Journal of Molecular Evolution

, Volume 72, Issue 3, pp 315–325

The Influence of Historical Geneflow, Bathymetry and Distribution Patterns on the Population Genetics of Morphologically Diverse Galápagos’ Opuntiaechios

Authors

    • Evolutionary Ecology GroupUniversity of Antwerp
  • P. Verdyck
    • Evolutionary Ecology GroupUniversity of Antwerp
  • S. Van Dongen
    • Evolutionary Ecology GroupUniversity of Antwerp
Article

DOI: 10.1007/s00239-011-9434-7

Cite this article as:
Helsen, P., Verdyck, P. & Van Dongen, S. J Mol Evol (2011) 72: 315. doi:10.1007/s00239-011-9434-7

Abstract

Throughout history, remote archipelagos have repeatedly been designated natural laboratories to study evolutionary processes. The extensive, geographically structured, morphological variation within Galápagos’ Opuntia cacti has been presumed to be another example of how such processes shape diversity. However, recent genetic studies on speciation and potential effects of plasticity within this system failed to confirm earlier classification and hypothesized radiation on both global and single island levels. Detailed population genetic information, however, is crucial in conserving these semi-arid ecosystem keystone species. In this article, we re-evaluate the genetics of Opuntia echios inhabiting one of the most taxon rich places on the archipelago: Santa Cruz and its surrounding satellite islands, using microsatellite data. Our analysis revealed high genetic variability within all sampled locations, providing little support for the hypothesis of clonal reproduction. Inter-island gene flow patterns appear to be largely influenced by bathymetry and sea levels during last ice ages. Although O. echios from Seymour Norte are morphologically recognized as being a separate taxon, Daphné Major’s cacti are the most differentiated. In addition, we found a potential barrier for gene flow along the ring-like distribution of Opuntias at the western side of Santa Cruz, suggesting potential links with geology.

Keywords

Population geneticsOpuntiaGeologyGalapagosGene flowBathymetryPolyploidy

Introduction

Historically, extensive morphological differentiation between endemic species inhabiting oceanic archipelagos (i.e. Hawaii, Canary and Galápagos Islands) inspired biologists studying speciation (i.e. Darwin 1849; Wallace 1880; Grant 1998). The discreteness, replicate nature, isolation, simple ecological settings and limited size of these archipelagos result in what might be seen as “natural laboratories” for evolutionary research (Wright 1980). It is no wonder that studies on island radiations have increased our understanding of how speciation takes place, explaining the importance of processes such as ecological release, adaptive radiations, genetic drift, founder effects, and isolation (reviewed in Whittaker and Fernández-Palacios 2007). Impressive examples of plant species assemblages on islands include the Hawaiian silver sword alliance (i.e. Baldwin and Sanderson 1998) and the Asteraceae on the Galápagos (Sønderberg and Adersen 2007). Another, although poorly understood example of such a radiation is found within the long-lived and currently threatened giant Opuntia cacti species of Galápagos.

Opuntias, more commonly known as prickly pears, are keystone species of Galápagos’ arid ecosystem (Hicks and Mauchamp 1996) and display a substantial amount of morphological variation (Wiggins and Porter 1971). Although mainland Opuntia species occupy large geographic ranges exhibiting weak species barriers (Goettsch and Hernandez 2006), species distributions on Galápagos are reported to be limited. With each large island possessing its own taxon (Hicks and Mauchamp 1996), six species, which are divided into 14 subspecific categories, are currently recognized (Wiggins and Porter 1971). Since the majority of variability is found between islands, allopatric island radiation (with open sea acting as the main dispersal barrier) appears as the most likely hypothesis to explain the evolution of these cacti. However, the limited differentiation reported in recent molecular studies (Browne et al. 2003; Griffith 2004; Helsen et al. 2009b) combined with striking inter-island differences in abiotic factors and the putative plasticity of morphological characters used to distinguish species (Gibson and Nobel 1986; Racine and Downhower 1974; Nobel 1981; Hicks and Mauchamp 1996, 2000), raises questions as to the correctness of this current classification and radiation. Since overlapping and continuous morphological characters are used to discriminate taxa (Wiggins and Porter 1971), taxonomic classification is problematic and is currently based primarily on locality.

In order to test which factors affect speciation (gene flow–allopatry–ring-like distribution pattern–plasticity–hybridization), we studied the large island of Santa Cruz (986 km²) in conjunction with a number of smaller satellite islands which encircle it (Fig. 1). Four subspecific varieties of O. echios with non-overlapping distribution ranges were described within this region (Wiggins and Porter 1971) making it an ideal place to test the role of fine-scale allopatric differentiation. However, the current classification seems to be in discordance with the islands’ topography. Whereas plants growing on relatively isolated islets (i.e. >10 km for Daphné Major) are catalogued as being identical to those inhabiting the central region (Santa Cruz and Baltra), cacti from Seymour Norte (isolated by a narrow <1.4 km and shadow strait of ocean) were described as being a separate taxon (var. zacana). Such classification is not in agreement with historical bathymetry (Bard et al. 1990) and suggests that other factors, such as westward ocean currents or animal-mediated seed dispersal, are important in the formation of subspecific groups of Galapagos Opuntias.
https://static-content.springer.com/image/art%3A10.1007%2Fs00239-011-9434-7/MediaObjects/239_2011_9434_Fig1_HTML.gif
Fig. 1

