Ecological interactions
Myrmecophiles of the ants of Vachellia (Acacia) drepanolobium
Naomi E. Pierce Museum of Comparative Zoology Harvard University
(2006) Molecular prospecting: Ants and symbiotic bacteria
Alex Wild Bacteria from some ant groups are specialized and closely related
Tetraponera associates are monophyletic
Cephalotini associates are monophyletic C. Moreau
Key: Jake Ant associates Russell Symbionts of plants Symbionts of invertebrates Ants with Rhizobiales have evolved independently
At least 5 separate origins of Rhizobiales associations with herbivorous ants
Key: Herbivore Predator Rhizobiales host Phylogeny from Moreau et al. 2006
. PNAS (2009) 106: 21236 Symbiotic gut bacteria are likely to have facilitated the evolution of herbivory in ants
7 Cephalotes
Jon Sanders
Coevolution of ants and their microbiota ... Characterizing gut bacteria for Cephalotes qPCR corroborates visual evidence
16S copies / ng DNA: ant guts Tetraponera penzigia 1E+07
1E+06
1E+05
1E+04 16S copies / ng DNA 1E+03
1E+02 0.1 1.0 10.0 100.0 extracted DNA concentration Other Guts Cephalotes Guts Polyrachis gasters
Cephalotes rohweri 10µm 0.1mm 1mm 1m Host nuclei Bacteria Abundances per sample
Verrucomicrobiae Gammaproteobacteria Alphaproteobacteria Betaproteobacteria Other
1
1
1
0
0
bohlsi atratus pellans rohweri pusillus setulifur minutus placidus cordatus mompox spinosus eduarduli persimilis clypeatus simillimus scutulatus maculatus borgmeieri pallidoides a ff . pallens peruviensis crenaticeps a ff .angustus grandinosus umbraculatus Procryptocerus sp. Procryptocerus 0.01 Legend:
Dataset species_dots
fdr_sig_nodes
total_nodes
Dataset per_species_heatmap
0
16
32
50
3366
6682
10000
Color ranges:
c__Clostridia
p__GN02
c__Opitutae
k__Bacteria
p__Bacteroidetes
c__Flavobacteria
c__Bacteroidia
c__Bacilli
c__Chloroplast
c__Alphaproteobacteria
p__Proteobacteria
c__Gammaproteobacteria
c__Epsilonproteobacteria
c__Betaproteobacteria Many OTUs codiversify
204
117 33 26
65
92
350
63 71
136 75 121
60 262 189 67
34 27 113 695
170 57 10
35 896
66 138
361 0 426 275
725 209
208 156
45 174 197 78
391 64 515 74 166 286 738 28 14 119 332 183 817
39
47
13
1
20 888 131 Cephalotes 852
31 195
43 574 pOTU width p < 0.01 (tested) 158 37 468 46
2 487
11 6
99 69 89 19 (83) 23 72 899 44
699 289
272 32
148 115 93 19 (80) 529 186 400 790 191 527
16 706
7 12
36 95 15 (123) 409 836 30 29 368 22 48 405 165 671 900 9 21 97 9 (166) 772 766 880 492 965
162 644
566 3
153 125 90 374
829 15 54 709
76 137 489 323 Molecular Ecology (2014) 23: 1268 18 809 288
8 815 520 394
89 993 5
25 83 58
59
88 122
17
967
157 Pollination biology: the evolution of orchid bees
Phylogeny of Euglossini (stingless bees) Fossils used to “calibrate” age of tree
Santiago Ramírez
Penalized likelihood rate smoothing =10 CO1 + Ef1-alpha (F2) 75 50 25 0 My Mechanical floral isolation in euglossine-pollinated orchids
