bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Title: On the evolution of host specificity: a case study of helminths
Authors: Alaina C. Pfenning-Butterworth1*, Sebastian Botero-Cañola1 and Clayton E. Cressler1
1School of Biological Sciences, University of Nebraska-Lincoln, NE 68588
*Corresponding Author: A. C. Pfenning-Butterworth, [email protected]
Competing Interests: The authors have no competing interests to declare.
Keywords: host specificity, phylogeny, helminth, parasite evolution, zoonoses bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
1 ABSTRACT
2 The significant variation in host specificity exhibited by parasites has been separately linked to
3 evolutionary history and ecological factors in specific host-parasite associations. Yet, whether
4 there are any general patterns in the factors that shape host specificity across parasites more
5 broadly is unknown. Here we constructed a molecular phylogeny for 249 helminth species
6 infecting free-range mammals and find that the influence of ecological factors and evolutionary
7 history varies across different measures of host specificity. Whereas the phylogenetic range of
8 hosts a parasite can infect shows a strong signal of evolutionary constraint, the number of hosts a
9 parasite infects does not. Our results shed new light on the evolution of host specificity in
10 parasites, suggesting that phylogenetic breadth may capture the evolutionary potential of a
11 parasite to jump between hosts, whereas the number of hosts may reflect ecological opportunity.
12 Finally, we show parasite phylogenies can also provide an alternative perspective on zoonosis by
13 identifying which hosts are infected by a broad phylogenetic range of parasites.
1 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
14 INTRODUCTION
15 Parasites vary considerably in their host specificity, or the range of hosts they can infect, and
16 previous work has identified a large number of ecological and evolutionary factors that may lead
17 to variation in host specificity (Bernays and Graham 1988; Poulin 1992; Desdevises et al. 2002;
18 Mouillot et al. 2006; Clark and Clegg 2017). In some systems, host specificity appears to be
19 more strongly shaped by local environment (e.g., host diversity/abundance and abiotic factors,
20 Krasnov et al. 2005; Loiseau et al. 2012; Dallas and Presley 2014), whereas in other systems,
21 host specificity appears to be constrained by evolutionary history (e.g., outcomes of adaptive
22 evolution, Desdevises et al. 2002; Clark and Clegg 2017). These contrasting results are perhaps
23 not surprising, given the phylogenetic, phenotypic, and ecological diversity of parasites –
24 perhaps there are no general patterns that underlie the evolution of specificity across parasite
25 lineages. However, there is value in exploring that question, as a broader understanding of the
26 ecological and evolutionary factors that shape host specificity may help to identify hosts and
27 parasites that are the most likely reservoirs of novel zoonoses.
28 Studies like those referenced above have assessed the determinants of variation in host
29 specificity using taxonomically restricted datasets that focus on a single group of parasites (e.g.,
30 fleas, avian malaria, etc.). This taxonomic restriction can allow researchers to test specific
31 hypotheses (e.g., whether host specificity varies with geography; Krasnov et al. 2005; Krasnov et
32 al. 2008; reviewed in Poulin et al. 2011) and to more readily identify ecological and
33 physiological factors likely to be driving differences in specificity among parasite species (e.g.,
34 whether host specificity is shaped by host phylogeny; McCoy et al. 2001; Fallon et al. 2005;
35 Clark and Clegg 2017). However, the widespread variation in host specificity among groups of
36 parasites suggests that inferences that apply to one group will likely be of limited value in other
2 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
37 groups. For instance, monogeneans, ecto-parasitic flat worms, have high host-specificity (~74%
38 infect a single host, Bikhovski 1957) and are thought to have diversified during the Cretacious
39 period (~100 mya, Kearn 1994). Monogeneans diverged from Platyhelminthes (which emerged
40 ~270 mya, Dentzien-Dias et al. 2013), and the oldest known helminth lineages emerged ~550
41 mya (Zhang et al 2020). Given such ancient divergences, it would be unwise to conclude that all
42 helminths are highly host-specific, or that the factors that drive variation in specificity among
43 monogeneans will also be important to Platyhelminthes. Nevertheless, helminths like these are
44 perhaps the best group of parasites to study to determine whether local environment or
45 evolutionary history are the primary determinants of host specificity across the broadest
46 phylogenetic scale studied to date.
