Combining phylogeography and the geography of adaptation to understand population history - Ruddy Ducks in the new world

Texto completo

(1)1 . Combining Phylogeography and the Geography of Adaptation to. 2 . Understand Population History: Ruddy Ducks in the New World. 3 . Maria Lozano-Jaramillo1*, Kevin McCracken2, Carlos Daniel Cadena1. 4 5 . 1. 6 7 . 2. 8 . *Corresponding author, e-mail: [email protected]. 9 . Abstract. Laboratorio de Biología Evolutiva de Vertebrados, Departamento de Ciencias Biológicas, Universidad de. Los Andes Institute of Arctic Biology & Department of Biology and Wildlife University of Alaska Fairbanks. 10 . Phylogeography largely focuses on finding associations between geography and. 11 . putatively neutral genetic variation, but variation in genes of functional importance can. 12 . also be used to examine the phylogeographic history of species. Species with populations. 13 . exposed to a variety of environments, such low- and high-elevation habitats, are expected. 14 . to show local adaptations in different populations, such that individuals will exhibit better. 15 . performance in their native environments relative to others. Here, we test a demographic. 16 . hypothesis of historical gene flow of Ruddy Ducks (Oxyura jamaicensis), a widely. 17 . distributed species in the New World, into South America from North America, using. 18 . multilocus analyses, including genes of functional importance, and for the first time. 19 . including samples from the Colombian Andes. We also examined whether the Colombian. 20 . Ruddy Duck population is of hybrid origin, as previously suggested, and evaluated signs. 21 . of adaptive signatures of hemoglobin adaptation to life at high elevations. Our results. 22 . suggest that the biogeographic and population genetic history of Ruddy Ducks is more. 23 . complex than previously elucidated, with genes moving most likely from the Southern. 24 . Andes to Colombia. We also find strong evidence to support that the Colombian. 25 . population might be of an old hybrid origin. Taken together, our results indicate that it is. 26 . crucial to include samples from Colombia when evaluating phylogeographic and. 27 . populations genetics questions.. 28 . Keywords: Functional importance, selection, adaptation, plasticity.

(2) 29 30 . Introduction. 31 . Understanding species’ distributions and the historical and evolutionary processes. 32 . shaping them is a central goal of biogeography. Phylogeography uses information on the. 33 . geographic distribution of genetic variation within species to make inferences about the. 34 . processes involved in patterns of population differentiation (Avise, 2000; Hickerson et al.,. 35 . 2010; Lee et al., 2008; Wang et al., 2008). Generally, phylogeographic studies focus on. 36 . finding associations between geography and the distribution of presumably neutral. 37 . genetic variation (i.e. microsatellites, introns, mitocondrial DNA; Hickerson et al., 2010),. 38 . but variation in genes of functional importance can also be used to examine the. 39 . phylogeographic history of species while also evaluating the role of natural selection in. 40 . population differentiation. However, few phylogeographic studies have been developed. 41 . focusing on the examination of geographic variation in functionally important genes. 42 . (Hoekstra et al., 2004; Londo et al., 2006; Muñoz-Fuentes et al., 2013), and few studies. 43 . have taken into account the possibility that genes widely used to make inferences of. 44 . population history may be of functional importance and therefore exhibit non-neutrality. 45 . (Dowling et al., 2008; Peters et al., 2012a). For example, in Rufous-collared Sparrows. 46 . (Zonotrichia capensis) mitochondrial genes may be under strong selection along. 47 . elevational gradients owing to their functional role in oxidative phosphorylation; thus,. 48 . selection (and not processes such as drift in isolated populations) likely accounts for. 49 . patterns of mtDNA differentiation across elevations (Cheviron and Brumfield, 2009).. 50 . When a species has populations distributed across environmental gradients, individuals. 51 . are exposed to different habitats and selective pressures (Sexton et al., 2009). This will. 52 . promote local adaptations in different populations, such that individuals will exhibit. 53 . better performance in their native environments relative to others (Kawecki and Ebert,. 54 . 2004). Organisms living at high elevations need to cope with environmental stress factors,. 55 . such as hypoxia. In high-elevation regions above 4000 m, such as in the high Andes, the. 56 . partial pressure of oxygen (pO2) is approximately 60% of that at sea level (Beall, 2006;. 57 . Hopkins and Powell, 2001). Under such conditions, transport of O2 to tissues may be. 58 . reduced, altering metabolism and the ability to perform physical activity (McCracken et.

(3) 59 . al., 2009b; Storz, 2010; Storz and Moriyama, 2008). Genetic changes increasing the. 60 . affinity of hemoglobin for O2, and higher concentration of hemoglobin and red blood. 61 . cells are two important mechanisms known to enable high-altitude populations of birds. 62 . and mammals to cope with hypoxia (Beall et al., 1998; Storz and Moriyama, 2008;. 63 . Weber, 2007). Genetic adaptations that increase the affinity of hemoglobin for O2 in. 64 . different bird species have been shown to result from independent substitutions at the. 65 . same intersubunit contact sites (Jessen et al., 1991; Weber et al., 1993), or from. 66 . substitutions altering the sensitivity of the allosteric effectors (McCracken et al., 2009a).. 67 . Therefore, a well-characterized protein like hemoglobin is not only appropriate to explore. 68 . the role of selective pressures in different environments, but it may also allow one to. 69 . make inferences about the history of genetic adaptations if patterns of variation are. 70 . studied in a biogeographic context (McCracken et al., 2009a; Muñoz-Fuentes et al., 2013).. 71 . The Ruddy Duck Oxyura jamaicensis (Anatidae) is broadly distributed in the Americas.. 72 . Three subspecies are recognized based on variable plumage patterns: the New World. 73 . Ruddy Duck (O. j. jamaicensis) limited to lowland North America and Middle America,. 74 . the Colombian Ruddy Duck (O. j. andina) restricted to high elevations in the Colombian. 75 . Andes, and the Andean Ruddy Duck (O. j. ferruginea) ranging from southern Colombia. 76 . to Tierra del Fuego in both highlands and lowlands (Fig.1; Adams and Slavid, 1984;. 77 . Fjeldså, 1986; McCracken and Sorenson, 2005). Based on plumage coloration, namely. 78 . the black-and-white mottled cheek of andina, several authors have suggested that this. 79 . geographically and phenotypically intermediate population might be the result of genetic. 80 . admixture between ferruginea (all-black head) and jamaicensis (all-white cheek; Fjeldså,. 81 . 1986; Siegfried, 1976).. 82 . The broad distribution of Ruddy Ducks in contrasting environments of the Americas and. 83 . patterns of phenotypic variation have resulted in considerable interest in studying the. 84 . historical biogeography of the species (Fjeldså, 1986; McCracken and Sorenson, 2005;. 85 . Muñoz-Fuentes et al., 2013; Muñoz-Fuentes et al., 2006). A phylogenetic study of. 86 . stifftail ducks based on a relatively small sample of mitochondrial DNA sequences. 87 . obtained results consistent with two alternative biogeographic hypotheses that may. 88 . account for the broad distribution of Ruddy Ducks in the New World (McCracken and. 89 . Sorenson 2005). First, wintering birds from a migratory population in North America.

