1. Evolutionary Biology
Download icon

Convergent changes in muscle metabolism depend on duration of high-altitude ancestry across Andean waterfowl

  1. Neal J Dawson  Is a corresponding author
  2. Luis Alza
  3. Gabriele Nandal
  4. Graham R Scott
  5. Kevin G McCracken
  1. Department of Biology, McMaster University, Canada
  2. Department of Biology University of Miami, United States
  3. University of Alaska Museum and Institute of Arctic Biology, University of Alaska Fairbanks, United States
  4. Centro de Ornitología y Biodiversidad - CORBIDI, Peru
  5. Department of Marine Biology and Ecology, Rosenstiel School of Marine and Atmospheric Sciences, University of Miami, United States
  6. Human Genetics and Genomics, Hussman Institute for Human Genomics, University of Miami Miller School of Medicine, United States
Research Article
  • Cited 0
  • Views 580
  • Annotations
Cite this article as: eLife 2020;9:e56259 doi: 10.7554/eLife.56259

Abstract

High-altitude environments require that animals meet the metabolic O2 demands for locomotion and thermogenesis in O2-thin air, but the degree to which convergent metabolic changes have arisen across independent high-altitude lineages or the speed at which such changes arise is unclear. We examined seven high-altitude waterfowl that have inhabited the Andes (3812–4806 m elevation) over varying evolutionary time scales, to elucidate changes in biochemical pathways of energy metabolism in flight muscle relative to low-altitude sister taxa. Convergent changes across high-altitude taxa included increased hydroxyacyl-coA dehydrogenase and succinate dehydrogenase activities, decreased lactate dehydrogenase, pyruvate kinase, creatine kinase, and cytochrome c oxidase activities, and increased myoglobin content. ATP synthase activity increased in only the longest established high-altitude taxa, whereas hexokinase activity increased in only newly established taxa. Therefore, changes in pathways of lipid oxidation, glycolysis, and mitochondrial oxidative phosphorylation are common strategies to cope with high-altitude hypoxia, but some changes require longer evolutionary time to arise.

Introduction

Given a common set of environmental challenges, evolution often converges upon a phenotype that maximizes fitness in that environment (i.e., fitness optimum). Many studies have explored the phenomenon of convergent evolution at molecular and biochemical levels by focusing on a single protein or gene across a broad number of taxa (Storz, 2016; Storz et al., 2010). However, we know relatively little about convergence of biochemical pathways, or how long convergent adaptations to an environment may take to evolve. Moreover, when evolution converges upon the same predictable phenotypes, does this process occur quickly over short evolutionary time scales or does it take longer durations to evolve?

Birds that have adapted to the challenges of high altitude present a compelling system in which to explore the convergence of metabolic pathways in response to common environmental challenges. The cold and hypoxic environment at high altitude requires that endotherms maintain high rates of O2 consumption for locomotion and thermogenesis in O2-thin air (Bishop et al., 2015; Hayes, 1989). Flying birds face the additional challenge of maintaining lift with reductions in air density, which more than offset the metabolic savings from reductions in drag, such that birds flying at high altitude must flap their wings harder and maintain higher metabolic rates to stay aloft (Bishop et al., 2015). Both evolved and phenotypically plastic changes in respiratory physiology and metabolism are believed to help mitigate the challenges posed by the cold and hypoxic environment at high altitude (Beall, 2000; Lague et al., 2017; Monge and León-Velarde, 1991). In the bar-headed goose (Anser indicus), for example, evolutionary adaptations to high altitude appear to have arisen throughout the O2 transport pathway, including increases in effective ventilation, vital capacity and air-sac volume, haemoglobin-O2 affinity, capillarity of the flight muscle and heart, and oxidative capacity of the flight muscle (Jessen et al., 1991; McCracken et al., 2009a; Natarajan et al., 2015; Petschow et al., 1977; Scott et al., 2009a; Scott and Milsom, 2006; Scott and Milsom, 2007; Scott et al., 2011; Weibel, 1984; York et al., 2017; Zhang et al., 1996). However, except for studies of a few key proteins like hemoglobin (Natarajan et al., 2018; Natarajan et al., 2015; Projecto-Garcia et al., 2013; Storz et al., 2010), we still know little about whether convergent phenotypic changes have arisen across independent high-altitude lineages, particularly for the pathways of energy metabolism that support locomotion and thermogenesis. Metabolic genes have been outliers in genome scans of selection in high-altitude taxa (Qu et al., 2015), and recent studies in high-altitude populations of mice and humans point towards skeletal muscle as a common target of selection (Lundby and Calbet, 2016; Scott et al., 2018). However, the extent to which convergent reorganization of metabolic pathways has occurred across high-altitude taxa to help sustain locomotion and thermogenesis in hypoxia remains unclear, particularly across species that independently colonized high altitude in the same geographic region.

The activities of enzymes involved in energy metabolism are important determinants of capacity and flux of metabolic pathways (Kurata et al., 2007; Madhukar et al., 2015; Vogt et al., 2002a; Vogt et al., 2002b). Flux capacity is an emergent property of the contributions of several enzymes in a pathway that is set via hierarchical regulation, which determines the limits of metabolic fluxes that can be achieved via metabolic regulation (Suarez and Moyes, 2012), and the maximal activities of key enzymes can be valuable markers of flux capacity. Some previous studies suggest that the activities of some enzymes in major energy producing pathways differ in high-altitude natives compared to their low-altitude counterparts (Dawson et al., 2016; Leon-Velarde et al., 1993; Reynafarje, 1962; Rosser and Hochachka, 1993). However, similar differences in enzymatic activities have not been observed in other studies of high-altitude natives (Mathieu-Costello, 2001; Scott et al., 2018), and most previous work has been limited to a small number of enzymes and/or a single or a handful of species.

Here, we surveyed variation in 13 metabolic enzymes and myoglobin content across seven species encompassing four genera of high-altitude waterfowl (Family Anatidae) (Figure 1). Established markers of key metabolic pathways were chosen to provide a holistic view of energy metabolism during locomotion and thermogenesis, including aerobic and anaerobic glycolysis, fatty acid oxidation and mitochondrial function. We relied primarily on paired-lineage tests to make comparisons between high- and low-altitude taxa within a phylogenetic framework (Storz et al., 2010), but we also complemented these tests with standard ANOVA and phylogenetically independent contrasts (Felsenstein, 1985; Garland et al., 2005). We uncovered significant patterns of convergence in the remodeling of energy metabolism pathways in the major locomotor and thermogenic muscle, the pectoralis, between high- and low-altitude populations (summarized in Figure 1). Furthermore, by integrating population genetic data to infer how long each species has been established at high altitude, we show that some high-altitude phenotypes arose quickly whereas others required much longer evolutionary time to arise.

Enzyme pathway diagram illustrating where we observed differences in metabolic enzyme activity and myoglobin content in high-altitude waterfowl compared to their close low-altitude relatives.

In addition to the observed increases in myoglobin content, increases in the activities of hexokinase, ATP synthase, HOAD, and complex II (succinate dehydrogenase), and decreases in activities of pyruvate kinase, lactate dehydrogenase, creatine kinase and complex IV (cytochrome c oxidase), we observed no changes in activity for the enzymes citrate synthase, isocitrate dehydrogenase, malate dehydrogenase, complex I (NADH-ubiquinone oxidoreductase), and adenylate kinase.

Results and discussion

Diversity in the duration of high-altitude ancestry

We collected muscle samples of birds from a broad range of high-altitude sites in the Andes and from paired low-altitude sites, from species in the genera Anas (n = 76 specimens), Lophonetta (n = 21 specimens), Chloephaga (n = 20 specimens) and Oxyura (n = 16 specimens). Our sampling effort included seven Andean waterfowl species, subspecies, or populations that independently colonized high altitude at different times in geological history, and exhibit a range of divergence from their corresponding low-altitude population (Figure 2; Table 1). The species include strictly high-alpine specialists such as Andean goose (C. melanoptera, high-altitude range = 2000–5000 m above sea level) and puna teal (Anas puna, syn. Spatula puna, 3500–4600 m), which have diverged sufficiently from their low-altitude counterparts, Magellan goose (C. picta) and silver teal (Anas versicolor, syn. Spatula versicolor), to be classified as separate species (Fjdelsa and Krabbe, 1990). More recently diverged high-altitude populations of three dabbling duck species are classified as distinct subspecies, including crested duck (L. specularioides alticola, high-altitude range = 2000–5000 m; L. s. specularioides, low-altitude resident), speckled teal (A. flavoristris oxyptera, high-altitude range = 2500–4500 m; A. f. flavirostris, low-altitude resident), and cinnamon teal (Anas cyanoptera orinomus, syn. Spatula cyanoptera orinomus, high-altitude range = 2500–5000 m; Anas cyanoptera, syn. Spatula cyanoptera, low-altitude resident). The yellow-billed pintail (A. georgia spinicauda) has distinct populations occupying high altitude (up to 3500–4600 m) and low altitude, but they are not considered to be separate subspecies. Finally, we sampled a highly specialized diving duck that feeds extensively on aquatic insects, the ruddy duck, which has distinct subspecies at high altitude (O. jamaicensis ferruginea, high-altitude range = 2500–4500 m) and low altitude (O. j. jamaicensis).

Figure 2 with 1 supplement see all
Simplified phylogenetic tree, generated using maximum parsimony and constrained to the same topology as the global waterfowl phylogeny published by Gonzalez et al., 2009.

(see Figure 2—figure supplement 1). Branch lengths are measured as the total number of nucleotide substitutions in the 5’ end of the mtDNA control region.

Table 1
Seven species of Andean ducks showing classification level, ΦST, time since divergence (t/site), and the approximate time (T) ago in years they became established at high altitude based on coalescent analysis.