47 Helminths are perhaps the most successful parasites in the world, in terms of global
48 prevalence (Dallas et al. 2018). They are a diverse group consisting of four major groups of
49 parasitic worms—acanthocephalans, cestodes, nematodes, and trematodes—that differ in their
50 morphology, transmission, and host specificity (Mackiewicz 1988; Hayunga 1991; Kennedy
51 2006). They also differ in their evolutionary history (Weinstein and Kuris 2016); for example,
52 nematodes coevolved alongside vertebrates (Poinar 2011), whereas cestodes have emerged more
53 recently (Baer 1952). Previous studies indicate that variation in life history among the helminth
54 groups may help explain key differences in host specificity (Stunkard 1957; Pedersen et al.
55 2005). For instance, the generalism observed in acanthocephalans is thought to be due to the
56 presence of a free-living stage in their life cycle (Stunkard 1957). The phylogenetic and
57 ecological diversity of helminths presents a rare opportunity to discern whether there are
58 fundamental processes that determine host specificity across broad taxonomic and phylogenetic
59 scales.
3 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
60 Here, we construct a molecular helminth phylogeny for 249 species of helminths that
61 parasitize free-living mammals. This novel phylogeny is the largest ever built for this group, and
62 thus provides unprecedented insight into the factors that shape the evolution of host specificity.
63 We use this phylogeny to test the evidence for two hypotheses in shaping variation in host
64 specificity: (1) evolutionary history, where closely related helminths would always have similar
65 host specificities, and (2) local ecology, where closely related helminths living in different
66 environments would have very different host specificities. We tested these hypotheses using two
67 metrics of host specificity: the number of hosts infected (taxonomic breadth) and the mean
68 pairwise phylogenetic distance among hosts (MPD, Webb et al 2002). We find that the mean
69 pairwise phylogenetic distance among hosts is shaped by evolutionary history, whereas
70 taxonomic breadth is not. Using this phylogeny, we also assessed which mammal species may be
71 more likely to serve as reservoirs for emerging infectious diseases finding that Old and New
72 World monkeys host a phylogenetically diverse set of helminths making them potential sources
73 of zoonotic disease.
74 75 MATERIAL AND METHODS
76 Host-parasite databases
77 We obtained records of mammal-helminth interactions from two datasets: Global Mammal
78 Parasite Database (GMPD, Stephens et al. 2017) and a novel dataset assembled from
79 parasitology museums (Botero-Cañola 2020). The GMPD includes taxonomic and trait data
80 obtained from the literature for over 10,000 host-helminth associations of wild populations from
81 four Orders of mammals: carnivores (Carnivora), primates (Primates), and ungulates
82 (Artiodactyla and Perissodactyla). Some limitations of the GMPD are the exclusion of rodent
83 hosts and the lack of expert identification of helminths, which could lead to an overestimation of
4 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
84 the number of specialist helminths due to missing hosts or the misidentification of rare species.
85 To account for these potential sources of bias we also ran all analyses with the dataset assembled
86 from voucher specimen-verified museum collections. This dataset includes host-parasite
87 associations for Nearctic mammals in the Orders Artiodactyla (even-toed ungulates), Carnivora
88 (carnivores), Didelphimorphia (opposums), Eulipotyphla (moles and shrews), Lagomorpha
89 (rabbits), and Rodentia (rodents) obtained from the H.W. Manter Parasitology Collection, the
90 former United States National Parasitology Collection, the Museum of Southwestern Biology,
91 and the Canadian Museum of Nature.
