and is cleared only by the kidney. Since bladder urine returns to the blood, creatinine cannot be excreted during hibernation and accumulates in the blood. However, this elevation remains relatively modest because creatinine generation is probably reduced in winter due to the lack of muscle activity. In most cases, when kidney function is reduced in humans and other mammals, serum creatinine and urea concentrations usually rise in parallel. In the case of hibernation, we observe a remarkable divergence in serum urea and creatinine (Fig. 3) that is explained by the possibility to re-use urea nitrogen, but not creatinine, through symbiotic gut bacteria.

The mechanism(s) responsible for the reduction in urea during hibernation probably involves a combination of factors. First, a reduction in metabolic rate probably leads to lesser urea synthesis in the liver. During winter bears obtains most of their energy from metabolizing stored fat producing only CO₂ and H₂O as end products. Second, as the urea that is still produced is recycled back into skeletal muscle and other body proteins [13] it has been speculated that (like in ruminants and some other herbivorous mammals) urea is hydrolyzed by urease-expressing gut bacteria into ammonia and CO₂. Ammonia is then used by enterocytes to synthetize glutamine which may be incorporated into proteins [14]. In ruminants, a facilitated urea transporter is expressed in the rumen epithelium [15]. A similar urea transporter is also expressed in the colon of other mammals [16] including in humans [17]. Most likely, bears express a similar urea transporter in the colon and/or some subsegment of the intestinal wall. Ahlqvist et al [18] showed that by serving as a carbon source for amino acid formation, glycerol released from fats might help prevent azotemia during winter denning. The observation by Nelson et al [19] that urea levels decrease in the autumn, when food is still available, suggests that metabolic changes occur prior to the hibernation state. This occur as a consequence of increased intake of fruits and berries prior to hibernation [20]. As berries provide abundant carbohydrates with little proteins, it reduces the need to synthetize and excretes urea.

Nutritional Parameters

Bears increase their food intake in the late summer and autumn to increase their fat stores before hibernation, and then will survive hibernation primarily by burning these fat stores [21]. In this study, the hibernating bears showed evidence for enhanced fat stores as noted by the higher levels of serum triglycerides and cholesterol levels (Table 1), which confirms findings by Arinell et al [22]. We observed higher levels of both the total protein and albumin in hibernation samples. This may be due to a modest dehydration during hibernation because bears (although inactive) cannot prevent some water loss through the airways in the expired air. However, this finding also indicates preserved protein stores and confirms Lohuis et al [23], who showed that protein synthesis and breakdown were in balance during winter anorexia. As bears spare most of their crucial skeletal muscle proteins during hibernation anorexia, it seems conceivable that reserves from other organs may be used as sources for nitrogen during denning. Complicated protein kinetic studies of other organ metabolism, such as liver and kidney, are needed to resolve this issue.

Several mammals, including hibernating marmots (Marmota) and ground squirrels (Sciuridae) as well as long distance migratory birds, develop insulin resistance in preparation for the period of fasting associated with hibernation or migration [24], [25]. In this study the hibernating bears had slightly higher glucose levels suggesting the presence of insulin resistance. In many hibernating animals, the brain might utilize the release of ketones, such as β-hydroxybutyric acid instead of glucose during hibernation [26]. Whether this is also occurring in bears is not known. In the present study we also noted a two-fold higher uric acid level in the active period in comparison to hibernation. Serum uric acid of hibernating mammals tends to be elevated during the active period and to decrease during torpor [27]. Nelson et al [12] studied two captive American black bears (Ursus americanus) and noted higher urinary uric acid excretion in the autumn before hibernation. Previous studies in hibernating squirrels have shown that there is a rapid drop in both inosine and uric acid in the liver during hibernation [28], [29], consistent with an inhibition of AMP deaminase. We have recently found evidence that AMP deaminase activity is low in the liver of the 13-lined ground squirrel (Ictidomys tridecemlineatus) during hibernation (R Johnson, unpublished) and a decrease in AMP deaminase may be important for the activation of AMP kinase and the burning of fat [30]. In addition, higher uric acid levels in summer could represent the generation of uric acid during the metabolism of purines and fruits. Fruits contain fructose, a monosaccharide that generates uric acid during its catabolism [31]. Fructose has been shown to increase fat stores in a variety of animals as well as to induce insulin resistance [32], and the mechanism is likely mediated in part by the effect of uric acid to induce mitochondrial oxidative stress [33]. For bears, fruit intake increases particularly in the autumn where it is a food source used to help fatten the animal. The typical bear diet during late summer and autumn includes enormous amount of berries, such as from the Vaccinum family [i.e. bilberries (V. myrtilus), huckleberrries (V. parvifolium) and lingonberries (V. vitis-idea)], which are rich in both fructose [34] and resveratrol [35]. It has been reported that large bears in captivity can eat as much as 260,000 huckleberries/day when fed ad libitum [36]. In fact, in a feeding and foraging trial using captive and wild American black bears, maximum intake ranged from 30 g/min for berries to amazing >200 g/min for fruits [20]. It should also be taken into account that during the ripening process of berries during the summer and autumn their content of fructose, flavonol, abscisic acid (a plant hormone) and anthocyanins increases [37] whereas vitamin C content decreases [38]. Considering the enormous amount of berries consumed by bears in the summer and autumn, their effects on metabolism and renal function need further studies.

