1. Biochemistry and Chemical Biology
  2. Ecology

The energy savings-oxidative cost trade-off for migratory birds during endurance flight

  1. Scott McWilliams  Is a corresponding author
  2. Barbara Pierce
  3. Andrea Wittenzellner
  4. Lillie Langlois
  5. Sophia Engel
  6. John R Speakman
  7. Olivia Fatica
  8. Kristen DeMoranville
  9. Wolfgang Goymann
  10. Lisa Trost
  11. Amadeusz Bryla
  12. Maciej Dzialo
  13. Edyta Sadowska
  14. Ulf Bauchinger
  1. Department of Natural Resources Science, University of Rhode Island, United States
  2. Department of Biology, Sacred Heart University, United States
  3. Max Planck Institute for Ornithology, Germany
  4. Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, China
  5. Institute of Biological and Environmental Sciences, University of Aberdeen, United Kingdom
  6. Institute of Environmental Sciences, Jagiellonian University, Poland
  7. Nencki Institute of Experimental Biology PAS, Poland
https://doi.org/10.7554/eLife.60626
Share this article
Cite this article
  1. Scott McWilliams
  2. Barbara Pierce
  3. Andrea Wittenzellner
  4. Lillie Langlois
  5. Sophia Engel
  6. John R Speakman
  7. Olivia Fatica
  8. Kristen DeMoranville
  9. Wolfgang Goymann
  10. Lisa Trost
  11. Amadeusz Bryla
  12. Maciej Dzialo
  13. Edyta Sadowska
  14. Ulf Bauchinger
(2020)
The energy savings-oxidative cost trade-off for migratory birds during endurance flight
eLife 9:e60626.
https://doi.org/10.7554/eLife.60626
  2. Download BibTeX
  3. Download .RIS

Abstract

Elite human and animal athletes must acquire the fuels necessary for extreme feats, but also contend with the oxidative damage associated with peak metabolic performance. Here, we show that a migratory bird with fuel stores composed of more omega-6 polyunsaturated fats (PUFA) expended 11% less energy during long-duration (6 hr) flights with no change in oxidative costs; however, this short-term energy savings came at the long-term cost of higher oxidative damage in the omega-6 PUFA-fed birds. Given that fatty acids are primary fuels, key signaling molecules, the building blocks of cell membranes, and that oxidative damage has long-term consequences for health and ageing, the energy savings-oxidative cost trade-off demonstrated here may be fundamentally important for a wide diversity of organisms on earth.

Introduction

Endurance exercise challenges the physiology of athletes because high metabolism must be fueled and sustained while avoiding the build-up of metabolites that cause oxidative stress and fatigue. Given that flying is very energetically costly compared to running or swimming (Butler, 2016), such a physiological trade-off may be especially acute for migrating birds, among the best high-performance endurance athletes on the planet. During long-duration flights, birds expend energy at a rate >10 times above their basal metabolic rate (BMR) (Butler, 2016), whereas among the most extreme and competent human endurance athletes - Tour de France riders - have sustained metabolic rates of around five times BMR (Hammond and Diamond, 1997). Furthermore, birds use fats as their primary fuel (about 95%) for high-intensity endurance exercise such as migratory flights (Jenni and Jenni-Eiermann, 1998; Guglielmo, 2010; Guglielmo, 2018), and these fats are highly susceptible to oxidative damage (Skrip and McWilliams, 2016). Regulating oxidative balance is important for all air-breathing organisms because reactive pro-oxidant molecules can cause considerable cellular damage and so affect performance, health and potentially longevity (Halliwell and Gutteridge, 1999; Lane, 2005; Cooper-Mullin and McWilliams, 2016). Here, we experimentally demonstrate that the fatty acid composition of fat stores in a migratory bird and especially the acquisition of a few essential fatty acids, can reduce the energy cost of endurance flight; however, such beneficial energy savings comes at the cost of longer term oxidative damage. This energy savings-oxidative cost trade-off has important implications for the ecology and physiology of birds, as well as potentially for other, non-avian, athletes.

Results and discussion

One of the more remarkable features of fat metabolism in vertebrates is that in general ‘you are what you eat’, that is the fatty acid composition of diet has a predominant influence on that of the body-fat stores (West and Meng, 1968; Thomas and George, 1975; West and Peyton, 1980; Phetteplace and Watkins, 1989; Pierce et al., 2004; Pierce and McWilliams, 2005; Price and Guglielmo, 2009; Ben-Hamo et al., 2011; Abbott et al., 2012; Pierce and McWilliams, 2014). We took advantage of this feature of fat metabolism and used diet manipulations (see Table 1, Table 2) to produce European starlings (Sturnus vulgaris) with distinct differences in certain essential fatty acids (Figure 1).

Figure 1
Download asset Open asset
Fatty acid composition of stored fat in European starlings was largely determined by their diet.

Hand-raised European starlings were fed over 4+ months one of two isocaloric diets (MUFA or PUFA) that differed only in the relative amounts of mono- and polyunsaturated fats (Table 1, Table 2), specifically the amounts of omega-9 (18:1), and the so-called ‘essential’ omega-6 (18:2) and omega-3 (18:3) (no. carbons in fatty acid backbone: no. double bonds, and the ‘omega’ designation identifies the location of the first double bond from the terminal end). The stored fat (in the furcular region) of MUFA-fed birds was 75% 18:1% and 10% 18:2, whereas that of PUFA-fed birds was 60% 18:1% and 20% 18:2 and 18:3. Importantly, these three fatty acids primarily composing the fat stores of our hand-raised starlings (i.e. 16:0, 18:1, 18:2) are the same three fatty acids that predominate in the fat stores of free-living passerines sampled during their migration (Pierce and McWilliams, 2005).

Table 1
Ingredients and composition of the two semi-synthetic diets fed to European starlings used in Experiments I and II.

The two diets were isocaloric and composed of 42% carbohydrates, 23% protein, and 20% fat. Different amounts of soybean and olive oil were used to produce two diets that differed only in their fatty acid composition (see Table 2).

MUFAPUFA
Ingredients% wet mass% dry mass% wet mass% dry mass
Glucose*16.8739.3516.8739.19
Casein†8.2319.208.2319.12
Cellulose‡2.144.992.144.97
Salt mixture§2.064.802.064.78
Olive oil¶7.8218.244.119.60
Soybean oil**0.410.964.119.60
Amino acid mix††1.152.691.152.68
Vitamin mix‡‡0.160.380.160.38
Mealworms§§2.656.192.656.16
Agar¶¶1.373.201.373.19
Water57.1457.14

  1. *Glucose, VWR International GmbH, Darmstadt, Germany;.

    †Casein, Affymetrix UK Ltd., High Wycombe, UK.

  2. ‡Alphacel, MP Biomedicals, Solon, OH, USA.

    §Brigg’s salt mix, MP Biomedicals, Solon, OH, USA.

  3. ¶Tip Native brand Olive oil (glass bottle, Vandemoortele Deutschland GmbH).

    **Soya oil, Sojola-brand Soja Oil; Vandemoortele Deutschland GmbH.

  4. ††Amino Acid Mix, Sigma-Aldrich, St. Louis, MO, USA.

    ‡‡AIN-76 vitamin mix, MP Biomedicals, Solon, OH, USA.

  5. §§Freeze-dried mealworms: Futtermittel Hungenberg Brand, Germany; ca. 45% protein, 33% fats, 7% carbohydrates, and 15% indigestible fiber and ash (Finke, 2002).

    ¶¶Agar, Ombilab-laborenzentrum GmbH, Bremen, Germany.

Table 2
Fatty acid composition (% ± SE) of the monounsaturated (MUFA) and polyunsaturated (PUFA) fatty acid diets plus mealworms, and of furcular fat from European starlings fed each diet for 4+ months.

Fatty acid concentration was directly measured by gas chromatography in lipids extracted from the diets and the furcular fat.

MUFAPUFA
Fatty acid*Diet w/mealwormsFurcular fat
(n = 13)		Diet w/mealwormsFurcular fat
(n = 16)
16:012.5216.61 ± 0.8112.7316.63 ± 0.65
16:10.003.34 ± 0.340.003.35 ± 0.32
18:01.382.69 ± 0.741.813.74 ± 0.56
18:1†70.6567.81 ± 1.1851.7056.73 ± 0.90
18:2†14.659.54 ± 1.3229.3417.70 ± 1.17
18:3†0.500.00 ± 0.004.111.75 ± 0.34

  1. *Fatty acid nomenclature = C:D where C refers to the number of carbon atoms in the chain and D refers to the number of double bonds present. Other fatty acids found in <1% of the lipid portions of diet were 12:0, 14:0, 20:1, 22:6, 24:1.

    †Fatty acid composition (% ± SE) of furcular fat from birds fed MUFA diets was significantly different from that of birds fed PUFA diets, in 18:1, 18:2, and 18:3 fatty acids. .

Such differences in fatty acid composition of fat stores in starlings, specifically the relative amounts of 18:1, 18:2, and 18:3, are also among the primary longer chain fatty acids that compose the fat stores of wild songbirds especially during migration (Blem, 1990; Pierce and McWilliams, 2005; Pierce and McWilliams, 2014). We used these two groups of starlings with different fatty acid composition of their fat stores to directly test the hypothesis that birds with more essential omega-6 and −3 PUFA (18:2 and 18:3) comprising their fat depot have enhanced exercise performance during long-duration flights, as previously shown for short-duration flights of a few minutes (Pierce et al., 2005; Price and Guglielmo, 2009). After 15 days of ramping up flight-training, we flew starlings (n = 33) for 6 hr (±5 min) in a windtunnel set at a fixed speed of 12–12.5 m/s, the equivalent of a ca. 260 km non-stop flight, and used doubly labeled water to measure energy expended during the 6 hr flight (see Materials and methods for details).

PUFA-fed birds composed of more essential omega-6 (18:2) and omega-3 (18:3) PUFA expended 11% less energy compared to MUFA-fed birds during the 6 hr endurance flight (Figure 2a; Diet effect: F1,28 = 8.88, p=0.006; Body mass covariate: F1,28 = 13.75, p=0.001; Diet × Body mass interaction: F1,28 = 0.25, p=0.62). As expected from the 11% energy savings, PUFA-fed birds lost substantially less body mass during the 6 hr flight (7.04 ± 0.30%) compared to MUFA-fed bird (9.01 ± 0.41%; Figure 2—figure supplement 1). We replicated this experiment with a second group of hand-raised starlings (n = 36) fed the same PUFA or MUFA diets, trained in the same way in the same wind-tunnel, to measure basal metabolic rate (BMR) and oxidative status of flight-trained as well as control, sedentary birds at different time points during exercise training (see Materials and methods, Experiment II). Exercise training, in general, and the transition to the migratory-state for birds, specifically, might result in upregulation of many fundamental aspects of physiology (McArdle et al., 2010; Guglielmo, 2010; Piersma and Gils, 2011) which may require an increase in BMR. However, we found that BMR measured 36–40 hr after the longest flight was similar for MUFA- and PUFA-fed birds (Figure 2b) and was not influenced by flight-training (Training x Diet interaction: F1,25 = 0.04, p=0.85; Training effect: F1,25 = 0.02, p=0.88; Diet effect: F1,25 = 0.61, p=0.44).

Figure 2 with 1 supplement see all
Download asset Open asset
Dietary fatty acid manipulation affects flight costs but not costs of self-maintenance.

Box plots (mean, 5% and 95% CI, range) for (a) Energy expenditure (measured with doubly labeled water) during a 6 hr ca. 260 km flight in a windtunnel for starlings (n = 33) fed one of two diets (n = 16, 17) and after 15 days of flight-training. Body mass was used as a covariate in the ANCOVA comparison of diet effects on energy expenditure, and (b) Basal metabolic rate (BMR; measured using open-flow respirometry) of MUFA- (n = 19) and PUFA-fed (n = 17) European starlings that were flight-trained or sedentary. Means with different letters within each panel are significantly different (p<0.05).

The 11% energy savings (Figure 2a) achieved by starlings composed of more essential omega-6 PUFA during long-duration flights is quite substantial compared to other performance-enhancing contexts for other organisms. For example, mutant strains of E. coli bacteria that did not incur the biosynthetic costs of the amino acid tryptophan used 0.01% less energy than wild types (Dykhuizen, 1978). Naked mole rats which inhabit the dark world of borrows saved 2% of their energy budgets by not developing a visual system (Cooper et al., 1993). Differences in the energy costs of the top-placing elite athletes such as Tour de France riders (Saris et al., 1989; Hammond and Diamond, 1997; Santalla et al., 2012) are on the order of <5%. Thus, avian migratory performance is not only extreme in terms of the amount of energy burned per unit time (Butler, 2016), but also in the potential savings of energy during long-duration exercise associated with having their fat stores composed of more essential omega-6 (18:2) and omega-3 (18:3) PUFA.