2 1 5 7
8 6 4 1 – Head 7 2 – Antennae 3 – Leg 1 4 – Ventral thorax 5 – Scutum 3 6 – Scutellum 7 – Thorax-abd 9 8 – Abdomen
sequenced nrITS, ycf1 from ~150 pollinaria Pollinaria sampling A 40 30 20 10 0 Eoc Oli Mio Gen sp. Galeandra
Cycnoches
Dressleria Catasetinae Clowesia Epidendroideae Higher Catasetum
Zygopetalinae
Macroclinium Dichaea Warczewiczella Kefersteinia
Stenia Chondrorhyncha Polycycnis
Stanhopea
Soterosanthus Sievenkingia
Coryanthes Stanhopeinae C Cymbidiinae 0.93 Catasetinae Bromheadiinae Eulophiinae Eriopsidinae Oncidiinae Gongora Zygopetalinae 1.0 Maxillariinae
0.92 Stanhopeinae (+Coeliopsidinae)
(My) Tertiary 40 30 20 10 0 Orchids A B 40 30 20 10 0 0 10 20 30 40 Eoc Oli Mio Mio Oli Eoc Gen sp. Galeandra Exaerete
Cycnoches 0.99
Catasetinae Dressleria Clowesia Eufriesea fragrance collection Origin of Epidendroideae Higher Catasetum
Zygopetalinae Eulaema Macroclinium Dichaea Warczewiczella Kefersteinia
Stenia Chondrorhyncha Polycycnis
Stanhopea Euglossa Soterosanthus D Sievenkingia
Coryanthes Stanhopeinae C Cymbidiinae 0.93 Catasetinae Bromheadiinae Eulophiinae Eriopsidinae Oncidiinae Gongora Zygopetalinae 1.0 Maxillariinae
0.92 Stanhopeinae (+Coeliopsidinae)
(My) (My) Tertiary Tertiary 40 30 20 10 0 0 10 20 30 40 Orchids Bees (Science, 2011) Leonora Bittleston
Fluid and insects collected from 3 Nepenthes species in 3 different sites in Singapore. Counts of macroscopic arthropods: 8 common ‘morphospecies’ including flies, mites and ants
Frog-biting midge: Corethrella Mosquitoes: Toxorhynchites, Scuttle fly: Endonepenthia Culex, Tripteroides Anoetid mites: Zwickia
Gall midge: Lestodiplosis Insect prey: mostly ant heads Biting midge: Dasyhelea Counts by eye versus sequencing diversity
Counts Metabarcoding • 8 arthropod • 189 arthropod “OTUs” in “morphospecies” rarefied table • Plus parasites, algae, protozoa, etc.
However, it’s difficult to connect sequences to individuals, especially since reference sequences for most of these organisms are not yet in sequencing databases. Inquiline counts and sequences generally match, but on different scales Mosquitoes and mites seem to be underrepresented in count data
ln (Counts + 1) 1 2 3 4 0 0 1.125 ln (sqrt (Rarefiedln (sqrt Seqs)+1) 2.25 3.375 4.5 Corethrella Lestodiplosis Dasyhelea Endonepenthia Mite Mosquito Phylogenetic trees of dipteran inquilines
18S (ribosomal) CO1 (mitochondrial)
“Endonepenthia”
“Dasyhelea”
“Corethrella”
“Lestodiplosis”
“Mosquitoes” Relative sequence abundance
N. ampullaria communities N. gracilis communities N. rafflesiana communities
Obligate insect parasites! Bipartite network of arthropods and pitcher plants
N. rafflesiana N. gracilis N. ampullaria Networks can reveal which organisms are shared across different host species and samples.