(4) 90 . could have colonized and established in the Northern Andes, and then spread north-to-. 91 . south all the way to Tierra del Fuego, top-down from high elevations in the Andes to. 92 . lowland habitats in southern South America. Alternatively, South American Ruddy. 93 . Ducks could have recently colonized the lowlands in North America. McCracken and. 94 . Sorenson (2005) also found that O.j.andina shares mtDNA haplotypes with the other two. 95 . populations, but only five samples from Colombia were analyzed.. 96 . A study examining geographic variation in hemoglobin genes of several waterfowl. 97 . species found evidence of putative genetic adaptations to high-elevations in the αA and βA. 98 . hemoglobin genes occurring in parallel across species (McCracken et al., 2009a).. 99 . Relative to other species, Ruddy Ducks did not exhibit presumably adaptive variation in. 100 . the αA gene, but showed a fixed substitution from serine to threonine in aminoacid. 101 . position 69 (βA-69) both in highland and lowland populations in South America. Because. 102 . βA-69 was absent in Ruddy Ducks from North American lowlands and in other stifftail. 103 . ducks, and considering the hypothesis of north-to-south colonization and subsequent top-. 104 . down dispersal, McCracken et al. (2009a) hypothesized that this substitution could. 105 . increase O2-affinity, a possibility requiring confirmation via functional studies.. 106 . A recent, more comprehensive study using both putatively neutral (mtDNA, nuclear. 107 . introns) and functional (hemoglobin) gene markers to characterize patterns of genetic. 108 . variation in Ruddy Ducks found population-genetic evidence for north-to-south dispersal. 109 . (Muñoz-Fuentes et al., 2013). In addition, based on comprehensive sampling across. 110 . North America and in South America from Peru to Argentina, this study supported the. 111 . scenario of amino acid replacements in the βA hemoglobin gene described above (no. 112 . geographic variation was observed in alpha globin genes). Taken together, the neutral and. 113 . functional genetic data are consistent with a biogeographic history involving colonization. 114 . of the Andes from the north (possibly by formerly migrant individuals), putative. 115 . adaptation to hypoxia in the highlands of the Andes, and subsequent colonization of the. 116 . lowlands of southern South America (McCracken et al., 2009a; Muñoz-Fuentes et al.,. 117 . 2013). However, such inferences were still made based on insufficient sampling from. 118 . Colombia, a critical region for the study of biogeographic hypotheses involving widely. 119 . distributed New World birds (Cadena et al., 2007; Chapman, 1917). Specifically, no. 120 . information on nuclear DNA variation was obtained from Colombian populations, and.

(5) 121 . the only mtDNA sequences analyzed were those of the same five individuals considered. 122 . by the earlier study (McCracken and Sorenson, 2005). Additionally, the inferences about. 123 . genetic adaptation to high elevations following colonization of northern South America. 124 . were based solely on patterns of nonsynonymous variation in the βA hemoglobin gene.. 125 . Because the putatively adaptive substitution involves a change from serine to threonine. 126 . and these two amino acids have similar physiochemical properties (Mathews et al., 2002). 127 . its true functional consequence remains to be established. Moreover, no relevant. 128 . information on the high-altitude physiology of Ruddy Ducks is available.. 129 . Owing to the lack of sufficient data from Colombia, central predictions of the north-to-. 130 . south colonization hypothesis have not yet been fully evaluated. This biogeographic. 131 . hypothesis predicts that alleles from Colombia are derived from North American alleles.. 132 . Similarily, if South American populations spread south along the Andes from high to. 133 . lower elevations (top-down), lowland populations from North America and South. 134 . America would be expected to form a paraphyletic group, with lowland populations from. 135 . South America being derived from South American highland populations. However,. 136 . assuming that colonization of the Colombian Andes from lowland populations of North. 137 . America required the evolution of genetic adaptations in hemoglobin genes to life at high. 138 . elevations, historical gene flow into Colombia from the north would be expected to be. 139 . reduced owing to the influence of selection (Bulgarella et al., 2012; McCracken et al.,. 140 . 2009c; Muñoz-Fuentes et al., 2013). In addition, birds from high elevations from. 141 . Colombia should show the putative genetic adaptation to high altitude documented in. 142 . earlier studies (McCracken et al., 2009a; Muñoz-Fuentes et al., 2013) or evidence of. 143 . physiological responses involving hemoglobin concentration and hematocrit levels (Storz. 144 . et al., 2010). On the other hand, the hypothesis that the Colombian population is of. 145 . hybrid origin predicts that it should show an admixture of alleles; if the population is the. 146 . result of recent and perhaps episodic but recurrent hybridization, one would further. 147 . expect to find evidence of ongoing historical gene flow from its putative parental taxa.. 148 . Alternatively, if the population originated via hybridization but has since evolved as an. 149 . independent entity, one would expect it to have private alleles (Gompert et al., 2006).. 150 . Here, we revisit the hypothesis that Ruddy Ducks colonized the Colombian Andes from. 151 . North America and then dispersed south along the Andes and down to low elevations.

(6) 152 . from the highlands including, for the first time, multilocus analyses including a. 153 . considerable number of samples from the Colombian Andes, the area through which. 154 . Ruddy Ducks are supposed to have first established populations and colonized South. 155 . America. We also examine whether the Colombian Ruddy Duck population is of hybrid. 156 . origin, as previously suggested. To accomplish these goals, we use a novel combination. 157 . of traditional phylogeographic and population genetics approaches based on putatively. 158 . neutral loci (introns and mtDNA), coupled with analyses of variation in a functionally. 159 . important gene expected to be involved in adaptation to hypoxia (βA hemoglobin). We. 160 . also analyze physiological parameters (i.e., hematocrit and hemoglobin concentration). 161 . relevant to understanding the role that resolving the challenges of living at high. 162 . elevations could have had in the history of colonization of different elevational habitats in. 163 . South America by Ruddy Ducks. The inclusion of Colombian populations in analyses. 164 . reveals that the biogeographic and evolutionary history of Ruddy Ducks may be more. 165 . complex than previously thought.. 166 167 . Methods. 168 . To complete the geographic sampling of genetic variation in Ruddy Ducks documented. 169 . in earlier studies, we collected and studied a total of 26 samples from the Cordillera. 170 . Oriental of the Colombian Andes. Tissue samples from six individuals from Lake. 171 . Fúquene (2580 m), Cundinamarca, were obtained from the Museo de Historia Natural at. 172 . Universidad de los Andes. Twenty additional blood and tissue samples were obtained. 173 . from two adjacent artificial lakes (i.e., abandoned gravel pits) located near the town of. 174 . Guasca, Cundinamarca (2630 m and 2640 m elevation). In the field, total hematocrit (i.e.,. 175 . packed red blood cell volume) was separated using a portable centrifuge and measured. 176 . with digital calipers, and the hemoglobin concentration was measured with a Hemocue. 177 . Hb 201+ portable spectrophotometer (Simmons and Lill, 2006). Blood samples were. 178 . preserved in lysis buffer and EDTA for genetic analyses.. 179 . Genomic DNA was extracted from blood, tissue or feathers using a phenol/chloroform. 180 . protocol (Sambrook, 2001). For each individual, we amplified and sequenced a total of. 181 . six genes: (a) the mitochondrial control region (578 bp); (b) four autosomal nuclear.