Map of Galápagos (b) displaying its relative position to the South American mainland (a) and the sampling locations on Santa Cruz and its surrounding satellite islands (c). Locations abbreviations (for Santa Cruz localities) with numbers of samples between brackets: LB Las Bachas (39), V1 Venecia1 (11), V2 Venecia2 (30), PB Punta Bowditch (10), NE North of Eden (10), SCB South of Cerro Ballena (10), LP Las Palmas (25), LPC Las Palmas Chica (10), LC Los Corrales (10), CDRS1 Charles Darwin Research Station 1 (33), CDRS2 Charles Darwin Research Station 2 (7), LG La Garrapatera (40), PR Punta Rocafuerte (13), SC Saca Calzon (16), AT Arbol de Tarzan (13), AI Arbol de Ilorón, (13), EC El Cochón (17) and CI Canal Itabaca (17). Shaded areas represent (semi) arid ecosystems

Although early stage allopatric speciation is roughly characterized by the principal of “one island one taxon”, two geographically separated O. echios varieties inhabit the lowlands of Santa Cruz. Whilst southern slopes are dominated by var. gigantea, a high treelike cactus with a tall trunk and short spines, var. echios, a more scrub like cactus with long spines and a short trunk, is exclusively found on the northern side of the island. At a restricted area on the western coast, these two varieties meet and morphological intermediates are found (P. Verdyck and A. Tye pers. observ.), suggesting either hybridization or a morphological plastic response to a rapid environmental transition (Helsen et al. 2009a). Although an earlier genetic study contradicted this observed morphological pattern (Helsen et al. 2009a), it was focused exclusively on the western (and most visited) side of Santa Cruz. Whereas the arid zone (where almost all Opuntias are to be found) forms a ring around higher altitude ecosystems on Santa Cruz, a micro-evolutionary counterpart of a ring-speciation pattern (Mayr 1942; Wake 2001) might have shaped to the observed morphological distribution. A more comprehensive sample strategy including data of eastern individuals (thereby closing the Opuntia distribution ring) may therefore yield valuable information on factors shaping the population genetics of these plants on Santa Cruz.

In this study, we examined gene flow patterns within and between islands using highly variable genetic markers (Helsen et al. 2006, 2009b) in order to: (i) attain a more complete understanding of the genetic constitution of previously described morphological subdivisions and their genetic variability, (ii) assess processes (e.g. allopatric speciation) shaping this genetic distributional pattern, (iii) evaluate the effect of a ring-like distribution pattern on the genetics of cacti on Santa Cruz, and (iv) identify important concerns for future conservation planning.

Materials and Methods

Natural History and Field Samples

The isolated Galápagos archipelago (Ecuador) comprises 13 large islands (>10 km2) and over 130 smaller islands and rocks (Fig. 1) with ages ranging from 0.7 to <5 Myr, following a west-east ageing gradient (Hickman and Lipps 1985; White et al. 1993; Geist 1996). Submerged seamounts located east of the presently emerged island, however, extent timing of first colonisation up to 14 Myr (Christie et al. 1992; Werner et al. 1999). Santa Cruz is amongst the largest and highest island found within Galápagos’ central region and comprises most of the known vegetation zones (Itow 1971). Its 135 km long jagged coastline is characterized by coastal cliffs at the north–north-eastern side (van der Werff and Adsersen 1993). The surrounding islets (Plaza Sur, Baltra, Daphné Major, and Seymour Norte) are characterized by one uniform vegetation zone: the “arid zone”. On Santa Cruz this arid zone ecosystem, which is the most important habitat for Opuntia species on the archipelago, forms a “vegetation” ring around higher altitude regions. Moreover, due to the fact that the southern part of Santa Cruz receives considerably more precipitation than the northern part (Alpert 1946) this arid ring varies in width (Fig. 1).

Opuntia echios samples (n = 444 in total), representing four varieties, were collected during two sampling periods (225 and 219 individuals in 2002 and 2007, respectively) at 22 localities (Fig. 1). Since Santa Cruz is thought to house two geographically isolated varieties and their morphological intermediates, it was sampled most extensively (18 localities). Due to their small size, only one population was sampled on Plaza Sur, Baltra, Seymour Norte, and Daphné Major. Whenever possible, tissues were collected from 30 individuals at any given locality. To avoid clonal sampling, minimal inter-sample distances were set at 100 m (Browne et al. 2003). Using this protocol, many plants were left unsampled for the majority of locations (i.e. distance to other samples <100 m). However, for some locations (especially for the eastern side of Santa Cruz) distances between neighbouring Opuntias exceeded the preconceived 100 m. Here, inter-sample distances were between 150 and 500 m. For each plant a fresh, mature terminal cladode (a modified stem segment resembling a leaf) was separated at the joint with a machete, a process comparable to natural damage by wind, and transported in an individual paper bag to the Charles Darwin Research Station (CDRS) where they were stored at room temperature. Before shipping, pads collected in 2002 were cut into pieces of approximately 10 × 10 cm, and transported in paper bags at ambient temperature to both the Royal Belgian Institute of Natural Sciences (RBINS) and the University of Antwerp (UA) where they were placed in plastic bags and stored at −20°C. New Ecuadorian export regulations, which went into effect shortly after the start of our second collection trip, resulted in an extended preservation period at Galápagos, where moist conditions resulted in suboptimal conservation of fresh material. After the loss of 30 samples (due to rotting), the remaining samples were kept from decay by isolating 25 cm² of tissue (excluding both the cuticle and the parenchyma layer containing a mucilaginous secretion). These samples were stored in four individual plastic vials at −20°C at CDRS. Two of these vials were sent in a cooled container (dry ice) to the UA where they are stored at −20°C whilst the two remaining vials were kept at CDRS (−30°C), where they remain as reserve specimens.