In theory, selectively eating and hence storing certain long-chain unsaturated fatty acids may be advantageous because such fatty acids (i) may affect composition and key functions of lipid-rich cell membranes (membrane hypothesis), (ii) may be metabolized more quickly (fuel hypothesis), or (iii) may stimulate key facets of aerobic metabolism (signal hypothesis) (Price, 2010; Pierce and McWilliams, 2014). In rats and humans, high levels of essential omega-6 PUFA in muscle membrane phospholipids have been associated with improved endurance capacity (Ayre and Hulbert, 1996; Ayre and Hulbert, 1997; Andersson et al., 1998). Maximum running speed, a short-term performance measure, in 30 species of mammals was positively correlated with the omega-6 fatty acid content in skeletal muscle phospholipids (Arnold et al., 2015). Our own recent work and that of colleagues confirmed that two different passerines (Red-eyed vireos [Vireo olivaceous], White-throated sparrows [Zonotrichia albicollis]) with fat stores composed of more omega-6 PUFA (18:2) have improved performance during short-term intense exercise (Pierce and McWilliams, 2005; Pierce et al., 2005; Price, 2010). Yellow-rumped warblers (Setophaga coronata) with enriched omega-3 PUFA in their muscle phospholipids decreased muscle oxidative enzymes, although these changes in muscle metabolism were not associated with changes in flight performance (Dick and Guglielmo, 2019a). Unlike most of these studies that were largely mensurative or measured exercise performance over shorter flights (mostly 20–30 min), our experimental results here demonstrate that migratory birds with fat stores composed of more essential omega-6 PUFA had improved performance (i.e. less energy used per km) during endurance flights (6 hr, ca. 260 km) compared to birds composed of more MUFA, and a recent companion study (Carter et al., 2020) suggests that this is because of the signaling properties of omega-6 PUFA.

We also tested the hypothesis that this enhanced performance comes with metabolic costs. Such enhanced endurance performance is associated with more lipid peroxide production which must be quenched by upregulation of the endogenous antioxidant system or, if not adequately quenched, would increase oxidative damage. We measured plasma indicators of antioxidant status in flight-trained as well as sedentary (untrained) starlings at three different time points during training: before the start of flight training in the windtunnel (‘Pre-training’), immediately after a long-duration flight on Day 15 (‘Post-flight’), and 1.5 days afterwards (‘Recovery’).

Flight-training over more than 2 weeks did not affect baseline levels of oxidative damage (compare Pre-training and Recovery; Figure 3c) while antioxidant capacity decreased in flight-trained but not sedentary birds (Figure 3b) and plasma uric acid decreased over time in both trained and sedentary starlings (Figure 3a; see Table 3 for detailed statistical results). A long-duration flight on Day 15 (average flight times for PUFA: 161.6 ± 31.2 min, and MUFA: 180.2 ± 28.3 min) was associated with increased plasma uric acid (Figure 3a), a product of protein metabolism in birds and also a potent antioxidant (Sautin and Johnson, 2008), and a reduction in plasma triglycerides (Figure 3—figure supplement 1). Regardless of flight-training, birds fed more essential omega-6 PUFA had higher oxidative damage than MUFA-fed birds (main effect of flight; Figure 3c) suggesting a fundamental oxidative cost of being composed of more polyunsaturated fatty acids.

Figure 3 with 1 supplement see all
Download asset Open asset
Oxidative status of European starlings associated with flight-training and in relation to diet quality.

Oxidative status (see Materials and methods, Experiment II) of MUFA-fed or PUFA-fed European starlings was measured in blood plasma at three different time points for flight-trained (black symbols and dashed lines) and untrained, sedentary control (gray symbols and lines) birds: before the start of flight training in the windtunnel (Pre-training), immediately after a long-duration flight on Day 15 (Post-flight), and ca. 1.5 days afterwards (Recovery). Untrained sedentary birds were sampled on the same days but were never exposed to flight-training. Body mass and date of measurement were included as a fixed covariate so we report the results as least square means (LSM). The comparison of Recovery and Post-flight timepoints reveals the effect of the long-duration flight: birds had very similar total flight times over the 15 days of exercise training with the primary difference being whether we sampled the bird’s blood immediately after flight (Post-flight) or 2 days after their final flight (Recovery). The main effect of diet (MUFA vs. PUFA) on oxidative status is shown in the right panels. Means with different letters across the three timepoints, or for the main effect of diet, are significantly different (p<0.05).

Table 3
The effect of flight (flight-trained for 15 days in windtunnel or not; Trained or Sedentary), diet (MUFA or PUFA), and time (blood sampled at three different time points: before the start of flight training in the windtunnel (‘Pre-training’), immediately after a long-duration flight on Day 15 (‘Post-flight’), and 1.5 days afterwards (‘Recovery’)) on plasma metabolites and oxidative status in European starlings in Experiment II. Individuals that did not undergo flight training (i.e. control ‘sedentary’ birds) were sampled on the same days as flight-trained birds in their same cohort. Test statistics: F-value with denominator degrees of freedom (ddf) and significance level p-value for main factors and their interactions from the linear mixed models.
MarkerFlightDietTime pointFlight × DietTime × FlightTime × DietDiet × Flight × Time
F ddfpF ddfpF ddfpF ddfpF ddfpF ddfpF ddfp
β-Hydroxybutyrate4.72 32.90.0414.0 32.9<0.00114.0 62.3<0.0010.26 32.90.619.44 62.3<0.0010.52 62.30.600.32 62.30.73
Total triglicerydes0.32 32.60.570.13 32.60.729.62 60.0<0.0010.22 32.60.886.90 60.00.0020.40 60.00.670.26 60.00.97
Uric acid9.12 33.50.0040.65 33.50.4325.94 63.5<0.0011.44 33.50.244.32 62.50.020.34 62.50.710.4262.50.66
Antioxidant capacity
(Oxy adsorbent assay)		1.05 31.50.311.72 31.50.206.00 58.90.0041.9 31.50.184.34 58.90.170.90 58.80.410.36 58.80.70
Oxidative damage
(dROM assay)		0.26 33.10.615.71 33.10.020.27 62.00.771.86 33.10.181.2 62.00.311.50 62.00.230.27 62.00.80

Increased energy metabolism during exercise is often associated with increased production of pro-oxidants regardless of the fuel types used (i.e. carbohydrates, protein, fats) which causes oxidative damage if not quickly quenched by dietary antioxidants and/or by increased production of antioxidant enzymes (e.g. superoxide dismutase, glutathione peroxidase) (Halliwell and Gutteridge, 1999 Costantini et al., 2007; Costantini, 2008; Costantini et al., 2008; Monaghan et al., 2009; Jenni-Eiermann et al., 2014; Dick and Guglielmo, 2019b). Animals performing exercise such as birds during a long-duration flight, and those that rely on large amounts of fat as fuel such as migratory birds, are especially prone to oxidative stress because (1) an increase in activity may increase production of ROS in excess of the immediate capacity of antioxidants to quench them, and (2) stored and structural fats, particularly PUFAs, are especially vulnerable to oxidative damage because of their chemical structure, and may generate their own variety of pro-oxidants (Cooper-Mullin and McWilliams, 2016; Skrip and McWilliams, 2016). Importantly, we found no significant change in oxidative damage or oxidative capacity immediately after flights (compare Post-flight to Pre-training and Recovery in Figure 3) suggesting that birds were capable of contending with the oxidative costs associated with a given flight. Instead, migratory birds pay the oxidative costs of being composed of more polyunsaturated fatty acids (primarily 18:2 and 18:3) over the long-term (e.g. migration period of the annual cycle) while gaining some energy savings only during a given migratory flight.

Collectively, our study provides compelling evidence that avian athletes – just like human athletes (Mickleborough, 2013; Neubauer and Yfanti, 2015; Pingitore et al., 2015) – face considerable trade-offs when deciding what to eat to enhance their performance. Given that fat composition of diet largely determines that of fat stores, migratory birds that consume diets with more essential omega-6 PUFA (18:2) can substantially enhance their exercise performance (i.e. expend less energy) during long-duration flights without incurring increased self-maintenance (BMR) costs. However, this enhanced performance during flights has longer-term costs in terms of increased oxidative damage when fat stores are composed of more PUFA. Cafeteria-style diet choice experiments with migratory songbirds have shown that birds carefully discriminate between diets that differ only in their fatty acid composition (Pierce et al., 2004; Pierce and McWilliams, 2014), and on average prefer to consume a composite diet that contains a 1:2 ratio of 18:2 to 18:1 (Pierce and McWilliams, 2014). This indicates that birds may choose among diets to optimize the trade-off between enhanced flight performance (more 18:2), while reducing the long-term costs of being composed of more long-chain PUFA. Migratory birds can also optimize this energy savings-oxidative cost trade-off by being composed of more n-3 and/or n-6 PUFAs only during migration periods when energy demands and fat catabolism are most extreme, and then become more monounsaturated in composition during non-migration periods - such seasonal changes in fatty acid composition are commonly observed in migratory birds (Pierce and McWilliams, 2005; Santalla et al., 2012; Pierce and McWilliams, 2014). Such a trade-off may become especially detrimental if foods with different quantities of micronutrients (notably long-chain PUFAs and antioxidants) are not available in nature, in which case birds may be unable to ameliorate such a trade-off through careful choices of diet. Given the long-term consequences of oxidative damage for health and aging (Halliwell and Gutteridge, 1999), and the enhanced exercise performance associated with certain fatty acids that incur long-term oxidative costs (i.e. the energy savings-oxidative cost trade-off demonstrated here), strong evolutionary forces likely act on these diet choices for human and non-human vertebrates especially those that at times require enhanced exercise performance.

Materials and methods

Study species and overall experimental plan

Request a detailed protocol

We used hand-raised European starlings (Sturnus vulgaris) from southern Germany (n = 95 total) for this two-part study because starlings are common migratory birds in western Europe (Feare, 1984) and they have been successfully hand-raised and trained to fly in windtunnels (Engel et al., 2006a). European starlings from this southeastern German population are medium-distance diurnal migrants that leave for wintering grounds in October and November and return to their Bavarian breeding grounds in April (Bairlein, 2014). Migratory distances for this population vary from many hundreds of kilometers to several thousand kilometers, with some individuals overwintering in the Euro-Mediterranean region and others in northwest Africa (Bairlein, 2014). European starlings are also quite social and curious, and quickly learn to successfully fly together in a given windtunnel as demonstrated by several recent studies (Carter et al., 2020; Casagrande et al., 2020). We conducted two complimentary experiments that involved feeding starlings over many months two diets that differed only in the relative amounts of mono- and polyunsaturated fats (Table 1), specifically the amounts of omega-9 18:1 and omega-6 (18:2) and omega-3 (18:3) (no. carbons in fatty acid backbone: no. double bonds). Previous work has established that these are the primary longer chain fatty acids in wild songbirds especially during migration (Blem, 1990; Pierce and McWilliams, 2005; Pierce and McWilliams, 2014), and that the fatty acid composition of birds reflects that of their diet (Pierce and McWilliams, 2005; Pierce et al., 2005; Price and Guglielmo, 2009) which we also demonstrate here (Table 2).

All procedures adhered to the ethical guidelines of the North American Ornithological Council (Fair et al., 2010) and were approved by the University of Rhode Island IACUC (Protocols #AN09-09-009, AN08-02-014) and the Government of Upper Bavaria, Germany (AZ 55.2-1-54-2532-216-2014).

Starling care, aviaries, and experimental diets

Capture and maintenance of birds

Request a detailed protocol

We collected 40–45 nestling (5–11 day-old) European starlings from nest boxes in late-April to early-May 2005 and 2015 from a colony in Upper Bavaria, South Germany (47°58' N, 11°13' E). We hand-raised the chicks at the Max Planck Institute for Ornithology (MPIO), Seewiesen, Germany (Figure 4) on a high-protein diet of bee larvae, crickets, and beef heart with vitamin mixture until they were able to feed independently (ca. 35 days old). During this hand-raising, small groups (2-5 ) of hatchlings of similar age were housed in wooden boxes (25 × 25 × 25 cm high) lined with paper towels that were changed after every feeding. This same diet has been successfully used to raise starlings for windtunnel experiments (Engel et al., 2006b; Schmidt-Wellenburg et al., 2008) and growth rates, molt, and size at fledging for our starlings were similar to previous studies (Engel et al., 2006a). At ca. 35 days old, fledglings were able to feed independently and so they were housed in groups of 10–15 individuals per aviary (1.5 × 2 × 2.5 m) and offered ad libitum live mealworms, fresh fruits, and vegetables. At the age of ca. 50 days, groups of 10–15 starlings were then moved to larger aviaries (2 × 4 × 2 m) and offered fresh fruit and vegetables, and a mixed diet of beef heart, yogurt, rusk flour, eggs, mealworms, and other insects (e.g. crickets, bee larvae, wax worms, green bottle fly larvae), soy oil, nuts and dried fruits, and vitamins and minerals.