Spring-loaded network of core OTUs present in at least 10% of the samples from each host species. Austral Ecology (2015) Myrmecophiles of the ants of Vachellia (Acacia) drepanolobium Chris Baker Jack Boyle
Dino Martins
Julianne Lori Pelaez Shapiro Ant-plant symbioses
• 465+ myrmecophytic plant species
• 100+ phytoecious ant species
• Restricted to the tropics East Africa... Vachellia drepanolobium is a dominant component of the flora of East Africa
Vachellia drepanolobium with swollen thorn ant domatia
Up to 16 different ant species have been recorded inhabiting Vachellia drepanolobium:
Crematogaster nigriceps Crematogaster mimosae C. sjostedi Tetraponera penzigi
Camponotus rufoglaucus Unidentified spp. Nesomyrmex sp. Three primary ant species occupy trees throughout Kenya, typically one colony per tree
Crematogaster mimosae Crematogaster nigriceps Tetraponera penzigi CM CN TP
Two species of Crematogaster; One Tetraponera Illustrations by Dino Martins The Plants The Ants Domatia Defensive Extra-floral patrolling nectaries Experimental work by Truman Young, Mau Stanton, Todd Palmer, Rob Pringle and others have shown clear differences in patrolling behavior between different ants
C. mimosae C. nigriceps T. penzigi For example, Dino Martins showed that browsing giraffe spend less time feeding on plants inhabited by C. mimosae
(Martins, 2010) This system has many third party associates Expanding the network
A.drepanolobium fungal communities Chris Baker 4/19 Expanding the network
A.drepanolobium fungal communities Chris Baker 4/19 Microdon - a parasitic fly Zoraptera in Tetraponera domatium (feed on fungal spores and hyphae— < 3mm long) Third parties in this system appear to interact at both the mutualistic and parasitic end of the spectrum
A systematic survey? Lycaenid butterflies
Anthene usamba Lycaenid butterfly that lays eggs on the acacias. The larvae are chemically defended against ant attacks. Do females prefer the more aggressive ant mutualist? Preferences tested in bush house Oviposition Experiment Females were given a choice of saplings inhabited by 3 different ant species, or no ants: • C. mimosae • C. nigriceps • T. penzigi • No ants Females of Anthene usamba laid eggs on plants with C. mimosae or no ants
Biol. J. Linn. Soc. 2013 Stable isotope analysis indicates that Anthene usamba is carnivorous, and is likely to be eating ants inside the domatia.
Biol. J. Linn. Soc. 2013 45 species of lycaenids parasitize acacia-ant mutualisms Our survey has involved collecting and processing many individual ant colonies from each of the three species throughout their range Workflow Field collection and DNA extraction PCR + sequencing processing COI primers
aacccctattaggtggatta atgctacgctagcatgctgg actagtgctattcggatatc actgcatgcggggtagctag
OTU 2
statistical and centroid sequence OTU 1
ty ri genetic analyses la i im s % 97
treat each OTU other as a distinct taxon sequences group similar assign taxonomy sequences where possible into ‘OTU’s Numbers summary for myrmecophile data
480 trees sorted 176 trees without myrmecophiles (162 CM, 163 CN, 140 TP)
22,229 domatia 304 trees with myrmecophiles Suyian 6
2000 Suyian 2000 2500 0.5 4
2000 2361 myrmecophiles Kenya meters meters 2000 5000 2500 3000 2 0.0 4000 (median = 3 per tree)
2500 4000 3500 2500 3000 3000 3000 0
2000 2500 2000 2000
1000
2000 0
Kitengela 0 20 40 60 km 1091 with 0 200 km
2000 32 34 36 38 40 42 44 35.5 36.0 36.5 37.0 37.5 sequence data Kenya Kitengela 1270 without 82 different OTUs sequence data (6 high abundance taxa) Numbers summary for myrmecophile data
480 trees sorted 176 trees without myrmecophiles (162 CM, 163 CN, 140 TP)