(7) 182 . introns: phosphoenolpyruvate carboxykinase intron 9 (PCK1-9; 316 bp), N-methyl-D-. 183 . aspartate-1-glutamate receptor intron 11 (GRIN1-11; 308 bp), beta fibrinogen intron 7. 184 . (FGB-7; 248 bp), and ornithine decarboxylase intron 5 (ODC1-5; 345 bp); and (c) exon 1,. 185 . intron 1, exon 2, and part of intron 2 of the βA globin subunit (581 bp), encompassing the. 186 . three previously described amino acid polymorphisms at positions 13, 14, and 69. 187 . documented by (McCracken et al., 2009a). We used previously described primers. 188 . (McCracken et al., 2009c) and the following PCR conditions: 3 min preheat at 94 °C,. 189 . followed by 45 cycles of 45 s at 94 °C, 30 s at Ta °C (annealing temperature was primer-. 190 . specific, table S1), 1 min at 72 °C, and a final extension of 10 min at 72 °C. Sequences. 191 . were edited using Geneious Pro 4.8.3 (http://www.geneious.com) or Sequencher v.4.6. 192 . (Gene Codes, Ann Arbor, MI) and aligned by eye using Se-Al v.2.0a11 (Rambaut, 2002).. 193 . To resolve the gametic phase of nuclear genes we used PHASE v.2.1, a Bayesian. 194 . statistical method for haplotype reconstruction from genotype data that considers. 195 . recombination (Stephens and Scheet, 2005; Stephens et al., 2001). We used default. 196 . settings in 1,000 burn-in and 1,000 sampling iterations (-X10), applying the algorithm. 197 . multiple times (-X5) for each locus with different random seeds. Most genes showed high. 198 . posterior allele pair probabilities (p > 0.99); for GRIN1 two individuals had allele pairs. 199 . with p=0.70 and two other individuals with p=0.85. For PCK1, four individuals had allele. 200 . pairs with p= 0.61.. 201 . We combined our sequence data with previously published sequences for a total of 69. 202 . individuals of O. j. jamaicensis (North American lowlands), five of f O. j. andina. 203 . (Colombian highlands; only mtDNA data was available), and 42 of O. j. ferruginea. 204 . (lowlands and highlands from Ecuador to Argentina). For simplicity, in the following, we. 205 . refer to our three populations based on the distribution of the three Ruddy Duck. 206 . subspecies, not on formally defined geographic regions: North America corresponds to. 207 . areas occupied by O. j. jamaicensis, Colombia to areas occupied by O. j. andina, and. 208 . Southern Andes to areas occupied by O. j. ferruginea. To infer relationships among. 209 . haplotypes at each locus, we constructed unrooted parsimony allelic networks in TCS. 210 . v1.21 (Clement et al., 2000). Because samples sizes between populations differred, we. 211 . corrected allelic richness standardized to the smallest sample size using the Rarefaction. 212 . Calculator (http://www2.biology.ualberta.ca/jbrzusto/rarefact.php)..

(8) 213 . To examine the distribution of genetic variation within and among North America,. 214 . Colombia and Southern Andes, we calculated pairwise Φst values for each locus based on. 215 . the Tamura & Nei (1993) nucleotide substitution model in Arlequin v.3.5.1.3 (Excoffier. 216 . and Lischer, 2010). To test whether different geographic populations (North America,. 217 . Colombia, Southern Andes) are genetically distinct clusters, we used the multilocus. 218 . clustering algorithm in the program STRUCTURE v 2.2.3 (Pritchard et al., 2000), which. 219 . uses a Bayesian approach to calculate the likelihood of an a priori defined number of. 220 . populations and to assign individuals to these populations. We based this analysis on our. 221 . six-locus data set and included all the individuals for which no more than two of the. 222 . genes were missing (n=78). We used the admixture model (α=1) with correlated allele. 223 . frequencies and tested models for varying numbers of populations (K=2-4). Each. 224 . algorithm was run for 1,000,000 generations (burn-in 100,000 generations) and replicated. 225 . 10 times. To choose the best value of K we used the Evanno et al. (2005) method. The. 226 . outputs of the repeated runs at each K were summarized using the greedy algorithm in. 227 . CLUMPP v1.1.2 (Jakobsson and Rosenberg, 2007).. 228 . To estimate gene flow (M = m/u) between populations we used a population-genetic. 229 . approach based on isolation-with-migration coalescent models implemented in the. 230 . program IM (Hey, 2010). This analysis allowed us to examine the direction of gene. 231 . exchange based on estimates of historical gene flow (Hey, 2005, 2010; Hey and Nielsen,. 232 . 2004; Nielsen and Wakeley, 2001). Because IM assumes no intralocus recombination, we. 233 . tested for evidence of recombination in our data using the four-gamete test (Hudson and. 234 . Kaplan, 1985) in DnaSP v.5.10 (Librado and Rozas, 2009). GRIN1 and βA showed. 235 . evidence of recombination; for the former we excluded five individuals following. 236 . Muñoz-Fuentes et al. (2013) and for the latter, we removed the recombinant blocks by. 237 . eye. Because mtDNA is maternally inherited, we used an inheritance scalar of 0.25 and. 238 . used the HKY mutation model (Hasegawa et al., 1985). For nuDNA genes we used a 1.0. 239 . scalar because they are biparentally inherited, and implemented the infinite-alleles. 240 . mutation model (Kimura, 1969). Because mtDNA and βA may be under selection. 241 . associated with the occupation of high-elevation environments (Cheviron and Brumfield,. 242 . 2009; McCracken et al., 2009a), we treated these two genes and the other four nuclear. 243 . loci separately, and also ran the analysis with all genes combined. IM was initially run.

(9) 244 . with wide priors to explore the sensitivity of the parameters to varying upper bounds.. 245 . These analyses were then repeated with uniform priors including the entire posterior. 246 . distribution of each parameter observed in preliminary runs. For each run, we reported. 247 . values of gene flow corresponding to the 90% highest posterior density (HPD) estimated. 248 . by IM. To ensure that all the parameters converged, autocorrelation and effective sample. 249 . sizes (ESS) were monitored and runs were continued until the ESS for each parameter. 250 . was at least 100.. 251 . To determine whether individuals from the Colombian population show genetic. 252 . signatures of variation consistent with hypothesized physiological adaptation to hypoxia. 253 . at high elevations as predicted by the biogeographic hypothesis of north-to-south, high-. 254 . to-low colonization (McCracken et al., 2009a; Muñoz-Fuentes et al., 2013), we examined. 255 . whether individuals from this region carried the βA hemoglobin subunit alleles found in. 256 . lowland North America (Thr/Thr-69) or the alleles found further south along the Andes. 257 . (Ser/Ser-69). Additionally, to determine whether ruddy ducks sampled in Colombia at. 258 . 2600 m exhibited evidence of physiological (i.e. hematological) variation possibly related. 259 . to a response to hypoxic stress, we compared the hemoglobin concentration, total. 260 . hematocrit and mean cell hemoblobin concentration (MCHC; [Hb]*100/Hct)) of 15 blood. 261 . samples from the Gravel pits of Guasca in Colombia (2630 m) with measurements taken. 262 . at two localities in Peru, Lake Titicaca in the highlands (n = 51; 3824 m) and El Paraíso. 263 . Lagoon near sea level (n = 11; 0 m). Two-sample T-tests were conducted in R v 2.12.2 (R. 264 . Development Core Team, 2011).. 265 266 . Results. 267 . Genetic structure between populations of Ruddy Ducks varied considerably across the six. 268 . loci, with Φst values ranging from 0.12 to 0.61 (Table 1). Despite this variation, genetic. 269 . differentiation was significant for all population comparisons and loci, except between. 270 . the Southern Andes and Colombia for the FGB intron.. 271 . We found a total of 25 mtDNA haplotypes; 15 of these were present only in North. 272 . America, five occurred only in the Southern Andes, one was shared between North.