Figure 4
Download asset Open asset
Experimental design for both Experiments I and II from hand-raising of nestling European starlings, to acclimation to one of two experimental diets (both composed of 42% carbohydrates, 23% protein, and 20% fat but differing in the amount of polyunsaturated (PUFA) or monounsaturated (MUFA) fatty acids), and then to flight training in a windtunnel.

During fall, cohort groups of 2–3 starlings were flight trained in the windtunnel for 14 days and then flew on Day 15 a long-duration (usually 6 hr) flight (see Figure 5) during which energy expenditure and plasma indicators of metabolism and oxidative status were measured. For Experiment I, we flight-trained 36 starlings of which 33 completed their 6 hr long-duration flight (16 fed the MUFA diet, 17 fed the PUFA diet). For Experiment II, we added 1–2 untrained, sedentary (control) starlings in each training cohort which resulted in a total of 19 MUFA-fed birds (11 flight-trained, eight sedentary) and 17 PUFA-fed birds (nine flight-trained, eight sedentary).

We randomly assigned 70–75 day old starlings to one of two semi-synthetic agar-based diets (Murphy and King, 1982) that were isocaloric and composed of 42% carbohydrates, 23% protein, and 20% fat (Table 1). The two diets (hereafter, MUFA, PUFA) were identical in terms of macronutrient composition and differed only in the relative amounts of certain mono- and polyunsaturated fats (Table 2). The macronutrient composition of the two semi-synthetic diets simulates a natural high-lipid fruit diet (Johnson et al., 1985; Smith et al., 2007a), and the primary fatty acids in the diet (>90% 16:0, 18:1, and 18:2) are also the most common fatty acids in natural fruits (Pierce and McWilliams, 2014) and in songbirds that eat fruits during migration (Pierce and McWilliams, 2005; Pierce and McWilliams, 2014). European starlings are like many fall-migrating songbirds in that they switch to eating largely fruits (Feare, 1984), much to the chagrin of many vineyard owners. For Experiment I, the random assignment resulted in roughly equal numbers of males and females fed each of the two diets, whereas for Experiment II we randomly assigned only males to the two diets because the females were used for a different experiment.

Light cycles

Request a detailed protocol

The light schedule from May to December 2005 and 2016 was the same as natural day:night photoperiod at the latitude of Seewiesen, Germany. Large windows ensured exposure to natural light levels, although there were additional supplemental lights inside the aviaries on the same light cycle (bulbs were Osram LUMILUX T8 58 W/865). We did not directly verify that such decreases in light levels in fall induced starlings to increase food intake, although many other studies provide such evidence in migratory birds (Gwinner, 1996; Helm et al., 2009; Bulte and Bairlein, 2013), or to increase Zugunruhe (nocturnal activity), since starlings are diurnal migrants.

Flight training schedule

Request a detailed protocol

After starlings were fully acclimated to their respective diets (1+ months), we then flew starlings for prescribed amounts of time each day for 15 days in a recirculating windtunnel designed for studying bird flight at the Max Planck Institute for Ornithology (MPIO), Seewiesen, Germany (Schmidt-Wellenburg et al., 2007, Figure 5). During the flight training period during fall (Sept-Dec), every 3 days we randomly selected two to threeindividuals from a given diet group to start the ca. 2-week flight training, although this flight-group selection was stratified by extent of Pre-basic I molt (Ginn and Melville, 1983) - birds furthest along in molt were flight-trained earliest. This stratified random sampling of individuals ensured that all birds had completed their flight feather molt prior to flight training (i.e. molt scores were >70 in all cases based on Ginn and Mellville [Ginn and Melville, 1983]) and the 20 groups of starlings completed their flight training by early December. For Experiment II, we also randomly selected a subset of starlings fed each diet that were not exposed to the flight-training - hereafter, the untrained, sedentary ‘control’ group. While flight-trained birds were being trained each day to fly in the windtunnel, their companion untrained, sedentary control birds remained in their aviaries adjacent to the windtunnel. On Day 15, while the trained birds were flying for up to 6 hr, the sedentary birds were kept in individual cloth cages without access to food and water.

Figure 5
Download asset Open asset
Fifteen-day flight-training schedule used for European starlings that were flown in the Max Planck Institute for Ornithology (MPIO) windtunnel.

(a) Amount of flying time each day was increased until the final 6 hr flight on Day 15; BMR was measured overnight on the day after this longest flight (b) Schematic of the MPIO windtunnel showing the 2 × 1.5 m working section (closest to the person) that the birds were trained to fly into through the gap (which was then closed during flights). A large fan, shown directly across from the working section, created the wind which was made laminar with a series of screens just before the compression of the tunnel (bottom left). The wind velocity was accelerated with the 20:1 compression of the tunnel just before the working section.

The flight-training schedule and the duration of the longest (6 hr) flight were chosen based on logistics as well as ecological relevance. The 15-day flight-training schedule and flying conditions (always 12 m s−1 wind speed and 15°C) had been used in previously successful experiments at MPIO designed to fly barn swallows (Hirundo rustica) and starlings (Sturnus vulgaris, Sturnus roseus) for long durations (Engel et al., 2006a; Engel et al., 2006b; Engel et al., 2006c; Schmidt-Wellenburg et al., 2008; Schmidt-Wellenburg et al., 2007), and free-living songbirds including starlings typically complete their migration from breeding to wintering areas over many days of flying and stopovers (Feare, 1984; Newton, 2006). The duration of the final longest flight (6 hr, 260 km) was sufficiently long to provide adequate turnover of the isotope-labeled water and so ensure accurate measurements of energy expenditure using the doubly labeled water (DLW) technique (Speakman, 1997). The 6 hr duration of flight is also within the range of typical single-day migratory flights for many free-living songbirds including starlings (Feare, 1984; Newton, 2006). During flight-training, each group of two to three individuals was transferred to smaller aviaries (2 × 0.7 × 2 m) that surrounded the working section of the windtunnel (2 × 1.5 m octagon). During the four pre-training days before the onset of flight training (=day 1), all birds were habituated to the windtunnel; each day they spent up to 30 min perched and taking short flights in the wind-tunnel. We used a specifically experienced ‘trainer-bird’ during this initial 4-day pre-training period in order to provide guidance to the inexperienced birds. Pre-training time was not included in the total flight time. For Experiment I, flight-training for each group of three birds involved 15 consecutive days of specified amounts of flying in the windtunnel set to 12 ms−1 wind speed and 15°C (day 1 flight time = 10 min, day 2 = 10 min, day 3 = 20 min, day 4 = 30 min, day 5 = 30 min, day 6 = 90 min, day 7 = 60 min, day 8 = 90 min, day 9 = 30 min, day 10 = 120 min, day 11 = 180 min, day 12 = no flight training, day 13 = 60 min, day 14 = 30 min, day 15 = 360 min (Figure 5)). The same flight training schedule was used for Experiment II except for logistical reasons birds flew for 20 min on days 1 and 2, for 60 and 30 min on days 5 and 6, respectively, and we added a rest day (no flight training) on day 9. This resulted in a total flight-training time, excluding the longest flight on day 15, of 715 min and 720 min for Experiments I and II, respectively. For each bird, we recorded actual flight time each day and summed the total time spent flying for the entire 15-day period.

Doubly labeled water to measure energy expenditure during long-duration flight

Request a detailed protocol

We used the DLW technique (Speakman, 1997) in Experiment I to test the hypothesis that energy expenditure of birds during the long-duration (6 hr) flight was affected by the fatty acid composition of their fat stores. Food was removed from aviaries at lights off on Day 14. On Day 15, the three birds in a given flight-training cohort were captured from their small aviaries 15 min after lights on. At 5 min intervals thereafter, we measured body mass of each of the three birds and then intraperitoneally injected 0.2 µL DLW (BD 0.3 ml Micro-Fine U-100 Sterile Insulin syringe). Amount injected was estimated by differential weighing of the syringe to the nearest 0.0001 g using an analytical Sartorious balance. Following injection birds were kept in a small cloth cage with a perch but no water or food. Precisely 60 min after injection, we collected two 40 µL blood samples into non-heparinized capillary tubes from the jugular vein or after puncture of the brachial vein with a 27G needle. All blood samples in capillary tubes were flame sealed immediately upon collection and stored at room temperature until analysis. Birds were allowed to rest in the windtunnel aviaries for about 5 min before starting their 6 hr flight in the windtunnel set to 12 m s−1 wind speed and 15°C. Exactly 6 hr later, and in the same order of injections, birds were removed from the windtunnel mid-flight and bled again. We collected two 40 µL blood samples into heparinized capillary tubes from the brachial vein after puncture with a 17G needle, plus an additional 160 µL blood sample transferred to 0.5 mL Eppendorf tubes and centrifuged at 10,000 RPM. Separated plasma was stored at −80°C until subsequent plasma metabolite measures.

Daily energy expenditure (DEE in watts, J s−1) was measured using the DLW technique (Lifson, 1966; Butler et al., 2004). This method has been previously validated by comparison to indirect calorimetry in a range of small mammals (Speakman and Król, 2005). To estimate the background isotope enrichments of 2H and 18O, uninjected birds were weighed (±0.1 g Kern pocket balance) and a 20 µL blood sample was obtained from the jugular vein (Speakman and Racey, 1987). Blood samples were immediately heat sealed into 2 × 75 µL glass capillaries, which were stored at room temperature. Analysis of the isotopic enrichment of all blood samples was performed blind, using a Liquid Isotope Water Analyser (Los Gatos Research, USA) (Berman et al., 2012). Initially, the blood encapsulated in capillaries was vacuum distilled (Nagy, 1983), and the resulting distillate was used for analysis. Samples were run alongside five lab standards for each isotope and International standards to correct delta values to ppm. A single-pool model was used to calculate rates of CO2 production as recommended for use in animals less than 5 kg in body mass (Speakman, 1993). There are several approaches for the treatment of evaporative water loss in the calculation (Visser and Schekkerman, 1999). We detected no differences in body water pool before and after 6 hr flights. We assumed evaporation of 25% of the water flux (Equation 7.17 in Speakman, 1997), which minimizes error in a range of conditions (Visser and Schekkerman, 1999; van Trigt et al., 2001). Given that our experimental design involved feeding birds diets that were identical in macronutrient composition and differed only in their fatty acid composition, the dietary substrates (carbohydrates, fats, proteins) available as fuels were also identical. Empirically measured variation in the amount of stored fat or protein used as fuel (as reflected in the RQ) results in errors in estimates of energy expenditure not exceeding ±2% (Black et al., 1986; Schmidt-Wellenburg et al., 2008; Westerterp, 2017).

Basal metabolic rate

Request a detailed protocol

We used open-flow respirometry (Lighton, 2019) in Experiment II to test the hypothesis that BMR of flight-trained birds was affected by the fatty acid composition of their fat stores. We measured BMR during the night on Day 16 after the birds had recovered from their last and longest-duration flight ca. 1.5 days earlier. BMR was measured with a multi-channel open flow respiratory system. Starlings were weighed and placed in separate plastic respirometric chambers (800 ml) without water or food. All the chambers were additionally covered with dark non-transparent paper and were placed in a Peltier effect temperature-controlled portable cabinet (Sable system International, USA) at 25°C, within the thermoneutral temperature range for starlings (Lustick and Adams, 1977; Geluso and Hayes, 1999). Sufficient mixing of air in the chamber was achieved by plumbing the air intake in the lower part of the chamber and the outtake in the top part of the chamber. A fresh sample of intake air at standard pressure and room temperature was dried with silica gel driers and divided into two streams (flush and master flow) controlled by Intelligent Multiplexer V5 (Sable system International, USA). The flush air stream was pumped with a Mini Laboport Vacuum Pump, (KNF, Germany) at an air flow rate of 780 ml min−1 for each chamber (Brooks mass flow controller, USA). The master air stream was pumped (Vebreclerwerk, Germany) at an air flow rate of 1000 mL min−1 (Sierra Instruments mass flow controller, USA) with a mass flow controller V1.1 (Sable system International, USA).