DNA Extraction and Microsatellite Genotyping

Genomic DNA was isolated using DNAeasy Plant Mini Kits (Qiagen). In addition to standard procedures, an extra centrifugation step was executed to the lysate (5 min at 13,000 RPM) to remove most precipitates. To maximize overall DNA yield, extraction products were dissolved in 100 μl dH2O. All individuals were genotyped using the eight most variable microsatellite markers specifically developed for Galápagos’ Opuntias (Helsen et al. 2006, 2009a). Except for changing cy5 labels in FAM (Opuntia1, Opuntia2, Opuntia5), VIC (Opuntia10, Opuntia11, Opuntia15), and NED (Opuntia9, Opuntia13), amplifications were performed as described by Helsen et al. (2006). Resulting PCR products were analysed on ABI 3730 instruments (Applied Biosystem), genotyped using size calling in GeneMapper Software v3.7 (Applied Biosystems), and visually checked afterwards. Thirty individuals collected during the 2002 trip were reprocessed in 2007 to check for potential scoring differences between years. Due to poor quality of some of the material, a limited number of individuals could not be genotyped (even after repeated amplifications) for all loci. For those specimens, these loci were treated as missing data. We evaluated whether these non-amplifications resulted from the impact of amplicon size on PCR success [reviewed by Broquet et al. (2007)].

Analysis of Genetic Variation

Opuntia echios’ hexaploid nature (Helsen et al. 2009a), hampers genetic analysis since individuals may posses up to six different alleles for each locus. Partial heterozygosity (Bruvo et al. 2004) makes it impossible to score genotypes exactly. For example, individuals bearing two alleles for one specific locus (i.e. A and B) may have any of the following genotypes: AAAAAB, AAAABB, AAABBB, AABBBB and ABBBBB.

We used two methods to counter such scoring problems in polyploids: (i) score alleles as presence/absence data (Rodzen et al. 2004) and (ii) a relatively novel approach termed Microsatellite DNA Allele Counting using Peak Ratios (MAC-PR) (Nybom 2004; Esselink et al. 2004; Nybom et al. 2006), hereafter referred to as the P/A- and MAC-PR datasets, respectively. Since P/A methods score co-dominant microsatellites as being dominant, some information is lost and as such the power of the analysis is reduced (Bjorklund 2005). In the latter method, individual allele dosage at a specific locus is estimated based on the assumption that abundant alleles amplify more often than rare alleles do. Relative allele peak heights are subsequently used to evaluate individual allele dosage, thereby correcting for allele-specific amplification success by comparing peak heights of all combinations of alleles (see Esselink et al.2004 for further details). Due to the high genetic variability and hexaploid nature of some samples, we occasionally were unable to define unambiguous correction factors to neutralize this effect. In such cases, we used a more robust method (assigning <6 alleles) comparable to the method described by Jenneckens et al. (2001).

For each locality, the observed heterozygosity (HO) was calculated as the frequency of heterozygotes averaged over loci, whereas expected heterozygosities (HE) were evaluated using both SPAGeDi 1.2 software (Hardy and Vekemans 2002) and HE = 1 − ∑pi6 (Table 1). Alternatively, we evaluated HO accounting for the multiple heterozygosity states polyploid species might be in (i.e. ABBBBB vs. AAABBB) by attributing heterozygosity scores to all possible classes. Whilst such heterozygosity scores were previously described for tetraploid species (Bever and Felber 1992), to our knowledge no such scores were ever reported for hexaploids. Consequently, we recovered heterozygosity scores (1-probability that any two alleles drawn at random were identical by descent) of all possible classes (recombination with replacement) using R (AAAAAA = 0, AAAAAB = 0.33, AAAABB = 0.53, AAABBB = 0.6, AAAABC = 0.6, AAABBC = 0.73, AABBCC = 0.8, AAABCD = 0.8, AABBCD = 0.87, AABCDE = 0.93 and ABCDEF = 1). These scores were subsequently used to assign individual heterozygosity scores.
Table 1