ΦST and t/site were calculated using previously published mtDNA sequences. T in years was calculated using the substitution rate published by Peters et al., 2005 of 4.8 × 10−8 substitutions/site/year.

Cinnamon tealYellow-billed pintailRuddy duckCrested duckPuna teal (H)
Silver teal (L)
Speckled tealAndean goose (H)
Magellan goose (L)
NewNewNewIntermediateEstablishedEstablishedEstablished
SubspeciesPopulationsSubspeciesSubspeciesSpeciesSubspeciesSpecies
ΦST = 0.07ΦST = 0.05ΦST = 0.38ΦST = 0.85ΦST = 0.93ΦST = 0.77ΦST = 1.0
t/site =
0.000143116
t/site =
0.00052227
t/site =
0.000806087
t/site =
0.003174242
t/site =
0.017886364
t/site =
0.019886364
t/site =
0.04547956
T (years) =
2982
T (years) =
10,898
T (years) =
16,793
T (years) =
66,130
T (years) =
372,633
T (years) =
414,219
T (years) =
947,491
Capture range
HA = 3812 m
LA = 0–13 m
Capture range
HA = 3812 m
LA = 3 m
Capture range
HA = 3812 m
LA = 480–507 m
Capture range
HA = 4281–4655 m
LA = 760–1050 m
Capture range
HA = 3812 m
LA = 410–485 m
Capture range
HA = 4209–4657 m
LA = 760–1050 m
Capture range
HA = 4368–4806 m
LA = 0–27 m
HA (n = 8)
LA (n = 8)
HA (n = 8)
LA (n = 10)
HA (n = 6)
LA (n = 10)
HA (n = 12)
LA (n = 10)
HA (n = 11)
LA (n = 10)
HA (n = 11)
LA (n = 10)
HA (n = 12)
LA (n = 8)

In sum, two taxon pairs represent deeply diverged sister species living at high and low altitude, four taxon pairs (including the diving duck) represent intermediate divergence between subspecies within species, and the last pair represents shallow divergence between populations of the same subspecies (Figure 2, Table 1). To quantify duration of high-altitude ancestry, for each taxon pair we used previously published sequences from the mitochondrial DNA (mtDNA) control region to calculate population genetic parameters including: (a) the fixation index ΦST, which measures nucleotide diversity (π) differences reflecting population subdivision, and (b) time since divergence between high and low altitude, as measured using a coalescent model incorporating drift and gene flow (Hey, 2005; Hey and Nielsen, 2004). We thus were able to reconstruct the rank order in time that these populations separated from each respective ancestral low-altitude population and order them according to the time they may have first become established in the Andean highlands.

Convergence across multiple pathways of energy metabolism

There were convergent decreases in the activities of multiple glycolytic enzymes (LDH and PK) and in creatine kinase (CK) across high-altitude taxa (Figure 3). The reductions in CK activity in particular provided one of the strongest cases for convergent changes in high-altitude waterfowl (>50% reduction in highland populations in all species). High-altitude taxa exhibited a significant reduction in CK activity using Wilcoxon signed-rank test (p<0.001; Figure 3). There was also a significant main effect of altitude (p<0.0001; Supplementary file 1b) using two-factor ANOVA and a negative correlation between CK activity and altitude using phylogenetically independent contrasts (PICs) (p<0.0001; Supplementary file 1f). These changes could have been a plastic response to chronic hypoxia, as observed in humans exposed to high altitude (Levett et al., 2015; Viganò et al., 2008). Since most of the CK in muscle is cytosolic, reductions in CK activity likely reflect a reduction in cytosolic ATP buffering capacity and/or a shift towards a more oxidative phenotype. CK is also expressed in mitochondria of muscle, where it is specialized for high-energy phosphate transfer as part of the phosphocreatine shuttle, a process that could also be impaired if mitochondrial CK activity is reduced in high-altitude waterfowl. However, the activity of the phosphocreatine shuttle is augmented, not reduced, in high-altitude bar-headed geese, and mitochondrial CK expression is elevated in high-altitude deer mice (Lui et al., 2015; Scott et al., 2009b). These results suggest that decreases in CK activity, along with reductions in LDH and PK activities, may be part of a general strategy to down-regulate some contributors to substrate-level phosphorylation in the muscle of high-altitude waterfowl. However, hexokinase (HK) activity was elevated in some of the least established high-altitude taxa (Figure 3D; Supplementary file 1f), so the decreases in LDH and PK activities are not associated with general reductions in capacity across glycolysis.

Metabolic enzyme activities for (A) lactate dehydrogenase (LDH), (B) pyruvate kinase (PK), (C) creatine kinase (CK), (D) hexokinase (HK), and (E) 3-hydroxyacyl-CoA dehydrogenase (HOAD), measured in the pectoralis of high- and low-altitude waterfowl.

The diagonal represents the line of equality (x = y). Values are shown as mean ± SEM U/g tissue (n = 8–12). High-altitude values are significantly different overall from the corresponding low-altitude values when p<0.05 in Wilcoxon’s Signed-Rank Tests, which were carried out including (*) and excluding (†) ruddy ducks.

Several high-altitude waterfowl also exhibited increases in HOAD activity (1.2 to 2.2-fold), which likely increases the capacity for beta-oxidation of fatty acids (Figure 3). HOAD activity was significantly elevated in highland taxa using Wilcoxon’s signed-rank test (p<0.001; Figure 3), and there was a significant main effect of altitude using twofactor ANOVA (p<0.0001; Supplementary file 1b) and a positive correlation between HOAD activity and altitude using PIC (p=0.0001; Supplementary file 1g). The peroxisome proliferator-activated receptors (PPAR) are key regulators of the expression of genes encoding the mitochondrial trifunctional enzyme (the heterooctamer that catalyzes the HOAD step and two additional steps in beta-oxidation) and other enzymes in beta-oxidation (i.e., acyl-CoA dehydrogenases), as well as genes controlling fatty-acid transport into mitochondria (Fan and Evans, 2015), so increases in HOAD activity in the flight muscle could reflect a general increase in the capacity for lipid oxidation. However, fatty-acid oxidation is strongly regulated by the enzymes involved in mitochondrial lipid uptake (e.g., carnitine palmitoyl transferase), such that changes in fatty acid oxidation can arise without changes in HOAD activity (Morash et al., 2013). Nevertheless, if increases in HOAD activity are indeed associated with increased capacity for lipid oxidation in the flight muscle, such changes would amplify the already remarkable capacity of birds to support high rates of muscle metabolism and power on lipids alone (Guglielmo, 2010; O'Brien and Suarez, 2001; Suarez et al., 1986). Sustained thermogenesis relies heavily on lipid oxidation (Marsh and Dawson, 1989; Swanson and Thomas, 2007; Vaillancourt et al., 2009; Vaillancourt et al., 2005), such that cold temperatures could have increased lipid metabolism and stimulated a corresponding rise in beta-oxidation capacity in high-altitude birds. Furthermore, there seems to be a positive association between altitude and body lipid content in insects (Parkash et al., 2008; Sømme et al., 1996), which could increase dietary lipid availability at high altitude for waterfowl (most of which forage on aquatic insects). There is also evidence showing a positive correlation between dietary lipids and HOAD activity in the pectoralis of both migratory and non-migratory birds (Guglielmo, 2010; Maillet and Weber, 2007; Nagahuedi et al., 2009). If high-altitude birds eat more lipid-rich foods than their low-altitude counterparts, then a corresponding increase in lipid metabolism might have stimulated the rise in beta-oxidation capacity.

There were also convergent changes in complexes II and IV of the electron transport system and in ATP synthase (F1FO-ATPase, complex V) (Figure 4). The activities of all TCA cycle enzymes assayed, including citrate synthase, isocitrate dehydrogenase, and malate dehydrogenase, were similar between high- and low-altitude populations (Supplementary file 1a). Citrate synthase in particular is a common marker of mitochondrial volume density in muscle tissue (Boushel et al., 2007; Dawson et al., 2018; Larsen et al., 2016; Mahalingam et al., 2017; Mogensen et al., 2006), so this result suggests that muscle mitochondrial content was similar between high- and low-altitude populations. The observed changes in complexes II, IV, and V activities could therefore reflect a change in mitochondrial quality affecting the function of a given amount of mitochondria. Complex II activity was higher in multiple high-altitude species (1.05- to 1.51-fold increases, with no change in complex I activity; Supplementary file 1a) using Wilcoxon’s signed-rank test (p<0.05; Figure 3), and there was a significant main effect of altitude using two-factor ANOVA (p=0.0060; Supplementary file 1b) and a positive correlation between complex II activity and altitude using PIC (p=0.0213; Supplementary file 1f). ATP synthase activity was also higher in multiple (but not all) high-altitude taxa using Wilcoxon’s signed-rank test (p<0.05; Figure 4), supported by results of two-factor ANOVA (altitude effect, p<0.0001; Supplementary file 1b) and the positive correlation between ATP synthase activity and altitude using PIC (p=0.0009; Supplementary file 1f). In contrast, the terminal acceptor for oxygen, cytochrome c oxidase (COX; complex IV), had ~50% lower activity when compared to low-altitude sister taxa across nearly all high-altitude species except the ruddy duck. In fact, there seemed to be a narrow optimum for the activity of COX across high-altitude waterfowl, as all species converged on a strikingly similar value (Figure 4B). COX activity showed a significant reduction in highland taxa using Wilcoxon’s signed-rank test (p<0.05; Figure 3), and there were a significant main effect of altitude in two-factor ANOVA (p<0.0001; Supplementary file 1b) and a negative correlation between COX activity and altitude using PIC (p=0.0001; Supplementary file 1f). Unique specializations in the activity, structure and function of COX have been observed in the locomotory muscles of several high-altitude taxa (Dawson et al., 2016; Lui et al., 2015; Scott et al., 2011; Sheafor, 2003). A similar reduction in COX activity (~50% less) was also observed in the cardiac muscle of bar-headed goose compared to low-altitude geese, in association with an increased affinity for cytochrome c (Scott et al., 2011). Similarly, COX of some hypoxia-tolerant fish has decreased activity but a greater affinity for O2 (Lau et al., 2017). Therefore, hypoxia may drive convergent changes in COX and mitochondrial function across vertebrates.