92
93 Phylogenetic construction
94 We searched Genbank for DNA sequences (CO1, 18S, and 28S) of all 920 species of helminths
95 included in the GMPD (Stephens et al. 2017), resulting in 249 species. We combined slow
96 evolving sequences (18S and 28S) that resolve deep phylogenetic relationships with a fast-
97 evolving sequence (CO1) to resolve recent relationships. Sequences were aligned using Multiple
98 Sequence Alignment (MUSCLE v. 3.8.425, Edgar 2004) and manually edited in Geneious Prime
99 2019.0.4 (Kearse et al. 2012). The sequences for each species included 8848 base pairs—1707,
100 3550, 3591 base pairs for CO1, 18S, and 28S respectively. We implemented a Bayesian
101 likelihood analysis on the combined data set with a GTR + G + I substitution model
102 (PartitionFinder v.2.1.1; Lanfear et al. 2016) and a relaxed lognormal molecular clock model
103 selected in BEASTv.2.5.2 (Bouckaert et al. 2014). We ran three replicates of the Bayesian
104 Markov chain Monte Carlo (MCMC) analysis for 60 million generations each, sampling every
105 1000 generations (BEASTv.2.5.2; Bouckaert et al. 2014). We assessed the MCMC log files for
106 convergence (effective sample size, ESS > 200, Tracer v.1.7.1; Rambaut et al. 2018) and
5 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
107 removed the first 44% as burn-in (TreeAnnotator v.2.5.2, Bouckaert et al. 2014). The constructed
108 tree is consistent with the topology of published trees for smaller helminth groups
109 (Acanthocephala—Vereweyen et al. 2011; García-Varela et al. 2013, Nematoda—Blaxter et al.
110 1998; Nadler and Hudspeth 2000, Platyhelminthes—Knapp et al. 2015; Sharma et al. 2016) and
111 clearly assigns recognized Orders to monophyletic clades (Supplemental Figure 1). Focusing
112 only on the host-parasite associations for which we have phylogenetic information for the
113 parasite, the analyses of the GMPD data included 971 associations among 197 mammal species
114 and 249 helminth species; the analyses of the Nearctic museum data included 363 associations
115 among 67 mammal species and 90 helminth species.
116
117 Phylogenetic signal in host specificity metrics
118 Metrics of host specificity used here—taxonomic breadth and mean pairwise phylogenetic
119 distances (MPD)—were calculated following Park et al. (2018). Taxonomic breadth refers to the
120 number of hosts a parasite can infect (Rhode 1980); whereas mean pairwise phylogenetic
121 distance refers to the mean branch length of all pairs of host species infected by a given parasite
122 (Harnos et al. 2017). We used the Phylogenetic Atlas of Mammal Macroecology’s mammal
123 phylogeny (PHYLACINE; Faurby and Svenning 2015; Faurby et al 2018) to determine the MPD
124 of hosts for each helminth using the R package picante (Kembel et al. 2010). For single host
125 parasites, we set host taxonomic breadth equal to one and MPD equal to 0. However, these
126 parasites may have a disproportionate influence on our phylogenetic conclusions: for example,
127 moving from one host to two can have a dramatic effect on MPD, in particular, and high host
128 specificity may be a result of under sampling rather than true specificity. To determine how
6 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
129 sensitive the analyses were to the inclusion of single host parasites we reran all analyses,
130 excluding single host parasites.
131 We assessed whether MPD and taxonomic breadth exhibited phylogenetic signal by
132 testing whether a non-phylogenetic model (white noise) or Pagel’s lambda best explained the
133 data (Pagel 1999). The non-phylogenetic model (white noise) treats each species MPD or
134 taxonomic breadth as an independent sample drawn from null distribution with shared mean and
135 variance. High support for the non-phylogenetic model would indicate that the trait is not
136 constrained by phylogenetic relationships. Under Pagel’s lambda, the strength of MPD or
137 taxonomic breadth’s phylogenetic signal is estimated between 0 and 1 (where 0 indicates that the
138 measure has been evolving as if the species were related by a “star” phylogeny, and 1 indicates
139 that the measure has been evolving under Brownian motion, BM; Felsenstein 1973). High
140 support for a Pagel’s lambda model greater than 0 would indicate that closely related helminths
141 have more similar MPD or taxonomic breadth than distantly related helminths.