Subsamples of air were collected from the chamber at rate of 150 mL min−1. Before passing the CO2 and O2 analyzers the subsample was pre-dried with PC-1 non-chemical drier (Sable system International, USA) and then chemically dried with a magnesium perchlorate (Anhydrone, J.T. Baker, USA) column. Samples were analyzed with a FOXBOX (Sable System International, USA) connected to a computer through UI3 (Sable system International, USA) and data were recorded continuously with ExpeData software (Sable System International, US). Dried samples of air were taken sequentially from the measurement chambers and the reference every 18 min. In each cycle, each measurement chamber was active for 110–120 s. We used data from the last 6 hr of measurement (ca. 4 hr after starting measurements) to be sure that metabolic rate was measured in the post-absorptive state. Rates of O2 consumption and CO2 production were calculated from the values recorded in the last 20 s before switching channels, and the mean of the two lowest recordings were used for further calculations. BMR in watts was calculated based on respirometric equations from Lighton, 2019. Because of failure of the O2 analyzer for 21 out of 53 available measurements, we calculated O2 consumption for all birds based on (a) measured CO2 values, and (b) the estimated average RQ of 0.789 for the 32 birds with complete measurements.

Blood sampling and analyses: antioxidant status and plasma metabolites

Request a detailed protocol

At each of the three time points (Pre-flight and Post-flight on Day 15) in both Experiments I and II and just before sacrifice on Day 17 in Experiment II, within 3 min of bird capture we drew ca. 200 µL blood from the brachial vein after puncture with a 27 G needle and collected it in heparinized capillary tubes. All birds were sampled at each time point while fasted for at least 12 hr (i.e. after an overnight without food and before offered food on the Pre-training and Recovery days, and after an overnight without food plus their longest flight on the Post-flight day). Within 10 min of sampling, blood samples were centrifuged at 214 g for 5 min to separate plasma from the RBCs. Plasma was stored at −80°C until lab analyses of antioxidant status (i.e. antioxidant capacity, oxidative damage) and plasma metabolites (beta-hydroxybutyrate, triglycerides, and uric acid) were conducted.

Non-enzymatic antioxidant capacity was measured with the OXY-adsorbent test (OXY) in the plasma diluted 1:100 dH2O (concentration unit = mmol L−1 of HClO neutralized; Diacron International, Grosseto, Italy). OXY directly measures the ability of a plasma sample to neutralize the oxidant hypochlorous acid and provides an index of non-enzymatic antioxidant capacity, without being complicated by the interaction of uric acid (Alan and McWilliams, 2013; Skrip and McWilliams, 2016; Costantini, 2016). Oxidative damage was measured with the d-ROMs test in plasma diluted 1:3 0.9%NaCl (concentration unit = mmol L−1 H2O2 equivalents; Diacron International). ROMs measured in this test are primarily hydroperoxides, which are produced when ROS interact with many different biological macromolecules (Costantini, 2016), but in plasma, are primarily produced during lipid oxidation events (Davies, 2016; Ito et al., 2017).

Total triglyceride (TRIG) is a marker of fatty acid anabolism and was measured by sequential endpoint assay (Sigma, St. Louis, Missouri; 5 μL plasma, 240 μL reagent A, 60 μL reagent B) by first measuring free glycerol and then subtracting free glycerol concentration from measured total triglyceride concentration. Beta-hydroxybutyrate is a marker of fatty acid catabolism and was measured by kinetic assay (Cayman Chemical Assay; 5 μL plasma, 50 μL of developer solution which was a mix of 2.4 mL enzyme solution and 100 μL colorimetric detector). Uric acid is a product of protein catabolism and was measured by endpoint assay (TECO Diagnostics, Anaheim, California; 5 μL, sample, 300 μL reagent) using a modified protocol from Smith et al., 2007b.

Fatty acid composition of diet and furcular fat

Request a detailed protocol

Fatty acid composition of the two semi-synthetic diets (Table 2) was measured by gas chromatography in lipids extracted using a modified Folch method (Carter et al., 2020) from composited subsamples of each diet taken over the course of the experiments. We collected adipose tissue from the visible fat stores in the furcular area of starlings used in Experiment I (n = 29) 5–7 days after their longest flight in the windtunnel. We conducted the biopsies of this furcular fat using the methods outlined in Rocha et al., 2016. Briefly, upper breast feathers were wetted and moved aside to expose the skin and visible subcutaneous yellow-colored fat stores. We selected an area without visible capillaries, disinfected the area with antiseptic solution, and applied a topical anaesthetic gel to this area. After ca. 10 min, we then pinched the skin in this area to ensure no pain response by the bird. We then made an ca. 3 mm-long incision in the skin using a sharp scalpel, pulled a small piece of adipose tissue (ca. 10–20 mg) through the incision using sterile forceps, and cut the sample under the forceps using sterile scissors. The incision area was then realigned and a thin layer of veterinary tissue glue was applied to seal the incision. All birds were checked weekly thereafter and the wound was completely healed within 2–3 weeks.

Dietary fat and furcular fat samples were stored at −80°C until fatty acid composition was measured by gas chromatography. We thawed samples, extracted total lipids using a modified Folch method (Carter et al., 2020), and then the extracted lipids were esterified into fatty acid methyl esters (FAMEs) by heating at 70°C for 2 hr in 1M acetyl chloride in methanol. Duplicate 1 µL aliquots of sample FAMEs (I mg/mL in dichloromethane) were injected into a Shmadzu Scientific Instruments QP2010S GC-MS linked to a 2010 FID (Shmadzu Scientific Instruments, Kyoto, Japan) at Sacred Heart University (Fairfield, CT). Peaks were identified by retention times established by analysis of GLC standard FAME mixes (Nu-Chek Prep, Elysian, MN USA) run every 15 samples and visual inspection of all chromatograms. Concentrations of individual FAs were calculated as a percent by mass (FA peak area/total chromatogram area).

Statistical analysis

Request a detailed protocol

All the analyses were performed in R (R Core Team 2018, R version 3.5.1) and prior to interpreting the output we checked model assumptions including inspection of residuals for homoscedasticity, normality and independence (lack of pattern). For Experiment I, we used general linear models to infer the fixed effect of Diet (MUFA vs. PUFA) on energy expenditure and changes in body mass and plasma metabolites during the 6 hr flight. Body mass was included as a fixed covariate to control for associated variation in energy expenditure. For Experiment II, we used linear mixed models (Bates et al., 2015) with restricted maximum likelihood (REML) estimation to infer the fixed effect of Diet (MUFA vs. PUFA), Flight training (trained vs. non-trained), and their interactive effect on BMR. Body mass and date of measurement were included as a fixed covariate and the number of measurement chamber as a random effect. Full models always included the interaction of main effects and covariates; however, whenever possible, we simplified the model by removing non-significant interactions. We also used linear mixed models with restricted maximum likelihood (REML) estimation to infer the fixed effect of Diet (MUFA vs. PUFA), Flight training (trained vs. non-trained) and time-point (‘Pre-training’ vs. ‘Post-flight’ vs. ‘Recovery’) on plasma metabolites and oxidative status, with individual bird included as a random effect. We used the Type III test with Kenward-Roger approximation method for the effective degrees of freedom, from the package ‘lmerTest’ (Kuznetsova et al., 2017). Descriptive statistics show adjusted least squares means with 95% confidence intervals (LSM ± 95% CI).

Data availability

All data are available in the main text or the supplementary materials.

References

    1. Alan RR
    2. McWilliams SR
    (2013) Oxidative stress, circulating antioxidants, and dietary preferences in songbirds
    Comparative Biochemistry and Physiology Part B: Biochemistry and Molecular Biology 164:185–193.
    https://doi.org/10.1016/j.cbpb.2012.12.005
  7. Book
    1. Bairlein F
    (2014)
    Atlas des Vogelzugs: Ringfunde Deutscher Brut- und Gastvögel
    Aula Verlag.
    1. Black AE
    2. Prentice AM
    3. Coward WA
    (1986)
    Use of food quotients to predict respiratory quotients for the doubly-labelled water method of measuring energy expenditure
    Human Nutrition. Clinical Nutrition 40:381–391.
    1. Butler PJ
    (2016) The physiological basis of bird flight
    Philosophical Transactions of the Royal Society B: Biological Sciences 371:20150384.
    https://doi.org/10.1098/rstb.2015.0384
  31. Book
    1. Fair J
    2. Paul E
    3. Jones J
    (2010)
    Guidelines to the Use of Wild Birds in Research
    Ornithological Council.
  32. Book
    1. Feare CJ
    (1984)
    The Starling
    Oxford University Press.
  35. Book
    1. Ginn HB
    2. Melville DS
    (1983)
    Moult in birds
    Thetford, UK: British Trust for Ornithology.
    1. Gwinner E
    (1996)
    Circadian and circannual programs in avian migration
    The Journal of Experimental Biology 199:39–48.
  47. Book
    1. Lane N
    (2005)
    Power, Sex, Suicide: Mitochondria and the Meaning of Life
    Oxford University Press.
  51. Book
    1. McArdle WD
    2. Katch FI
    3. Katch VL
    (2010)
    Exericise Physiology: Energy, Nutrition, and Human Performance
    Lippincott Williams & Wilkins.
  55. Book
    1. Nagy KA
    (1983)
    The Doubly Labeled Water (3HH18O) Method: A Guide to Its Use
    Laboratory of Biomedical and Environmental Sciences.
  56. Book
    1. Neubauer O
    2. Yfanti C
    (2015)
    Antioxidants in Sport Nutrition
    CRC Press, Taylor and Francis.
  57. Book
    1. Newton I
    (2006)
    The Migration Ecology of Birds
    Academic Press, Inc.
  63. Book
    1. Piersma T
    2. Gils JAV
    (2011)
    The Flexible Phenotype: A Body-Centred Integration of Ecology, Physiology, and Behaviour
    Oxford University Press.
    1. Speakman J
    (1993)
    How should we calculate CO2 production in doubly labeled water studies of animals?
    Functional Ecology 7:746–750.
  77. Book
    1. Speakman JR
    (1997)
    Doubly labeled water: theory and practice
    London, United Kingdom: Chapman and Hall.
    1. West GC
    2. Peyton LJ
    (1980)
    Fatty acids of depot lipids in migrating Lapland Longspurs
    Journal of Field Ornithology 51:138–143.

Decision letter

  1. Chima Nwaogu
    Reviewing Editor; University of Cape Town, South Africa
  2. Christian Rutz
    Senior Editor; University of St Andrews, United Kingdom

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

This study reports that migratory birds that consume omega-6 and -3 fats spend less energy during endurance flights; however, such short-term energy savings come at the cost of longer-term oxidative damage. This is the first demonstration of such an energy savings, oxidative cost trade-off in a migratory bird. An understanding of the ecological factors that mediate this trade-off has implications for the conservation physiology of migratory birds whose numbers are rapidly declining.

Decision letter after peer review:

Thank you for submitting your article "The energy savings-oxidative cost trade-off for birds during migration" for consideration by eLife. Your article has been reviewed by three peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the evaluation has been overseen by Christian Rutz as the Senior Editor. The reviewers have opted to remain anonymous.

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this decision letter to help you prepare a revised submission.

We would like to draw your attention to changes in our revision policy that we have made in response to COVID-19 (https://elifesciences.org/articles/57162). Specifically, we are asking editors to accept without delay manuscripts, like yours, that they judge can stand as eLife papers without additional data, even if they feel that they would make the manuscript stronger. Thus the revisions requested below only address clarity and presentation.

Summary:

Your study used diet manipulations to raise two groups of European starlings Sturnus vulgaris with different fatty acid composition in their fat stores. These starlings were used to test two complementary hypotheses: (1) that birds with more essential omega-6 and -3 polyunsaturated fatty acids will have an enhanced flight performance; and (2) that enhanced flight performance leads to upregulation of the endogenous antioxidant system or incurs oxidative damage. Your results showed that birds composed of more essential omega-6 and -3 polyunsaturated fatty acids expended 11% less energy compared to birds with more monounsaturated fatty acids. These groups did not differ in Basal Metabolic Rate, however, birds with more essential omega-6 and -3 polyunsaturated fatty acids had higher longer-term oxidative damage. You, therefore, conclude that migrants face an "energy savings-oxidative cost" trade off.

All reviewers found the hypotheses interesting, the experiment well designed, the data analyses sound and the results compelling. However, although we do not think that additional data are required to validate your conclusions, we have major concerns about some of your claims and these will need to be carefully addressed during revision, especially with respect to migration and migratory birds.