(10) 273 . America and the Southern Andes, one was shared between Colombia and the Southern. 274 . Andes, and three were shared between North America and Colombia (Fig. 2). Of the four. 275 . haplotypes found in Colombia, three (including the most common haplotype in North. 276 . America) were shared with North America, and the other one was the most common. 277 . haplotype of the Southern Andes.. 278 . Varying patterns of genetic variation were observed in nuclear introns (Fig. 3). The most. 279 . common FGB allele was found in the Southern Andes and also occurred at high. 280 . frequency in Colombia; six different FGB alleles were found in North America, and two. 281 . of these were found in Colombia as well. For GRIN1, a private allele and a shared allele. 282 . were found in Colombia, and the most common allele was shared among the three. 283 . populations. Of the remaining eight GRIN1 alleles, one was private to the Southern. 284 . Andes, five were private to North America, one was shared between North America and. 285 . the Southern Andes, and one was shared between North America and Colombia. Two. 286 . ODC1 alleles were shared between Colombia and the Southern Andes, and between. 287 . North America and Colombia; three alleles were private to North America and one was. 288 . private to Colombia. The most common PCK1 allele was found in all three populations;. 289 . one additional allele was private to the Southern Andes, and two were private to North. 290 . America. Allelic richness standardized to the smallest sample size was greater in North. 291 . America than in Colombia for all but one locus (ODC1), and more allelic diversity was. 292 . found in Colombia than in the southern Andes for three out of five loci (FGB, ODC1, and. 293 . HBB). However, mtDNA haplotype diversity was greater in the southern Andes than in. 294 . Colombia (Table 2).. 295 . The βA globin showed high levels of variation, with 18 diffferent alleles. The Colombian. 296 . population exhibited one private allele, one allele shared with the Southern Andes, and. 297 . four alleles shared with North America. There were 10 alleles private to North America,. 298 . and two allleles private to the Southern Andes (Fig. 4). As with nuclear introns, allelic. 299 . richness was greater in North America.. 300 . The STRUCTURE analysis showed strong support for the existence of three genetic. 301 . clusters (K = 3) corresponding to the three geographic distinct populations of Ruddy. 302 . Ducks. Most individuals had high probabilities of being assigned (p > 0.93) to their.

(11) 303 . respective population based on their multilocus genotypes, except for two individuals. 304 . from the Colombian cluster that are genetically admixed with North America (p = 0.60. 305 . and p = 0.16; Fig. 5). The H’ value estimated by CLUMPP was very high (p = 0.99),. 306 . indicating that all STRUCTURE runs were congruent.. 307 . Pairwise estimates of gene flow between populations indicated similar patterns in all gene. 308 . combinations (Fig. 6). The mtDNA control region indicated positive gene flow from. 309 . Colombia into North America (m/u=0.35), but the posterior distribution for gene flow. 310 . from North America to Colombia was flat, indicating that it could not be estimated. 311 . reliably (Fig. 6a). The mtDNA also indicated gene flow from the Southern Andes to. 312 . Colombia (m/u = 0.51), whereas gene flow on the opposite direction was probably zero. 313 . (Fig. 6b). For nuDNA introns, the peak estimate of gene flow from Colombia to North. 314 . America was positive (m/u = 0.38), but gene flow into Colombia from North America. 315 . could not be estimated (Fig. 6c). Gene flow from the Southern Andes to Colombia. 316 . estimated based on introns was positive (m/u = 1.94), but these markers revealed no. 317 . evidence of gene flow in the opposite direction from Colombia to the southern Andes. 318 . (Fig. 6d). For the βA hemoglobin, gene flow from North America to Colombia peaked. 319 . sharply at zero, but it could not be estimated in the other direction (Fig. 6e). Hemoglobin. 320 . gene flow from Colombia to the Southern Andes was also zero, but the posterior. 321 . distribution of gene flow in the opposite direction was flat (Fig. 6f). For all loci combined,. 322 . gene flow from Colombia to North America was inferred to be positive (m/u = 0.31),. 323 . whereas gene flow in the opposite direction was probably zero (Fig. 6g). In turn, gene. 324 . flow from Colombia to the Southern Andes appears to be zero, but gene flow into. 325 . Colombia from the Southern Andes appears to have been positive (m/u = 1.52; Fig. 6h).. 326 . Finally, for the βA hemoglobin 28% of the Colombian individuals were homozygous for. 327 . the North American low-elevation genotype ((Thr-69; Muñoz-Fuentes et al., 2013), 20%. 328 . were homozygous for the high-altitude Andean genotype (Ser-β69), and 52% were. 329 . heterozygous (Thr/Ser-β69; Fig. 7).. 330 . Hemoglobin concentration and mean cell hemoblobin concentration in the Colombian. 331 . population was significantly higher (p < 0.001) than in the other two populations. 332 . (highland Peru at Lake Titicaca at 3,824 m and lowland Peru at El Paraíso Lagoon; Fig..

(12) 333 . 8a). Total hematocrit did not differ between the Colombian population and the lowland. 334 . population (El Paraíso; p = 0.65), but was significantly higher in the highland Lake. 335 . Titicaca population (p < 0.001; Fig. 8b).. 336 . Discussion. 337 . Biogeography of Ruddy Duck Populations. 338 . The phylogeographic analysis of Muñoz-Fuentes et al. (2013) was consistent with the. 339 . hypothesis that Ruddy Ducks experienced a stepwise colonization of the northern and. 340 . southern Andes from North America, followed by a colonization of high-to-low. 341 . elevations. However, this study did not include samples from Colombia, the area. 342 . through which Ruddy Ducks would have entered before expanding in South America,. 343 . and where genetic adaptations to cope with hypoxia are inferred to have been acquired.. 344 . If historical occupation indeed occurred north-to-south, then one should find South. 345 . American alleles to be derived from North American alleles in all genes except the. 346 . hemoglobin gene. If colonization involved genetic adaptations in the hemoglobin gene,. 347 . one should find a paraphyletic group formed by lowland alleles from South and North. 348 . America, with lowland alleles from South America derived from South American. 349 . highland alleles. Furthermore, if colonization indeed occurred north-to-south and. 350 . involved genetic adaptation in the hemoglobin gene, then no gene flow from North. 351 . America into Colombia should be observed in hemoglobin sequences because of the. 352 . influence of selection, and Colombian individuals should exhibit hemoglobin genotypes. 353 . previously hypothesized to indicate putative genetic adaptations to life at high. 354 . elevations (McCracken et al., 2009a; Muñoz-Fuentes et al., 2013).. 355 356 . Our analyses, the first to include nuclear DNA sequences (i.e. introns, hemoglobin) from. 357 . Colombia and the first to consider a large number of Colombian individuals sampled for. 358 . mtDNA, show there is likely more complexity to the historical scenario of north-to-. 359 . south colonization than postulated by earlier studies (Muñoz-Fuentes et al., 2013).. 360 . Based on coalescent analyses of multilocus data, gene flow was inferred to occur from. 361 . the Southern Andes to Colombia and from Colombia to North America. Gene flow from. 362 . North America to Colombia could not be estimated, but gene flow from Colombia to the. 363 . Southern Andes was clearly zero. In contrast to the scenario proposed by Muñoz-.

(13) 364 . Fuentes et al., (2013), this pattern could actually be considered consistent with a south-. 365 . to-north colonization of highland areas from the southern South American lowlands, a. 366 . pattern seen in different waterfowl species (Fjeldså, 1985; McCracken et al., 2009b;. 367 . McCracken et al., 2009c; Vuilleumier, 1986; Wilson et al., 2012). In addition, these. 368 . results are consistent with a hypothesis of phenotypic evolution proposed by Siegfried. 369 . (1976), who considered the black cheek of ferruginea to be an ancestral character and. 370 . the white plumage of andina (and jamaicensis) to be secondarily developed, because all. 371 . species except jamaicenis and the Eurasian leucocephala have black heads. The. 372 . southern origin of Ruddy Ducks suggested by our data is also consistent with the. 373 . distribution of the sister species of O. jamaicensis: the Argentine Lake Duck (Oxyura. 374 . vittata) occupies lowland elevations of southern South America (McCracken and. 375 . Sorenson, 2005). However, because the Colombian population is genetically admixed. 376 . (see below), with lineages from both North America and the Southern Andes a plausible. 377 . interpretation of historical gene flow might be that this population has received. 378 . immigrants from both the Southern Andes and North America and such gene flow could. 379 . have obscured our ability to recover the true biogeographic pattern of colonization of. 380 . different parts of the Americas by Ruddy Ducks. Indeed, because the Colombian. 381 . population is genetically admixed, we could not reliably estimate splitting parameters. 382 . measuring the fraction of individuals in the ancestral population contributing to one. 383 . population (Hey, 2005) which would have provided valuable insights on the direction of. 384 . colonization and allowed for a more direct comparison with results of the earlier study. 385 . (Muñoz-Fuentes et al., 2013).. 386 . Another discrepancy between our results and those of earlier analyses is evident in. 387 . terms of patterns of variation in the hemoglobin gene. Muñoz- Fuentes et al. (2013). 388 . found a clear association between genotype and elevation when comparing North. 389 . American and southern South American populations, which led them to propose the. 390 . hypothesis that colonization of South America via highland environments involved the. 391 . origin of possible adaptations in the hemoglobin to cope with hypoxia. This hypothesis. 392 . implied that Colombian individuals would show genetic signatures of possible. 393 . adaptation in the hemoglobin gene. However, we found that not all Colombian. 394 . individuals have the "high- elevation" genotype identified by the earlier study: only. 395 . 20% of the individuals from Colombia were homozygous for the "high-elevation".