142
143 Evolutionary model fitting
144 Based on the results of the phylogenetic signal analyses, we used phylogenetic comparative
145 analysis to further explore the evolutionary history of MPD, with and without single host
146 parasites. Specifically, we fit the MPD data to the parasite phylogeny under different models of
147 trait evolution: Brownian motion (BM), a single regime Ornstein-Uhlenbeck (OU1), and a
148 multiple regime OU based on Phyla. The BM model assumes neutral trait evolution, with a
149 variance between taxa that is proportional to the phylogenetic distance (i.e., the branch lengths)
150 between the taxa (Felsenstein 1973; Felsenstein 1985). In contrast, the OU1 model incorporates
151 a deterministic trend towards a single trait value, often interpreted as stabilizing selection
7 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
152 towards an optimum (Hansen 1997; Butler and King 2004). The OU model can also account for
153 multiple selective regimes, allowing for different trait optima in each regime (Hansen 1997;
154 Butler and King 2004; Beaulieu et al 2012); this is often used to test whether ecological or
155 environmental differences among species influence trait evolution after accounting for
156 phylogeny (e.g., the influence of trophic transmission on the life histories of helminth parasites:
157 Benesh et al. 2011, 2014) . We tested an OU model with three regimes corresponding to the
158 different parasite Phyla. All models were fit using the R package ouch, with fit assessed using
159 휔AICc (R v.3.5.2, R Core Team 2018; Burnham and Anderson 2004; Butler and King 2004;
160 Cressler et al. 2015).
161 To assess the support for the best-fitting evolutionary model, we simulated 1000 datasets
162 with the best-fitting parameter estimates for each model and then re-fit the competing models to
163 the simulated data to create distributions of likelihood values under different generating models
164 (Boettiger et al. 2012). We used a Phylogenetic Monte Carlo (Boettiger et al. 2012) approach to
165 calculate distributions of the test statistic
166 훿 = −2 (푙표푔ℒ0 − 푙표푔ℒ1)
167 where ℒ0 is the likelihood of the simpler model and ℒ1 is the likelihood of the more complex
168 model. We computed this test statistic for the BM:OU3 comparison and the OU1:OU3
169 comparison for the GMPD and Nearctic datasets with and without specialist parasites.
170 To determine whether the observed value of 훿̂ from fitting the real data was significantly
171 different from a null expectation, we approximated a p-value (the probability of observing 훿 if
172 the simpler model were true). We simulated 1000 datasets using the simpler model at its MLE
173 parameter estimates and then we fit both the simple model and OU3 to the simulated dataset and
174 computed the values of 훿. This produces a null distribution of 훿 under the simpler model. We
8 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
175 calculate an approximate p-value as the fraction of 훿 values in this distribution that are as
176 extreme as the observed value, 훿̂.
177 To determine the power of each comparison, we computed the probability of rejecting the
178 simpler model when the data were generated by the more complex (OU3) model. We simulated
179 1000 datasets using the OU3 model at its MLE parameter estimates and then we fit both models
180 to the simulated dataset and computed the values of 훿. The fraction of δ values that are larger
181 than the 95th percentile of the null distribution generated under the simpler model above gives an
182 estimate of power.
183
184 Phylogenetic Generalized Least Squares
185 We assessed whether variation host specificity (MPD) is shaped by parasite and host life history
186 traits. Specifically, we used phylogenetic generalized least squares (PGLS) to test for if parasite
187 length (Benesh et al. 2017), transmission mode (close-contact, environmental, and trophic
188 transmission; Stephens et al. 2017), and average host mass (in grams, PHYLACINE, Faurby and
189 Svenning 2015; Faurby et al 2018) explain the variation in MPD for both datasets with and
190 without single host helminths. PGLS analyses were conducted with 52 species for which parasite
191 body length was well documented using the R package caper (Orme et al. 2013). To account for
192 phylogenetic signal, we fit each dataset’s PGLS using Pagel’s 휆 (Pagel 1999).
193
194 Helminth phylogenetic diversity within mammal clades
195 We calculated the mean pairwise phylogenetic distances (MPD) of helminths that infect each
196 host to identify patterns of helminth diversity within host clades. The helminth phylogeny
197 constructed in this study was used to determine the MPD of helminths that infect each host,
9 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
198 using the R package picante (Kembel et al. 2010). Host MPD was visualized on the mammal
199 phylogeny using the R package phytools (Revell 2012). We also determined the pairwise
200 distance (PD) of each mammal host to humans to assess which hosts are closely related to
201 humans and are also infected by a broad range of helminths, potentially identifying host species
202 at high risk for emerging infectious disease.