Revisions:

1) The content of the study is more about endurance exercise rather than migration, and this makes the title restrictive and the conclusion somewhat unsubstantiated. This issue is even more apparent because nothing is stated about the species' migratory ecology. Please pay attention to the following specific comments:

a) It seems a bit daring to speak about "a trade-off which migrants face". You did not show whether the starlings were in a migratory state (e.g., increase in body mass and/or fat score, Zugunruhe). Therefore it is not clear, whether the starlings did prepare for migratory state and went through "up-regulation of many aspects of physiology", such as the up-regulation of oxidative enzymes (Lundgren and Kiessling, 1985, Oecologia 66). Since the starlings were not feeding their natural diet, it might be possible that they did not have the same high oxidative capacity as free-living migrants. It would be better to speak about "endurance flight" instead of migration. Another aspect which was not considered is that European starlings are short-distance migrants, or sedentary in southern Germany. It was shown in many studies that they differ in their migration strategies and physiological adaptions from long-distance migrants. Short-distance migrants probably catabolize more protein and/or carbohydrates than long-distance migrants. Hence, one might question whether short-distance migrants, such as the starlings, which have the opportunities for frequent stop-overs to recover and to refuel, do actually face this trade-off. Maybe they do not choose a PUFA-rich diet. These aspects should be discussed and relevant literature which is missing in the manuscript should be included.

b) The basis for using European starlings for testing the hypotheses was not stated nor justified in the study. In addition, there was no explanation for using 15 days of flight-training or 6 hours (260 km) of non-stop flight for the experiment. Is this equivalent to the standard migratory distance of the species or of the population from which the nestlings were collected for the experiment? The striking absence of this information obscures the ecological relevance of the study and makes it seem like a purely physiology curiosity-driven study, with little ecological application. We would like to see some information on the ecology of the species and why it is interesting/suitable for testing these hypotheses, which apply to animals beyond birds.

c) The title of the study is restricted to migration or migratory birds even though the study is about endurance exercise in animals. Nonetheless, it is still necessary to highlight the implication of the findings for migration.

d) Can you expand this paragraph a little to substantiate the statement "Collectively, our study provides compelling evidence that avian athletes – just like human athletes – face considerable trade-offs when deciding what to eat to enhance their performance", with human examples of diet optimisation or its implications for enhanced exercise performance versus oxidative damage? Similarly, can you expand the said paragraph by discussing the implication(s) of this trade-off for migratory and other high-performance bird species?

2) The methodological limitations of the study were not addressed in the Discussion. Please pay attention to, and address, these specific comments:

a) We found the main strength and novelty of the study to be the first experiment. The experiment is well designed and uses some complicated and impressive methodology (DLW, wind tunnel). DLW methodolgy has much improved over the last 20 years and become a gold standard for activity metabolic rate measurement. But, DLW is still based on some assumptions for RQ and water loss. Is it possible that 11% difference (on average) is related to a difference in RQ, water loss and lean mass? We are not sure, but it should be considered and discussed. In a perfect design, we would expect to see a comparison between BMR recorded by respirometry and DLW, which would eliminate this doubt.

b) The second experiment is less clear and refers to recent previous work by the group (Carter et al., 2020) where an effect of diet and time (training) on BMR was found. We suggest this should be discussed, as no such effect was found in the current study.

3) Please provide methodological details about the experimental design and the data analyses that will help readers understand Figure 3, which is the basis of your conclusions about the "energy savings-oxidative cost" trade-off. Please pay attention to the comments below:

a) You hypothesize that higher PUFAs in biological membranes lead to elevated risk of oxidative damage, and report higher oxidative damage for the high PUFA group. Here, we had some difficulties to understand the results. We found Figure 3C to be confusing - in which of the three time points in the experiment is there a difference in oxidative stress? We would expect that the most significant effect of diet should be right after intensive aerobic activity, but from the data, we could not understand when the oxidative damage was recorded or if it was a summary of the three time points. Maybe we missed something, but this requires clarification, as this is the second most important finding of this work that should support the proposed trade-off hypothesis.

b) The Materials and methods and Results are not clearly presented. The terms are highly confusing, the legends do not explain all terms. For example, it is unclear how the mean values of the MUFA and PUFA groups in Figure 3 were calculated.

c) The experimental design should be described in more detail. Please give the physiological state of the birds (fasted, post-resorptive) when sampled, and the time elapsed after flight until the end of blood sampling; both factors crucially affect plasma metabolites. Please explain what happened to the starlings after the experiment and how you analysed the fatty acids (tissue, method, etc).

d) Results and Discussion paragraph six: The terms are confusing. The "trained" group includes the untrained control birds. Therefore, it would be better to call the untrained birds "controls" or "sedentary" throughout the manuscript. The "post-flight" and "trained" birds were both sampled after the 6 hours flight. Since "trained" birds were sampled 2 days "post-flight", the term "recovery" might be clearer. Another possibility would be to call them Flyers/Controls day 0, day 16, and day 18.

Results and Discussion paragraph seven, Figure 3A: According to Figure 3A, uric acid of the flyers was different between the two groups, for the flyers and the controls. Please correct.

Figure 3B: It is strange that the controls showed a decrease of antioxidant capacity within the 2 days post-flight. Please explain and discuss this result.

Figure 3, right panel: Did you take the average per diet over all 3 groups? Please explain this here and/or in the Materials and methods.

Figure 3: Please explain LSM in the legend.

Right panels with the main effect of diet on oxidative status: To which group do these values refer, pre-training, post-flight, or trained?

Figure 3: Sedentary birds: How do you explain the change of uric acid and AO in sedentary birds over time? They were just sitting in their cages. AO and uric acid in sedentary birds should remain at the same level, assuming they were in the same physiological state. Uric acid reflects diet composition and physiological state. Since the birds received the same diet throughout the experiment and did not do flight training, this is hard to explain.

4) Please revise the Discussion to focus on the core finding of the experiment, which is the occurrence of an "energy savings - oxidative cost" trade-off during endurance flight due to fatty acid composition of body fat stores. You can highlight the implications of this finding for understanding migratory birds and other animals that are likely to undertake such endurance flights. Please pay attention to the comments below:

a) Since birds may choose among food items to optimize the trade-off between enhanced flight performance and long-term oxidative damage in natural conditions, it may be important to highlight the ecological conditions under which this "energy savings - oxidative cost" trade-off will be detrimental. This would be interesting to a readership interested in subjects like the ecological factors (e.g, foraging) responsible for the survival of many declining long-distance migrants during migration.

b) Relevant literature is missing. There are other studies that measured oxidative stress in birds during endurance flight and on migration (e.g., Costantini et al., 2008; Jenni-Eiermann et al., 2014).

c) Relevant aspects regarding migrating birds are missing in the Discussion:

Introduction: Above you mention that birds during endurance flight derive their energy mainly from fats and that these fats are highly "susceptible to oxidative damage". Since other non-avian athletes derive their energy mainly from carbohydrates (e.g., humans) and varying amounts of protein, the effect of the catabolism of these other fuels on oxidative stress should also be considered. Literature data showed a relationship between anti-oxidants and ROS with muscle score for free-living European robins during migratory flight, indicating the importance of muscle degradation (Jenni-Eiermann et al., 2014).

"Differences in the energy costs…". If known, please give the cause why the cyclists differ in their energy costs. Otherwise this sentence can be omitted.

d) Flying birds: Why should uric acid in trained birds be lower than pre-training? During flight, birds catabolize proteins and uric acid increases; during the recovery days, the UA levels should return to basal values. This is very strange. Please discuss this result. It is also hard to understand why the oxidant capacity does not recover, especially because the birds did not experience oxidative damage.

e) The composition of the diet may change between resting places and it also differs between spring and autumn migration. Maybe the migrants "return" to a less PUFA-rich diet after migratory bouts. In that case, the oxidative damage would be reduced. Please consider and discuss this aspect. There is a vast literature on this.

https://doi.org/10.7554/eLife.60626.sa1

Author response

Revisions:

1) The content of the study is more about endurance exercise rather than migration, and this makes the title restrictive and the conclusion somewhat unsubstantiated. This issue is even more apparent because nothing is stated about the species' migratory ecology. Please pay attention to the following specific comments:

a) It seems a bit daring to speak about "a trade-off which migrants face". You did not show whether the starlings were in a migratory state (e.g., increase in body mass and/or fat score, Zugunruhe). Therefore it is not clear, whether the starlings did prepare for migratory state and went through "up-regulation of many aspects of physiology", such as the up-regulation of oxidative enzymes (Lundgren and Kiessling, 1985, Oecologia 66). Since the starlings were not feeding their natural diet, it might be possible that they did not have the same high oxidative capacity as free-living migrants. It would be better to speak about "endurance flight" instead of migration. Another aspect which was not considered is that European starlings are short-distance migrants, or sedentary in southern Germany. It was shown in many studies that they differ in their migration strategies and physiological adaptions from long-distance migrants. Short-distance migrants probably catabolize more protein and/or carbohydrates than long-distance migrants. Hence, one might question whether short-distance migrants, such as the starlings, which have the opportunities for frequent stop-overs to recover and to refuel, do actually face this trade-off. Maybe they do not choose a PUFA-rich diet. These aspects should be discussed and relevant literature which is missing in the manuscript should be included.

Thank you for carefully considering the most appropriate context for the results we report. As requested, we have added information about the migratory disposition of the population of starlings from which the birds we used were sampled, and we have made some substantial changes to the text as outlined below. We have also modified the title, as requested, to be less restrictive.

Regarding the specific comments of the reviewers, we acknowledge that any controlled experiment such as ours by definition controls for many potentially confounding variables at the potential expense of some extrapolation to natural conditions. Here we make the case that we went to great lengths to make our experiment as relevant to migratory birds as possible, although the reviewers comments also helped to improve the manuscript along these lines. Specifically, as detailed below and in the revised manuscript, the design of our experiment involved hand-raising nestlings taken from a population of free-living, migratory birds. Furthermore, we used prepared diets that had the same macronutrient composition as natural fruits regularly eaten by wild birds – such prepared diets ensured that the diet composition was entirely known and consistent, that it was ecologically relevant and, most importantly for our study, that the two experimental diets differed only in their fatty acid composition. As we have shown, this difference in fatty acid composition of diets produced starlings with fat stores that were similar in fatty acid composition to that observed in free-living songbirds during migration. In addition, we trained birds to fly in one of three windtunnels in the world designed for flying songbirds for long durations so that we could accurately measure energy expenditure during flight and the associated oxidative costs. We agree that some readers may question the extent to which our results from this controlled experiment apply to free-living migratory birds; however, we emphasize in the revised manuscript these many aspects of the experiment that were designed to increase the ecological relevance of the work. Thank you to the reviewers for raising these important points and for the opportunity to revise the original work along these lines.

Below we outlined in detail how we revised the article to address these important points.

We knew from earlier banding studies that the starling population in southern Germany from which our nestlings were taken was migratory. Thus, we have added the following to the Materials and methods section to emphasize that this population of starlings is migratory and so appropriate for our study:

“European starlings from this southeastern German population are medium-distance diurnal migrants that leave for wintering grounds in October and November and return to their Bavarian breeding grounds in April (48). Migratory distances for this population vary from many hundreds of kilometers to several thousand kilometers, with some individuals overwintering in the Euro-Mediterranean region and others in northwest Africa (48). European starlings are also quite social and curious, and quickly learn to successfully fly together in a given windtunnel as demonstrated by several recent studies (34, 49).”

We also carefully formulated the two experimental diets to have the macronutrient composition of high-lipid fruits that are often eaten by fall-migrating songbirds including European starlings. We added the following text related to the point:

“The macronutrient composition of the two semi-synthetic diets simulates a natural high-lipid fruit diet (53, 54), and the primary fatty acids in the diet (>90% 16:0, 18:1, and 18:2) are also the most common fatty acids in natural fruits (21) and in songbirds that eat fruits during migration (21, 45). European starlings are like many fall-migrating songbirds in that they switch to eating largely fruits (46), much to the chagrin of many vineyard owners.”

We also knew from previous studies, including some of our own, that the fatty acid composition of fat stores in starlings that we produced using the two experimental diets was similar to that of freeliving migratory songbirds. Thus, we added the following to the main text and the Materials and methods section to emphasize the ecological relevance of this fatty acid composition of starlings used in our study:

“Such differences in fatty acid composition of fat stores in starlings, specifically the relative amounts of 18:1, 18:2, and 18:3, are also among the primary longer-chain fatty acids that compose the fat stores of wild songbirds especially during migration (18, 21, 22).”

“Previous work has established that these are the primary longer-chain fatty acids in wild songbirds especially during migration (18, 21, 22), and that the fatty acid composition of birds reflects that of their diet (13, 18, 19) which we also demonstrate here (Table 2).”

We certainly agree with the reviewer(s) that the migration strategy of birds (i.e., whether short- or long-distance migrants) can influence important aspects of how often individuals must rest and refuel, and how far they may travel in a given day/night. However, we respectfully disagree that birds with different migration strategies may mobilize different amounts of proteins, carbohydrates, and lipids during a given long-duration flight. Guglielmo, 2018, recently reviewed this topic and describes how glycogen and amino acid stores can power flight for only very short periods – approximately 5 minutes for glycogen and slightly longer for amino acids. Furthermore, migratory birds have been shown to upregulate fatty acid transport proteins to promote fatty acid uptake within muscle cells where fat is also stored – these stored lipids can thus be quickly mobilized to fuel flight. Therefore, birds regardless of migration strategy use primarily lipids to fuel flights longer than ~30 minutes, something that makes them exceptional endurance athletes. Furthermore, others have compelling argued that distinguishing between short- and long-distance migrants does not result in clear differences when considering most behavioural and physiological traits, rather migratory traits appear as a continuum with no clear distinction necessary (Piersma et al. 2005 pp. 282-293 in Bauchinger et al., editors. Bird Hormones and Bird Migrations: Analyzing Hormones in Droppings and Egg Yolks and Assessing Adaptations in Long-Distance Migration.)