Mitochondrial enzyme activities for (A) Complex II, (B) Complex IV, and (C) ATP synthase measured in the pectoralis of high- and low-altitude waterfowl.

The diagonal represents the line of equality (x = y). Values are shown as mean ± SEM U/g tissue (n = 8–12). High-altitude values are significantly different overall from the corresponding low-altitude values when p<0.05 in Wilcoxon’s Signed-Rank Tests, which were carried out including (*) and excluding (†) ruddy ducks.

Several high-altitude waterfowl exhibited elevated myoglobin in the pectoralis muscle when compared to low-altitude ducks. Myoglobin content showed a significant main effect of altitude (p=0.0077; Supplementary file 1b) using twofactor ANOVA and this effect was nearly significant in Wilcoxon’s signed-rank test (0.1 > P > 0.05; Figure 5B) and in PIC correlations between myoglobin content and altitude (p=0.0706; Supplementary file 1g). Myoglobin content was variable across species and tended to be greatest in ducks with the highest body mass (Supplementary file 1a), and the effect of body mass was nearly significant (p=0.0585; Supplementary file 1d). Body mass, however, was not significantly greater in high-altitude waterfowl (p=0.9707; Supplementary file 1b), suggesting that the elevated levels of myoglobin in some high-altitude taxa is not simply due to increased body size. Elevated myoglobin content in flight muscle may serve to increase cellular O2 stores and to facilitate intracellular O2 diffusion (Kanatous et al., 2009; Wittenberg and Wittenberg, 2003). Elevated myoglobin content could potentially augment intracellular lipid transport as well, as there are some suggestions that myoglobin may bind and facilitate fatty acid diffusion through the sarcoplasm (Gros et al., 2010). Elevated myoglobin content or transcript expression have been previously observed in some other taxa that reside at and/or were acclimatized to high altitude, including torrent ducks (Mergantta armata) (Dawson et al., 2016), dogs (Gimenez et al., 1977), rats (Vaughan and Pace, 1956), and Tibetan humans (Moore et al., 2002). Therefore, elevated myoglobin levels in the locomotory muscle appear to be an important strategy across high-altitude taxa for augmenting mitochondrial O2 availability (and possibly intracellular lipid transport) and thus sustaining thermogenesis and locomotion in hypoxia at high altitude.

Myoglobin content measured in the pectoralis of high- and low-altitude waterfowl.

The diagonal represents the line of equality (x = y). Values are shown as mean ± SEM mg/g tissue (n = 8–12). High-altitude values are significantly different overall from the corresponding low-altitude values when p<0.05 in Wilcoxon’s Signed-Rank Tests, which were carried out including (*) and excluding (†) ruddy ducks.

Idiosyncratic changes in high-altitude ruddy ducks

In many cases, the ruddy duck, the most distantly related and the only diving species studied herein, showed a contrasting pattern of changes compared to the other species that forage on the surface of water (dabbling ducks) or graze on land (sheldgeese). Low-altitude ruddy ducks seem to have fundamentally different physiological and biochemical characteristics than the other low-altitude taxa in this study, with relatively high activities of HOAD, complex II, and HK, relatively low activities of complex IV (Figures 36), and a relatively high haemoglobin-O2 affinity atypical of low-altitude populations but more similar to high-altitude waterfowl populations (Natarajan et al., 2015). Compared to this low-altitude relative, high-altitude ruddy ducks had decreased activities of both complex II and HOAD in the flight muscle, in contrast to most/all other high-altitude taxa in which the activities of these enzymes were elevated (Figure 3C; Figure 4A), and they also do not show the typical and expected increase in Hb-O2 affinity that is seen in the other high-altitude waterfowl from this study system (Natarajan et al., 2015) and in many other high-altitude birds (Storz, 2016). The distinct direction of the differences between high- and low-altitude ruddy ducks appeared to strongly contribute to the significant species × altitude interactions that were detected in two-factor ANOVAs for PK (p=0.0003), HOAD (p<0.0001), complex II (p<0.0001), and complex IV (p<0. 001) activities (Supplementary file 1b). It is possible that diving created a unique set of physiological challenges for this species as it invaded high-altitude habitats; however, we do not see similar changes in enzyme activities in the pectoralis (flight muscle) or gastrocnemius (swimming muscle) of torrent ducks at high altitude, another diving duck species native to the Andes (Dawson et al., 2016). It is more likely that the unique features of low-altitude ruddy ducks – related to diet, physiology, biochemistry, etc. – favoured distinct mechanisms of adaptation or plasticity during the process of high-altitude colonization.

Changes over evolutionary time at altitude of (A) ATP synthase activity and (C) hexokinase activity measured in thepectoralis of seven high- and low-altitude waterfowl pairs.

Values are given as the mean ± SEM U/g tissue (n = 8–12). * - Significantly different activity in high-altitude ducks compared to low-altitude ducks (two-factor ANOVA followed by the Bonferroni post-tests; p<0.05). (B) ATP synthase and (D) Hexokinase activities in each high-altitude taxon plotted against the t/site value between each high-low pair.

Some metabolic changes arise only after prolonged evolutionary time at high altitude

One of the clear advantages to these particular high-altitude waterfowl as a study system is the ability to infer and rank order the evolutionary time these populations have lived at high altitude, and thus provide insight into metabolic changes that take longer evolutionary times to arise. Using population genetic data for the mtDNA control region, we determined that the high-altitude taxa studied here most likely diverged from their low-altitude sister taxa for evolutionary times that differed by as much as three orders of magnitude, from within the last several thousands of years (i.e., more recently established high-altitude populations) to approximately a million years in the case of species that are now established high-altitude endemic species (Table 1). In each case, all available evidence indicates that high-altitude taxa were derived from low-altitude ancestral populations and not the reverse (Bulgarella et al., 2014; Bulgarella et al., 2012; Fjeldså, 1985; Graham et al., 2018; McCracken et al., 2009b; McCracken et al., 2009c; Muñoz-Fuentes et al., 2013; Wilson et al., 2013). This approach allowed us to broadly classify high-altitude taxa into three groups based on their mtDNA divergence: newly established highland populations (~3,000–17,000 years; cinnamon teal, yellow-billed pintail and ruddy duck), intermediate (66,000 years; crested duck), and long established subspecies/species (372,000–947,000 years; puna teal, speckled teal, Andean goose) (Table 1). These time since divergence (T) calculations, while approximations, are derived from t/site (τ) values representing the average tree height of all genealogies in each coalescent analysis divided by the mutation rate (μ), assuming a mutation rate of 4.8 × 10–8 substitutions per site per year (Peters et al., 2005) as previously applied to these same species. In support of our inferences about duration of high-altitude ancestry, a very similar rank order is also seen with ΦST (Table 1), which provides an independent estimate of divergence not dependent on any particular mutation rate. Using this rank order, we predicted that some metabolic pathways would only change in the more established lineages (i.e., deeper T, higher ΦST) that have occupied the high-altitude environment for longer and have therefore experienced hypoxia for longer evolutionary time, therefore with more time for adaptation to proceed.

Following this classification, ATP synthase activity was elevated in only the most established high-altitude taxa, whereas no differences were observed in the newly established populations (Figure 6A,B). The magnitude of these increases in established high-altitude taxa were appreciable, ranging from 2.0- to 3.1-fold, and they appeared to strongly contribute to the significant species × altitude interaction that was detected in two-factor ANOVA (p=0.0342; Supplementary file 1b). These changes would have led to a strong increase in the capacity for ATP synthesis relative to electron transport. Such a change could reduce mitochondrial membrane potential and attenuate the production of reactive oxygen species (ROS), which may be advantageous at high altitude for reducing oxidative stress (Brand et al., 1999; Korshunov et al., 1997; Skulachev, 1996). Increasing ATP synthesis capacity may also reduce the magnitude of phosphorylation control over mitochondrial respiration, shifting more control towards the electron transport system (Gnaiger et al., 2000; Jacobs et al., 2012; Pesta and Gnaiger, 2012).

Our results also show that newly established populations at high altitude show increased HK activity in comparison to low-altitude populations (Figure 3D; Figure 6C,D). Most of the low-altitude taxa had very similar HK activities in the flight muscle (with the exception of ruddy ducks). Only newly established and intermediate high-altitude populations had elevated HK activities, ranging from 2.20- to 3.56-fold, which likely contributed to the significant species × altitude interaction that was detected in two-factor ANOVA (p<0.0001; Supplementary file 1b). These transient increases in HK activity along with the convergent reductions in PK and LDH activity suggest that there are broad changes across glycolysis in high-altitude taxa, potentially associated with changes in carbohydrate oxidation and/or lactate production. Hypoxia exposure can lead to plastic increases in the reliance on carbohydrate oxidation in mammals (Hochachka et al., 1998; McClelland et al., 1998; Robin et al., 1984) and may be useful at high altitude by generating more ATP per molecule of O2 than lipid oxidation. However, carbohydrate oxidation may be constrained over time by glycogen stores if carbohydrate fuels cannot be adequately supplied by the circulation (McClelland, 2004), and we observed that the rise in HK activity in newly-established populations returned to low-altitude levels in the most established high-altitude taxa (Figure 6C,D). These particular high-altitude taxa may instead take advantage of more plentiful but O2-costly fuels (i.e., lipids), provided that sufficient tissue O2 supply is maintained by evolved or plastic changes throughout the O2 transport pathway (e.g., increases in haemoglobin O2 affinity, expansion of pulmonary O2 diffusing capacity) (Maina et al., 2017; Natarajan et al., 2015).