203
204 RESULTS
205 Phylogenetic signal in host specificity metrics
206 We assessed whether the host specificity metrics, MPD and taxonomic breadth, exhibited
207 phylogenetic signal and found that MPD had phylogenetic signal and taxonomic breadth did not,
208 regardless of the dataset or the inclusion of single host parasites. For MPD, datasets that
209 excluded single host parasite had higher 휆 values, indicating stronger phylogenetic signal
210 (Nearctic with single host parasites: 휆 = 0.38; Nearctic without single host parasites: 휆 = 0.92;
211 GMPD with single host parasites: 휆 = 0.61; GMPD without single host parasites: 휆 = 0.90). For
212 taxonomic breadth, the 휆 value did not change with the exclusion of single host parasites
213 (Nearctic with single host parasites: 휆 = 0.09; Nearctic without single host parasites: 휆 = 0.09;
214 GMPD with single host parasites: 휆 = 0.00; GMPD without single host parasites: 휆 = 0.00).
215 Given that MPD showed phylogenetic signal, MPD values were visualized on the
216 helminth phylogeny, including single host helminths, for the GMPD and Nearctic dataset
217 (Figures 1 and 2, respectively). Within the GMPD phylogeny, there are clades of specialist
218 helminths (e.g., Mansonella sp. and Cylicocyclus sp., Figure 1) and clades of generalist
219 helminths (e.g., Alaria sp. and Trichinella sp., Fig. 1). There are also clades in which one or a
220 few species’ host specificity varies from the majority of the clade (e.g., Taenia madoquae is a
10 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
221 specialist in a predominantly generalist clade, Figure 1). Overall, there is no obvious sign of
222 evolution either towards or away from specialism across the tree. The Nearctic phylogeny show
223 similar variation, including clades of only specialist helminths (e.g., Trichuris sp., Figure 2) and
224 clades with both extreme specialists and generalists (e.g., acanthocephalans, Figure 2).
225
226 Evolutionary model fitting
227 We modelled the evolutionary history of host specificity (MPD) and established that an OU3
228 model with regimes for each of the parasite Phyla best described the Nearctic and GMPD
229 dataset, with and without the exclusion of single host parasites (using 휛AICc; Table 1; full
230 details in Supplemental Tables 1-4). We assessed the support for the OU3 model using a
231 Phylogenetic Monte Carlo approach. There was high support for the OU3 model compared to the
232 BM model for both datasets, with and without single host parasites (Supplemental Figure 2). The
233 OU3 model had high support compared to the OU1 model for both datasets, with single host
234 parasites. However, the GMPD dataset without single host parasites had lower power, but still
235 maintained a high p-value, suggesting that it was unlikely that the data was generated by an OU1
236 model (Supplemental Figure 3).
237 The OU3 model parameter estimates for the GMPD and Nearctic datasets, with and
238 without single host parasites, can be found in Table 2. For both datasets, with and without single
239 host parasites, the nematode optimum is at a lower MPD value, indicating higher host specificity
240 for nematodes than Platyhelminthes and Acanthocephala. In the GMPD, with and without single
241 host parasites, the Platyhelminthes optimum is the largest, whereas in the Nearctic dataset, the
242 Acanthocephala optimum is the largest. However, regardless of dataset, the stationary variance
11 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
243 of the OU process (given by 휎2/2훼) is very large, indicating that, within each taxonomic clade,
244 there is considerable variation in MPD among parasite species.
245
246 Life-history traits and host specificity
247 For each dataset, with and without specialist helminths, we assessed whether variation host
248 specificity (MPD) was shaped by parasite and host life history traits (parasite length, close-
249 contact transmission, environmental transmission, trophic transmission, and average host mass)
250 using PGLS. In the Nearctic dataset, with specialists, the best model only included host mass as a
251 predictor (R2 = 0.18, p = 0.01). We found a significant negative effect of host mass indicating
252 that more specialist helminths are found in large-bodied mammals (p = 0.01, See Supplemental
253 File). In the Nearctic dataset, without specialists, the best model included parasite length, close
254 transmission, environmental transmission, and trophic transmission (R2 = 0.07, p = 0.27);
255 however, none of these predictors were significant. In the GMPD, with and without specialists,
256 the best model only included close transmission (R2 = 0.01, p = 0.24; R2 = 0.05, p = 0.09). In
257 neither model did close transmission have a significant effect on MPD.