We emphasize this point in the text:

“birds use fats as their primary fuel (about 95%) for high-intensity endurance exercise such as migratory flights (8-10)”

We have also carefully considered each instance in the article where we use the terms “migratory flight(s)” to determine when it is more appropriate and accurate to use “endurance flight(s)”. We thank the reviewer for drawing our attention to this inappropriate usage. For example, we have modified the title so it now reads:

“The energy savings-oxidative cost tradeoff for migratory birds during endurance flight.” as opposed to the original “The energy savings-oxidative cost tradeoff for birds during migration.”

Finally, for logistical reasons we did not directly measure food intake of starlings over the course of the fall experiments. The use of Zugunruhe (nocturnal activity) as an indicator of “migration state” is not appropriate as starlings are diurnal migrants. We have emphasized this point in the text:

“We did not directly verify that such decreases in light levels in fall induced starlings to increase food intake, although many other studies provide such evidence in migratory birds (55-57), or to increase Zugunruhe (nocturnal activity), since starlings are diurnal migrants.”

b) The basis for using European starlings for testing the hypotheses was not stated nor justified in the study. In addition, there was no explanation for using 15 days of flight-training or 6 hours (260 km) of non-stop flight for the experiment. Is this equivalent to the standard migratory distance of the species or of the population from which the nestlings were collected for the experiment? The striking absence of this information obscures the ecological relevance of the study and makes it seem like a purely physiology curiosity-driven study, with little ecological application. We would like to see some information on the ecology of the species and why it is interesting/suitable for testing these hypotheses, which apply to animals beyond birds.

As noted above, we acknowledge that any controlled experiment such as ours by definition controls for many potentially confounding variables at the potential expense of some extrapolation to natural conditions. Here we make the case that we went to great lengths to make our experimental results as relevant to migratory birds as possible, and we have revised the article (specified above) to include text that better makes this case so that readers can better judge the extent to which our results are ecologically relevant. In short, the design of our experiment involved hand-raising nestlings taken from a population of free-living, migratory birds, and we fed starlings prepared diets that simulated the macronutrient composition of natural fruits regularly eaten by wild birds including starlings. Furthermore, we trained birds to fly in one of three windtunnels in the world designed for flying songbirds for long durations so that we could accurately measure energy expenditure during flight and oxidative costs. We trust that this additional information in the revised manuscript on the ecology of these European Starlings will convince readers of the ecological applications of this work.

Regarding the explanation for the training schedule and long-duration flight, we have added the following text:

“The flight-training schedule and the duration of the longest (6 hr) flight were chosen based on logistics as well as ecological relevance. The 15-day flight-training schedule and flying conditions (always 12 m s-1 wind speed and 15 °C) had been used in previously successful experiments at MPIO designed to fly barn swallows (Hirundo rustica) and starlings (Sturnus vulgaris, Sturnus roseus) for long durations (47, 50, 51, 58, 60), and free-living songbirds including starlings typically complete their migration from breeding to wintering areas over many days of flying and stopovers (46, 61). The duration of the final longest flight (6-hrs, 260 km) was sufficiently long to provide adequate turnover of the isotope-labelled water and so ensure accurate measurements of energy expenditure using the doubly-labelled water (DLW) technique (62). The 6-hr duration of flight is also within the range of typical single-day migratory flights for many free-living songbirds including starlings (46, 61).”

In sum, we thank the reviewers for requesting this additional information because it does make clearer the ecological relevance of the work.

c) The title of the study is restricted to migration or migratory birds even though the study is about endurance exercise in animals. Nonetheless, it is still necessary to highlight the implication of the findings for migration.

Thank you for this suggestion. We have modified the title so it now reads “The energy savings-oxidative cost tradeoff for migratory birds during endurance flight.”

d) Can you expand this paragraph a little to substantiate the statement "Collectively, our study provides compelling evidence that avian athletes – just like human athletes – face considerable trade-offs when deciding what to eat to enhance their performance", with human examples of diet optimisation or its implications for enhanced exercise performance versus oxidative damage? Similarly, can you expand the said paragraph by discussing the implication(s) of this trade-off for migratory and other high-performance bird species?

Thank you for the opportunity to extend our Discussion of the implications of our work. There is a large literature on the importance of nutrition in determining human exercise performance, and especially in the last 10 yrs quite a bit of attention on dietary fatty acid composition and antioxidants on human exercise performance. We have added several references to this sentence that substantiate this statement and point the readers to key literature on this topic. Also as requested, we have expanded this paragraph to include other implications of this trade-off for migratory birds. Specifically, this section of the paragraph now reads as followed:

“ …. birds may choose among diets to optimize the trade-off between enhanced flight performance (more 18:2) while reducing the long-term costs of being composed of more longchain PUFA. Migratory birds can also optimize this energy savings-oxidative cost tradeoff by being composed of more n-3 and/or n-6 PUFAs only during migration periods when energy demands and fat catabolism are most extreme, and then become more monunsatured in composition during non-migration periods – such seasonal changes in fatty acid composition are commonly observed in migratory birds (21, 45).”

2) The methodological limitations of the study were not addressed in the Discussion. Please pay attention to, and address, these specific comments:

a) We found the main strength and novelty of the study to be the first experiment. The experiment is well designed and uses some complicated and impressive methodology (DLW, wind tunnel). DLW methodolgy has much improved over the last 20 years and become a gold standard for activity metabolic rate measurement. But, DLW is still based on some assumptions for RQ and water loss. Is it possible that 11% difference (on average) is related to a difference in RQ, water loss and lean mass? We are not sure, but it should be considered and discussed. In a perfect design, we would expect to see a comparison between BMR recorded by respirometry and DLW, which would eliminate this doubt.

Thank you for this inciteful comment. As pointed out by the reviewer(s), differences in rates of water loss during flights can significantly affect the DLW estimates of energy expenditure. This was unlikely in our study given that the relative humidity and temperature and duration of the 6-hr flights in the windtunnel were tightly controlled. We have stated the following to make sure such a concern is alleviated:

“A single-pool model was used to calculate rates of CO2 production as recommended for use in animals less than 5 kg in body mass (69). There are several approaches for the treatment of evaporative water loss in the calculation (70). We detected no differences in body water pool before and after 6-hr flights. We assumed evaporation of 25% of the water flux (equation 7.17 in Speakman (62), which minimizes error in a range of conditions (70, 71).”

Also as pointed out by the reviewer(s), a key assumption of the DLW that is most relevant to our study is that the substrates used as fuel (carbohydrate, fat, protein) are known – this is important because the DLW technique estimates carbon dioxide production (not oxygen consumption) and carbon dioxide production depends on fuel type (unlike oxygen consumption). Given that our experimental design involved feeding birds diets that were identical in macronutrient composition and differed only in their fatty acid composition, the dietary substrates (carbohydrates, fats, proteins) available as fuels were also identical. Although it was not possible to directly measure RQ or change in lean and fat mass of individuals during the 6-hr flights, the effect of variation in the amount of stored fat vs. protein used as fuel (as reflected in the RQ) results in errors in estimates of energy expenditure not exceeding ± 2% (Black et al., 1986, Schmidt-Wellenburg et al., 2008, Westerterp, 2017). We have added the following to the description of the DLW method to emphasize this point:

“Given that our experimental design involved feeding birds diets that were identical in macronutrient composition and differed only in their fatty acid composition, the dietary substrates (carbohydrates, fats, proteins) available as fuels were also identical. Empirically measured variation in the amount of stored fat or protein used as fuel (as reflected in the RQ) results in errors in estimates of energy expenditure not exceeding ± 2% (50, 72, 73).”

b) The second experiment is less clear and refers to recent previous work by the group (Carter et al., 2020) where an effect of diet and time (training) on BMR was found. We suggest this should be discussed, as no such effect was found in the current study.

We trust that the substantial clarifications we have made to the description of the experimental design and methods (outlined throughout this letter), plus the revised main text describing the results will make the results from Experiment II much clearer. Once again, thank you for the opportunity to make these clarifying changes.

The reviewer is correct that we recently published results from another experiment conducted in Canada by our group (i.e., Carter et al., 2020) – we had referenced this study (#34) in the original draft of this manuscript. In that study, we fed Canada-born European Starlings one of two diets that differed primarily in 18:2n-6 and this in turn produced starlings with corresponding differences in fatty acid composition of fat stores and muscle membranes that were consistent over the 4-month fall experiment. Furthermore, birds with higher concentrations of 18:2n-6 in membranes and fat stores had higher BMR and peak metabolic rates, although this pattern was evident only early in the fall and not later in the fall experiment. This study did not include starlings that flew under controlled conditions in a wind-tunnel for 6-hrs at a time, unlike the present manuscript that we have submitted for your consideration. The change through time in these performance measures but not membrane composition led Carter et al., 2020, to conclude that their results were most consistent with the signal hypothesis compared to other competing hypotheses – we had also discussed this result in the original draft of this manuscript. Given that we had already discussed these results in the original version of the manuscript, we have not added any further discussion of this other published work in the revised draft of this manuscript.

3) Please provide methodological details about the experimental design and the data analyses that will help readers understand Figure 3, which is the basis of your conclusions about the "energy savings - oxidative cost" trade-off. Please pay attention to the comments below:

a) You hypothesize that higher PUFAs in biological membranes lead to elevated risk of oxidative damage, and report higher oxidative damage for the high PUFA group. Here, we had some difficulties to understand the results. We found Figure 3C to be confusing, in which of the three time points in the experiment is there a difference in oxidative stress? We would expect that the most significant effect of diet should be right after intensive aerobic activity, but from the data, we could not understand when the oxidative damage was recorded or if it was a summary of the three time points. Maybe we missed something, but this requires clarification, as this is the second most important finding of this work that should support the proposed trade-off hypothesis.

Thank you for making sure that this result is clear to readers, as we agree this is an important finding of our work.

The experimental design included two diet groups (MUFA and PUFA) and two exercise groups (flight-trained and untrained, “sedentary” control individuals). Birds in each of these four groups were blood sampled at three times over the course of the flight training period (prior to any training, immediately after a long-duration flight, and two days later after the birds had recovered from any acute effect of flight). Importantly, sedentary control birds were blood sampled on the same days as the flight-trained birds in their cohort to control for any coincident time-of-fall effects.

To make this clearer, we have revised the legend to Table 3 (where the main statistical results are reported) so that it now reads:

“Table 3. The effect of flight (flight-trained for 15 days in wind-tunnel or not; Trained or Sedentary), diet (MUFA or PUFA), and time (blood sampled at three different time points: before the start of flight training in the wind-tunnel (“pre-training”), immediately after a long duration flight on Day 15 (“post-flight”), and ca. 1.5 days afterwards (“recovery”)) on plasma metabolites and oxidative status in European starlings in Experiment II. Note that individuals that did not undergo flight training (i.e., control “untrained” birds) were sampled on the same days as flight-trained birds in their same cohort. Test statistics: F-value with denominator degrees of freedom (ddf) and significance level p-value for main factors and their interactions from the linear mixed models.”

We have also revised Figure 3 to make clearer the comparisons and key results. We corrected a small typo in the reporting of the multiple comparisons test in Figure 3B which now makes clear that the sedentary control birds did not show a decrease in antioxidant capacity over the flight training period. We also added a new header to the right panel in Figure 3 (“Main effect of diet”) that should make crystal clear that these panels are presenting the key diet main effects.

As outlined elsewhere in this letter, we have also made substantial changes to the text of the Materials and methods section to make clearer the experimental design and methods.

b) The Materials and methods and Results are not clearly presented. The terms are highly confusing, the legends do not explain all terms. For example, it is unclear how the mean values of the MUFA and PUFA groups in Figure 3 were calculated.

We of course did not intend for the Materials and methods and Results to be confusing. In order to more clearly present the results, we have adopted the suggestions of the reviewer for the renaming of the groups. In short, we revised the manuscript terms throughout to now refer to “trained” or “flight-trained” birds versus “sedentary” control birds (instead of the original “untrained” term). Likewise, the three time points are now referred throughout as “Pre-training”, “Post-flight”, and “Recovery” (instead of the original “trained” term for the latter). Thank you for these suggestions.