Conclusions

Convergent changes have occurred in many pathways of metabolism across high-altitude waterfowl, with increases in capacity for beta oxidation of lipids and adjustments in the activity of oxidative phosphorylation (OXPHOS) enzymes that likely fine-tune mitochondrial function (Figure 1). However, some changes required longer evolutionary time at high altitude to arise, suggesting that adaptive changes in high-altitude taxa may involve several steps, such that some changes are only observed in the longest-established highland taxa. Indeed, some changes in the activities of enzymes involved in metabolizing lipids and carbohydrates may be convergent across taxa (increased HOAD activity, decreased PK and LDH activities), whereas others are more time-dependent. In the latter case, HK activity is elevated in relatively new high-altitude colonists, but is subsequently blunted over evolutionary time. Similar distinctions between convergent (e.g., reduced complex IV activity) and time-dependent (increase in complex V activity in only the most established highland taxa) changes exist for mitochondrial OXPHOS enzymes. It is likely that high-altitude animals rely upon hypoxia acclimatization (Hochachka et al., 1998) when first colonising high altitude, followed by evolved physiological specializations that adjust the capacity and flux of metabolic pathways, along with evolved improvements in mitochondrial O2 supply that arise from increases in tissue capillarity (Leon-Velarde et al., 1993; Mathieu-Costello et al., 1998; Scott et al., 2009a), Hb-O2 affinity (Galen et al., 2015; Natarajan et al., 2015; Projecto-Garcia et al., 2013), circulatory O2 delivery, and/or pulmonary O2 uptake (Calbet et al., 2003; Maina et al., 2017; McClelland and Scott, 2019). Our data suggest that increases in capacity for beta oxidation, changes in capacity across glycolysis, and adjustments in mitochondrial function are common strategies to cope with the challenges of high altitudes, but that longer time scales of evolutionary adaptation can be required to fully converge upon the ultimate high-altitude phenotype.

Materials and methods

Key resources table
Reagent type
(species) or resource
DesignationSource or referenceIdentifiersAdditional
information
Chemical compound, drugGlucoseSigma
G8270
D-(+)-Glucose
≥99.5% (GC)
Enzyme assay reagent
Chemical compound, drugATPSigma
A2383
Adenosine 5’-triphosphate disodium salt hydrate
Grade I,≥99%, from microbial
Enzyme assay reagent
Chemical compound, drugMgCl2Sigma
M8266
Magnesium Chloride anhydrous,≥98%Enzyme assay reagent
Chemical compound, drugNADP+BioShop Canada NAD007B-NADP, Disodium trihydrate,>95%Enzyme assay reagent
Chemical compound, drugG6PDHRoche
10127655001
Glucose-6-Phosphate Dehydrogenase (G6P-DH) grade I, from yeastEnzyme assay reagent
Chemical compound, drugLDHRoche
10127876001
L-Lactate Dehydrogenase (L-LDH) from rabbit muscleEnzyme assay reagent
Chemical compound, drugPEPSigma
P7002
Phosphoenolpyruvic acid trisodium salt hydrate
≥97% (enzymatic)
Enzyme assay reagent
Chemical compound, drugADPSigma
A5285
Adenosine 5’-diphosphate monopotassium salt dehydrate bacterial,≥95%, powderEnzyme assay reagent
Chemical compound, drugPyruvateSigma
P2256
Sodium pyruvate
ReagentPlus,≥99%
Enzyme assay reagent
Chemical compound, drugNADHBioShop Canada NAD002NADH ß-NICOTINAMIDE ADENINE REDUCEDEnzyme assay reagent
Chemical compound, drugOxaloacetateSigma
O4126
Oxaloacetic acid
≥97% (HPLC)
Enzyme assay reagent
Chemical compound, drugAcetyl CoABioShop Canada ACO201ACETYL COENZYME A, Trilithium SaltEnzyme assay reagent
Chemical compound, drugDTNBSigma
D218200
5,5’-Dithiobis(2-nitrobenzoic acid)
ReagentPlus, 99%
Enzyme assay reagent
Chemical compound, drugIsocitrateSigma
I1252
DL-Isocitric acid trisodium salt hydrate
≥93%
Enzyme assay reagent
Chemical compound, drugCoQ10Sigma
C9538
Coenzyme Q10,
≥98% (HPLC)
Enzyme assay reagent
Chemical compound, drugRotenoneSigma
R8875
Rotenone,
≥95%
Enzyme assay reagent
Chemical compound, drugBSASigma
A6003
Bovine Serum Albumin lyophilized powder, essentially fatty acid free,≥96% (agarose gel electrophoresis)Enzyme assay reagent
Chemical compound, drugKCNSigma
60178
Potassium cyanide
BioUltra,≥98.0% (AT)
Enzyme assay reagent
Chemical compound, drugSuccinateSigma
S2378
Sodium succinate dibasic hexahydrate
ReagentPlus,≥99%
Enzyme assay reagent
Chemical compound, drugDCPIPSigma
D1878
2,6-Dichloroindophenol sodium salt hydrate,
BioReagent
Enzyme assay reagent
Chemical compound, drugDUBSigma
D7911
Decylubiquinone,
≥97% (HPLC)
Enzyme assay reagent
Chemical compound, drugCytcCH2Sigma
C7752
Cytochrome c from equine heart
≥95% based on Mol. Wt. 12,384 basis
Enzyme assay reagent
Chemical compound, drugOligomycinSigma
O4876
Oligomycin from Streptomyces diastatochromogenes
≥90% total oligomycins basis (HPLC)
Enzyme assay reagent
Chemical compound, drugHKRoche
11426362001
Hexokinase (HK)Enzyme assay reagent
Chemical compound, drugAcetoacetyl CoASigma
A1625
Acetoacetyl coenzyme A sodium salt hydrate
Cofactor, for acyl transfer
Enzyme assay reagent
Chemical compound, drugCreatineSigma
C3630
Creatine monohydrate,
≥98%
Enzyme assay reagent
Chemical compound, drugPKRoche
PK-RO
Pyruvate Kinase (PK) from rabbit muscleEnzyme assay reagent
Chemical compound, drugKH2PO4P5378Potassium phosphate monobasic,
ReagentPlus
Assay buffer reagent
Chemical compound, drugEGTASigma
E4378
Ethylene glycol-bis(2-aminoethylether)-N,N,N’,N’-tetraacetic acid,
≥97.0%
Assay buffer reagent
Chemical compound, drugEDTASigma
EDS
Ethylenediaminetetraacetic acid
BioUltra, anhydrous,≥99% (titration)
Assay buffer reagent
Chemical compound, drugTriton-X 100Sigma
X100
Triton X-100
laboratory grade
Assay buffer reagent
Software, algorithmGeneiousBiometters Ltd., Auckland, NZUsed for sequence alignment
Software, algorithmPAUPVersion 4, Sinauer Associates, Sunderland, Massachusetts, USAUsed to generate branch lengths
Software, algorithmMESQUITEhttps://www.mesquiteproject.org/Used to analyze phylogenetic contrasts
Software, algorithmPDAP modulehttp://mesquiteproject.org/pdap_mesquite/Used to analyze phylogenetic contrasts
Software, algorithmIMhttps://bio.cst.temple.edu/~hey/softwareUsed to calculate divergence

Tissue sampling

Request a detailed protocol

Waterfowl were captured at high altitudes (3822–4806 m) or at low altitudes (0–1050 m) in various locations in and near the Andes across South America and from low-altitude sites in North America. Tissues from some birds were sampled immediately on site, whereas others were sampled after birds were transported to a nearby field lab where they were provided with unlimited access to water for 12–18 hr, prior to being euthanized. In all cases, samples of pectoralis muscles were quickly dissected and frozen in liquid N2 and stored at −80°C for enzyme analysis (see below). Muscle samples were taken at three depths in the middle of the tissue (surface, intermediate, and deep), in order to account for heterogeneity of muscle fibers throughout the pectoralis This heterogeneity is important to consider, because flight muscle tends to become more oxidative at deeper depths from the ventral surface (Scott et al., 2009a), as confirmed by the variation observed here (Supplementary file 1j).

Samples were imported to Canada with authorization from the Canadian Wildlife Service (Scientific Possession No. 369) and collected with authorization from the Servicio Nacional de Area Naturales Protegidas del Peru (004–2014-SERNANP-DGANP-RNT/J), Dirección General Forestal y de Fauna Silvestre del Peru (RD 169–2014 MIN AGRI-DGFFS/DGEFFS, 190–2015-SERFOR-DGGSPFFS), Ministerio de Industria, Agricultura, y Ganaderia Chubut (No. 24/07 y 1636/14), Ministerio de Asuntos Agrarios Buenos Aires, Ministerio de Producción de Entre Rios, Oregon Department of Fish and Wildlife (101-15), and USFWS Region 1 Migratory Bird Permit Office (MB68890B-0). All protocols were carried out in accordance with guidelines that were approved by the Institutional Animal Care and Use Committee at the University of Miami or University of Alaska.