22,229 domatia 304 trees with myrmecophiles 200 80 Kitengela 2013 2361 myrmecophiles 150 60 full dataset (median = 3 per tree) taxa
40 Suyian 2013 100 chao2 estimator Kitengela 2012 20 50 0
0 1091 with 0 100 200 300 400 500 K12 K13 S13 all sequence data trees collection location/year
1270 without 82 different OTUs sequence data (6 high abundance taxa)
Trees occupied by C. mimosae have more myrmecophiles than those inhabited by C. nigriceps or T. penzigi
All samples 2013 samples only 1.0 1.0 2012 Kitengela 2013 Suyian 0.8 0.8 0.6 0.6 0.4 0.4 myrmecophile incidence myrmecophile incidence 0.2 0.2 0.0 0.0 CM CN TP CM CN TP
ant species ant species Different ants have different myrmecophile communities
2012−13 samples / all myrmecophiles
otu.42otu.16otu.67otu.68otu.6 otu.45otu.51
● ●
− 3 CM otu.30 ● CN ● TP otu.39otu.59otu.80 otu.70 ● ●● − 2 otu.74 otu.19 ● ● otu.0●●● C. nigriceps ●● ●● ● ● ● ● ● ● ● − 1 ● ● otu.3 ● ● ●● ●otu.10 ● ● ● ● ● otu.55 otu.24 ● ● ● ●otu.64● ●● ● otu.15●● ● ● ● ●● ● ● ●●● ● ●●● ●otu.25otu.76 ●● ● ● ● otu.63 otu.72● ●● ●● ● ● ●● ● ● ● otu.71 ●●● otu.46 ● ● ● ●●● ●●● ●● ●● ● ● ● otu.21●otu.27●otu.28 ●
0 ● otu.37 ● ● ● ● ●● otu.17 otu.23● ● ● ● ● otu.38 ●● ● ●● ●● ●● otu.60 CCA2 otu.85 ● ● otu.8● ●● otu.1 ● ●● ● ●● otu.84 ● ● ●●●●● ● ● ●● ● otu.12 ●●● ●●● ● ● ● ●● ● ● ● ● ●● ● otu.57otu.7otu.5 ● ●● ● ● ● otu.54● C. mimosae ● otu.82 1 ● ● ● ● otu.18 otu.33otu.26 otu.43otu.22otu.78 ● otu.40 otu.35 otu.14otu.11otu.31otu.62otu.66otu.9 ● otu.83● otu.73 ● otu.65 otu.34 T. penzigi ● otu.41 otu.32otu.20otu.48otu.53otu.77otu.86 2 ● ●
● otu.50 ● otu.69 ● ● 3 otu.29otu.13otu.56otu.61otu.58otu.81
3 2 1 0 −1 −2 −3
CCA1 Different ants have different myrmecophile communities
2012−13 samples / all myrmecophiles
●
− 3 CM ● CN ● TP − 2 otu.19 C. nigriceps
− 1 otu.3 ● otu.15 otu.64 otu.25 otu.63 otu.72 otu.71 otu.21otu.27otu.28 0 ● CCA2 otu.85 ●
C. mimosae
1 otu.82
otu.65 T. penzigi 2 3
3 2 1 0 −1 −2 −3
CCA1 Specialists and generalists
2012−13 samples / all myrmecophiles
●
− 3 CM ● CN ● TP ant mimic (generalist!) − 2 otu.19 C. nigriceps Dichomeris sp.
(generalist) − 1 otu.3 ● otu.15 otu.64 otu.25 otu.63 otu.72 otu.71 otu.21otu.27otu.28 0 ● CCA2 otu.85 ● Hystrichophora griseana C. mimosae (CM + CN)
1 otu.82
otu.65 T. penzigi 2 3
3 2 1 0 −1 −2 −3
CCA1 Gastropoda (TP) scale insects (CM) Photographs and barcodes can be used to establish life histories
2012−13 samples / all myrmecophiles
●
− 3 CM ● CN ● TP − 2 otu.19 C. nigriceps Dichomeris sp.
(generalist) − 1 otu.3 ● otu.15 otu.64 otu.25 otu.63 otu.72 otu.71 otu.21otu.27otu.28 0 ● CCA2 otu.85 ● Hystrichophora griseana C. mimosae (CM + CN)
1 otu.82
otu.65 T. penzigi 2 3
3 2 1 0 −1 −2 −3
CCA1 Agassiz 2011 Myrmecophiles
Overall incidence • Varies strongly with year and location • C. mimosae more likely to have myrmecophiles
Taxonomic breakdown • Lepidoptera dominate, then spiders • Domatium communities distinct from canopy communities
Community composition • Strong effect of location, year, and ant species • Strong ant specificity not common, but does occur
Additional potential for barcoding • Allows analysis of broad communities by non-taxonomists • Link up life history information • Population genetics analysis Expanding the network
A.drepanolobium fungal communities Chris Baker 4/19 Expanding the network
Fungal communities
How many species? • What species? •
What governs diversity? • What role in system? •
A.drepanolobium fungal communities Chris Baker 6/19 Why fungal communities?
Dino Martins
Showed that workers of Tetraponera penzigi seemed to be cultivating and harvesting white fungus in nest tubes in the lab
A.drepanolobium fungal communities Chris Baker 7/19 Why fungal communities?