258
259 Helminth phylogenetic diversity within mammal clades
260 For each dataset, we calculated the mean pairwise phylogenetic distance of helminths infecting
261 each mammal and assessed patterns of helminth diversity within mammal clades. In the Nearctic
262 dataset, 28% of the mammals were only infected by one helminth (MPD = 0), and thirteen
263 species (19%) were infected by multiple phylogenetically distant helminths (MPD value > 0.9;
264 since the helminth tree is scaled to a height of one, a host with parasite MPD > 0.9 indicates that
265 its parasites diverged deep in the tree). Of mammals with high parasite MPDs, 15% are
12 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
266 ungulates, 54% are carnivores, and 31% are rodents (Supplemental Figure 4). In the GMPD,
267 38% of the mammals were only infected by one helminth (MPD = 0), and 27 species (14%) were
268 infected by multiple phylogenetically distant helminths (MPD value > 0.9). Of these, 22% are
269 ungulates, 63% are carnivores, and 11% are primates (Supplemental Figure 5). There are five
270 carnivores with high MPD ( > 0.9) values in both datasets (Lynx rufus, Urocyon
271 cinereoargenteus, Lontra canadensis, Procyon lotor, and Ursus americanus). Urocyon
272 cinereoargenteus (gray fox) and Procyon lotor (raccoon) have MPD values greater than 0.95 in
273 both datasets.
274 We plotted the pairwise distance (PD) of each mammal host to humans against the MPD
275 of helminths infecting each mammal to identify mammals that might pose a higher risk for a
276 spillover event. In the GMPD dataset, which does not include rodents, we found that new world
277 monkeys, which are closely related to humans, have the highest MPD values (Figure 3). In the
278 Nearctic dataset, the species with the highest MPD values (i.e., Procyon lotor and Peromyscus
279 gossypinus) are not closely related to humans (Figure 3).
280 281 DISCUSSION 282 283 Prior work has assessed the variation in host specificity in taxonomically restricted groups of
284 parasites (Desdevises et al. 2002; Krasnov et al. 2005, 2008; Hellgren et al. 2009; Loiseau et al.
285 2012; Dallas and Presley 2014; Clark and Clegg 2017). These and other studies on host
286 specificity are necessary to address specific hypotheses about the determinants of host specificity
287 within a parasite group (e.g., determining whether monogenean host specificity varies with host
288 body size; Desdevises et al. 2002). Krasnov et al. have studied the effects of specific
289 environmental factors on the host specificity of fleas that infect mammals finding that fleas with
290 a larger geographic range also have a larger host breadth and that these increase with latitude
13 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
291 (2005; 2008). Additionally, work on two species of fleas, Listropsylla agrippinae and
292 Chiastopsylla rossi, has indicated that host breadth varies with time spent on the host, suggesting
293 that life history traits shape host specificity (Van der Mescht et al. 2015). In this study, we asked
294 whether we could identify any general patterns of variation in host specificity. The general
295 patterns indicated in this study would benefit from additional work within specific clades to
296 identify the specific evolutionary or ecological interactions shaping host specificity.
297 We used a novel helminth phylogeny and two helminth-mammal databases to test
298 competing hypotheses about the determinants of variation in host specificity. The first hypothesis
299 we considered suggests that host specificity is determined by local environmental factors, such as
300 host diversity and abundance or abiotic factors. Under this hypothesis, closely related parasites
301 living in different environments would have very different host specificities, whether measured
302 using the phylogenetic distance between hosts (MPD) or the number of hosts (taxonomic
303 breadth). The second hypothesis suggests that host specificity is determined by evolutionary
304 constraint. Under this hypothesis, closely related parasites would always have similar host
305 specificities. Finally, host specificity could be determined by some combination or neither of
306 these hypotheses.