(14) 396 . genotype, 52% were heterozygous, and 28% had the lowland genotype found in North. 397 . America. A possible implication of this result is that the Thr/Ser-69 substitution does. 398 . not increase O2-affinity as hypothesized by earlier studies (McCracken et al., 2009a;. 399 . Muñoz-Fuentes et al., 2013); accordingly, such a genetic change would not have. 400 . facilitated the colonization of high-elevation environments by Ruddy Ducks, implying. 401 . that the previously proposed evolutionary scenario may need reevaluation. In any case,. 402 . to determine whether this substitution is of functional importance, isolation of the. 403 . specific hemoglobin isoforms and measurements of the concentration of the allosteric. 404 . effectors should be obtained.. 405 406 . Hybrid Origin of Colombian Ruddy Ducks?. 407 . Our results are consistent with the hypothesis suggested by several authors (Fjeldså,. 408 . 1986; McCracken and Sorenson, 2005; Siegfried, 1976) that the Colombian population of. 409 . Ruddy Ducks might be of hybrid origin. Variation in all genes indicated that the. 410 . Colombian population shares alleles with populations from both North America and the. 411 . Southern Andes as expected for an admixed population. However, variation in two. 412 . introns (GRIN1 and ODC) and in the βA globin gene indicates that the Colombian. 413 . population has likely had enough time to diverge independently and acquire private. 414 . alleles not found in other populations. In addition, pairwise estimates of genetic structure. 415 . and STRUCTURE analysis showed that the Colombian population is well differentiated. 416 . from the others. Furthermore, we observed no relationship between the genetic. 417 . assignment probability of different individuals and their pattern of cheek coloration,. 418 . which suggests that the phenotypic variability existing among Colombian individuals is. 419 . not the result of recent influx of individuals from other populations. Taken together, our. 420 . data are consistent with the hypothesis that during periods of glaciation (Fjeldså, 1986;. 421 . Van Geel and Van der Hammen, 1973) Ruddy Ducks from the highlands of the Southern. 422 . Andes could have extended north to the Bogotá plateau in the Colombian Cordillera. 423 . Oriental, where they could have hybridized with formerly migratory individuals from. 424 . North America giving rise to a new geographic population which later diverged in. 425 . isolation following range fragmentation..

(15) 426 427 . Significance of Environmentally-mediated Physiological Traits. 428 . Genetic changes in hemoglobin genes at high elevations are expected to evolve in. 429 . populations that have been exposed to selective pressures for a sufficiently long period. 430 . of time such that adaptive substitutions can arise and become fixed. Alternatively, in. 431 . populations more recently exposed to hypoxia one would more likely expect plastic. 432 . physiological responses involving alterations in hemoglobin concentration and. 433 . hematocrit levels. Because increased hematocrit may cause a detrimental increase in. 434 . blood viscosity, theory predicts that in populations in which genetic changes increasing. 435 . O2 affinity have already occurred, hematocrit should be reduced to values similar to. 436 . those seen at low elevations (Storz, 2010; Villafuerte et al., 2004). Considering that. 437 . Colombian Ruddy Ducks may not have genetic adaptations to increase O2 affinity,. 438 . there is some evidence of physiological changes that arise as a result of a possible. 439 . plastic response that is associated with a recent history of life at high elevations.. 440 . Compared to the low-altitude population at Laguna El Paraiso, Colombian populations. 441 . do exhibit high hemoglobin concentrations, but they also show low hematocrit. This. 442 . pattern is unexpected because hemoglobin concentrations and hematocrit are usually. 443 . positively correlated as an acclimatization response to hypoxia, however the adaptive. 444 . response of high hematocrit might me questionable, this is why low values of. 445 . hematocrit are typical of populations showing adaptations increasing O2-affinity (Storz,. 446 . 2010). The combination of high Hb concentration and low Hct results in high mean. 447 . cell hemoglobin concentrations in Colombian Ruddy Ducks, and this may allow them. 448 . to cope with hypoxia. In agreement with the genetic data showing that the Colombian. 449 . population has differentiated from others to the point that it has acquired private alleles. 450 . in nuclear genes (including hemoglobins), our results showing low hematocrit levels in. 451 . Colombia (equivalent to those from sea-level populations, as expected for long-. 452 . established populations) might be consistent with a relatively long-term occupation of. 453 . the Colombian highlands. To better understand the physiological changes involved. 454 . with the occupation of this region, it would be of great interest to compare. 455 . physiological data between South American and North American populations (such. 456 . data for North American populations are currently lacking), and also to perform.

(16) 457 . transplantation experiments to examine the acclimatization responses (Tufts et al.,. 458 . 2013).. 459 460 . Conclusions. 461 . Our results indicate that examining patterns of genetic and phenotypic (i.e. physiological,. 462 . plumage) variation in Colombian populations is critical to gain a comprehensive. 463 . understanding of the biogeographic history of Ruddy Ducks and other species with broad. 464 . distribution ranges in the Americas. As highlighted by Chapman (1917), Colombia is of. 465 . great interest not only because it is located "at the crux of intercontinental relationships",. 466 . but because it holds diverse climatic and topographic conditions that may considerably. 467 . influence patterns and processes of population differentiation. Therefore, it is imperative. 468 . for studies focusing on Neotropical organisms to include detailed sampling in Colombia. 469 . when addressing phylogeographic questions and examining patterns of geographic. 470 . variation using genetic and phenotypic data. To the extent that this study and others (e.g.,. 471 . Cadena et al., 2007; Gutiérrez-Pinto et al., 2012; Isler et al., 2012) demonstrate the. 472 . importance of including samples from Colombia when studying phylogeography and. 473 . geographic variation, we can only wonder to what extent might studies lacking material. 474 . from this area could have missed some of the complexity inherent to avian biogeographic. 475 . histories.. 476 . Finally, the Colombian Ruddy Duck is endemic to isolated high-Andean wetlands in the. 477 . Bogotá region, and it is currently considered endangered (Botero J. E, 2002). A recent. 478 . study highlighted several factors leading to the decline of populations of endemic and. 479 . threatened birds living in wetlands of the Bogotá region (Rosselli and Stiles, 2012).. 480 . Increased urbanization, competition by coots, poor water quality and poaching are the. 481 . main causes of population declines. Our work provides evidence that the Colombian. 482 . population of Ruddy Ducks is genetically distinct from other populations of this. 483 . widespread species. Along with its morphological diagnosability, this population may be. 484 . considered a full species under different species concepts (Baum and Shaw, 1995;. 485 . Cracraft, 1983.; De Queiroz, 2005, 2007; Templeton, 1989), and minimally, an. 486 . evolutionary significant unit for conservation purposes (Moritz, 2002)..