We have also substantially revised the legends to Figure 3 and Table 3 to make clearer the comparisons and key results. In short, the experimental design included two diet groups (MUFA and PUFA) and two exercise groups (flight-trained and untrained, “sedentary” control individuals). Birds in each of these four groups were blood sampled at three times over the course of the flight training period (prior to any training, immediately after a long-duration flight, and two days later after the birds had recovered from any acute effect of flight). Importantly, sedentary control birds were blood sampled on the same days as the flight-trained birds in their cohort to control for any coincident time-of-fall effects. The key results are shown in Figure 3 and the details of the statistical analyses are reported in Table 3. This right panel in Figure 3 reports the main effect of diet (MUFA vs. PUFA) on the three measures. To make this crystal clear, we have added a new header in Figure 3 associated with this right panel, “Main effect of diet”.

The legend to Table 3 has also been revised and now reads:

“Table 3. The effect of flight (flight-trained for 15 days in wind-tunnel or not; Trained or Sedentary), diet (MUFA or PUFA), and time (blood sampled at three different time points: before the start of flight training in the wind-tunnel (“pre-training”), immediately after a long duration flight on Day 15 (“post-flight”), and ca. 1.5 days afterwards (“recovery”)) on plasma metabolites and oxidative status in European starlings in Experiment II. Note that individuals that did not undergo flight training (i.e., control “untrained” birds) were sampled on the same days as flight-trained birds in their same cohort. Test statistics: F-value with denominator degrees of freedom (ddf) and significance level p-value for main factors and their interactions from the linear mixed models.”

c) The experimental design should be described in more detail. Please give the physiological state of the birds (fasted, post-resorptive) when sampled, and the time elapsed after flight until the end of blood sampling; both factors crucially affect plasma metabolites. Please explain what happened to the starlings after the experiment and how you analysed the fatty acids (tissue, method, etc).

Thank you for the opportunity to provide more details about the experimental methods.

First, all blood samples used for metabolites were taken after birds had been fasted for at least 12 hrs (i.e., after an overnight without food and before offered food on the Pre-training and Recovery days, and after an overnight without food plus their longest flight on the Post-flight day). We have included this information in the Materials and methods in section “7. Blood sampling and analyses”.

Second, all blood samples used for metabolites were taken within 3 min of capture, including within 3 min of completing the longest duration flight. We have also included this information in the Materials and methods in section “7. Blood sampling and analyses”.

Lastly, apologies for this omission about how the fat stores were sampled from starlings, and about the fatty acid analyses. We have added a section “8. Fatty acid composition of diet and furcular fat” to the Materials and methods in which we describe these methods, as follows:

“Fatty acid composition of diet and furcular fat

Fatty acid composition of the two semi-synthetic diets (Table 2) was measured by gas chromatography in lipids extracted using a modified Folch method (34) from composited subsamples of each diet taken over the course of the experiments. We collected adipose tissue from the visible fat stores in the furcular area of starlings used in Experiment I (n = 29) 5-7 days after their longest flight in the wind-tunnel. We conducted the biopsies of this furcular fat using the methods outlined in Rocha et al. (82). Briefly, upper breast feathers were wetted and moved aside to expose the skin and visible subcutaneous yellow-colored fat stores. We selected an area without visible capillaries, disinfected the area with antiseptic solution, and applied a topical anaesthetic gel to this area. After ca. 10 min, we then pinched the skin in this area to ensure no pain response by the bird. We then made an ca. 3 mm-long incision in the skin using a sharp scalpel, pulled a small piece of adipose tissue (ca. 10-20 mg) through the incision using sterile forceps, and cut the sample under the forceps using sterile scissors. The incision area was then realigned and a thin layer of veterinary tissue glue was applied to seal the incision. All birds were checked weekly thereafter and the wound was completely healed within 2-3 weeks.

Dietary fat and furcular fat samples were stored at -80 C until fatty acid composition was measured by gas chromatography. We thawed samples, extracted total lipids using a modified Folch method (34), and then the extracted lipids were esterified into fatty acid methyl esters (FAMEs) by heating at 70 C for 2 hr in 1M acetyl chloride in methanol. Duplicate 1 ul aliquots of sample FAMEs (I mg/ml in dichloromethane) were injected into a Shmadzu Scientific Instruments QP2010S GC-MS linked to a 2010 FID (Shmadzu Scientific Instruments, Kyoto, Japan) at Sacred Heart University (Fairfield, CT). Peaks were identified by retention times established by analysis of GLC standard FAME mixes (Nu-Chek Prep, Elysian, MN USA) run every 15 samples and visual inspection of all chromatograms. Concentrations of individual FAs were calculated as a percent by mass (FA peak area/total chromatogram area).”

d) Results and Discussion paragraph six: The terms are confusing. The "trained" group includes the untrained control birds. Therefore, it would be better to call the untrained birds "controls" or "sedentary" throughout the manuscript. The "post-flight" and "trained" birds were both sampled after the 6 hours flight. Since "trained" birds were sampled 2 days "post-flight", the term "recovery" might be clearer. Another possibility would be to call them Flyers/Controls day 0, day 16, and day 18.

Apologies for the confusion of terms. Of course for us these short-hand terms made total sense as we conducted the experiment, although we now see how the naming of these experimental groups could be improved. We have adopted the suggestions of the reviewer for the renaming of the groups. In short, we revised the manuscript terms throughout to now refer to “trained” or “flight-trained” birds versus “sedentary” birds (instead of the original “untrained” term). Likewise, the three time points are now referred throughout as “Pre-training”, “Post-flight”, and “Recovery” (instead of the original “trained” term for the latter). Thank you for these suggestions.

Results and Discussion paragraph seven, Figure 3A: According to Figure 3A, uric acid of the flyers was different between the two groups, for the flyers and the controls. Please correct.

Figure 3B: It is strange that the controls showed a decrease of antioxidant capacity within the 2 days post-flight. Please explain and discuss this result.

Thank you for catching these two errors – we have corrected what we wrote about the uric acid results, and we realized there was a small typo in the reporting of the multiple comparisons test in Figure 3B which now makes clear that the sedentary control birds did not show a decrease in antioxidant capacity over the flight training period. The sentence now reads:

“Flight-training over more than two weeks did not affect baseline levels of oxidative damage (compare Pre-training and Recovery; Figure 3C) while antioxidant capacity decreased in flight-trained but not sedentary birds (Figure 3B) and plasma uric acid decreased over time in both trained and sedentary starlings (Figure 3A; see Table 3 for detailed statistical results).”

Figure 3, right panel: Did you take the average per diet over all 3 groups? Please explain this here and/or in the Materials and methods.

This right panel in Figure 3 reports the main effect of diet (MUFA vs. PUFA) on the three measures. The detailed statistical results are shown in Table 3. We have cited this table and revised the legend to this figure to make this clearer. We have also included a new header in Figure 3 associated with this right panel, “Main effect of diet”.

Figure 3: Please explain LSM in the legend.

Right panels with the main effect of diet on oxidative status: To which group do these values refer, pre-training, post-flight, or trained?

This right panel in Figure 3 reports the main effect of diet (MUFA vs. PUFA) on the three measures. The detailed statistical results are shown in Table 3. We have cited this table and revised the legend to this figure to make this clearer. We have also included a new header in Figure 3 associated with this right panel, “Main effect of diet”. Thank you for pointing out that we needed to define/explain the “LSM” acronym in the legend – we have done so.

Figure 3: Sedentary birds: How do you explain the change of uric acid and AO in sedentary birds over time? They were just sitting in their cages. AO and uric acid in sedentary birds should remain at the same level, assuming they were in the same physiological state. Uric acid reflects diet composition and physiological state. Since the birds received the same diet throughout the experiment and did not do flight training, this is hard to explain.

Thank you for carefully considering these results – as noted above, we have corrected what we wrote about the uric acid results, and we realized there was a small typo in the reporting of the multiple comparisons test in Figure 3B which now makes clear the sedentary control birds did not show a decrease in antioxidant capacity over the flight training period. The sentence now reads:

“Flight-training over more than two weeks did not affect baseline levels of oxidative damage (compare Pre-training and Recovery; Figure 3C) while antioxidant capacity decreased in flight-trained but not sedentary birds (Figure 3B) and plasma uric acid decreased over time in both trained and sedentary starlings (Figure 3A; see Table 3 for detailed statistical results).”

Like the reviewer, we did not expect the decrease in plasma uric acid over time in both sedentary and flight-trained birds, although now this remains the only one of the plasma metabolites that shows this trend. We have no good explanation for why plasma uric acid decreased in sedentary (and flight-trained) birds over time, although such evidence makes clear the importance of including this sedentary control group.

4) Please revise the Discussion to focus on the core finding of the experiment, which is the occurrence of an "energy savings - oxidative cost" trade-off during endurance flight due to fatty acid composition of body fat stores. You can highlight the implications of this finding for understanding migratory birds and other animals that are likely to undertake such endurance flights. Please pay attention to the comments below:

a) Since birds may choose among food items to optimize the trade-off between enhanced flight performance and long-term oxidative damage in natural conditions, it may be important to highlight the ecological conditions under which this "energy savings - oxidative cost" trade-off will be detrimental. This would be interesting to a readership interested in subjects like the ecological factors (e.g, foraging) responsible for the survival of many declining long-distance migrants during migration.

Thank you for this suggestion. We have further discussed the ecological conditions under which this tradeoff can be detrimental as follows:

“…birds may choose among diets to optimize the trade-off between enhanced flight performance (more 18:2) while reducing the long-term costs of being composed of more longchain PUFA. Migratory birds can also optimize this energy savings-oxidative cost tradeoff by being composed of more n-3 and/or n-6 PUFAs only during migration periods when energy demands and fat catabolism are most extreme, and then become more monounsaturated in composition during non-migration periods – such seasonal changes in fatty acid composition are commonly observed in migratory birds (21, 45). Such a tradeoff may become especially detrimental if foods with different quantities of micronutrients (notably long-chain PUFAs and antioxidants) are not available in nature across the seasons, in which case birds may be unable to ameliorate such a tradeoff through careful choices of diet.”

b) Relevant literature is missing. There are other studies that measured oxidative stress in birds during endurance flight and on migration (e.g., Costantini et al., 2008; Jenni-Eiermann et al., 2014).

Thank you for allowing us the opportunity to emphasize that there have been a few other studies that have measured the oxidative challenges of endurance exercise such as migratory flights. We were quite aware of the original contributions provided by these two papers (Costantini et al., 2008; Jenni-Eiermann et al., 2014) and we were sure to cite their contributions in the revised manuscript – see, for example:

“Increased energy metabolism during exercise is often associated with increased production of pro-oxidants regardless of the fuel types used (i.e., carbohydrates, protein, fats) which causes oxidative damage if not quickly quenched by dietary antioxidants and/or by increased production of antioxidant enzymes (e.g., superoxide dismutase, glutathione peroxidase).”

c) Relevant aspects regarding migrating birds are missing in the Discussion:

Introduction: Above you mention that birds during endurance flight derive their energy mainly from fats and that these fats are highly "susceptible to oxidative damage". Since other non-avian athletes derive their energy mainly from carbohydrates (e.g., humans) and varying amounts of protein, the effect of the catabolism of these other fuels on oxidative stress should also be considered. Literature data showed a relationship between anti-oxidants and ROS with muscle score for free-living European robins during migratory flight, indicating the importance of muscle degradation (Jenni-Eiermann et al., 2014).

Thank you for allowing us the opportunity to emphasize that the oxidative challenges of endurance exercise such as migratory flights of birds are evident independent of fuel type. We were quite aware of the original contributions provided by these two papers (Costantini et al., 2008; Jenni-Eiermann et al., 2014, and we were sure to cite their contributions in the revised manuscript:

“Increased energy metabolism during exercise is often associated with increased production of pro-oxidants regardless of the fuel types used (i.e., carbohydrates, protein, fats) which causes oxidative damage if not quickly quenched by dietary antioxidants and/or by increased production of antioxidant enzymes (e.g., superoxide dismutase, glutathione peroxidase) (5, 36-41).”

"Differences in the energy costs…". If known, please give the cause why the cyclists differ in their energy costs. Otherwise this sentence can be omitted.

When we have presented this work to audiences at meetings there is often a question about the relevance of an 11% energy savings. Thus, we would prefer to retain this one sentence that points out that the top-placing Tour de France cyclists differ by <5% in energy costs in hopes of emphasizing to readers the relevance of an 11% energy savings for birds composed of more PUFA.

d) Flying birds: Why should uric acid in trained birds be lower than pre-training? During flight, birds catabolize proteins and uric acid increases; during the recovery days, the UA levels should return to basal values. This is very strange. Please discuss this result. It is also hard to understand why the oxidant capacity does not recover, especially because the birds did not experience oxidative damage.

