key: cord-0294886-ts91ivwf
authors: Becker, Daniel J.; Speer, Kelly A.; Brown, Alexis M.; Fenton, M. Brock; Washburne, Alex D.; Altizer, Sonia; Streicker, Daniel G.; Plowright, Raina K.; Chizhikov, Vladimir E.; Simmons, Nancy B.; Volokhov, Dmitriy V.
title: Ecological and evolutionary drivers of hemoplasma infection and bacterial genotype sharing in a Neotropical bat community
date: 2020-03-18
journal: bioRxiv
DOI: 10.1101/2019.12.21.885921
sha: d778ed7c4c1b3e298a3612c94a2a71b94a70e50b
doc_id: 294886
cord_uid: ts91ivwf

Most emerging pathogens can infect multiple species, underscoring the importance of understanding the ecological and evolutionary factors that allow some hosts to harbor greater infection prevalence and share pathogens with other species. However, our understanding of pathogen jumps is primarily based around viruses, despite bacteria accounting for the greatest proportion of zoonoses. Because bacterial pathogens in bats (Order: Chiroptera) can have conservation and human health consequences, studies that examine the ecological and evolutionary drivers of bacterial prevalence and barriers to pathogen sharing are crucially needed. We here studied hemotropic Mycoplasma spp. (i.e., hemoplasmas) across a species-rich bat community in Belize over two years. Across 469 bats spanning 33 species, half of individuals and two-thirds of species were hemoplasma positive. Infection prevalence was higher for males and for species with larger body mass and colony sizes. Hemoplasmas displayed high genetic diversity (21 novel genotypes) and strong host specificity. Evolutionary patterns supported co-divergence of bats and bacterial genotypes alongside phylogenetically constrained host shifts. Bat species centrality to the network of shared hemoplasma genotypes was phylogenetically clustered and unrelated to prevalence, further suggesting rare—but detectable—bacterial sharing between species. Our study highlights the importance of using fine phylogenetic scales when assessing host specificity and suggests phylogenetic similarity may play a key role in host shifts for not only viruses but also bacteria. Such work more broadly contributes to increasing efforts to understand cross-species transmission and epidemiological consequences of bacterial pathogens.

12 missing values (n=323), we fit the phylogenetic GLMMs using the brms package, default priors, 235 and infection status as a Bernoulli-distributed response. We included random effects for bat 236 species and phylogeny, the latter of which used the phylogenetic covariance matrix (Bürkner, 237 2017). We ran four chains for 20,000 iterations with a burn-in period of 10,000, thinned every 10 238 steps, for a total 4,000 samples. We compared GLMMs using the leave-one-out cross-validation 239 (LOOIC) and assessed fit with a Bayesian R 2 , including the total modeled variance and that To identify species trait correlates of prevalence, we fit 11 PGLS models (weighted by 258 sampling variance) with body mass, annual fecundity (litters per year * pups per litter), dietary 259 guild, quantitative diet, foraging strata, aspect ratio, roost type, roost flexibility, colony size, 260 geographic range size, and evolutionary distinctiveness as predictors. We also fit PGLS models 261 with only sample size or an intercept. We compared models with Akaike information criterion 262 corrected for small sample sizes (AICc) and estimated R 2 (Burnham & Anderson, 2002) . 263 264

We compared our 16S rRNA sequences to those in GenBank (Volokhov et al., 2017; Volokhov 266 et al., 2011) . Briefly, we aligned sequences using Clustal X, and inter-and intra-species 267 similarity values were generated using BioEdit. Genetic distances were calculated with the 268 Kimura two-parameter and Tamura-Nei models, and the phylogeny was constructed using 269 MEGA X with the minimum evolution algorithm (Kumar, Stecher, Li, Knyaz, & Tamura, 2018). 270

We assigned hemoplasma genotypes to positive bats based on analysis of the 16S rRNA 271 partial gene (860-1000 bp) sequences in GenBank and their clustering on the phylogeny. 272

Genotypes were designated as novel if (i) sequences differed from the closest hemoplasma 273 sequences in GenBank by ≥ 1.5% and/or (ii) if sequence similarity was <1.5% but genotype-274 specific reproducible mutations (at least two per sequence) were observed between hemoplasma 275 sequences from at least two independent bat samples and the nearest GenBank hemoplasma 276 sequences. These genotype-specific mutations were further used to differentiate closely related 277 hemoplasma genotypes from our sample. We caution that genotype is not synonymous with 278 species, as analysis of the 16S rRNA gene alone is insufficient for accurate species identification 279 14 of Mycoplasma spp. (Volokhov et al., 2012) . Future studies using genomics or housekeeping 280 genes may identify independent but closely related hemoplasma species in our genotypes. 281

To assess if hemoplasma genotype assignments were associated with site and year, we 282 used χ 2 tests with p values generated through a Monte Carlo procedure. Prior to our phylogenetic 283 and network analyses of genotype distributions across bat species (see below), we used another 284 χ 2 test to assess the association between hemoplasma genotype identify and bat host identity. 285 286