(17) 487 488 . References. 489 490 . Adams, J., Slavid, E.R., 1984. Cheek plumage pattern in Colombian Ruddy Duck Oxyura jamaicensis. Ibis 126, 405-407.. 491 . Avise, J., 2000. Phylogeography: The history and formation of species, Cambridge.. 492 493 . Baum, D.A., Shaw, K.L., 1995. Genealogical perspectives of the species problem. Experimental and molecular approaches to plan biosystematics, 289-303.. 494 495 . Beall, C.M., 2006. Andean, Tibetan, and Ethiopian patterns of adaptation to high-altitude hypoxia. Integrative and Comparative Biology 46, 18-24.. 496 497 498 499 . Beall, C.M., Brittenham, G.M., Strohl, K.P., Blangero, J., Williams-Blangero, S., Goldstein, M.C., Decker, M.J., Vargas, E., Villena, M., Soria, R., Alarcon, A.M., Gonzales, C., 1998. Hemoglobin concentration of high-altitude Tibetans and Bolivian Aymara. American Journal of Physical Anthropology 106, 385-400.. 500 501 502 503 . Botero J. E, 2002. Oxyura jamaicensis. en: Renjifo, L. M., A. M. Franco-Maya, J.D. AmayaEspinel, G. Kattan y B.López-Lanús (eds.). 2002. Libro rojo de aves de Colombia. Serie Libros Rojos de Especies Amenazadas de Colombia. Instituto de Investigación de Recursos Biológicos Alexander von Humboldt y Ministerio del Medio Ambiente. Bogotá, Colombia.. 504 505 506 . Bulgarella, M., Peters, J.L., Kopuchian, C., Valqui, T., Wilson, R.E., McCracken, K.G., 2012. Multilocus coalescent analysis of haemoglobin differentiation between low- and high-altitude populations of crested ducks (Lophonetta specularioides). Molecular Ecology 21, 350-368.. 507 508 509 . Cadena, C.D., Klicka, J., Ricklefs, R.E., 2007. Evolutionary differentiation in the Neotropical montane region: Molecular phylogenetics and phylogeography of Buarremon brush-finches (Aves, Emberizidae). Molecular Phylogenetics and Evolution 44, 993-1016.. 510 511 . Chapman, F.M., 1917. The distribution of bird-life in Colombia: a contribution to a biological survey of South America. Pub. by order of the trustees.. 512 513 514 . Cheviron, Z.A., Brumfield, R.T., 2009. Migration-Selection Balance and Local Adaptation of Mitochondrial Haplotypes in Rufous-Collared Sparrows (Zonotrichia capensis) Along an Elevational Gradient. Evolution 63, 1593-1605.. 515 516 . Clement, M., Posada, D., Crandall, K., 2000. TCS: a computer program to estimate gene genealogies. . Molecular Ecology 9, 1657–1660.. 517 . Cracraft, J., 1983. Species concepts and speciation analysis. . Current Ornithology 1, 159-187.. 518 519 . De Queiroz, K., 2005. A Unified Concept of Species and its Consequences for the Future of Taxonomy. Proceedings of the California Academy of Sciences 56, 196-215.. 520 521 . De Queiroz, K., 2007. Species Concepts and Species Delimitation. Systematic Biology 56, 879886.. 522 523 . Dowling, D.K., Friberg, U., Lindell, J., 2008. Evolutionary implications of non-neutral mitochondrial genetic variation. Trends in Ecology & Evolution 23, 546-554..

(18) 524 525 . Evanno, G., Regnaut, S., Goudet, J., 2005. Detecting the number of clusters of individuals using the software structure: a simulation study. Molecular Ecology 14, 2611-2620.. 526 527 528 . Excoffier, L., Lischer, H.E.L., 2010. Arlequin suite ver 3.5: a new series of programs to perform population genetics analyses under Linux and Windows. Molecular Ecology Resources 10, 564567.. 529 530 . Fjeldså, J., 1985. Origin, evolution, and status of the avifauna of Andean wetlands. Ornithological monographs, 85-112.. 531 532 . Fjeldså, J., 1986. Color variation in the Ruddy Duck (Oxyura jamaicensis andina). Wilson Ornithological Society.. 533 534 . Gompert, Z., Fordyce, J.A., Forister, M.L., Shapiro, A.M., Nice, C.C., 2006. Homoploid hybrid speciation in an extreme habitat. Science 314, 1923-1925.. 535 536 537 538 . Gutiérrez-Pinto, N., Cuervo, A.M., Miranda, J., Pérez-Emán, J.L., Brumfield, R.T., Cadena, C.D., 2012. Non-monophyly and deep genetic differentiation across low-elevation barriers in a neotropical montane bird (Basileuterus tristriatus; Aves: Parulidae). Molecular Phylogenetics and Evolution.. 539 540 . Hasegawa, M., Kishino, H., Yano, T.-a., 1985. Dating of the human-ape splitting by a molecular clock of mitochondrial DNA. Journal of Molecular Evolution 22, 160-174.. 541 542 . Hey, J., 2005. On the Number of New World Founders: A Population Genetic Portrait of the Peopling of the Americas. PLoS Biol 3, e193.. 543 544 . Hey, J., 2010. Isolation with Migration Models for More Than Two Populations. Molecular Biology and Evolution 27, 905-920.. 545 546 547 . Hey, J., Nielsen, R., 2004. Multilocus methods for estimating population sizes, migration rates and divergence time, with applications to the divergence of Drosophila pseudoobscura and D. persimilis. Genetics 167, 747-760.. 548 549 550 . Hickerson, M.J., Carstens, B.C., Cavender-Bares, J., Crandall, K.A., Graham, C.H., Johnson, J.B., Rissler, L., Victoriano, P.F., Yoder, A.D., 2010. Phylogeography's past, present, and future: 10 years after. Molecular Phylogenetics and Evolution 54, 291-301.. 551 552 553 . Hoekstra, H.E., Drumm, K.E., Nachman, M.W., 2004. Ecological genetics of adaptive color polymorphism in pocket mice: geographic variation in selected and neutral genes. Evolution 58, 1329-1341.. 554 555 . Hopkins, S., Powell, F., 2001. Common themes of adaptation to hypoxia. Hypoxia: from genes to the bedside. Plenum Publishers, New York, pp. 153-167.. 556 557 . Hudson, R.R., Kaplan, N.L., 1985. Statistical properties of the number of recombination events in the history of a sample of dna sequences. Genetics 111, 147-164.. 558 559 560 . Isler, M.L., Cuervo, A.M., Bravo, G.A., Brumfield, R.T., 2012. An Integrative Approach to Species-Level Systematics Reveals the Depth of Diversification in an Andean Thamnophilid, the Long-Tailed Antbird. The Condor 114, 571-583..