Studied islands and locations, island sizes (km²), sample sizes (N), the number of private alleles (PA) and their frequency, observed heterozygosity (HO) calculated as the frequency of heterozygotes averaged over loci, HO* accounting for the multiple heterozygosity states within polyploid species, expected heterozygosity calculated both for the MAC-PR dataset average over loci/locality (HE) and HE* = 1 − ∑pi6, genetic diversity (GD) calculated by Nei’s unbiased measure

Island

Locality

Island size (km²)

# Samples

# PA (freq)

HO

HO*

HE

HE*

Nei’s genetic diversity

Santa Cruz

 

986

341

59 (0.40)

     
 

El Colchón

 

17

0

0.8996

0.56

0.7274

0.98

0.1502

 

Arbol ILorón

 

13

0

0.9092

0.61

0.7113

0.98

0.1333

 

Arbol de Tarzan

 

13

1 (0.007)

0.8610

0.57

0.7154

0.97

0.1642

 

Saca Calzon

 

16

1 (0.007)

0.8461

0.58

0.6840

0.97

0.1405

 

Punta Rocafuerte

 

13

0

0.9168

0.57

0.7083

0.97

0.1527

 

La Garapatera

 

40

6 (0.042)

0.8815

0.58

0.7516

0.99

0.1530

 

CDRS 2

 

7

0

0.9036

0.62

0.7726

0.99

0.1483

 

CDRS 1

 

33

5 (0.035)

0.8938

0.56

0.7549

0.99

0.1584

 

Los Corrales

 

10

2 (0.014)

0.8656

0.57

0.7545

0.97

0.1699

 

Las Palmas Chica

 

10

2 (0.014)

0.8785

0.53

0.7573

0.98

0.1604

 

Las Palmas

 

25

0

0.8822

0.56

0.7217

0.98

0.1417

 

South of Cerro Ballena

 

10

0

0.7320

0.42

0.6754

0.96

0.1166

 

North of Eden

 

10

1 (0.007)

0.8958

0.53

0.7087

0.96

0.1430

 

Punta Bowditch

 

10

1 (0.007)

0.8500

0.49

0.7386

0.99

0.1436

 

Venecia 2

 

30

1 (0.007)

0.8552

0.53

0.7423

0.98

0.1512

 

Venecia 1

 

11

1 (0.007)

0.8306

0.47

0.6912

0.97

0.1344

 

Las Bachas

 

39

2 (0.014)

0.8502

0.51

0.7355

0.99

0.1771

 

Canal Itabaca

 

17

1 (0.007)

0.8977

0.54

0.7315

0.98

0.1459

Baltra

 

27

24

0

0.8433

0.53

0.7054

0.97

0.1366

Seymour Northe

 

1.9

27

0

0.8837

0.50

0.7196

0.98

0.1285

Plaza Sur

 

0.13

8

0

0.8393

0.51

0.6376

0.94

0.0978

Daphné Major

 

0.32

20

1 (0.007)

0.8668

0.55

0.6939

0.97

0.1198

Genetic diversity of the P/A dataset was estimated for each locality and putative variety using Nei’s unbiased diversity estimator on the individual marker level (Nei and Roychoudhury 1974). Before calculating total genetic diversity for both the locality and island levels, we evaluated the effect of differences in sampling effort on estimates of genetic diversity. Making use of the theoretical logarithmic relationship between sampling effort and genetic diversity, we used the most thoroughly sampled locations (n ≥ 20) to estimate the number of individuals needed to capture most (i.e. 95%) of the total genetic diversity. For each of these more extensively sampled locations, we calculated genetic diversity of repeatedly (n = 100) composed subpopulations containing one to the total number of sampled individuals, thereby randomly picking unique individuals from the original dataset (resampling procedure in R). The result of this simulation was used in later analysis of genetic diversity by either omitting locations containing less individuals than this threshold value and/or adjusting diversity estimates.

Differences in genetic diversity between varieties were evaluated using a mixed-model ANOVA with variety as fixed effect and locality (nested within group) and allele as random effects. We also tested for inter-locality differences within SAS release 9.1 (SAS Institute Inc). Numbers of private alleles were calculated on both the island and locality level.

Genetic differentiation at different hierarchical levels (within locations, between locations and between islands) was calculated on the P/A dataset using analysis of molecular variance (AMOVA) in ARLEQUIN 3.01 (Excoffier et al. 2005). Since AMOVA analysis assumes Hardy–Weinberg equilibrium, the P/A dataset was tested for deviations from this equilibrium using a Bayesian-based deviance information criterion (DIC) selection statistics model implemented in Hickory v1.0.4 (Holsinger et al. 2002; Holsinger and Wallace 2004). The P/A data of all individuals and those of Santa Cruz exclusively were fitted to four models: (i) a “full model” which allows for inbreeding and includes priors for f, πi (the mean of the beta distribution (Holsinger et al. 2002)), and θ; (ii) “f = 0” assuming no inbreeding; (iii) “θB = 0” implying a zero-valued Bayesian FST estimate; and (iv) “f free” model allowing the incorporation of uncertainty about f into the analysis. Default settings were used for burn-in (50000), sampling (250000), and thinning (50), running five independent MCMC runs to test consistency of the outcome. The best fit of these models was evaluated using measures of DIC. Resulting best fitted models were subsequently used to estimate parameters related to genetic structure (θB: an estimate of FST under random-effects model population sampling).