Phylogenetic tree generation

Request a detailed protocol

Mitochondrial DNA (mtDNA) sequences were obtained for each population in this study. Most were available from previously published data sets including McCracken et al., 2009c for A. georgica, Bulgarella et al., 2012 for Lophonetta specularioides, Wilson et al., 2013 for A. cyanoptera (syn. Spatula cyanoptera), Muñoz-Fuentes et al., 2013 for Oxyura jamaicensis, Bulgarella et al., 2014 for Chloephaga spp., and Graham et al., 2018 for A. flavirostris. To this, we also supplemented unpublished mtDNA sequences from A. puna and A. versicolor (syn. Spatula spp.) (Wilson et al., 2013. Dissertation 2010). The sequence we utilized comprise ~684 bp of the mtDNA control region corresponding to previously published primers L78-H774 (Johnson and Sorenson, 1999; Sorenson and Fleischer, 1996). GenBank accession number can be found in the referenced articles and in supplementary materials (Supplementary file 1i). As each species possessed numerous small indels, alignment was performed in Geneious (Biomatters Ltd., Auckland, NZ). Indels were treated as missing data, and the resulting alignment was refined by eye to correct ambiguities. All sequences were obtained using PCR and capillary DNA sequencing protocols as described in McCracken et al., 2009a. Next, we generated a tree with branch lengths constrained to match the Gonzalez et al., 2009 topology, which is the most recent phylogenetic analysis of waterfowl to include all of these species. Branch lengths for this tree were generated using maximum parsimony in the software PAUP (Version 4, Sinauer Associates, Sunderland, Massachusetts, USA).

Enzyme activities and myoglobin assays

Request a detailed protocol

The maximal activities of 13 enzymes as well as myoglobin content were assayed as previously described (Dawson et al., 2016). The reported enzyme activity content for each individual bird is the average value from the samples taken across the three depths of muscle (which are fully reported in Supplementary file 1j). Wilcoxon signed-rank test (paired lineage tests; Figures 35) as well as two-factor ANOVA followed by Bonferroni post-tests (Figure 6; Supplementary file 1b, Supplementary file 1c) were used to make statistical comparisons between high-altitude versus low-altitude taxa. The maximal activities of 13 enzymes as well as myoglobin content were assayed as previously described (Dawson et al., 2016) using a Spectramax Plus 384 spectrophotomer (Molecular Devices, Sunnyvale, CA, USA) at avian body temperature of 41°C. Samples were homogenized in 10 volumes of ice-cold homogenizing buffer [100 mM KH2PO4 buffer, pH 7.2, containing 1 mM EGTA, 1 mM EDTA, 0.1% Triton-X 100, and 1 mM phenylmethylsulfonyl fluoride (PMSF)]. Homogenates were then centrifuged at 1,000 × g at 4°C and the supernatant was collected for use in assays. Enzyme activities were determined for each sample as the difference between the rate measured using all assay components (assayed in triplicate) and the background reaction rate, all measured at avian body temperature (41°C). Measurements were carried out as described in Supplementary file 1h. Preliminary experiments determined that all substrate concentrations were saturating. Enzyme activities are expressed in units of micromole substrate per gram tissue per minute, with protein concentrations determined using the Bradford method (BioRad, Mississauga, ON, Canada). Myoglobin content was assayed in triplicate using a modified version of the original method (Reynafarje, 1962), as described in Dawson et al., 2016. Biochemicals were from Sigma-Aldrich (Oakville, ON, Canada) unless otherwise stated.

Wilcoxon signed-rank test (paired lineage tests; Figures 35) as well as two-factor ANOVA followed by Bonferroni post-tests (Figure 6; Supplementary file 1b, Supplementary file 1c) were used to make statistical comparisons between high-altitude versus low-altitude taxa. Data are presented as means ± SE. p<0.05 was considered significant. When significant interactions occurred, they were often attributed primarily to opposing patterns observed in the ruddy ducks. The exceptions were HK and ATPase, for which the significant interactions were attributed to changes that appeared to be associated with differences in evolutionary time at high altitude.

The linear relationship between mass and enzyme activity or myoglobin content was assessed in order to determine if enzyme activity or myoglobin content varies with the mass of individual waterfowl independent of altitude. Enzyme activity or myoglobin content was plotted on the y-axis against mass on the x-axis from both high- and low-altitude waterfowl and the slope, as well as, goodness of fit (r2) were determined (Supplementary file 1d, Supplementary file 1e). A significant relationship between mass and enzyme activity was determined when p<0.05.

Phylogenetically independent contrasts

Request a detailed protocol

To conduct analysis of phylogenetically independent correlations (PIC), the maximum parsimony tree constrained to match the same topology as Gonzalez et al., 2009 global waterfowl phylogeny (Fig. S1) and branch lengths were imported into MESQUITE (Maddison and Maddison, 2016). The PDAP module (Midford, 2010) was used to carry out PIC analyses to assess whether the relationship between altitude and enzyme activity persisted after taking into account the effects of phylogeny (Supplementary file 1f, Supplementary file 1g). Significant relationships between raw contrasts and their standard deviations were not generally observed, and only in a few cases when ruddy ducks were included in the analysis. In such cases, we used exponential transformation of branch lengths to eliminate these significant relationships before carrying out PIC correlations (Garland et al., 1992). We observed similar results in PIC analyses using branch lengths that were set to one (data not shown).

Time at altitude and time since divergence

Request a detailed protocol

To quantify duration of high-altitude ancestry, we used two methods for each taxon pair to calculate population genetic parameters indicative of the evolutionary time each highland population has likely been at high altitude, and to examine the time dependence of activity of enzymes. First, we calculated the pair-wise fixation index ΦST between each pair of low- and high-altitude sister populations. We utilized the same previously published mtDNA control region data sets for these populations (Bulgarella et al., 2014; Bulgarella et al., 2012; Graham et al., 2018; McCracken et al., 2009c; Muñoz-Fuentes et al., 2013; Wilson et al., 2013). This index of population differentiation, comparable to Wright’s (1965) fixation index FST, is bounded by 0 and 1 (Wright, 1965). Thus, whereas pairs of populations with ΦST closer to 0 are expected to be recently diverged, ΦST closer to one is indicative of populations with much older divergence, as is the case for fully differentiated species that have ceased gene flow. Populations with intermediate ΦST are expected to fall somewhere in the middle of this continuum. This approach allowed us to relate metabolic distinctiveness between high- and low-altitude populations to their relative magnitude of genetic divergence. Second, we used a two-population coalescent model in the software IM (Hey Lab, Temple University) to calculate the population divergence parameter, t/site, which represents the average tree height of all genealogies in each coalescent analysis (Hey, 2005; Hey and Nielsen, 2004). Multiplied by the mutation rate (μ) this parameter can then be used to calculate time since divergence (T) in years. This has advantages over ΦST because it allowed us to incorporate evolutionary processes in our model including divergence due to genetic drift, and therefore uncertainty in our estimates, as well as gene flow (i.e., migration in both directions) for which some of these populations have been shown to exhibit more connectivity than others. Starting parameters for these analyses were conditioned on the data by first making several preliminary runs using wide priors and described as published previously (Bulgarella et al., 2014; Bulgarella et al., 2012; Graham et al., 2018; McCracken et al., 2009c; Muñoz-Fuentes et al., 2013; Wilson et al., 2013). Finally, we were able to convert t/site to time in years using the substitution rate for the 5’ end of the control region published by Peters et al., 2005 of 4.8 × 10−8 substitutions/site/year (Peters et al., 2005). This provided an approximation to the date at which these high-altitude populations might have experienced a founder event following the colonization of high altitude for the first time, whereas the rank order of ΦST also likely corresponds to their rank order newest to oldest high-altitude residents. Finally, it should be noted that the Andes have been rising throughout the Cenozoic for approximately 80 million years (Ramos, 2005; Royden et al., 2008), and while some parts of the Andes uplifted more recently than others, the oldest of these waterfowl populations (i.e., South American sheldgeese) are probably diverged no more than several millions years ago (Johnson and Sorenson, 1999). Therefore, in all cases the Andes had already uplifted to close to their present height when these high-altitude populations first became established.