(19) 561 562 563 . Jakobsson, M., Rosenberg, N.A., 2007. CLUMPP: a cluster matching and permutation program for dealing with label switching and multimodality in analysis of population structure. Bioinformatics 23, 1801-1806.. 564 565 . Jessen, T.H., Weber, R.E., Fermi, G., Tame, J., Braunitzer, G., 1991. Adaptation of bird hemoglobins to high altitudes: demonstration of molecular mechanism by protein engineering.. 566 567 . Kawecki, T.J., Ebert, D., 2004. Conceptual issues in local adaptation. Ecology Letters 7, 12251241.. 568 569 . Kimura, M., 1969. The Number of Heterozygous Nucleotide Sites in a Finite Population due to Steady Flux of Mutations. Genetics 61, 893-903.. 570 571 572 . Lee, J.Y., Edwards, S.V., Webster, M., 2008. Divergence Across Australia's Carpentarian Barrier: Statistical Phylogeography of the Red-Backed Fairy Wren (Malurus melanocephalus). Evolution 62, 3117-3134.. 573 574 . Librado, P., Rozas, J., 2009. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics 25, 1451-1452.. 575 576 577 . Londo, J.P., Chiang, Y.C., Hung, K.H., Chiang, T.Y., Schaal, B.A., 2006. Phylogeography of Asian wild rice, Oryza rufipogon, reveals multiple independent domestications of cultivated rice, Oryza sativa. Proceedings of the National Academy of Sciences 103, 9578-9583.. 578 . Mathews, C.K., Van Holde, K.E., Ahern, K.G., 2002. Bioquímica.. 579 580 581 . McCracken, K.G., Barger, C.P., Bulgarella, M., Johnson, K.P., Sonsthagen, S.A., Trucco, J., Valqui, T.H., Wilson, R.E., Winker, K., Sorenson, M.D., 2009a. Parallel evolution in the major haemoglobin genes of eight species of Andean waterfowl. Molecular Ecology 18, 3992-4005.. 582 583 584 585 . McCracken, K.G., Barger, C.P., Bulgarena, M., Johnson, K.P., Kuhner, M.K., Moore, A.V., Peters, J.L., Trucco, J., Valqui, T.H., Winker, K., Wilson, R.E., 2009b. Signatures of HighAltitude Adaptation in the Major Hemoglobin of Five Species of Andean Dabbling Ducks. Anglais 174, 20.. 586 587 588 589 . McCracken, K.G., Bulgarella, M., Johnson, K.P., Kuhner, M.K., Trucco, J., Valqui, T.H., Wilson, R.E., Peters, J.L., 2009c. Gene Flow in the Face of Countervailing Selection: Adaptation to HighAltitude Hypoxia in the bA Hemoglobin Subunit of Yellow-Billed Pintails in the Andes. Molecular Biology and Evolution 26, 815-827.. 590 591 592 . McCracken, K.G., Sorenson, M.D., 2005. Is Homoplasy or Lineage Sorting the Source of Incongruent mtDNA and Nuclear Gene Trees in the Stiff-Tailed Ducks (Nomonyx-Oxyura)? Systematic Biology 54, 35-55.. 593 594 . Moritz, C., 2002. Strategies to protect biological diversity and the evolutionary processes that sustain it. Systematic Biology 51, 238-254.. 595 596 597 . Muñoz-Fuentes, V., Cortazar-Chinarro, M., Lozano-Jaramillo, M., McCracken, K., 2013. Stepwise colonization of the Andes by Ruddy Ducks and the evolution of novel β-globin variants. Molecular Ecology 22, 1231-1249..

(20) 598 599 600 . Muñoz-Fuentes, V., Green, A.J., Sorenson, M.D., Negro, J.J., Vila, C., 2006. The ruddy duck Oxyura jamaicensis in Europe: natural colonization or human introduction? Molecular Ecology 15, 1441-1453.. 601 602 . Nielsen, R., Wakeley, J., 2001. Distinguishing migration from isolation: a Markov chain Monte Carlo approach. Genetics 158, 885-896.. 603 604 605 . Peters, J.L., Roberts, T.E., Winker, K., McCracken, K.G., 2012a. Heterogeneity in genetic diversity among non-coding loci fails to fit neutral coalescent models of population history. PloS one 7, e31972.. 606 607 . Pritchard, J.K., Stephens, M., Donnelly, P., 2000. Inference of population structure using multilocus genotype data. Genetics 155, 945-959.. 608 609 . R Development Core Team, 2011. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. (http://www.R-project.org).. 610 611 . Rambaut, A., 2002. Se-Al: Sequence Alignment Editor. Department of Zoology, University of Oxford, Oxford, UK.. 612 613 . Rosselli, L., Stiles, F.G., 2012. Local and Landscape Environmental Factors are Important for the Conservation of Endangered Wetland Birds in a High Andean Plateau. Waterbirds 35, 453-469.. 614 615 . Sambrook, J., D. Russell, 2001. Molecular cloning: A laboratory manual. In: old Spring Harbor Laboratory Press, N.Y. (Ed.).. 616 617 . Sexton, J.P., McIntyre, P.J., Angert, A.L., Rice, K.J., 2009. Evolution and Ecology of Species Range Limits. Annual Review of Ecology, Evolution, and Systematics 40, 415-436.. 618 . Siegfried, W.R., 1976. Social organization in ruddy and Maccoa ducks The Auk 93, 560-570.. 619 620 621 . Simmons, P., Lill, A., 2006. Development of parameters influencing blood oxygen carrying capacity in the welcome swallow and fairy martin. Comparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology 143, 459-468.. 622 623 . Stephens, M., Scheet, P., 2005. Accounting for Decay of Linkage Disequilibrium in Haplotype Inference and Missing-Data Imputation. The American Journal of Human Genetics 76, 449-462.. 624 625 . Stephens, M., Smith, N.J., Donnelly, P., 2001. A New Statistical Method for Haplotype Reconstruction from Population Data. The American Journal of Human Genetics 68, 978-989.. 626 . Storz, J.F., 2010. Genes for High Altitudes. Science 329, 40-41.. 627 628 . Storz, J.F., Moriyama, H., 2008. Mechanisms of Hemoglobin Adaptation to High Altitude Hypoxia. High Altitude Medicine & Biology 9, 148-157.. 629 630 631 . Templeton, A.R., 1989. The meaning of species and speciation: a genetic perspective. In: Speciation and its consequences (Ed.). ed. D. Otte and J. A. Endler. Sunderland,Massachussetts, Sinauer Associates, pp. Pp. 3-27. 632 633 634 . Tufts, D.M., Revsbech, I.G., Cheviron, Z.A., Weber, R.E., Fago, A., Storz, J.F., 2013. Phenotypic plasticity in blood-oxygen transport in highland and lowland deer mice. The Journal of Experimental Biology 216, 1167-1173..

(21) 635 636 637 . Van Geel, B., Van der Hammen, T., 1973. Upper Quaternary vegetational and climatic sequence of the Fuquene area (Eastern Cordillera, Colombia). Palaeogeography, Palaeoclimatology, Palaeoecology 14, 9-92.. 638 639 640 . Villafuerte, F.C., Cárdenas, R., Monge-C, C., 2004. Optimal hemoglobin concentration and high altitude: a theoretical approach for Andean men at rest. Journal of Applied Physiology 96, 15811588.. 641 642 643 . Vuilleumier, F., 1986. Origins of the tropical avifaunas of the high Andes. Pp. 586–622 in F. Vuilleumier and M. Monasterio, eds. High altitude tropical biogeography. Oxford Univ. Press and American Museum of Natural History, New York.. 644 645 646 . Wang, I.J., Crawford, A.J., Bermingham, E., 2008. Phylogeography of the Pygmy Rain Frog (Pristimantis ridens) across the lowland wet forests of isthmian Central America. Molecular Phylogenetics and Evolution 47, 992-1004.. 647 648 . Weber, R.E., 2007. High-altitude adaptations in vertebrate hemoglobins. Respiratory Physiology & Neurobiology 158, 132-142.. 649 650 651 . Weber, R.E., Jessen, T.H., Malte, H., Tame, J., 1993. Mutant hemoglobins (alpha 119-Ala and beta 55-Ser): functions related to high-altitude respiration in geese. Journal of Applied Physiology 75, 2646-2655.. 652 653 . Wilson, R.E., Peters, J.L., McCracken, K.G., 2012. Genetic and phenotypic divergence between low- and high-altitude populations of two recently diverged Cinnamon teal subspecies. Evolution.. 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 .