The relationship between geographical and pair wise genetic distances on the locality level (calculated as FST/(1 − FST)) was evaluated to determine whether isolation by distance (IBD, Wright 1943) explains the observed distributions of phenotypic differences. Since Opuntias are restricted to the arid ecosystem, which roughly follows the coastline within high altitude islands, we used shortest coastline distances as an alternative of straight in line pairwise geographical distances. These coastline distances were calculated based on a map displaying sample location as revealed by their coordinates using ImageJ software (Abramoff et al. 2004). Whilst it remains to be tested how Opuntias genes spread within Galápagos, both the effect of geographical distances and their log transformation (one vs. two dimensional, Rousset 1997) were implemented in a partial Mantel test (with 3.105 bootstrap randomizations) within IBDWS (Jensen et al. 2005).

We evaluated abrupt changes in genetic variation using both BARRIER software (Manni et al. 2004) and a method described by Koizumi et al. (2006). This latter method makes use of biased residual values of the regression between pairwise genetic and geographic distances to define divergent populations (Koizumi et al. 2006). We, therefore, verified whether populations with the highest mean residuals contained zero within their 95% CI, as described by Koizumi et al. (2006). If the relevant CI did not include zero, the population was excluded and the IBD procedure was rerun.

Since genetic differentiation between islands (and locations within them) is potentially small and as a result of differences in allele frequencies rather than the presence or absence of alleles, we used a model-based Bayesian clustering approach on the MAC-PR dataset to define the most likely number of clusters or subpopulations (K) (STRUCTURE v. 1.0: Pritchard et al. 2000). This model accounts for the presence of Hardy–Weinberg or linkage disequilibrium by introducing population structure and attempts to find population groupings that (as far as possible) are not in disequilibrium. Based on preliminary analysis, we evaluated the likelihood for K = 1–12 for the whole dataset. For each analysis, ten independent runs of the Gibbs sampler were evaluated for each K to check potential mixing. Because of the potential for being closely related, due to migration and possibly shared recent ancestry, a combination of the “admixture model with no prior population (geographical) information” and the “correlated allele frequency model” was used to analyse our data (Pritchard et al. 2000). Based on stationary in several statistics, burn-in length was set to 5.104 after which 2.106 MCMC replicates were run. The most likely number of clusters was evaluated using ΔK values (Evanno et al. 2005). In order to unravel substructuring, analyses were rerun on partial datasets which represent population genetic clusters as defined by the outcome of the first Structure analysis (Coulon et al. 2008).

Results

No amplifying or scoring differences were revealed between 2002 and 2007 analysis after reprocessing 30 individuals collected in 2002. Moreover, microsatellite amplification and peak patterns proved to be reproducible, as indicated by the reprocessing of 25 individuals. The average number of missing loci per individual was low ≤5% (ranging from 2.7 to 6.9%). Differences in amplification success between loci was significantly correlated with the amplicon length (r = 0.77, df = 5, P = 0.04). With all individuals possessing unique haplotypes according to the P/A dataset (the most conservative dataset), clonal reproduction on higher (≥100 m according to sampling protocol) geographical scale seemed to be negligible. The number of alleles varied from 8 to 29 per locus, with mean observed heterozygosity ranging from 0.63 to 0.77 using the frequency of observed heterozygotes (HO) and 0.47–0.62 using the method correcting for multiple allelic states (HO*). Expected heterozygosity fell within the 0.73 to 0.92 interval (HE) or the 0.96–0.99 interval using HE = 1 − ∑pi6 (HE*) (Table 1). The frequency of private alleles within localities never exceeded 4% of the total number of alleles. A relatively high number of unique alleles were found within plants collected at Santa Cruz compared to other islands, which most likely resulted from a higher sampling effort here (correlating the number of collected samples per island vs. number of PA: r = 0.65, P = 0.0033). For private alleles found in more than 2% of all individuals, only 17 alleles were unique for Santa Cruz.

Genetic diversity, calculated using Nei’s unbiased measure on both the total and the size corrected dataset, was significantly different amongst localities, islands, and between Santa Cruz and its satellite islands (ANOVA, F = 20.0, 14.8 and 12.5; P = 3.10−4, 10−3 and 2.10−3, respectively), with diversity on the central island Santa Cruz significantly higher than on the satellite islands. Moreover, genetic diversity correlated with the natural logarithm of an islands’ arid zone size (km²) (r = 0.940; df = 5; P = 0.022). Within Santa Cruz the highest genetic variability was found on the northern slopes of the island (Las Bachas) and the lowest variability on and western side (South of Cerro Ballena).