Fungus removal
n = 12 1.0 0.8 0.6
n = 13 n = 5
0.4 Showed that when ● offered fungi cultured ● 0.2 ● from the domatia, workers of T. penzigi Proportion of plate area removed by ants Proportion by of plate area removed 0.0 harvested them, CM CN TP whereas C. mimosae Ant species and C. nigriceps were not as interested Chris Baker
A.drepanolobium fungal communities Chris Baker 7/19 Pyrosequencing workflow
gDNA amplicons sequences
GGTTAGTACCTCCATTGT ACTGGTCGCGTGTCATCG GTTGAGATCAAACGTAAC TCCTAGTGACACTTCTTT GGTTAGTACCTCCATTGT
TCCTAGTGACACTTCTTT ACTGGTCGCGTGTCATCG DNA PCR 454 TCCTAGTGACACTTCTTT TCCTAGTGACACTTCTTT extraction GTTGAGATCAAACGTAAC ACTTTCGAGCAAGACCGG ACTGGTCGCGTGTCATCG
Each sample ITS =onetree =onedomatium
OTU clustering
SampleSample 1 Sample 2 Sample 3 Sample 4 Sample 5 Sample 6 Sample 7 Sample 8 Sample 9 Sample 10 Sample 11 12 GGTTAGTACCTCCATTGT OTU 1 GGTTAGTACCTCCATTGT
OTU 2 100 OTU 3 OTU 4 TCCTAGTGACACTTCTTT Mpala leaves 80 OTU 5 taxonomy OTU 6 TCCTAGTGACACTTCTTT
60 OTU 7 TCCTAGTGACACTTCTTT
Mpala domatia OTU 8 Chao1 index Chao1 40 assignment TCCTAGTGACACTTCTTT Kitengela domatia ● OTU 9 ● Kitengela domatia OTU 10 ● ● ● ● greenhouse domatia●
20 0.2 1000 ● ● ● Mpala domatia ● ● 800 OTU 11 ● ● ● GTTGAGATCAAACGTAAC 600 ● ● ● ● x● ● ● 400 ● ● OTU 12 ● ●
● ● 0 ● ● ● ● ¯
200 ● ● ● ● ● GTTGAGATCAAACGTAAC 0 0.0 Rarefaction depth (no of sequences) ● OTU 13
● ● ● PCoA2 ● ● ● ● ● ● OTU 14 ● ● ● ● ● ● ● ● OTU 15 ● ● ● ● ● ● ●
● ● −0.2 ● ACTTTCGAGCAAGACCGG ● ● ● greenhouse domatia ● ● OTU 16 ● OTU 17 ●
● ● Mpala leaves OTU 18 ● ●
−0.4 ● ● ● ● OTU 19 ACTGGTCGCGTGTCATCG −0.4 −0.2 0.0 0.2 statistical OTU 20 ACTGGTCGCGTGTCATCG PCoA1 OTU 21 OTU 22 ACTGGTCGCGTGTCATCG analysis OTU 23
OTU table clusters A.drepanolobium fungal communities Chris Baker 8/19 Fungal sampling
Kitengela Mpala KENYA
CN 8 9 Mpala 0°
CM 7 12 Nairobi
Kitengela TP 7 13
0 200 400 km
CN CM TP
A.drepanolobium fungal communities Chris Baker 9/19 Fungal sampling
Kitengela Mpala greenhouse CN 8 9 8 CM 7 12 3 domatia TP 7 13 2
CN CM TP
A.drepanolobium fungal communities Chris Baker 10 / 19 Fungal sampling
Kitengela Mpala greenhouse CN 8 9 8 CM 7 12 3 domatia TP 7 13 2 CN – CM 3 leaves TP 3
A.drepanolobium fungal communities Chris Baker 11 / 19 Location and sample type a↵ect diversity 100 80
Mpala leaves 60 Chao1 index 40
Mpala domatia
20 Kitengela domatia
greenhouse domatia 0
0 200 400 600 800 1000
Rarefaction depth (no of sequences)
A.drepanolobium fungal communities Chris Baker 12 / 19 Ants have limited e↵ect on diversity in domatia 100 80
Mpala leaves 60 Chao1 index 40
CN domatia 20 CM domatia TP domatia
greenhouse domatia 0
0 200 400 600 800 1000
Rarefaction depth (no of sequences)
A.drepanolobium fungal communities Chris Baker 13 / 19 Community composition varies at class level
1.0 Class breakdown Dothideomycetes Eurotiomycetes Sordariomycetes Tremellomycetes 0.8 other unknown 0.6 0.4 proportion of sequences 0.2 0.0 greenhouse CM CN TP leaves
A.drepanolobium fungal communities Chris Baker 14 / 19 Community composition varies at class level
1.0 Class breakdown
0.6 Dothideomycetes ● ● ● ●EurotiomycetesCN 1.0 ● ● ● ● Sordariomycetes 0.4 ● Tremellomycetes● 0.8 ● ● ● other ● ●● 0.5 ●● ● ● ● ● unknown ● 0.2 ● ● ●● ● ● ● ● ● ● ●●● ● ● ●● ● ● ● ● ● ● ● 0.0 ● ●