(22) 670 671 672 673 674 675 676 677 . Table 1. Φst values calculated for pairwise comparisons between three populations of Ruddy Ducks for 6 loci. Bold values are significant at the 0.05 level.. 678 . 679 . North America vs Colombia Southern Andes vs Colombia North America vs Southern Andes Overall. Control Region (mtDNA). FGB. GRIN1. ODC1. PCK1. βA hemoglobin. 0.26. 0.12. 0.52. 0.06. 0.32. 0.19. 0.31. 0.01. 0.81. 0.47. 0.18. 0.37. 0.36 0.32. 0.13 0.12. 0.44 0.61. 0.63 0.41. 0.27 0.28. 0.65 0.42. 680 681 682 . Table 2. Allelic richness corrected to the smallest population simple size (±SD) for six loci. North America. 683 684 685 . Colombia. Southern Andes. Locus. n. Allelic richness. n. Allelic richness. n. Allelic richness. mDNA. 69. 12.2 ± 1.6. 31. 4. 41. 8.8 ± 0.9. FGB. 50. 5.8 ± 0.4. 52. 3.0 ± 0.0. 40. 1. GRIN1. 48. 7.9 ± 0.3. 52. 3.0 ± 0.0. 46. 3. ODC1. 52. 5.2 ± 0.7. 52. 5.0 ± 0.2. 34. 2. PCK1. 48. 3.0 ± 0.0. 52. 1.0 ± 0.0. 40. 2. HBB. 42. 14. 50. 5.8 ± 0.4. 46. 2.9 ± 0.3.

(23) 686 . Figure 1. Current geographic distribution and sampling locations of Ruddy Ducks.. 687 688 689 . Figure 2. Haplotype network for the control region (mtDNA). The three populations are indicated by different colors and abbreviations (CO=Colombia, PE=Peru, BO=Bolivia, EC=Ecuador, AR=Argentina). North American localities are abbreviated according to U.S. and Canadian postal codes.. 690 691 692 . Figure 3. Allelic networks for four nuclear loci (FGB, GRIN1, ODC1 and PCK1). The three populations are indicated by different colors and abbreviations (CO=Colombia, PE=Peru, BO=Bolivia, EC=Ecuador, AR=Argentina). North American localities are abbreviated according to U.S. and Canadian postal codes.. 693 694 695 . Figure 4. Allelic network for the first 578 bp of the βA globin gene. The three populations are indicated by different colors and abbreviations (CO=Colombia, PE=Peru, BO=Bolivia, EC=Ecuador, AR=Argentina). North American localities are abbreviated according to U.S. and Canadian postal codes.. 696 697 . Figure 5. Posterior probability of assignment to each of the three populations for six loci using STRUCTURE. From right to left clusters read, North America, Colombia and Southern Andes.. 698 699 700 . Figure 6. Gene flow estimates from IM analysis for pairwise population comparisons (North America vs Colombia and Colombia vs Southern Andes) for separate genes (a-f) and all genes combined (g-h). 90 % HPDs are shown in parenthesis.. 701 702 703 . Figure 7. Genotype frequencies for one amino acid replacement observed in the βA hemoglobin subunit from North American, Colombian and Southern Andes populations Ruddy ducks. Thr/Thr-69, Ser/Thr-69 and Ser/Ser-69.. 704 705 706 707 . Figure 8. Box plot of a) hemoglobin concentration, b) hematocrit percentage and c) mean corpuscular hemoglobin concentration (MCHC) in three Ruddy duck populations, Lowland Peru (0 m), Highland Colombia (2630 m) and Highland Peru (3824 m). Significant differences between populations are marked by a line and an asterisk (p<0.001).. 708 709 710 711 712 .

(24) 713 714 . 715 716 717 . Figure 1..

(25) 718 719 720 . Figure 2..

(26) 721 . Figure 3.. 722 723 . GRIN1 FGB. 724 725 726 727 728 729 730 731 732 733 . ODC1. 734 735 736 737 738 739 740 741 742 743 744 745 . PCK1.

(27) 746 747 . 748 . Figure 4..

(28) 749 . Figure 5.. 750 . Probability of assigment. 751 . 752 753 754 755 756 757 758 . . Population. . .

(29) Figure 6. 0.024. Posterior density. 0.006. a) mtDNA (control region). b) mtDNA (control region). Colombia to North America 0.35 (0.00-3.86). 0.0045. 0.018. 759 . North America to Colombia 11.98 (1.66-12.0) 0.003. 760 0.0015. 761 . 0 0. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 12 762 . 11. 0.004. Posterior density. c) nuDNA (introns) 0.003. Colombia to North America 0.38 (0.01-10.58). 0.002. North America to Colombia 1.03 (0.01-14.06). 763 . 764 765 . 0.001. 766 . 0 0. 2. 4. 6. 8. 10. 12. 14. Posterior density. 0.012. 0.006. 0 0.0032. 0. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. d) nuDNA (introns). ---- Colombia to Southern Andes 0.07 (0.01-14.76) ---- Southern Andes to Colombia 1.94 (0.01-12.94). 0.0024 0.0016 0.0008 0 0. 2. 4. 6. 8. 10. 12. 14. 0.008. e) beta globin 0.009. f) beta globin 0.006. Colombia to North America 0.01 (0.01-4.39). ---- Colombia to Southern Andes 0.00 (0.00-6.43) ---- Southern Andes to Colombia 0.2 (0.01-6.82). America to Colombia 10.10 (2.05-12.00) e) betaNorth globin 0.006. 0.004. 0.003. 0.002. 0 0. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 0.012. Posterior density. ---- Colombia to Southern Andes 0.01 (0.01-1.77) ---- Southern Andes to Colombia 0.51 (0.01-5.85). 0.012. 0 0. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 0.024. g) 6 loci. 0.009. Colombia to North America 0.31 (0.01-2.93). 0.006. 0.006. 0 0. 2. 4. 6. 8. Gene flow (M=m/u). 10. 12. ---- Colombia to Southern Andes 0.01 (0.01-1.67) ---- Southern Andes to Colombia 0.7 (0.01-6.31). 0.012. North America to Colombia 0.41 (0.01-16.93). 0.003. h) 6 loci. 0.018. 14. 0 0. 1. 2. 3. 4. 5. 6. 7. Gene flow (M=m/u). 8. 9. 10. 11. 12.

(30) 767 Figure 7.. North America. Colombia. Frequency. 768 . Southern Andes.

(31) 769 770 771 . Figure 8. a). *. 775 776 *. 777 778 779 780 781 782 b). 785 786 787 788 789 790 791 792 793 794 795 . *. *. *.

(32) 796 797 *. *. c). 800 801 802 803 804 805 806 807 808 809 . Supporting Information. 810 811 . Table S1.. 812 . PCR annealing temperature (Ta) for locus specific primers. Locus mtDNA (control region) ODC1 FGB GRIN1 PCK1 HBB. 813 . Primer L78. Sequence (5´-3´) GTTATTTGGTTATGCATATCGTG. H774. CCATATACGCCAACCGTCTC. ODC1-5.F. TCGTTCAAGCCATTTCTGATGCC. ODC1.6.R. CCAGGRAAGCCACCACCAATRTC. FGB-7.F. GTTAGCATTATGAACTGCAAGTAATTG. FGB-7.R. TTTCTTGAATCTGTAGTTAACCTGATG. GRIN1-11.F. CTGGTGGGGCTGTCTGTG. GRIN1-11a.R. ACTTTGAASCGKCCAAATG. PCK1-9.F. CAGCCATGAGATCTGAAGCA. PCK1-9.R. TTGAGAGCTGGCTTTCATTG. StartF1. GCCACACGCTACCCTCCACCCGACACC. In2R3Oxy. CCTGCCCATCCTTCTGGATCTGCC. Ta 55°C 62.9°C 55.9°C 55°C 55.8°C 71.3°C.

(33)

(34)

(35)

(36)

(37)

(38)

(39)