In testing for the most likely population model, all Bayesian Hickory runs resulted in comparable outcomes for both the complete (Table 2a) and Santa Cruz (Table 2b) P/A dataset, displaying lowest DIC and Dbar values for the “f = 0” or “no inbreeding model” (DIC = 6685 and 5549 respectively). Whilst these DIC scores where at least six units lower than all other models (Table 2), only this “no inbreeding model” deserved further consideration (Spiegelhalter et al. 2002), indicating that inbreeding was unlikely to be a major driving force in determining the genetic structure revealed by other methods. Moreover, high DIC and Dbar scores of the “θB = 0” model (DIC = 7838 and 6191 respectively) indicated some genetic differentiation within the dataset. According to the most likely model (f = 0), θB (Bayesian estimate of FST) was 0.030 ± 0.002 for the whole central region and 0.022 ± 0.002 within Santa Cruz.
Table 2

Hickory results including all models tested on the P/A dataset

Model

Dbar

Dhat

dP

DIC

(a) All individuals

Full

5918.41

5144.351

774.0595

6692.47

f = 0

5909.521

5134.243

775.278

6684.799

θB = 0

7699.925

7561.895

138.03

7837.955

f free model

5950.424

5084.46

865.9642

6816.388

(b) Only Santa Cruz individuals

Full

4973.48

4398.03

575.42

5548.87

f = 0

4966.18

4389.98

576.21

5542.34

θB = 0

6054.00

5916.94

137.06

6191.06

f free model

4988.63

4318.36

670.26

5658.89

Bold values indicates the lowest DIC score

Geographic Patterns in Genetic Differentiation

Within-locality variation explained 92.52% of the total molecular variance (AMOVA), whereas the residual molecular variance was equally partitioned between the variance amongst island (3.68%) and amongst localities within Santa Cruz (3.81%) (Table 3).
Table 3

Analysis of molecular variance (AMOVA) of the P/A data for Opuntia echios varieties on Santa Cruz and surrounding satellite islands

Source of variation

df

Sum of squares

Variance components

Percentage of variation

Between islands

4

88.72

0.2753

3.68

Amongst locations within islands

17

207.2

0.2850

3.81

Within localities

398

2756.5

6.926

92.52

Total

419

3052.4

7.48619

 

All estimated P-values (tested with 10,000 permutations) were highly significant (P < 0.005)

The genetic data from this study appeared to support the prior prediction that IBD is the most important driving force in explaining differentiation at the central region. Mean inter-island genetic distances (calculated as FST/(1 − FST)) were highest between Daphné Major and Plaza Sur, the geographically most distant islands within our dataset (29 km), and likewise the genetic distances from the satellite islands to Santa Cruz peaked for Daphné Major, the most distant satellite island (≥8.5 km). There was a general tendency of significant correlation between genetic and geographical distances for the island dataset (r = 0.82, df = 8, P < 0.016 and r = 0.81, df = 8, P = 0.05 for log transformed geographic distances). However, analyses at the individual island level displayed no such correlation for Daphné Major (r = 0.04, df = 19, one-sided P > 0.4) (Fig. 2).
https://static-content.springer.com/image/art%3A10.1007%2Fs00239-011-9434-7/MediaObjects/239_2011_9434_Fig2_HTML.gif
Fig. 2

Pairwise differentiation (represented as FST/(1 − FST)) versus the logarithm of the geographical distances (km) for all islands under study

Within the more thoroughly sampled island of Santa Cruz, pairwise genetic distances between adjacent sample localities peaked at the western coast (from Punta Bowditch to Las Palmas). Although IBD was usually positive, the effect was not significant (r = 0.0814, df = 16, P > 0.08 under 30,000 permutations). Three different categories were found within the IBD graphs when plotted per locality, pointing out: (i) significant positive correlations (r ≥ 0.5) (CDRS, CDRS2, LG, PR, SC, AT, AI, EC and LC), (ii) no significant relationships (0.5 ≥ r ≥ −0.5) (CI, NE, SCB, LP and LPC), and (iii) negative correlations (r ≤ −0.5) (LB, V1, V2 and PB). Notably, these three groups had a geographical structure. Positive correlations were found within southern to north-eastern localities, whilst negative relationships characterized the north-western side of the island. All midlying locations showed no correlation. Subsequently, we excluded one location a time from the whole dataset and recalculated geographical distances, considering each location being a barrier for gene flow. Using this approach, the effect of IBD increased substantially when western populations (LB, V1, V2, PB, NE, SCB, LP and LPC) were excluded and peaked whilst excluding NE and PB individuals. At a finer scale, plants from “South of Cerro Ballena” showed a remarkable feature: The 95% credibility interval (CI) of residuals of the IBD regression for this locality was different from all other CI’s (0.06 vs. 0.03), indicating the relative effect of genetic drift was stronger than gene flow at this locality (Koizumi et al. 2006). Pairwise genetic distances between adjacent localities at Santa Cruz peaked at the western side of the island (from Punta Bowditch to Las Palmas), with a similar pattern found using BARRIER software.

Bayesian inference (STRUCTURE software) indicated that cacti from the central region could be divided into multiple subclusters (K). Although Ln probabilities decreased with increasing K values, differences in succeeding Log(P(X|K)), values were too small to make any decision on the most likely number of cluster. ΔK values (Evanno et al. 2005), however, peaked for K = 3, which was geographically consistent since one of the clusters status a “satellite island cluster” (See Supplementary Figure S1). Remaining Santa Cruz samples scored between the two remaining clusters. With each island analysed separately, the presence of a single population best explained the data, with the exception of Santa Cruz. The two clusters found on this island suggest that most individuals have mixed membership in two groups. This, however, makes little biological sense and most likely resulted from the fact that the underlying structure model is not well suited to handle situations were allele frequencies vary gradually (Structure Software documentation).