Thank you for carefully considering these results – as noted above, we realized there was a small typo in the reporting of the multiple comparisons test in Figure 3b which now makes clear the sedentary control birds did not show a decrease in antioxidant capacity over the flight training period. The sentence now reads:

“Flight-training over more than two weeks did not affect baseline levels of oxidative damage (compare Pre-training and Recovery; Figure 3C) while antioxidant capacity decreased in flight-trained but not sedentary birds (Figure 3B) and plasma uric acid decreased over time in both trained and sedentary starlings (Figure 3A; see Table 3 for detailed statistical results).”

Thus, there was no significant change in oxidative capacity in sedentary, control birds over time.

Like the reviewer, we did not expect the decrease in plasma uric acid over time in both sedentary and flight-trained birds. We have no good explanation for why plasma uric acid decreased in sedentary (and flight-trained) birds between the Pre-Flight and Recovery time points. However, we note that Table 3, which is referenced in the text, reveals that this is the only one of the three measures shown in Figure 3 with a significant Flight X Time interaction term. It is the remarkable increase in plasma uric acid during a given long-duration flight (in only flight-trained birds) that we have chosen to emphasize.

e) The composition of the diet may change between resting places and it also differs between spring and autumn migration. Maybe the migrants "return" to a less PUFA-rich diet after migratory bouts. In that case, the oxidative damage would be reduced. Please consider and discuss this aspect. There is a vast literature on this.

Thank you for pointing this out. We whole-heartedly agree that availability of PUFA-rich diets may differ over space and time, and especially between spring and autumn migration. We have briefly discussed this important point as follows:

“...birds may choose among diets to optimize the trade-off between enhanced flight performance (more 18:2) while reducing the long-term costs of being composed of more longchain PUFA. Migratory birds can also optimize this energy savings-oxidative cost tradeoff by being composed of more n-3 and/or n-6 PUFAs only during migration periods when energy demands and fat catabolism are most extreme, and then become more monounsaturated in composition during non-migration periods – such seasonal changes in fatty acid composition are commonly observed in migratory birds (21, 45). Such a tradeoff may become especially detrimental if foods with different quantities of micronutrients (notably long-chain PUFAs and antioxidants) are not available in nature across the seasons, in which case birds may be unable to ameliorate such a tradeoff through careful choices of diet.”

https://doi.org/10.7554/eLife.60626.sa2

Article and author information

Author details

  1. Scott McWilliams

    Department of Natural Resources Science, University of Rhode Island, Kingston, United States
    Contribution
    Conceptualization, Resources, Data curation, Formal analysis, Supervision, Funding acquisition, Investigation, Methodology, Writing - original draft, Project administration, Writing - review and editing
    For correspondence
    srmcwilliams@uri.edu
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-9727-1151

  2. Barbara Pierce

    Department of Biology, Sacred Heart University, Fairfield, United States
    Contribution
    Conceptualization, Resources, Funding acquisition, Methodology, Project administration, Writing - review and editing
    Competing interests
    No competing interests declared

  3. Andrea Wittenzellner

    Max Planck Institute for Ornithology, Starnberg, Germany
    Contribution
    Resources, Investigation, Methodology, Project administration, Writing - review and editing
    Competing interests
    No competing interests declared

  4. Lillie Langlois

    Department of Natural Resources Science, University of Rhode Island, Kingston, United States
    Contribution
    Investigation, Methodology, Project administration, Writing - review and editing
    Competing interests
    No competing interests declared

  5. Sophia Engel

    Max Planck Institute for Ornithology, Starnberg, Germany
    Contribution
    Investigation, Methodology, Writing - review and editing
    Competing interests
    No competing interests declared

  6. John R Speakman

    1. Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing, China
    2. Institute of Biological and Environmental Sciences, University of Aberdeen, Scotland, United Kingdom
    Contribution
    Formal analysis, Investigation, Methodology, Writing - review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-2457-1823

  7. Olivia Fatica

    Department of Biology, Sacred Heart University, Fairfield, United States
    Contribution
    Investigation, Methodology, Writing - review and editing
    Competing interests
    No competing interests declared

  8. Kristen DeMoranville

    Department of Natural Resources Science, University of Rhode Island, Kingston, United States
    Contribution
    Data curation, Investigation, Methodology, Project administration, Writing - review and editing
    Competing interests
    No competing interests declared

  9. Wolfgang Goymann

    Max Planck Institute for Ornithology, Starnberg, Germany
    Contribution
    Formal analysis, Investigation, Methodology, Writing - review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-7553-5910

  10. Lisa Trost

    Max Planck Institute for Ornithology, Starnberg, Germany
    Contribution
    Investigation, Methodology, Project administration, Writing - review and editing
    Competing interests
    No competing interests declared

  11. Amadeusz Bryla

    Institute of Environmental Sciences, Jagiellonian University, Kraków, Poland
    Contribution
    Data curation, Formal analysis, Investigation, Methodology, Writing - review and editing
    Competing interests
    No competing interests declared

  12. Maciej Dzialo

    Institute of Environmental Sciences, Jagiellonian University, Kraków, Poland
    Contribution
    Formal analysis, Investigation, Methodology, Writing - review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-3632-8572

  13. Edyta Sadowska

    1. Institute of Environmental Sciences, Jagiellonian University, Kraków, Poland
    2. Nencki Institute of Experimental Biology PAS, Warszawa, Poland
    Contribution
    Data curation, Formal analysis, Investigation, Methodology, Writing - review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-1240-4814

  14. Ulf Bauchinger

    Nencki Institute of Experimental Biology PAS, Warszawa, Poland
    Contribution
    Conceptualization, Resources, Formal analysis, Supervision, Funding acquisition, Investigation, Methodology, Project administration, Writing - review and editing
    Competing interests
    No competing interests declared

Funding

National Science Foundation (IBN-9983920)

  • Scott McWilliams
  • Barbara Pierce

U.S. Department of Agriculture (RIAES-538748)

  • Scott McWilliams

National Science Foundation (UMO-2015/19/B/NZ8/01394)

  • Ulf Bauchinger

National Science Foundation (IOS-1354187)

  • Scott McWilliams
  • Barbara Pierce

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

We are thankful to the many members of the McWilliams lab who provided stimulating discussions on key topics. Major logistical support was provided by the animal care staff, technical staff, and scientists at the Max Planck Institute for Ornithology, Seewiesen, Germany. Special thanks to Carola Schmidt-Wellenburg and Ninon Ballerstaedt who helped with the windtunnel experiments, B Helm, B Kempenaers, H Gwinner, B Biebach, and M Starck for regular stimulating discussions, and H Biebach for having the foresight and fortitude to create the windtunnel at MPIO, Seewiesen. Funding: this research was supported by grants from the National Science Foundation to SRM and BJP (IBN-9983920 and IOS-1354187), the USDA to SRM (RIAES-538748), and an OPUS grant from the National Science Foundation (NCN), Poland, to UB (UMO-2015/19/B/NZ8/01394).

Ethics

Animal experimentation: All procedures adhered to the ethical guidelines of the North American Ornithological Council (Fair et al., 2010) and were approved by the University of Rhode Island IACUC (Protocols #AN09-09-009, AN08-02-014) and the Government of Upper Bavaria, Germany (AZ 55.2-1-54-2532-216-2014).

Senior Editor

  1. Christian Rutz, University of St Andrews, United Kingdom

Reviewing Editor

  1. Chima Nwaogu, University of Cape Town, South Africa

Version history

  1. Received: July 1, 2020
  2. Accepted: December 3, 2020
  3. Version of Record published: December 11, 2020 (version 1)

Copyright

© 2020, McWilliams et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 1,734
    Page views
  • 178
    Downloads
  • 18
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Scott McWilliams
  2. Barbara Pierce
  3. Andrea Wittenzellner
  4. Lillie Langlois
  5. Sophia Engel
  6. John R Speakman
  7. Olivia Fatica
  8. Kristen DeMoranville
  9. Wolfgang Goymann
  10. Lisa Trost
  11. Amadeusz Bryla
  12. Maciej Dzialo
  13. Edyta Sadowska
  14. Ulf Bauchinger
(2020)
The energy savings-oxidative cost trade-off for migratory birds during endurance flight
eLife 9:e60626.
https://doi.org/10.7554/eLife.60626

Categories and tags

Further reading

    1. Biochemistry and Chemical Biology
    2. Microbiology and Infectious Disease

    Indole produced during dysbiosis mediates host–microorganism chemical communication

    Rui-Qiu Yang, Yong-Hong Chen ... Cheng-Gang Zou
    Research Article Updated

    An imbalance of the gut microbiota, termed dysbiosis, has a substantial impact on host physiology. However, the mechanism by which host deals with gut dysbiosis to maintain fitness remains largely unknown. In Caenorhabditis elegans, Escherichia coli, which is its bacterial diet, proliferates in its intestinal lumen during aging. Here, we demonstrate that progressive intestinal proliferation of E. coli activates the transcription factor DAF-16, which is required for maintenance of longevity and organismal fitness in worms with age. DAF-16 up-regulates two lysozymes lys-7 and lys-8, thus limiting the bacterial accumulation in the gut of worms during aging. During dysbiosis, the levels of indole produced by E. coli are increased in worms. Indole is involved in the activation of DAF-16 by TRPA-1 in neurons of worms. Our finding demonstrates that indole functions as a microbial signal of gut dysbiosis to promote fitness of the host.

    1. Biochemistry and Chemical Biology
    2. Structural Biology and Molecular Biophysics

    The neuronal calcium sensor NCS-1 regulates the phosphorylation state and activity of the Ga chaperone and GEF Ric-8A

    Daniel Muñoz-Reyes, Levi J McClelland ... Maria Jose Sanchez-Barrena
    Research Article

    The Neuronal Calcium Sensor 1, an EF-hand Ca2+ binding protein, and Ric-8A coregulate synapse number and probability of neurotransmitter release. Recently, the structures of Ric-8A bound to Ga have revealed how Ric-8A phosphorylation promotes Ga recognition and activity as a chaperone and guanine nucleotide exchange factor. However, the molecular mechanism by which NCS-1 regulates Ric-8A activity and its interaction with Ga subunits is not well understood. Given the interest in the NCS-1/Ric-8A complex as a therapeutic target in nervous system disorders, it is necessary to shed light on this molecular mechanism of action at atomic level. We have reconstituted NCS-1/Ric-8A complexes to conduct a multimodal approach and determine the sequence of Ca2+ signals and phosphorylation events that promote the interaction of Ric-8A with Ga. Our data show that the binding of NCS-1 and Ga to Ric-8A are mutually exclusive. Importantly, NCS-1 induces a structural rearrangement in Ric-8A that traps the protein in a conformational state that is inaccessible to Casein Kinase II-mediated phosphorylation, demonstrating one aspect of its negative regulation of Ric-8A-mediated G-protein signaling. Functional experiments indicate a loss of Ric-8A GEF activity towards Ga when complexed with NCS-1, and restoration of nucleotide exchange activity upon increasing Ca2+ concentration. Finally, the high-resolution crystallographic data reported here define the NCS-1/Ric-8A interface and will allow the development of therapeutic synapse function regulators with improved activity and selectivity.

    1. Biochemistry and Chemical Biology
    2. Cell Biology

    Scaling of an antibody validation procedure enables quantification of antibody performance in major research applications

    Riham Ayoubi, Joel Ryan ... Carl Laflamme
    Research Advance

    Antibodies are critical reagents to detect and characterize proteins. It is commonly understood that many commercial antibodies do not recognize their intended targets, but information on the scope of the problem remains largely anecdotal, and as such, feasibility of the goal of at least one potent and specific antibody targeting each protein in a proteome cannot be assessed. Focusing on antibodies for human proteins, we have scaled a standardized characterization approach using parental and knockout cell lines (Laflamme et al., 2019) to assess the performance of 614 commercial antibodies for 65 neuroscience-related proteins. Side-by-side comparisons of all antibodies against each target, obtained from multiple commercial partners, have demonstrated that: (i) more than 50% of all antibodies failed in one or more applications, (ii) yet, ~50–75% of the protein set was covered by at least one high-performing antibody, depending on application, suggesting that coverage of human proteins by commercial antibodies is significant; and (iii) recombinant antibodies performed better than monoclonal or polyclonal antibodies. The hundreds of underperforming antibodies identified in this study were found to have been used in a large number of published articles, which should raise alarm. Encouragingly, more than half of the underperforming commercial antibodies were reassessed by the manufacturers, and many had alterations to their recommended usage or were removed from the market. This first study helps demonstrate the scale of the antibody specificity problem but also suggests an efficient strategy toward achieving coverage of the human proteome; mine the existing commercial antibody repertoire, and use the data to focus new renewable antibody generation efforts.