To determine the degree to which bat hemoplasma genotypes display host specificity and to 288 describe their evolutionary relationships with host species, we used our bat and hemoplasma 289 phylogenies to construct a binary association matrix. To test the dependence of the hemoplasma 290 phylogeny upon the bat phylogeny and thus assess evidence of evolutionary codivergence, we 291 applied the Procrustes Approach to Cophylogeny (PACo) using distance matrices and the paco 292 package ( We used hemoplasma genotype assignments to create a network, with each node representing a 299 bat species and edges representing shared genotypes among bat species pairs. We built an 300 adjacency matrix using the igraph package and used the Louvain method to assess the structure 301 of bat-hemoplasma communities within this network (Csardi & Nepusz, 2006) . To test whether 302 15 the distribution of hemoplasma genotypes across our Neotropical bat species is shaped by host 303 phylogeny, we used two GLMs to predict counts of shared genotypes (Poisson errors) and the 304 presence of sharing (binomial errors) by phylogenetic distance between bat species. We assessed 305 statistical significance with a quadratic assignment procedure via the sna package (Butts, 2008). 306

We calculated two metrics of network centrality to quantify different aspects of how 307 important a node (bat species) is to hemoplasma genotype sharing: degree and eigenvector 308 centrality (Bell, Atkinson, & Carlson, 1999). Whereas degree indicates the number of other 309 species with which a host shares bacterial genotypes (i.e., links per node), eigenvector centrality 310 indicates the tendency for a host to share genotypes with species that also share more genotypes 311 (i.e., connectivity). Eigenvector centrality is thus an extension of degree that can identify hubs of 312 parasite sharing (Gómez, Nunn, & Verdú, 2013). These two metrics were moderately correlated 313 (ρ=0.59), with many non-zero degree species displaying zero eigenvector centrality. To examine 314 spatial and temporal patterns in host centrality, we built separate adjacency networks per each 315 site and year. We fit separate GLMs to ask how hemoplasma sharing centrality was predicted by 316 site, year, and the two-way interaction. Degree was modeled as a Poisson-distributed response, 317 while eigenvector centrality was logit-transformed and used Gaussian errors. We next applied 318 phylogenetic factorization to both metrics and weighted the algorithms by the square-root sample 319 size per species (Garamszegi, 2014). We then fit the same PGLS models used in our prevalence 320 analysis to identify the most competitive trait predictors of bat species centrality to hemoplasma 321 sharing. To lastly assess whether network centrality is associated with hemoplasma prevalence, 322 we fit two weighted PGLS models with each centrality metric as a univariate predictor. 323 324

We detected sequence-confirmed hemoplasma infection in 239 of 469 individuals (51%; 95% 327 CI: 46-55%), with positive individuals in 23 of the 33 sampled bat species (Table S3) Pteronotus mesoamericanus, for which prevalence was greater than 58%. Although these three 346 species were also heavily sampled, other well-sampled species such as Sturnira parvidens and 347

Carollia sowelli had lower prevalence, and sample size did not predict prevalence (Table 1) . 348

Our phylogenetic analysis identified 29 Mycoplasma genotypes in the Belize bat community 351 (Table 2) After controlling for multiple comparisons, our 29 bacterial genotypes were associated 362 with site (χ 2 =47.11, p<0.01) and year (χ 2 =40.40, p<0.01). Genotype composition was more 363 diverse at LAR (Fig. S4) , and KK hemoplasmas were dominated by vampire bat genotypes 364 (VBG1-3). Genotype composition was more idiosyncratic by study year. However, these 29 365 bacterial genotypes were most strongly associated with bat species (χ 2 =3532, p<0.01; Fig. S4 ). 366 367

Although some hemoplasma genotypes were shared between bat species (i.e., VBG1, CS2, PPM, 369 EF1, AH1-2, MYE, PLU, SP; n=9), most showed strong host specificity (n=20; whereas six genotypes were each restricted to a single bat species (Fig. 5A ). GLMs showed that 381 both the frequency and presence of genotype sharing declined with phylogenetic distance 382 between bat species (Poisson: p<0.001, R 2 =0.08; binomial: p<0.001, R 2 =0.51; Fig. 5B ). 383

Bat species shared hemoplasma genotypes with zero to five other species (i.e., degree), 384 and most hosts were not central to the network of genotype sharing (i.e., eigenvector centrality of 385 zero). Six bat species had non-zero eigenvector centrality values that ranged from 37% to 100%, 386 indicating that these hosts generally shared more hemoplasma genotypes with other highly 387 connected hosts. Stratifying our hemoplasma genotype network across sites and years showed 388 that centrality measures varied by space but not time (Fig. S7, Table S5 ). We observed no 389 hemoplasma genotype sharing at KK, likely reflecting lower host diversity (Herrera et al., 2018) . 390