We also tested for potential population genetic structure within all satellite islands (roughly one of the three proposed clusters of the initial structure outcome) by rerunning the analysis excluding plants from Santa Cruz. This analysis suggests the existence of two discrete Opuntia populations which may shed new light on earlier described varieties (i.e. var. echios and var. zacana) (See Supplementary Figure S1). In concordance with the IBD analysis, a first cluster corresponded to the Daphné Major population, whilst the second cluster contained all other prickly pears from Santa Cruz’s surrounding islets.

Discussion

Island archipelagos have repeatedly been referred to as “theatres of evolution” (Hutchinson 1957) promoting rapid species radiations. However, not all phylogenetic groups display such a clear differentiation of species (i.e. Ricklefs and Bermingham 2007). Although three morphologically “very” distinct O. echios varieties (Wiggins and Porter 1971) were described within Galápagos’ central region, we retrieved only two genetic clusters. The first cluster includes all prickly pears of Santa Cruz (O. echios var. echios and var. gigantea), whereas the second cluster is limited to cacti from Santa Cruz’s satellite islands (var. echios and zacana). Contrary to the previous classification (where plants from Seymour Norte are described as a distinct group), this second cluster is largely defined by cacti from Daphné Major. This new insight appears to make sense from a geological point of view. Sea levels today are 125 m higher than during the last glacial period (20,000 years ago) when Seymour Norte was interconnected to Baltra and Santa Cruz (Bard et al. 1990, Geist unpubl. data). Deeper waters, separating Daphné Major from all other central island, however, did not result in land bridges, at least not within that historical period (Bard et al. 1990). In addition, Daphné Major’s current jagged and steep shoreline makes it difficult for Opuntia pads to recolonize the island. However, Darwin’s finches are reported to breed or visit this island regularly (Millington and Price 1982; Grant and Grant 1980), making inter-island pollination and/or seed dispersal potentially possible (i.e. Ridley 1930). We suggest that the present data do not violate the allopatric model of speciation (in combination with IBD), when evaluated in a historical context. Whilst these new insights in genetic differentiation follow currently accepted phylogeographical theories, they contradict the currently accepted morphological classification. Although we can not exclude that described morphological subdivisions result from differentiation in unsampled genetic regions, this is the fourth study (in addition to Browne et al. 2003; Griffith 2004; Helsen et al. 2009b), that fails to support this earlier classification. Bearing in mind methodological problems and uncertainties of this former morphological subdivision, we suggest caution when using earlier descriptions and distribution patterns as the sole method to discriminate species or varieties.

According to morphology an exception to the “one island one taxon” rule occurs for Santa Cruz where two taxa (O. echios var. echios and var. gigantea) and their potential morphological intermediates have been reported to coexist (Wiggins and Porter 1971, A. Tye and P. Verdyck pers. obs.). A preliminary study, which focused on the western part of the island, revealed little genetic evidence for such a subdivision (Helsen et al. 2009a). In this study, we used a more detailed dataset (covering the whole of Santa Cruz) to attain a more complete understanding of the population genetic structure. This approach proved to be successful in that it revealed a more heterogeneous gene flow pattern than expected from both the cacti’s continuous distribution range and from earlier genetic insights (Helsen et al. 2009a). More specifically, the north-western side of the island, where potential hybridization events have been reported earlier (P. Verdyck and A. Tye pers. obs.), is characterized by lower levels of gene flow. In this specific case, dispersal patterns and/or the presence of a local (historical) barrier for gene flow are therefore amongst the most plausible factors causing this genetic heterogeneity. Low colonization rate of native plant species at Galápagos (Porter 1983), and fast intra-island invasion of Opuntias, make multiple colonisations (as described for tortoises Caccone et al. 2002) less likely here.

We did not find any records describing geological barriers in Santa Cruz’s lower elevation arid zone. On the other hand, we found several other examples for taxa where there are unusual biogeographical gaps between northern and western populations on Santa Cruz. Bursera graveolens (Porter 1973), Alternanthera (A. Tye pers. obs.) and giant tortoises (displaying an isolated north-western saddleback group: i.e. Snow 1964; Pritchard 1996; Caccone et al. 2002) indeed exhibit either phenotypic and/or unexpected population genetic breaks within this region. Also a specialist weevil (Gerstaeckeria galapagoensis Van Dyke 1953), feeding exclusively on Opuntias, is partitioned in a southern and northern monophyletic group at Santa Cruz (Helsen et al. unpublished data).

Inquiring which factors may have created these multiple subdivisions, we re-examined Santa Cruz’s well-documented geological records. Volcanism at Santa Cruz peaked between 2.2 and 3.6 Myr. Although most of Galápagos’ islands are characterized by one central volcano, Santa Cruz displays a series of volcanoes stretching from the south-eastern to north-western side of the island. Taking into account that Galápagos was formed by an eastern movement of the Nazca plate over a stationary and periodically active volcanic hotspot (Simkin 1984), we hypothesize that Santa Cruz’s latest volcanic activity must have taken place at its north-western side. Accordingly, the last barren lava fields should have been situated at this side of the island and may have acted as barrier for geneflow in times when the rest of the island was already colonized.

