sign in sign up
<<
Home , Desert finch, Finch, Greenfinch, Rhodopechys

Studies in Avian Biology No. 22:98-103, 2001.

PHYLOGENETIC PLACEMENT OF THE POOULI,‘ MELAMPROSOPS PHAEOSOMA, BASED ON MITOCHONDRIAL DNA SEQUENCE AND OSTEOLOGICAL CHARACTERS

ROBERT C. FLEISCHER, CHERYL L. TARR, HELEN E JAMES, BETH SLIKAS, AND CARL E. MCINTOSH

Abstract. The Po‘ouli (Melamprosops phaeosoma) is a small oscine first discovered on Maui in the early 1970s and originally described as a member of the (Hawaiian honey- creepers). A recent study suggested that the Po‘ouli may not be a drepanidine because it lacks most of a small set of drepanidine synapomorphies (e.g., specialized tongue morphology and distinctive odor). We conducted phylogenetic analyses of the Po‘ouli and a number of drepanidine and potentially related songbird taxa. Our character sets included mitochondrial DNA sequences (obtained for Me- Zamprosops via PCR of DNA isolated from museum specimens) and osteological characters. Analyses support the placement of the Po‘ouli within the drepanidine clade, although the position of the Po‘ouli within the clade is not strongly supported by either data set. Our results indicate that the Po‘ouli is relatively distinct phylogenetically among drepanidines. If a goal of biodiversity conservation is to retain as much genetic diversity as possible then the Po‘ouli should be considered a species of very high priority for conservation efforts. Kev Words: ancient DNA: Dreoanidini: Melamprosops phaeosoma; mitochondrial DNA; osteology; phylogeny; Po‘ouli. L

In 1973 a new and species of Hawaiian relationships and phylogenetic uniqueness of the was discovered by a group of student re- Poouli’ will help in deciding how much effort searchers in a small area of rainforest on the should be expended to recover the species (Faith north slope of Haleakala Volcano on Maui. It 1992, Krajewski 1994). Here we present cladis- was described from two collected specimens as tic analyses of mitochondrial DNA sequences the first new, living species of Hawaiian hon- and skeletal morphology that indicate that this eycreeper (Drepanidini) to be found in over 50 troubling (and troubled) little bird is a Hawaiian years (Casey and Jacobi 1974). Later, however, honeycreeper, albeit an extremely distinctive doubts arose concerning whether this small, one. brown, snail-eating bird is a drepanidine or some METHODS other type of songbird (Pratt 1992a). It was giv- SAMPLED TAXA en the scientific name Melarnprosops phaeoso- ma, and the common Hawaiian name Poouli‘ We compared DNA and skeletal characters of Me- (which means “black-faced” in reference to its lamprosops to a sampling of taxa from within the Dre- panidini, Carduelini, Fringillini, Emberizinae, and oth- prominent black mask). The Poouli‘ is now on er outgroups. Common and scientific names of North the verge of extinction. Recent and intensive ef- American and Hawaiian taxa follow the AOU Check- forts to locate the species has resulted in detec- list (1998). Common and scientific names of other tion (and marking) of only three individuals (S. taxa, and subfamily classifications, are from Monroe Reilly and M. Collins, pers. comm.; Reynolds et and Sibley (1993). al. this volume). It is possible that this number Drepanidini analyzed for mtDNA sequence or os- represents the entire living population for the teology (see Figs. 1 and 2) include Nihoa , Te- species. lespiza ultima; , T. cantans; , Lox- ioides bailleui; 05,‘ Psittirostra psittacea; Lana‘i Although the Poouli‘ differs in morphology, Hookbill, Dysmorodrepanis munroi; , behavior and ecology from other living Hawai- Pseudonestor xanthophrys; Kaua‘i Creeper, Oreomys- ian (Pratt et al. 1997b), its phylogenetic tis bairdi; Hawai‘i Creeper, 0. mana; Maui ‘Alauahio, uniqueness and closest relatives remain uncer- montana; ‘Akeke‘e, caeruleiros- tain (Bock 1978, Pratt 1992a). According to tris; ‘Akepa, L. coccineus; ‘Akiapola‘au, Pratt (1992a), Melamprosops completely lacks munroi; Lesser ‘, H. obscurus; ‘Anianiau, H. the few synapomorphies that define the Drepan- parvus; Kaua‘i ‘Amakihi, H. kauaiensis; O‘ahu ‘Ama- idini, most notably the unique musty odor and kihi, H. javus; Maui ‘Amakihi, H. virens wilsoni; Ha- specialized tongue characteristics. It also differs wai‘i ‘Amakihi, H. v. virens; ‘I‘iwi, Vestiaria cocci- nea; Hawai’i , paci$ca; ‘Apapane, from all known drepanidines in color sanguinea; and ‘Akohekohe, Palmeria do- and pattern, bill morphology, vocalizations, diet lei. (i.e., specialization on snails), and other aspects Carduelini analyzed include the White-browed Ro- of behavior (Pratt 1992a). Knowledge about the sefinch (Curpodacus thura, Genbank number