307 We found, regardless of the dataset used or the inclusion of single host parasites, that
308 MPD is constrained by evolutionary history and taxonomic breadth is not. This suggests that the
309 phylogenetic breadth of hosts a parasite can infect is shaped by evolution, and possibly
310 coevolution (Hadfield et al. 2014), whereas the number of hosts a parasite can infect is not.
311 Similar results have been found in other helminths (monogeneans, Desdevises et al. 2002;
312 trematodes, cestodes, and nematodes; Mouillet et al. 2006). This difference between host
313 specificity metrics suggests that host specificity is determined by local environment and
14 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
314 evolutionary history. One possible interpretation of these results is that the phylogenetic range of
315 hosts a parasite can infect is constrained its life history traits (given the results of PGLS, it
316 appears that the life history traits that effect MPD likely vary with parasite species), whereas the
317 number of hosts that a parasite can infect within that range is determined by the availability of
318 those hosts within the extent of the parasite’s geographic range. Some studies have shown
319 support for this idea, finding that the host range of a parasite varies across different areas of their
320 geographic distribution (Thompson 2005; Ricklefs 2010; Huang et al. 2018). Additional
321 evidence for this interpretation could come from studies counting how many hosts within a
322 parasite’s phylogenetic range have an overlapping geographic range with the parasite: this
323 number should correspond to the parasite’s taxonomic breadth. Together, these results suggest
324 that, like free-living species, parasites have a fundamental (shaped by MPD) and a realized
325 (taxonomic breadth) niche.
326 MPD was best described by an OU3 model with regimes specified by Phyla for both
327 datasets, with and without single host helminths, indicating that host specificity has evolved
328 differently within clades of helminths. The OU3 model parameters for each dataset indicate that
329 Nematodes are the most host specific phylum of helminths. However, the most generalist
330 phylum of helminths varies with the dataset used; in the GMPD Platyhelminthes are the most
331 generalist and in the Nearctic dataset the Acanthocephala are the most generalist. The
332 Acanthocephala are trophically transmitted, with both a free-living life stage and life stages
333 inside their intermediate and definitive hosts (Kennedy, 2006); given this variability in habitats
334 and hosts, it makes some sense that they would be generalist, able to infect a broad range of hosts
335 (Pedersen et al. 2005). Given that both Platyhelminthes and Nematodes contain a diverse range
15 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
336 of life cycles, habitats, and hosts, further research is needed to understand why Platyhelminthes
337 appear more generalist and Nematodes appear more specialist.
338 Mammals are considered common reservoirs for zoonotic pathogens because they are
339 closely related to and have frequent interactions with humans (Brook and Dobson 2015; Han et
340 al. 2015). Prior research has used mammal traits and parasite host specificity to predict which
341 mammal-parasite interactions are most likely to lead to spill-over events (Stephens et al. 2016;
342 Olival et al. 2017). These and other research indicate that parasites with low host specificity, i.e.,
343 generalists, are a greater zoonotic threat because they are more likely to infect novel hosts than
344 specialists (Woolhouse and Gowtage-Sequeria 2005; Johnson et al. 2015).
345 Here we used the helminth phylogeny to add a new perspective to this discussion, by
346 considering the phylogenetic diversity of parasites infecting the host, in addition to the host
347 range of the parasites. Hosts that are closely related to humans and are infected by
348 phylogenetically distance parasites may be more likely to harbor zoonotic pathogens. We found
349 that both Old and New World monkeys serve as hosts to many distantly related helminths
350 (Figure 3). This is consistent with prior work indicating that zoonotic diseases ‘jump’ from hosts
351 that are closely related to humans (Pederson and Davies 2009; Han et al. 2016). This helminth
352 phylogeny provides a unique approach for addressing which hosts are likely to harbor zoonotic
353 diseases by allowing for the identification of host clades that are infected by phylogenetically
354 distant parasites.
16 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
355 Acknowledgements
356 We thank the members of the Macroecology of Infectious Disease Research Coordination
357 Network (NSF-DEB 131223), especially J. Davies, M. Farrell, J. Herrera, A. Park, and P.
358 Stephens, for helpful discussion and development of methods, and S. Huang for their comments
359 on the manuscript.