The distribution of Opuntias on Santa Cruz may also have played part in the observed geneflow pattern. Opuntias almost exclusively inhabit the arid zone (Jackson 1993), which has a remarkable ring-like shape (encircling higher, more humid, zones) on Santa Cruz (Alpert 1946). Conceptually, geneflow along such distribution pattern might be thought of as a micro-evolutionary counterpart of ring speciation (cfr. Jordan 1905). Although it should not be confused with “real” ring speciation (geographical distances are limited and Opuntia speciation rates tend to be low) geneflow patterns are comparable in that genetic differentiation converges with the geographical distance as measured along the ring-like distribution. Following the hypothesis that the north-western side of Santa Cruz was available for colonization much later than other parts of the arid zone, the original Opuntia distribution had a horse-shoe shape. Genetic differentiation likewise peaked between the two tips of this distribution range. Later, when ecological settings changed, this north-western region was colonized by these most differentiated cacti, resulting in the currently observed IBD pattern. As previously noted, Gerstaeckeria galapagoensis shows the same, but more pronounced, genetic structure at the western coast of Santa Cruz. We hypothesize that the combination of their short generation times (≤10 generations a year on the mainland) and host specialisation make these beetles promising candidates for ring speciation along Santa Cruz’s arid zone.

Genetic studies are increasingly a part of planning in conservation programs. Although extinctions or serious population declines may occur naturally, historical records indicate most extinction on Galápagos took place after human colonisation (Steadman et al. 1991). Introduced mammals (Kastdalen 1982; Jackson 1993; Tye 2005) cause enormous devastation within fragile endemic plant and animal populations on remote islands archipelagos such as Galápagos’ Opuntias (Brockie et al. 1988; Grant and Grant 1989; Snell et al. 1994; Coronel 2002). Whilst recently much effort has been made to remove introduced animals, goats and donkeys (the primary threats for Opuntias) still are found throughout Santa Cruz. On the other hand, the success of recent repatriation programs of endangered species, such as giant tortoises on Española (Gibbs et al. 2008), indirectly affect the survival of Opuntias in that they restrict natural regeneration of these cacti (Coronel 2002; Tye 2005).

Attempts to conserve and/or restore Galápagos’ Opuntias are still in their infancy and until now could not rely on genetic insights. Nevertheless, negative effects of small population sizes and/or the combined lack of genetic variability may affect the success of these programs. As indicated by the distance between neighbouring Opuntias (as described in the sampling procedure and calculated using geographical coördinates), population sizes locally can be low. Nevertheless, neutral genetic variability within all sample locations was high and evenly distributed, making it difficult to assign locations that need high conservation priority. Our data, however, do suggest a higher priority for conservation for cacti on Daphné Major and a lower priority for prickly pears on Seymour Norte than previously suggested. In addition, our analysis on the distribution of genetic variability on the individual level did not support the hypothesis that clonal reproduction plays an important role in the distribution, and thereby success, of Galápagos’ Opuntias (Hicks and Mauchamp 2000), which could have a major impact in planning future conservation strategies.

Whilst this is another study describing weak or no molecular support for earlier described taxonomic categories, the accumulating evidence indicates that we should be sceptical about previously described taxonomic subdivisions. However, more experiments (e.g. common garden) are needed before definitive conclusions can be drawn? Moreover, we should bear in mind that Opuntias are keystone species of the semi-arid ecosystem. Therefore, genetic structuring of species that largely or completely depend on these Opuntias (i.e. giant tortoises, land iguana’s, mocking birds and Gerstaeckeria galapagoensis) should be included in defining Opuntia conservation units.

Acknowledgments

Our study was financially supported by: a BOF-NOI project (FAO70400/4) at the Evolutionary Ecology Group of the UA and the Belgian Society Policy. This is a contribution of the CDRS, the institute that provided cooperation and field logistics support. Sincere thanks go out to Robert Browne (Wake Forest State University) for his help on this final manuscript and Xavier Arturo, Daniel Segura, Jorge-Luis Renteria, Fredy Nugra and Luis Molina (all CDRS) for their assistance on the field. Alan Tye, Frank Bungartz, Frauke Ziemmeck, Anne Guezou, Rachell Atkinson, Ivan Aldaz and Patty Jaramillo (all CDRS) for their support and comments at and far away from Galápagos. Finally, we would like to thank reviewers for their constructive work on this manuscript.

Supplementary material

239_2011_9434_MOESM1_ESM.jpg (1.7 mb)
Population structure of O. echios varieties as revealed by Structure analysis for a all localities (K = 3) and b satellite islands (K = 2) based on eight microsatellite loci. Each individual is represented by a single line which is partitioned in K coloured segments according to the individuals estimated membership fractions in each of the K clusters. Localities displayed counter clockwise starting at CDRS (the most southern locations)

Copyright information

© Springer Science+Business Media, LLC 2011