References

  1. 1
    Tibetan and andean patterns of adaptation to high-altitude hypoxia
    1. CM Beall
    (2000)
    Human Biology 72:201–228.
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10
  11. 11
  12. 12
    Birds of the High Andes: A Manual to the Birds of the Temperate Zone of the Andes and Patagonia, South America
    1. J Fjdelsa
    2. N Krabbe
    (1990)
    Apollo Books.
  13. 13
  14. 14
  15. 15
  16. 16
  17. 17
  18. 18
  19. 19
  20. 20
  21. 21
  22. 22
  23. 23
  24. 24
  25. 25
  26. 26
  27. 27
  28. 28
  29. 29
  30. 30
  31. 31
  32. 32
  33. 33
  34. 34
  35. 35
  36. 36
  37. 37
  38. 38
  39. 39
  40. 40
    Mesquite: a modular system for evolutionary analysis, version 3.04
    1. W Maddison
    2. D Maddison
    (2016)
    Mesquite: a modular system for evolutionary analysis.
  41. 41
  42. 42
  43. 43
  44. 44
  45. 45
    Avian adjustments to cold
    1. R Marsh
    2. W Dawson
    (1989)
    In: LCH Wang, editors. Animal Adaptation to Cold. Springer. pp. 205–253.
    https://doi.org/10.1007/978-3-642-74078-7_6
  46. 46
  47. 47
  48. 48
  49. 49
    Fat to the fire: the regulation of lipid oxidation with exercise and environmental stress
    1. GB McClelland
    (2004)
    Comparative Biochemistry and Physiology Part B: Biochemistry and Molecular Biology 139:443–460.
    https://doi.org/10.1016/j.cbpc.2004.07.003
  50. 50
  51. 51
  52. 52
  53. 53
  54. 54
    PDAP: PDTREE: a translation of the PDTREE application of Garland, et al.'s phenotypic diversity analysis programs, version 2.75
    1. P Midford
    (2010)
    PDAP: PDTREE: a translation of the PDTREE application of Garland, et al.'s phenotypic diversity analysis programs.
  55. 55
  56. 56
  57. 57
  58. 58
  59. 59
  60. 60
  61. 61
  62. 62
  63. 63
  64. 64
  65. 65
  66. 66
  67. 67
  68. 68
  69. 69
  70. 70
  71. 71
  72. 72
  73. 73
    Metabolic capacity of muscle fibers from high-altitude natives
    1. BW Rosser
    2. PW Hochachka
    (1993)
    European Journal of Applied Physiology and Occupational Physiology 67:513–517.
    https://doi.org/10.1007/BF00241647
  74. 74
  75. 75
  76. 76
  77. 77
  78. 78
  79. 79
  80. 80
    Control of breathing and adaptation to high altitude in the bar-headed goose
    1. GR Scott
    2. WK Milsom
    (2007)
    American Journal of Physiology-Regulatory, Integrative and Comparative Physiology 293:R379–R391.
    https://doi.org/10.1152/ajpregu.00161.2007
  81. 81
  82. 82
  83. 83
  84. 84
  85. 85
  86. 86
  87. 87
    Metabolic sources of energy for hummingbird flight
    1. RK Suarez
    2. GS Brown
    3. PW Hochachka
    (1986)
    American Journal of Physiology-Regulatory, Integrative and Comparative Physiology 251:R537–R542.
    https://doi.org/10.1152/ajpregu.1986.251.3.R537
  88. 88
  89. 89
  90. 90
  91. 91
  92. 92
  93. 93
  94. 94
  95. 95
  96. 96
    The Pathway for Oxygen: Structure and Function in the Mammalian Respiratory System
    1. ER Weibel
    (1984)
    Harvard University Press.
  97. 97
  98. 98
  99. 99
  100. 100
  101. 101

Decision letter

  1. Kevin Campbell
    Reviewing Editor; University of Manitoba, Canada
  2. Christian Rutz
    Senior Editor; University of St Andrews, United Kingdom
  3. Kevin Campbell
    Reviewer; University of Manitoba, Canada
  4. Andrew Murray
    Reviewer; University of Cambridge, United Kingdom

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

Thank you for submitting your article "Convergent changes in muscle physiology depend on duration of high-altitude ancestry across Andean waterfowl" for consideration by eLife. Your article has been reviewed by three peer reviewers, including Kevin Campbell as Guest Editor and Reviewer #1, and the evaluation has been overseen by Christian Rutz as the Senior Editor. The following individual involved in the review of your submission has agreed to reveal their identity: Andrew Murray (Reviewer #2).

The reviewers have discussed their reviews with one another, and the Guest 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).

Summary:

This study employed 7 high- and low-altitude species pairs that spanned ~1 million years of evolution in hypoxic environments to identify convergent and time-dependent trajectories in muscle metabolic biochemistry to high altitude environments. Importantly, since the work examines fuel and energy metabolism in flight muscle, rather than oxygen transport, it provides a valuable perspective on what is happening at the level of metabolic biochemistry. Overall, the experiments appear to have been carefully conducted, analyses seem sound (except the construction of the phylogenetic tree, which requires particular attention during revision), and the manuscript is well written.

Essential revisions:

1) The precision of writing related to evolution, convergence, and phenotypic plasticity could be improved. The first paragraph doesn't fully account for the wide variety of evolutionary studies of convergence and other processes, and sets up a fairly weak hypothesis about whether convergence is always the same or idiosyncratic because it sets up a black and white dichotomy. Evolution/adaptation/drift are always happening and of course each lineage will have an idiosyncratic process. They may arrive at generally similar phenotypes, but do you ever expect exact matching? The second question about the speed of adaptation is more interesting. Accordingly, the first paragraph or most of it could be dropped and the paper could be started with paragraph 2.

2) The interpretation of the results pertaining to glycolytic capacity is handled somewhat inconsistently throughout the manuscript. The Abstract states that glycolytic capacity is decreased at altitude, whereas the data for HK activity in particular does not support this conclusion with Figure 1 indicating greater HK activity at high altitude. The argument made in the main text is more subtle, supporting an increase in HK activity in many of the more recent high-altitude populations, but no difference in the more established residents. In fact, HK activity is not lower in any highland waterfowl studied compared with its low altitude counterpart with the exception of the ruddy duck (Supplementary file 1A). Moreover, the difference between the low- and high-altitude Andean goose species is almost two-fold, though this appears not to be significantly different? Could the authors confirm this, as the data in Supplementary file 1A seem to suggest that it might be. The Abstract should be modified to better reflect the manuscript in regards to the above discussion, and would also benefit from the inclusion of additional details regarding what parameters were actually measured.

3) More information needs to be provided regarding the phylogenetic tree construction methodology, which is central for the phylogenetically independent contrast analysis outlined in the Materials and methods. For instance, what specific algorithm and settings were used in Geneious for the multiple sequence alignment? Was the resulting alignment refined by eye to correct ambiguities, and how were indels dealt with? Additionally, it is stated that the tree was constructed using "neighbor joining" in the Materials and methods (again, with key details/parameters missing), though it is noted that a "UPGMA tree and branch lengths" were used for the PIC analysis. Importantly, the tree presented in Figure 2 is poorly resolved, which may have affected the results of the PIC analyses (several of which were borderline significant). Since the results of the PIC analysis are only meaningful if the input tree is accurate/reliable, the authors should consider using more current and robust tree building methods (e.g., maximum likelihood) with the best-fitting models to better resolve the actual relationships among these species. Additionally, given the relatively low amount of sequence data employed for tree building, the authors should also consider including additional sequence data where available and constraining the positions of certain lineages based on the results of other studies with larger datasets. For example, constraining Loponetta as sister to Anas (see e.g. Bulgarella et al., 2010), and placing Oxyura as the outgroup (see e.g. Gonzalez et al., 2009), though we recommend using more recent/robust phylogenetic hypotheses if available.

Bulgarella, M., Sorenson, M.D., Peters, J.L., Wilson, R.E. and McCracken, K.G., 2010. Phylogenetic relationships of Amazonetta, Speculanas, Lophonetta, and Tachyeres: four morphologically divergent duck genera endemic to South America. Journal of Avian Biology, 41(2), pp.186-199.

4) The authors helpfully sampled from three sites in the muscle, at different depths, and took a mean of the activities from each location. Please report the degree of variation between the three locations.

5) It would be helpful for all data to be presented in the manner of Figures 3, 4, and 5. Figure 6 is useful in showing differences in HK activity and ATPase activity in relation to evolutionary time, but the data in panel 6D, which only shows absolute rates of HK activity for high altitude groups, suggest that HK activity decreases below low altitude values as a function of time, whereas it is higher than in lowland counterparts (albeit not always significantly higher) for almost all species studied.

6) One of the main themes of the paper was that fatty acid oxidation capacity is increased at high altitude. However, this is based on a single enzyme indicator, HOAD, whereas other aspects of metabolism were assessed with multiple enzymes. This potential limitation should be noted or discussed.

7) In the third paragraph of the subsection “Convergence Across Multiple Pathways of Energy Metabolism”, the change in complex II (succinate dehydrogenase) activity is attributed to a need to support greater capacities for fatty acid oxidation, though this is not correct. The free FADH2 from beta-oxidation does not donate electrons to complex II (which oxidizes succinate using a fixed FAD/FADH cofactor), but to the electron-transferring flavoprotein complex (CETF). Please clarify/revise.

8) Why are the results for Complex V (ATP synthase) not reported in the comparisons between populations/species, and included only in the analysis of time? They seem to be important.

9) The possible explanations for increased ATP-synthase activities at altitude are useful, but the second paragraph of the subsection “Some Metabolic Changes Arise Only After Prolonged Evolutionary Time at High Altitude” refers to anoxia, and does not appear to be relevant to non-anoxia tolerant waterfowl. The authors cite a review article here (Garcia-Aguilar and Cuevez) but the original work was carried out by St-Pierre and colleagues (2000, PNAS, 97(15), pp.8670-8674) in anoxia-tolerant frogs. The relevance to the current study is not clear, since it is not necessarily the case that membrane potential would fall in hypoxia exposed waterfowl (a different state to anoxia).

10) The authors speculate that the differences in metabolic function are related to hypoxia or cold at altitude, but given this is a flight muscle, how might the decreased air resistance alter the metabolic demand placed on this muscle? Would activity be expected to fall as there is less drag, or might the low air pressure increase the metabolic demand of lift?

11) In the statistical analyses presented in Supplementary file 1B and C, there are highly significant species by altitude interactions, but these are not addressed. When there are significant interactions, main effects should not be considered until the interactions are addressed, either by plotting them out to see if they are all in the same direction and just have different slopes, or by separating the species and looking at them individually if there is crossover of the effects. The authors need to discuss what the interactions were and how they handled them. Most of these are present in cases where a significant altitude effect was detected, so they are potentially crucial to the conclusions.

https://doi.org/10.7554/eLife.56259.sa1

Author response

Summary:

This study employed 7 high- and low-altitude species pairs that spanned ~1 million years of evolution in hypoxic environments to identify convergent and time-dependent trajectories in muscle metabolic biochemistry to high altitude environments. Importantly, since the work examines fuel and energy metabolism in flight muscle, rather than oxygen transport, it provides a valuable perspective on what is happening at the level of metabolic biochemistry. Overall, the experiments appear to have been carefully conducted, analyses seem sound (except the construction of the phylogenetic tree, which requires particular attention during revision), and the manuscript is well written.