Amino acids

We demonstrated higher total amino acid, EAA and NEAA levels in summer active bears compared to healthy humans. Interestingly, more than 3-fold higher taurine levels were observed in bears during their active period. Since taurine is key for the conjugation of the unique bear bile ursodeoxycholic acid it has been speculated that taurine deficiency in captive bears causes metabolic bone disease due to decreased absorption of vitamin D [39]. Previous studies on bears have shown discrepant results regarding changes in plasma amino acid levels; with constant [12], elevated [40] or decreased [41] levels reported during winter anorexia. Our amino acid analyses demonstrated no seasonal change in either total amino acids, EAA, NEAA or BCAA levels. This contrasts with the dramatic changes in amino acid levels that occur during prolonged starvation in humans [42], deep hibernators, such as hedgehogs (Erinaceinae spp) [43], and other mammals undergoing prolonged fasting, such as elephant seals (Mirounga spp) [44]. This suggests that bears maintain a remarkably stable whole amino acid balance despite several months of anorexia and inactivity. As inflammation contributes to low amino acid levels in CKD patients [45] the absence of inflammation during hibernation may contribute to maintained amino acid levels in bears.

Several specific changes in individual amino acid concentrations were observed during hibernation (Table 2). For the following discussion we assume that EAAs are identical in bears and humans. Although this has not been evaluated to our knowledge, a recent study suggests that the essential amino acids are the same in humans, rats, dogs and a few other species, with differences only in quantitative requirements among species [46]. In the present study, five of the EAAs did not change significantly between summer and winter. Since there is no food intake and since these AAs are assumed not to be synthetized in the body, it means that their balance is kept constant, and thus, that they are not at all degraded (since they could not be re-synthetized, being “essential”). BCAA (leucine, isoleucine and valine) usually serve as metabolic fuel in muscle and kidney. During hibernation, the metabolism is markedly reduced and the level of these amino acids is, thus, unchanged. It is notable that methionine, a sulfur amino acid, declined to almost half of its summer value. The increased concentration of lysine and histidine may result from some protein breakdown. The possibility that amino acids synthetized by the gut microflora might be used by the bears require further studies.

A larger fraction of the NEAAs showed seasonal changes since only three out of 12 did not change significantly. The three NEAA that exhibit a similar concentration during hibernation as during the active period are the three smallest amino acids (glycine, alanine and serine). Interestingly, the amino acids that are involved in nitrogen excretion in the form of urea and ammonia exhibit coordinated changes in winter. We observed about 50% higher levels of glutamine during hibernation compared to the active summer state. As glutamine is formed from glutamate and ammonium (NH₄⁺) by glutamine synthase, increased glutamine formation may provide a mechanistic explanation to counter unwanted NH₄⁺ buildup during conditions of reduced renal clearing capacity. Glutamine is the amino acid that serves to transport nitrogen between organs. Notably, glutamine is used to transfer nitrogen from the digestive tract to the kidney where NH₄ is produced and excreted [47]. Furthermore, glutamine exerts translational control of mitochondrial uncoupling protein (UCP)-2 by binding to an open reading frame [48] and UCP-2 mediates mitochondrial uncoupling by facilitating proton translocation across the mitochondrial inner membrane without the production of ATP [49]. Interestingly, increased UCP-2 activity results in defective glucose-induced insulin release from pancreatic beta-cells [50], which could explain the apparent hyperglycemia during hibernation. It is tempting to propose that the increased glutamine concentration during hibernation is a mechanism to signal increased mitochondrial uncoupling, and thus increased heat production, from peripheral tissue other than the UCP-1 containing brown adipose tissue. The normally devastating UCP-2-mediated increased kidney oxygen usage, resulting in intrarenal hypoxia and kidney disease [51], is likely prevented by the low GFR, and thus low oxygen demand, during hibernation. The observation that both glutamine and ornithine are significantly increased in winter also suggests a lower metabolic rate of these nitrogen excretion-related pathways. During normal feeding, dietary arginine is almost entirely degraded in the liver. The body's supply in arginine originates from the synthesis of arginine in the kidney, using citrulline that is released by the liver in the blood stream [52]. The opposite differences observed during hibernation vs. summer in citrulline (increase) and arginine (decline) is consistent with a reduced metabolic activity of the kidney.