98 PLACEMENT OF POOULI--FZeischer‘ et ul. 99

c I Buff-barred

FIGURE I. Phylogenetic tree constructed using a maximum parsimony criterion from mitochondrial DNA cytochrome h sequences. The phylogram is one of two maximum parsimony trees of (weighted) length 1255 and CT of 0.53. The numbers at particular nodes are the percentage of trees containing the node following a 500 repetition bootstrap. Nodes with percentages below 50% are not noted. These nodes are assumed to be unresolved and their branches collapse to a polytomy. See Methods for scientific names of taxa exhibited here.

AFO15765), (C. mexicuncts; Fleischer et bler (Protonotaria citrea; this study), Brown-headed al. 1998) Common (C. erythvinus), Purple Cowbird (Molothrus ater; this study), Northern Car- Finch (C. purpureus), Spot-winged (Mycer- dinal (Cardinalis cardinalis), Black-and-white Warbler obas melanozanthos), (Hesperi- (Mniotilta varia), Vesper Sparrow (Pooecetes grami- phona vespertina), ( obso- neus), White-lined (Tachyphonus rufus), Red- letu), Golden-winged Grosbeak (Rhynchostruthus so- winged Blackbird (Agelaius phoeniceus), and Saffron cotranus), European ( chloris), Finch (Siculisflaveola). Outgroups are the House Spar- (C. pinus) Red (Loxin curviros- row (Passer domesticus) and the Buff-barred Warbler tra), Yellow-fronted Canary ( mozambicus), (Phylloscopus pulcher; Y10732). Grey-headed Bullfinch ( erythracu), (Pinicola enucleator), and MITOCHONDRIAL DNA (Leucosticte arctoa). The (Frin- DNA was isolated from samples taken from the only gilla coelebs) is a fringilline outgroup. two Melamprosops museum specimens that exist. The Emberizines include the Green Honeycreeper (Chlo- tip of one small secondary feather was removed from rophanes spiza; Fleischer et al. 1998), Scarlet-rumped the B. P Bishop Museum specimen (holotype: BBM- Tanager (Ramphocelus passerinii; U15717), Summer X1471 12; under the care of C. Kishinami and A. Al- Tanager (Piranga rubra; U15725), Prothonotary War- lison), and a small piece of skin from the ventral open- 100 STUDIES IN AVIAN BIOLOGY NO. 22

L L___ >--LA ...... -win ed Grosbeak Evenin 8 rosbeak Desert f inch Golden-win ed Grosbe !ai n:_-European rT:-I.!- 8 reenfinch

Yellow-fronted Canary ; - Grey-headed Bullfinch ! Pine Grosbeak Puyple_Finch_.

FIGURE 2. Phylogenetic tree constructed using a maximum parsimony criterion from a matrix of osteological characters. A strict consensus of 128 optimal trees found by repeated random searches of these data (500 replicates, closest addition sequence with ten trees held at each step, initial tree improved upon with TBR branch swapping; optimal tree length 286 steps). See Methods for scientific names of species included in the tree.