0.6 ● 0.0 ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ●● ●
NMDS2 ● ● NMDS2 ● ●● ● ● ● ● ● ● ● ● ● − 0.2 ● ● ● ● ● ● ● ● ●● ● ● − 0.5 ● ● ● ● ● ● ● 0.4 ● TP ● ● ● − 1.0 ● − 0.6 proportion of sequences stress = 0.14 ● CM 0.2 −1.0 −0.5 0.0 0.5 1.0 −0.6 −0.2 0.0 0.2 0.4 0.6
NMDS1 NMDS3 0.0 greenhouse CM CN TP leaves Community composition varies by ant species
A.drepanolobium fungal communities Chris Baker 14 / 19 Location and sample type a↵ect community composition
A.drepanolobium fungal communities Chris Baker 15 / 19 Ant-specific fungal partners? otu404 otu370 otu551 otu022 otu280 otu450 otu099 otu212 otu216 otu673 otu274 otu150 otu303 otu323 otu524 otu220 otu112 otu535 otu318 otu432 otu404 otu370 otu551 otu022 otu280 otu450 otu099 otu212 otu216 otu673 otu274 otu150 otu303 otu323 otu524 otu220 otu112 otu535 otu318 otu432 TP CN CM TP CN CM
A.drepanolobium fungal communities Chris Baker 16 / 19 Summary
Fungal communities
How many species? • What species? •
What governs diversity? • What role in system? •
A.drepanolobium fungal communities Chris Baker 18 / 19 Ant-specific fungal partners? otu404 otu370 otu551 otu022 otu280 otu450 otu099 otu212 otu216 otu673 otu274 otu150 otu303 otu323 otu524 otu220 otu112 otu535 otu318 otu432 otu404 otu370 otu551 otu022 otu280 otu450 otu099 otu212 otu216 otu673 otu274 otu150 otu303 otu323 otu524 otu220 otu112 otu535 otu318 otu432 TP CN CM TP CN CM
A.drepanolobium fungal communities Chris Baker 17 / 19 ●
1.0 ●
●
Alates of C. nigriceps 0.8 carry most of the fungal diversity from 0.6
the domatia in their 0.4
●
infrabuccal pockets 0.2 recovered from CN alates recovered
●
proportion of domatium sequences ● 0.0
CM CN TP
ant Take home message: it’s important to consider ant-plant mutualisms in the context of third party interactions and the ‘extended phenotype’ of different ant species
For example, workers of C. mimosae may be the best defenders against browsing vertebrates, but these colonies harbor more herbivorous myrmecophiles Perhaps the true mutualism is between the plant and the fungi, and ants are simply a vector for fungal colonization? Summary
‘Barcodes’ as identifiers can uncover pattern and process because they provide both identification tools in biological prospecting and historical markers to analyze population processes, phylogeography and phylogeny
• “Ants as legumes” Bacteria may have played a key role in the evolution of herbivorous ants, enabling them to live in the canopy and other nitrogen-poor environments
• Mutualisms are often assymmetrical, with one partner providing the ecological template against which the other can radiate
• Cospeciation between bacterial symbionts and their hosts may be an indicator of mutualism
• Symbiosis is a source of evolutionary novelty `
Thanks! Templeton Foundation National Science Foundation Putnam Expeditionary Fund Baker Foundation
Army ant myrmecophile data from Rettenmeyer (2011) Eciton burchellii 300 obligate myrmecophiles: beetles, mites, spiders, bristletails, snails, flies, wasps, millipedes Ant-plant myrmecophiles: Vachellia drepanolobium: Lepidoptera, spiders, wasps, flies, beetles