Calcium-Phosphate Axis

In humans, long-term inactivity, as well as a reduction in kidney function, are associated with substantial changes in parameters of bone and mineral metabolism [5]. In contrast, a striking feature in bears is that calcium levels (as observed in the present study) and bone mineral density are unaltered during hibernation, indicating the presence of bone-preserving mechanisms [53], [54]. In this study, we found that serum inorganic phosphorous level was significantly lower during hibernation. This may reflect a state of starvation and reduced caloric intake with redistribution from extracellular to intracellular space when cellular phosphate demand is insufficient. This has previously been demonstrated in healthy humans and in CKD patients with protein energy wasting and/or reduced protein intake [55]. Low winter phosphate levels may also reflect bone-preserving mechanisms; i.e. no excess phosphate is released from bone. As renal and intestinal phosphate transport is modulated by amino acids and glucose it is possible that also other metabolic changes will off-set downstream alterations in phosphate metabolism. Although numerous studies in man and rodents support that FGF23 is stimulated by dietary phosphate intake and hyperphosphatemia [56], [57], [58], we observed higher levels of circulating FGF23 during hibernation, despite the presence of hypophosphatemia. Similarly, there is a gradual increase in FGF23 that parallels the decline in kidney function in patients with CKD [56], [59]. Thus, these data are consistent with the hypothesis that GFR per se is a strong regulator of FGF23, independent of serum inorganic phosphorous level or dietary phosphate intake.

Inflammation

In humans reduced renal function is commonly associated with low-grade persistent inflammation [60]. However, in the present study circulating levels of the inflammatory biomarkers CRP and PTX3 were markedly low both during hibernation and the active period (Table 1), which indicate that despite reduced renal function and prolonged inactivity bears do not develop systemic inflammation during hibernation. This may be due to a documented increase in antioxidant levels during hibernation, especially increases in vitamin C levels [61]. The high levels of vitamin C may act as a mitochondrial antioxidant and counter the effects of uric acid, thereby aiding the oxidation of fat [27], [33]. It could also be speculated that high intake of resveratrol via Vaccinum berries may contribute to the antioxidative milieu in bears [62]. We observed markedly higher levels of taurine of bears, which also may contribute to their overall anti-oxidant defense [63]. Notably, taurine depletion is a common feature of the pro-oxidant human uremic milieu [9]. Other possibilities include the effect of low body temperature in blocking the immune system [64], changes in plasma bile acid composition, such as ursodeoxycholic acid [65] known to have anti-inflammatory effects [66], or changes in mitochondrial function and the regulation of apoptosis [67], [68].

Strengths and limitations

The present study is, to our knowledge, the largest study to date on free-ranging bears. Sixteen bears were each studied twice while living in their normal environment and fed ad libitum their natural diet. Several coordinated metabolic pathways and markers were studied simultaneously (carbohydrate, lipid and nitrogen metabolism, calcium-phosphate balance and inflammation). The small number of sampled bears limited most previous studies and the majority was conducted under laboratory or enclosure conditions. Such studies may not mimic the natural circannual fattening cycle and eating habits as well as the natural winter hibernation profiles. However, some limitations remain. First, the analytic methods were developed for human assays and the accuracy for bear samples could not be unequivocally determined. Secondly, due to limited amount of serum/plasma, some analyses could not be performed in all bears. Three proteinogenic amino acids are missing in our analyses and we did not measure, hematocrit, plasma sodium and osmolality (which could have given information about the hydration status of the animals). It should also be acknowledged that as sampling was performed in February (or March) and June, respectively, our data are not representative of late hibernation or late summer-autumn conditions. Although body weight did not differ between hibernation and the active summer period significant differences in body composition would probably be observed if bears were instead captured at the end of the hibernation period (i.e. leanest) and late autumn (i.e. fattest). As the bear diet changes in the late summer when berries are abundant and start to ripe, it is possible that dietary changes may lead to a different metabolic milieu. Indeed, a previous study indicated that urea levels decline by about 50% between summer and autumn samples [19], which indicates that changes in the bear diet may reduce the intensity of nitrogen catabolism even before the fasting period.