We would like start by thanking the reviewers for their excellent suggestions. We feel as though the suggestions put forward by the reviewers have greatly improved the manuscript. We agree that the phylogenetic tree needed to be improved. We have reconstructed the phylogenetic tree by using maximum parsimony while constraining it to the topology depicted in Gonzalez, 2019. This has resolved the two polytomies present in our original tree and has resulted in minor changes in the reported statistics; however, there are no changes in the significance reported previously, largely due to the fact that each high-low pair was quite distantly related from all other pairs. Nevertheless, we believe the paired-lineage Wilcoxon tests are the most appropriate for examining the convergent differences that exist across high-low pairs. We have made some minor editorial changes throughout the Results to bring the results of the Wilcoxon tests to the forefront.

Essential revisions:

1) The precision of writing related to evolution, convergence, and phenotypic plasticity could be improved. The first paragraph doesn't fully account for the wide variety of evolutionary studies of convergence and other processes, and sets up a fairly weak hypothesis about whether convergence is always the same or idiosyncratic because it sets up a black and white dichotomy. Evolution/adaptation/drift are always happening and of course each lineage will have an idiosyncratic process. They may arrive at generally similar phenotypes, but do you ever expect exact matching? The second question about the speed of adaptation is more interesting. Accordingly, the first paragraph or most of it could be dropped and the paper could be started with paragraph 2.

We agree that the question surrounding the speed at which adaptation may take place is more interesting, and as such, we have pared down the initial paragraph to focus on this particular question and have removed the first question regarding similar or idiosyncratic changes.

2) The interpretation of the results pertaining to glycolytic capacity is handled somewhat inconsistently throughout the manuscript. The Abstract states that glycolytic capacity is decreased at altitude, whereas the data for HK activity in particular does not support this conclusion with Figure 1 indicating greater HK activity at high altitude. The argument made in the main text is more subtle, supporting an increase in HK activity in many of the more recent high-altitude populations, but no difference in the more established residents. In fact, HK activity is not lower in any highland waterfowl studied compared with its low altitude counterpart with the exception of the ruddy duck (Supplementary file 1A). Moreover, the difference between the low- and high-altitude Andean goose species is almost two-fold, though this appears not to be significantly different? Could the authors confirm this, as the data in Supplementary file 1A seem to suggest that it might be. The Abstract should be modified to better reflect the manuscript in regards to the above discussion, and would also benefit from the inclusion of additional details regarding what parameters were actually measured.

We agree with the reviewers that the Abstract would benefit from being more precise and from including greater detail of the exact parameters measured. We have made the appropriate changes to the Abstract to reflect these suggestions. We realize in hindsight that we were inconsistent in how we discussed the issue of glycolytic capacity, both in the Abstract and elsewhere, and we have made several other changes to correct this inconsistency (Abstract, subsections “Convergence Across Multiple Pathways of Energy Metabolism”, “Some Metabolic Changes Arise Only After Prolonged Evolutionary Time at High Altitude”, and “Conclusions”).

To clarify, differences in HK activity between high- and low-altitude geese are not statistically significant. Even when doing a simple pairwise comparison using a student’s t-test, there is no significant difference between high- and low-altitude geese.

3) More information needs to be provided regarding the phylogenetic tree construction methodology, which is central for the phylogenetically independent contrast analysis outlined in the Materials and methods. For instance, what specific algorithm and settings were used in Geneious for the multiple sequence alignment?

The default settings used included treatment of gaps as missing data, which is appropriate for a gene such as the mtDNA control region in which it becomes difficult to code gaps of a different length as character states. The search criterion used for the simplified tree for the PICs was maximum parsimony, as the UPGMA method cannot accommodate a constraint. The maximum parsimony tree with branch lengths (Figure 2) was constrained to the Gonzalez (2009) topology, which is the most comprehensive global phylogeny yet available for all of these waterfowl species (Gonzalez, Düttman and Wink, 2009). To clarify their taxonomic position among the many lineages of waterfowl we have added a supplementary figure (Figure 2—figure supplement 1) showing their position in the Gonzalez et al., 2009 topology. This new figure clearly emphasizes the relatively distant phylogenetic relationships between each high-low pair, although there is a much closer relationship within each high-low pair. We have made a small revision in the Introduction to remove a statement that was inconsistent with this information.

Was the resulting alignment refined by eye to correct ambiguities, and how were indels dealt with?

The reviewers are correct that the resulting alignment was refined by eye to correct any ambiguities. We treated any indels as missing data (see above). We have included this information in the Materials and methods subsection “Phylogenetic Tree Generation”.

Additionally, it is stated that the tree was constructed using "neighbor joining" in the Materials and methods (again, with key details/parameters missing), though it is noted that a "UPGMA tree and branch lengths" were used for the PIC analysis.

We would like to thank the reviewer for pointing out this mistake. As stated above, we have reconstructed our phylogenetic tree using maximum parsimony constrained to the topology depicted in Gonzalez et al., 2009, which we now describe accurately in both the main text and Figure 2—figure supplement 1 (subsection “Phylogenetic Tree Generation”).

Importantly, the tree presented in Figure 2 is poorly resolved, which may have affected the results of the PIC analyses (several of which were borderline significant).

Our newly built tree using maximum parsimony has fully resolved the two polytomies present in our original tree constrained to match Gonzalez et al., 2009. This has only resulted in minor changes in the reported statistics, however, such that there are no qualitative differences and no changes in the significance reported previously.

Since the results of the PIC analysis are only meaningful if the input tree is accurate/reliable, the authors should consider using more current and robust tree building methods (e.g., maximum likelihood) with the best-fitting models to better resolve the actual relationships among these species. Additionally, given the relatively low amount of sequence data employed for tree building, the authors should also consider including additional sequence data where available and constraining the positions of certain lineages based on the results of other studies with larger datasets. For example, constraining Loponetta as sister to Anas (see e.g. Bulgarella et al., 2010), and placing Oxyura as the outgroup (see e.g. Gonzalez et al., 2009), though we recommend using more recent/robust phylogenetic hypotheses if available.

We, again, thank the reviewers for their excellent suggestion to constrain our tree to the topology published in Gonzalez et al., 2009 (see Figure 2—figure supplement 1, which shows the topology). After implementing this change we have resolved the polytomies in question that were previously present in the discarded NJ tree. We are unable to use other available sequence data in generating our tree, because there are no other common genes that are available for all species and all populations. The only available gene sequence for all low- and high-altitude taxa is the mitochondrial control region. The data sets in the citations below that were suggested by the reviewers unfortunately possess either a single low-altitude taxon and are missing the corresponding high-altitude taxon or vice versa. Nevertheless, we also ran the PIC analysis with branch lengths set to 1 to help account for potential variation in branch lengths across the tree, and this yielded qualitatively the same pattern of variation as our analyses using the actual branch lengths. Nevertheless, as described in our earlier response, we believe the paired-lineage Wilcoxon tests (rather than the PIC analysis) are the most appropriate for examining the convergent differences that exist across high-low pairs, and we have made changes throughout the manuscript to bring the results of the Wilcoxon tests to the forefront.

Bulgarella, M., Sorenson, M.D., Peters, J.L., Wilson, R.E. and McCracken, K.G., 2010. Phylogenetic relationships of Amazonetta, Speculanas, Lophonetta, and Tachyeres: four morphologically divergent duck genera endemic to South America. Journal of Avian Biology, 41(2), pp.186-199.

Unfortunately, this data set, while including many genes, does not include both high- and low-altitude populations and is missing 5 species from our data set within the genus Anas and the ruddy duck. We have, however, made many changes that improve our phylogenetic comparisons as stated above.

Gonzalez, Düttmann and Wink, 2009.

We have included a figure to outline the topology that we have used to constrain our newly created tree. This is the correct global waterfowl phylogeny to use as it contained all 7 species sampled in our study (see Figure 2—figure supplement 1).

4) The authors helpfully sampled from three sites in the muscle, at different depths, and took a mean of the activities from each location. Please report the degree of variation between the three locations.

We have included an additional table (Supplementary File 1J.) containing the enzymatic activities obtained at each depth of tissue. We have also made a small revision to the main text to describe the general pattern of variation observed between the three locations (subsection “Tissue Sampling”).

5) It would be helpful for all data to be presented in the manner of Figures 3, 4, and 5. Figure 6 is useful in showing differences in HK activity and ATPase activity in relation to evolutionary time, but the data in panel 6D, which only shows absolute rates of HK activity for high altitude groups, suggest that HK activity decreases below low altitude values as a function of time, whereas it is higher than in lowland counterparts (albeit not always significantly higher) for almost all species studied.

We have added graphs from HK and ATP synthase in Figures 3 and 4 to show the same data consistently across all enzymes. We have also made changes to the text to discuss HK and ATP synthase in the context of convergence (subsections “Convergence Across Multiple Pathways of Energy Metabolism” and “Some Metabolic Changes Arise Only After Prolonged Evolutionary Time at High Altitude”).

6) One of the main themes of the paper was that fatty acid oxidation capacity is increased at high altitude. However, this is based on a single enzyme indicator, HOAD, whereas other aspects of metabolism were assessed with multiple enzymes. This potential limitation should be noted or discussed.

This is a great point. HOAD is a good marker of beta-oxidation capacity, but the overall capacity for lipid oxidation will of course depend on other key regulators as well (e.g., mitochondrial fatty acid uptake by CPT). We’ve added two new sentences to address this issue (subsection “Convergence Across Multiple Pathways of Energy Metabolism”).

7) In the third paragraph of the subsection “Convergence Across Multiple Pathways of Energy Metabolism”, the change in complex II (succinate dehydrogenase) activity is attributed to a need to support greater capacities for fatty acid oxidation, though this is not correct. The free FADH2 from beta-oxidation does not donate electrons to complex II (which oxidizes succinate using a fixed FAD/FADH cofactor), but to the electron-transferring flavoprotein complex (CETF). Please clarify/revise.