Phylogenetic factorization identified similar bat clades with significantly different 391 centrality compared to the paraphyletic remainder ( Fig. 6A-B Trait-based analyses showed that degree centrality was best predicted by diet (Table S6) ; 400 bat species feeding more heavily on fruit and nectar shared more bacterial genotypes with other 401 species (β=0.004, p<0.001, R 2 =0.20; Fig. 6C ). Similarly, eigenvector centrality was best 402 predicted by bat colony size and diet (Table S7 ); highly central species had small colonies 403 (β large =-1.93, p=0.05, R 2 =0.13) and fed more on plants (β=0.03, p<0.01, R 2 =0.10; Fig. 6D ). 404

As a final analysis, we assessed whether network centrality (i.e., a bat species' role in 405 hemoplasma genotype sharing) predicted contemporary infection prevalence (Fig. S8) . However, 406

we found no associations between species-level infection prevalence and centrality as measured 407 by degree (β=-0.13, R 2 =0.03, p=0.42) or eigenvector centrality (β=-0.20, R 2 <0.01, p=0.79). prevalence were distinct from those in genotype sharing centrality (e.g., large-colony species had 482 higher prevalence but lower connectivity), and prevalence accordingly did not predict centrality. 483 Second, we found general congruence between bat and hemoplasma phylogenies. Although this 484

shows codivergence is a strong evolutionary force, congruence can also stem from preferential cophylogeny plot shows the bat phylogeny on the left and the hemoplasma genotype phylogeny 933 on the right. We used the treespace package to collapse our complete hemoplasma phylogeny 934 ( Fig. S3) 

Phylogeny matters: revisiting 'a 683 comparison of bats and rodents as reservoirs of zoonotic viruses

Global Patterns of Zoonotic Disease in 686

Undiscovered bat hosts of filoviruses

Candidatus Mycoplasma haemohominis in Human, Japan. Emerging Infectious Diseases

Fragmented Bat Communities in Orange Walk District

Package 'geosphere

Density-dependent decline of host abundance 698 resulting from a new infectious disease

Molecular detection of vector-borne bacteria in bat ticks

Argasidae) from eight countries of the Old and New Worlds

Phylogenetically related and ecologically similar carnivores harbour similar parasite 706 assemblages

Parasite diversity declines 708 with host evolutionary distinctiveness: A global analysis of carnivores

711 implementing Procrustean Approach

Evidence and molecular characterization of Bartonella spp. and 715 hemoplasmas in neotropical bats in Brazil

Living fast and dying of infection: host life history drives interspecific variation in 718 infection and disease risk

Foraging and 724 roosting behaviour of the fringe-lipped bat

Prokaryotic taxonomy in the sequencing era-the 727 polyphasic approach revisited

Transmission pathways and spillover of an erythrocytic bacterial pathogen from 730 domestic cats to wild felids

Sexual 733 dimorphism in immunity across animals: a meta-analysis

Picante: R tools for integrating phylogenies and ecology

Evolutionary Genetics Analysis across Computing Platforms

Preliminary information on the home range and 742 movement patterns of Sturnira lilium (Phyllostomidae) in a naturally fragmented landscape 743 in Bolivia

The 745

Evolution and Genetics of Virus Host Shifts

Network analysis of host-virus communities in bats and rodents 749 35 reveals determinants of cross-species transmission

Infection with hemotropic Mycoplasma species in patients with 753 or without extensive arthropod or animal contact

Hemotropic mycoplasmas in little brown bats (Myotis lucifugus). Parasit Vectors

How should pathogen transmission be 759 modelled?

Host phylogeny, geographic overlap, and roost sharing shape parasite communities in 763

Hemotrophic mycoplasmas (hemoplasmas): a review and new insights 765 into pathogenic potential

rotl : an R package to interact with 768 the Open Tree of Life data

Detection and Characterization of Hemotropic Mycoplasmas in Bats in 772

Widespread 774 infection with hemotropic mycoplasmas in bats in Spain, including a hemoplasma closely 775 related to "Candidatus Mycoplasma hemohominis

Evolutionary patterns and processes in the 778 radiation of phyllostomid bats

Bats and Bacterial Pathogens: A Review

In vivo transmission studies of "Candidatus Mycoplasma turicensis" in 784 the domestic cat

An amniote life-history database to perform comparative analyses with birds, 787 mammals, and reptiles

Host and viral traits predict zoonotic spillover from mammals

APE: analyses of phylogenetics and evolution 792 in R language

Patterns of 794 host specificity and transmission among parasites of wild primates

Mycoplasma host specificity: Fact or fiction? The 797

Pathways to zoonotic spillover

A Field Guide to the Mammals of Central America and Southeast Mexico

Roosting Ecology and the 804 Evolution of Pelage Markings in Bats

American Society of Mammalogists for the use of wild mammals in research and education

Occurrence of canine hemotropic mycoplasmas in domestic dogs from urban and rural areas 811 of the Valdivia Province, southern Chile

Ecological and anthropogenic drivers of rabies exposure in vampire bats: 815 38 implications for transmission and control

Host phylogeny constrains cross-species emergence and 819 establishment of rabies virus in bats

Host and host-site specificity 821