ing was taken from the American Museum of Natural extractions and centrifugal dialysis to remove buffer History specimen (paratype: AMNH-8 10456; under and other solutes (as in Paxinos et al. 1997). the care of G. Barrowclough). Museum specimen We amplified and sequenced two regions of mtDNA DNA was isolated in a small laboratory dedicated to from the museum and modern specimens (Fig. 1) using ancient DNA analyses using “ancient DNA” proce- the polymerase chain reaction and specific primers: (1) dures (e.g., Cooper et al. 1996, Paxinos et al. 1997). 675 bp of the Cytochrome b (Cyt b) gene in two over- Modern DNA analyses were conducted in a laboratory lapping pieces (see Fleischer et al. 1998); and (2) 224 separated by >500 m from our ancient laboratory. bases of the 5’ end of the mitochondrial control region Briefly, DNA was isolated by digesting skin or feather (CR; Tarr 1995). Cyt b and CR sequences were also pulp overnight at 55” C in a DTT-SDS-EDTA buffer obtained for some non-drepanidine songbird species with proteinase K, followed by phenol and chloroform from Genbank (see Fig. 1). The Cyt b sequence was PLACEMENT OF PO‘OULI--Fleischer eb al. 101 amplified only from the AMNH specimen, and ana- gardless of whether transversions are weighted lyzed with 18 other drepanidine taxa as reported in 4.0: 1, 10.0: 1, or unweighted relative to transi- Fleischer et al. (1998). The CR segment was amplified tions (although weighting and additional out- from the BPBM specimen only, and from an additional group taxa does affect the topology of drepani- 6 drepanidine species. PCR controls were negative (i.e.. no apparent product produced) for the study skin dine relationships). Forcing the Poouli‘ from the amplifications for both Cyt b and CR. Sequences were Drepanidini to the Carduelini (in MacClade; produced either manually as in Fleischer et al (1998) Maddison and Maddison 1992) increases the or on an ABI-373 automated DNA sequencer as in length of the tree (unweighted) in Figure 1 by Greenberg et al. (1998), and were aligned with Se- 12 additional steps. This constrained tree is sig- quencher 3.0. Phylogenetic reconstructions and other nificantly longer than that of Figure 1 based on analyses utilized PAUP*4.0d64 (D. Swofford, pers. both parsimony (Kishino-Hawegawa test, t = comm.) and MacClade 3.01 (Maddison and Maddison 2.69, P = 0.0072; Kishino and Hasegawa 1989) 1992), and are described in the results section below. and maximum likelihood (G = 52.71, P < OSTEOLOGY 0.001; Felsenstein 1988) tests. Making Melam- A subsetof data from a separate study of cranial prosops the sister to each emberizine clade also osteology and phylogeny in the drepanidines (James significantly increases tree length (by 20-27 ad- 1998) was used to determine if the Po‘ouli is supported ditional steps; Kishino-Hasegawa test, t = 3.56, as part of the drepanidine clade. The original study P < 0.001). involved 72 characters and 55 species of drepanidines, Distance analyses further support a drepani- including 17 species that became extinct follow- dine relationship for Melamprosops. Kimura 2- ing human settlement of the archipelago less than two parameter and gamma-corrected distances were thousand years ago (James and Olson 1991). For the lower present study. the fossil taXa were excluded in order for comparisons of the Poouli‘ and dre- to specifically examine the phylogenetic placement of panidines (0.086 2 0.002, range 0.062-0.102) Melamprmops relative to extant or historically extinct than for comparisons of the Poouli‘ and cardue- drepanidines. Twenty-one other species of nine-pri- lines (0.147 + 0.005, range 0.142-o. 152) or em- maried oscines were included so that other potential berizines (0.196 i- 0.010, range 0.170-0.218). relationships might be revealed. Passer domesticus The CR sequence analyses also place the was includedas an outgroup.The resulting matrix had Poouli‘ within the Drepanidini. First, three sin- 45 terminal taxa and 57 informative characters. gle-base deletions found in the Fringillini and The osteological matrix was analyzed using a par- simony criterion. All characters were run as ordered Carduelini CR sequences do not occur in dre- characters except for seven multistate characters that panidines nor in Melamprosops CR sequence were run as unordered because the states were not (Table 1). Second, 1000 replication bootstraps of judged to be sequential. Ten charactershad an essen- maximum parsimony trees (with gaps and trans- tially binary distribution of states except that a few versions weighted 10: 1 or 5:1 over transitions; taxa showed intermediate conditions. In these in- heuristic search) reveal 88% and 90% support, stances, the intermediate condition was scored as a respectively, for monophyly of the drepanidines, third state, but the character was assigned a weight of including the Poouli.‘ Last, forcing the Poouli‘ 0.5 for the parsimony analyses, to prevent intermediate from the Drepanidini into the Carduelini (i.e., conditions from exerting an undue influence on tree length. All other characters were unweighted. sister to Curduelis chloris) or Emberizinae (i.e., as a sister to Melospizu georgiana) increases un- RESULTS weighted tree length by 4 and 10 steps, respec- MITOCHONDRIALDNA tively. The constrained trees are significantly longer (Kishino-Hasegawa test, t = 2.15, P = Cladistic parsimony analyses of the Cyt b se- 0.032 when sister to Carduelis; t = 2.32, P = quences consistently place the Poouli’ within the 0.021 when sister to Melospiza). Drepanidini (Fig. 1). We initially ran a heuristic search in PAUP* with replicated, random addi- OSTEOLOGY tion and no character weighting, and obtained seven equally most parsimonious trees for dre- Parsimony analysis produced 128 equally panidines and carduelines. A maximum likeli- most parsimonious trees from which we derive hood (ML) estimate of the transition-to-trans- a strict consensus tree (Fig. 2). The Poouli‘ is version ratio was then made using the tree with nested within the drepanidine clade in all of the the lowest ML score (ts:tv - 4.O:l). This ratio 128 trees. Moving the Poouli‘ outside the dre- was used to weight transversional changes, and panidine clade to a position as sister to either a heuristic search generated two maximum par- cardueline terminal taxa or cardueline resolved simony trees of length 1255 (unweighted for the clades adds 9 to 23 additional steps to the total same topology is 685 steps) and a consistency tree length. Making the Poouli‘ a sister to index of 0.53 (Fig. 1). Placement of Melampro- or any emberizine outgroup adds 13 to sops within the Drepanidini, however, occurs re- 20.5 steps. 102 STUDIES IN AVIAN BIOLOGY NO. 22