We did not originally intend to suggest that FADH2 produced via beta-oxidation would be utilized by complex II (TCA cycle), but we agree in hindsight that this point was not clear. Apologies for the confusion, we have deleted this sentence from the paper.

8) Why are the results for Complex V (ATP synthase) not reported in the comparisons between populations/species, and included only in the analysis of time? They seem to be important.

We have added a short section to report changes in ATP synthase in the context of species-specific changes (subsection “Convergence Across Multiple Pathways of Energy Metabolism”).

9) The possible explanations for increased ATP-synthase activities at altitude are useful, but the second paragraph of the subsection “Some Metabolic Changes Arise Only After Prolonged Evolutionary Time at High Altitude” refers to anoxia, and does not appear to be relevant to non-anoxia tolerant waterfowl. The authors cite a review article here (Garcia-Aguilar and Cuevez) but the original work was carried out by St-Pierre and colleagues (2000, PNAS, 97(15), pp.8670-8674) in anoxia-tolerant frogs. The relevance to the current study is not clear, since it is not necessarily the case that membrane potential would fall in hypoxia exposed waterfowl (a different state to anoxia).

We agree with the reviewers that the relevance of this work is not clear for this study and removed this section from the Discussion.

10) The authors speculate that the differences in metabolic function are related to hypoxia or cold at altitude, but given this is a flight muscle, how might the decreased air resistance alter the metabolic demand placed on this muscle? Would activity be expected to fall as there is less drag, or might the low air pressure increase the metabolic demand of lift?

This is a great suggestion. In fact, recent work on high-altitude flight in the bar-headed goose done by Bishop et al., 2015, show that the reduction in drag has been shown to be more than offset by the reduction in capacity to produce lift at low air densities, and thus increases the metabolic cost of flight at high altitudes. We have included a sentence outlining this in the Introduction.

11) In the statistical analyses presented in Supplementary file 1B and C, there are highly significant species by altitude interactions, but these are not addressed. When there are significant interactions, main effects should not be considered until the interactions are addressed, either by plotting them out to see if they are all in the same direction and just have different slopes, or by separating the species and looking at them individually if there is crossover of the effects. The authors need to discuss what the interactions were and how they handled them. Most of these are present in cases where a significant altitude effect was detected, so they are potentially crucial to the conclusions.

The significant interaction effects were attributed primarily to opposing patterns observed in the ruddy ducks, with the exception of HK and ATPase, where the interaction were attributable to changes observed in either species that are more recently established (HK) or established for longer periods of time at altitude (LDH, ATPsyn). We have added mention of the significant interactions where we discuss these patterns of variation in the Results/Discussion section (subsections “Idiosyncratic Changes in High-Altitude Ruddy Ducks”, “Some Metabolic Changes Arise Only After Prolonged Evolutionary Time at High Altitude”, as well as “Enzyme activities and myoglobin assays”).

https://doi.org/10.7554/eLife.56259.sa2

Article and author information

Author details

  1. Neal J Dawson

    1. Department of Biology, McMaster University, Hamilton, Canada
    2. Department of Biology University of Miami, Coral Gables, United States
    Contribution
    Conceptualization, Data curation, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing - original draft, Project administration, Writing - review and editing
    For correspondence
    neal.dawson@glasgow.ac.uk
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-5389-8692
  2. Luis Alza

    1. Department of Biology University of Miami, Coral Gables, United States
    2. University of Alaska Museum and Institute of Arctic Biology, University of Alaska Fairbanks, Fairbanks, United States
    3. Centro de Ornitología y Biodiversidad - CORBIDI, Lima, Peru
    Contribution
    Resources, Project administration, Writing - review and editing
    Competing interests
    No competing interests declared
  3. Gabriele Nandal

    Department of Biology, McMaster University, Hamilton, Canada
    Contribution
    Investigation, Writing - review and editing
    Competing interests
    No competing interests declared
  4. Graham R Scott

    Department of Biology, McMaster University, Hamilton, Canada
    Contribution
    Conceptualization, Resources, Supervision, Funding acquisition, Methodology, Project administration, Writing - review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-4225-7475
  5. Kevin G McCracken

    1. Department of Biology University of Miami, Coral Gables, United States
    2. University of Alaska Museum and Institute of Arctic Biology, University of Alaska Fairbanks, Fairbanks, United States
    3. Centro de Ornitología y Biodiversidad - CORBIDI, Lima, Peru
    4. Department of Marine Biology and Ecology, Rosenstiel School of Marine and Atmospheric Sciences, University of Miami, Miami, United States
    5. Human Genetics and Genomics, Hussman Institute for Human Genomics, University of Miami Miller School of Medicine, Miami, United States
    Contribution
    Conceptualization, Resources, Supervision, Funding acquisition, Methodology, Project administration, Writing - review and editing
    Competing interests
    No competing interests declared

Funding

Natural Sciences and Engineering Research Council of Canada (Discovery Grant)

  • Graham R Scott

National Science Foundation (IOS-0949439)

  • Kevin G McCracken

Canada Foundation for Innovation (John R. Evans Leaders Fund)

  • Graham R Scott

Ontario Ministry of Research and Innovation (Early Researcher Award)

  • Graham R Scott

Kushlan Endowment for Waterbird Biology and Conservation (Kushlan Chair)

  • Kevin G McCracken

Canada Research Chairs (Tier 2 - Comparative and Environmental Physiology)

  • Graham R Scott

Natural Sciences and Engineering Research Council of Canada (Postdoctoral Fellowship)

  • Neal J Dawson

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

Acknowledgements

This research was supported by funds from McMaster University, the Canadian Foundation for Innovation, the Ontario Ministry of Research and Innovation, and a Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grant to GRS. This research was also supported by a NSF grant (IOS-0949439) to KGM and the Kushlan Endowment for Waterbird Biology and Conservation at the University of Miami. Gratitude for all kinds of assistance is expressed to E Bautista, M Bulgarella, C, Kopuchian, D Lijtmaer, B Reishus, M Smith, M St. Louis, P Tubaro, and T Valqui. GRS is supported by the Canada Research Chairs Program, NJD was supported by an NSERC Postdoctoral Fellowship, and KGM is supported by the Kushlan Chair for Waterbird Biology and Conservation at the University of Miami.

Ethics

Animal experimentation: Samples were imported to Canada with authorization from the Canadian Wildlife Service (Scientific Possession No. 369) and collected with authorization from the Servicio Nacional de Area Naturales Protegidas del Peru (004-2014-SERNANP-DGANP-RNT/J), Dirección General Forestal y de Fauna Silvestre del Peru (RD 169-2014-MIN AGRI-DGFFS/DGEFFS, 190-2015-SERFOR-DGGSPFFS), Ministerio de Industria, Agricultura, y Ganaderia Chubut (No. 24/07 y 1636/14), Ministerio de Asuntos Agrarios Buenos Aires, Oregon Department of Fish and Wildlife (101-15), and USFWS Region 1 Migratory Bird Permit Office (MB68890B-0). All protocols were carried out in accordance with guidelines that were approved by the institutional animal care and use committee at the University of Miami or University of Alaska.

Senior Editor

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

Reviewing Editor

  1. Kevin Campbell, University of Manitoba, Canada

Reviewers

  1. Kevin Campbell, University of Manitoba, Canada
  2. Andrew Murray, University of Cambridge, United Kingdom

Publication history

  1. Received: February 21, 2020
  2. Accepted: July 23, 2020
  3. Accepted Manuscript published: July 30, 2020 (version 1)
  4. Accepted Manuscript updated: August 10, 2020 (version 2)
  5. Version of Record published: September 16, 2020 (version 3)
  6. Version of Record updated: September 17, 2020 (version 4)

Copyright

© 2020, Dawson 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

  • 580
    Page views
  • 108
    Downloads
  • 0
    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)

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

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

Further reading

    1. Evolutionary Biology
    2. Genetics and Genomics
    Tom Hill, Robert L Unckless
    Research Article Updated

    Hosts and viruses are constantly evolving in response to each other: as a host attempts to suppress a virus, the virus attempts to evade and suppress the host’s immune system. Here, we describe the recurrent evolution of a virulent strain of a DNA virus, which infects multiple Drosophila species. Specifically, we identified two distinct viral types that differ 100-fold in viral titer in infected individuals, with similar differences observed in multiple species. Our analysis suggests that one of the viral types recurrently evolved at least four times in the past ~30,000 years, three times in Arizona and once in another geographically distinct species. This recurrent evolution may be facilitated by an effective mutation rate which increases as each prior mutation increases viral titer and effective population size. The higher titer viral type suppresses the host-immune system and an increased virulence compared to the low viral titer type.

    1. Ecology
    2. Evolutionary Biology
    Naïma Jesse Madi et al.
    Research Article

    Microbes are embedded in complex communities where they engage in a wide array of intra- and inter-specific interactions. The extent to which these interactions drive or impede microbiome diversity is not well understood. Historically, two contrasting hypotheses have been suggested to explain how species interactions could influence diversity. 'Ecological Controls' (EC) predicts a negative relationship, where the evolution or migration of novel types is constrained as niches become filled. In contrast, 'Diversity Begets Diversity' (DBD) predicts a positive relationship, with existing diversity promoting the accumulation of further diversity via niche construction and other interactions. Using high-throughput amplicon sequencing data from the Earth Microbiome Project, we provide evidence that DBD is strongest in low-diversity biomes, but weaker in more diverse biomes, consistent with biotic interactions initially favoring the accumulation of diversity (as predicted by DBD). However, as niches become increasingly filled, diversity hits a plateau (as predicted by EC).