TABLE 1. LISTED ARE 67 VARIABLE NUCLEOTIDE SITES (OF 224 TOTAL)FROM THE 5'.END OR LEFT DOMAIN OF THEMITOCHONDRIAL CONTROLREGION ASSESSEIIFOR~NEEMBERIZINAE (MELOSPIZAGEOKGIANA;GREENBERGETAL. 1998) AND TEN FRINGILLINAE, INCLUDING THREE FRINGILLINI (FR/NGILLA; MARSHALL AND BAKER 1997), ONE CAR- DUELINI (CARDUEL& MARSHALL AND BAKER 1997), AND SIX EXTANT MEMBERS OF THE DREPANIDINI(TARR 1995)

1 2 3 4 5 6

1234567890123456789012345678901234567890123456789012345678901234567

IIIII/IIllIIIIIIIIlIII/IIIII/IIIlIIIIII/IIIl/IIl/IIIIIlIllIIlIIIIlI

Melospiza georgiana TAGCCACGACACCTTATTATGAA-CCACTAGTGA-A-AACACTCCCGTAGGTATATTCAATAGATAGATAG

Frinqilla teydea TGTA.-.T.....A.C..TA..T....A..-.-.G.TA..T...T....-.GCTTC.TA.C..

Frinqilla montifrinqilla ___.T.TAG-..AC...... -..T.CC.GA.-.-.G.TA...... T....-.GCTTC.TA.C..

Frinqilla coelebs C....G...T...... A..-A.T....A..-.-.G.TA...... T....-.GCTTC.TAGC.A

Carduelis chloris .CAAT.A....GT...... A.TM.CT...GA.-G-.GA.A..T....ACAT-GCCTGCCTAGC..

Paroreomyza montana .CA...A...GATC.....C.CTA.AC.AG.GAGG.TGG...... ACT...... C..C.T..C..

Loxioides balleui .CA.T.A....G....C..C.CCAA.T.AC.C..G.G.G...... GT...... C...C.T.....

Telespiza cantans .CA...A...GGT.C.C...ACCA.AC.A.....G.G.G.....NNNNN...... C...C.TT.C..

Hemiqnathus parvus .CA...A...GA...... G.ATCAARC.A..A..G.NNG...... AGT....G....TC.T..C..

Hemiqnathus kauaiensis .CA.T.A...GA...... ACCCAAC.A..A..A.A.G...C...CAC....AG.C.TC.T..CC.

Himatione sanguinea .C..TTA...G...... C.CTAAAC.ATCAC.G.NNG....A.T..T...CG..C.TC.T..C..

Melamprosops phaeosoma .CA.T.A...... C.....TTC.AT.A..A..G.NNG...NNNNNNN...... GCTTC.T..C..

Note: A4elampro.wp.~phaeosomo sequencr 1s from this study.A “." indicatesidentity of the nuclrotide to the topmost base and an “N” indicates a base that could not be called. A “-" indlcateril gap or deletion m the qucnce. Note the three insertions found in all drepanidines relative to fringllline\ (at cites 15, 37, and 54). In addition, there are three drepamdine transversional synapomorphies (22, 26, and 29). See Fig. 1for common names of drepanidine taxa

DISCUSSION not know what factors effected the of In spite of Pratts’ (1992a) assessment that the the brownish coloration and the black facial Poouli‘ might not be a drepanidine, we find con- mask, nor why Melamprosops (and apparently sistent evidence to the contrary. Pratt (1992a) Paroreomyza; Pratt 199213) lack the distinctive notes that the Poouli‘ should be considered a drepanidine odor. “nine-primaried oscine of uncertain affinities,” While our results indicate that the Poouli‘ is and that it “does not look, smell, act, or sound a , the relationships of like a Hawaiian honeycreeper.” Our DNA evi- the Poouli‘ within the drepanidines are not well dence places Melamprosops within the drepani- resolved by the mtDNA data (Fig. 1). Majority dines, and osteological characteristics indeed rule bootstrap analysis results in collapse of sup- make the Poouli‘ “look” like a honeycreeper. porting branches such that Melamprosops be- How does one reconcile the apparent morpho- comes a basal drepanidine lineage. On the strict logical, ecological, and behavioral distinctive- consensus for the morphological trees, the ness of the Poouli‘ (Pratt 1992a; Pratt et al. Poouli‘ joins at a node proximal to the finch-like 1997b, this volume) with our results? Two ex- species but distal to most other living drepani- planations may account for this: (1) some of the dines. It is not depicted as the sister group of phenotypic traits that Pratt emphasizes (i.e., any living drepanidine species. Thus, in both those associated with foraging mode and feed- mtDNA and osteological trees the Poouli‘ ap- ing) may be affected by and pears to represent a unique drepanidine lineage. thus we might expect to see wide diversity in Its lineage may have diverged from other dre- their character states; and (2) some of the 17 panidine lineages prior to evolution of the syn- extinct drepanidine species known only from apomorphic characters defined by Pratt (1992a). may have shared these traits with the How phylogenetically distinct is the Poouli‘ Poouli,‘ thus making it different only in the con- among living drepanidines? To answer this we text of living or historically extinct taxa. We do estimated the contribution of each taxon to the PLACEMENT OF POOULI-Heischer‘ et al. 103 total minimum evolution score for the Cyt b tree ago (fairly early in the drepanidine radiation; in Figure 1. In PAUP”, we constrained the tree Tarr and Fleischer 1995, Fleischer et al. 1998). topology, pruned a taxon from the tree, then re- Of course extinct fossil drepanidines (James and calculated the ME score. The process was re- Olson 1991) not included here, such as Xestos- peated for each drepanidine taxon; each ME piza, may turn out to be more closely related score was subtracted from the total ME score to genetically. Nonetheless, in comparison to other provide a phylogenetic “distinctiveness” score extant drepanidines, the Poouli‘ has had a long, (u) for the taxon (essentially that of Faith 1992). independent evolutionary history. This long pe- The Poouli‘ had the highest U (0.044) among riod of independent evolution can perhaps ex- the 19 drepanidines (mean and SE of U for the plain some of Melamprosops ’ unique phenotyp- other 18 taxa was 0.015 -C 0.002). To evaluate ic characteristics. Such phylogenetic distinctive- the Poouli‘ s’ distinctiveness in the osteology- ness also increases the Poouli‘ s’ conservation based tree we constrained the tree in Figure 2 in value, in that the species represents a significant MacClade 3.01. A drepanidine taxon was re- fraction of the genetic diversity of the drepani- moved and the length of the reduced tree was dines (Faith 1992, Krajewski 1994). Along with subtracted from the length of the total tree. The its singular ecological, behavioral, and morpho- procedure was repeated for each of the 23 dre- logical characteristics, the Poouli‘ s’ unique evo- panidines, and revealed that the Poouli‘ was the lutionary history convinces us that serious ef- fourth most distinctive taxon based on osteology forts should be undertaken to avoid its impend- (after Maui and Kauai creepers and the Akia-‘ ing extinction. polaau).‘ Thus we consider the Poouli‘ to be ACKNOWLEDGMENTS phylogenetically unique among the drepanidi- We gratefully acknowledgeC. Kishinamiand A. Al- nes, and the taxon that individually contributes lison of the B. P Bishop Museum, and G. Barrow- most to extant drepanidine phylogenetic diver- clough of the American Museum of Natural History, sity. for allowing us to sample from the precious Po‘ouli The closest corrected genetic distance be- specimens under their care. The Smithsonian Institu- tween the Poouli‘ and other drepanidines is tion’s Walcott and Wetmore Funds, the National Geo- 0.062. Applying a corrected internal rate cali- graphic Society, and the Friends of the National Zoo provided funding for our Hawaiian honeycreeper re- bration for Cyt b in honeycreepers of about search. We thank T Pratt and S. Reilly for providing 0.016 + O.O05/MY (from Fleischer et al. 1998) information and discussion, and D. Pratt, C. van Riper suggests that the Poouli‘ split from its nearest and S. Conant for reading and evaluating the results in living drepanidine relative about 3.8 ? 0.9 MY the manuscript.