360
17 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
361 References
362 Baer, J.G. (1952). Ecology of Animal Parasites. University of Illinois Press, Urbana, pp. 224.
363 Benesh, D.P., Lafferty, K.D. & Kuris, A. (2017). A life cycle database for parasitic
364 acanthocephalans, cestodes, and nematodes. Ecology, 98, 882.
365 Beaulieu, J.M., Jhwueng, D.C., Boettiger, C. & O’Meara, B.C. (2012). Modeling stabilizing
366 selection: expanding the Ornstein–Uhlenbeck model of adaptive evolution. Evolution, 66,
367 2369-2383.
368 Benesh, D.P., Chubb, J.C. & Parker, G.A. (2011). Exploitation of the same trophic link favors
369 convergence of larval life‐history strategies in complex life cycle
370 helminths. Evolution, 65(8), 2286-2299.
371 Benesh, D.P., Chubb, J.C. & Parker, G.A. (2014). The trophic vacuum and the evolution of
372 complex life cycles in trophically transmitted helminths. Proc. R. Soc. B., 281(1793),
373 20141462.
374 Bernays, E. & Graham, M. (1988). On the evolution of host specificity in phytohagous
375 arthropods. Ecology, 69, 886-892.
376 Bikhovski, B.E. (1957). Monogenetic trematodes, their classification and phylogeny. Moscow &
377 Leningrad: Isdatelsvo Akad, pp. 509
378 Blaxter, M.L., De Ley P., Garey, J.R., Liu, L.X., Scheldeman, P., Vierstraete, A. et al. (1998). A
379 molecular evolutionary framework for the phylum Nematoda. Nature, 392, 71-75.
380 Boettiger, C., Coop, G. & Ralph, P. (2012). Is your phylogeny informative? Measuring the
381 power of comparative methods. Evolution, 66(7), 2240-2251.
18 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
382 Botero-Cañola, S. (2020). Parasite geography: exploring drivers of parasite biodiversity
383 distribution at species and community scales. PhD Dissertation, College of Arts and
384 Science, University of Nebraska, Lincoln, Nebraska, USA
385 Bouckaert, R.R., Heled, J., Kühnert, D., Vaughan T, Wu, C-H, Xie, D., et al. (2014). BEAST 2:
386 A software platform for Bayesian evolutionary analysis. PLoS Comput. Biol., 10,
387 e1003537.
388 Brook, C.E. & Dobson, A.P. (2015). Bats as ‘special’ reservoirs for emerging zoonotic
389 pathogens. Trends in Microbiol., 23, 172-180.
390 Burnham, K.P. & Anderson, D.R. (2004). Multimodel inference: understanding AIC and BIC in
391 model selection. Sociol. Methods Res., 33, 261-304.
392 Butler, M.A. & King, A.A. (2004). Phylogenetic comparative analysis: a modeling approach for
393 adaptive evolution. Am. Nat., 164, 683-695.
394 Clark, N.J. & Clegg, S.M. (2017). Integrating phylogenetic and ecological distances reveals new
395 insights into parasite host specificity. Mol. Ecol., 26, 3074-3086.
396 Cressler, C.E., Butler, M.A. & King, A.A. (2015). Detecting adaptive evolution in phylogenetic
397 comparative analysis using the Ornstein-Uhlenbeck model. Syst. Biol., 64, 953-968.
398 Dallas, T.A., Aguirre, A.A., Budischak, S., Carlson, C., Ezenwa, V., Han, B., et al. (2018).
399 Gauging support for macroecological patterns in helminth parasites. Global Ecol.
400 Biogeogr., 27(12), 1437-1447.
401 Dallas, T. & Presley, S.J. (2014). Relative importance of host environment, transmission
402 potential and host phylogeny to the structure of parasite metacommunities. Oikos, 123,
403 866-874.
19 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
404 Dentzien-Dias, P.C., Poinar Jr, G., de Figueiredo, A.E.Q., Pacheco, A.C.L., Horn, B.L. &
405 Schultz, C.L. (2013). Tapeworm eggs in a 270 million-year-old shark coprolite. PLoS
406 One, 8, e55007.
407 Desdevises, Y., Morand, S. & Legendre, P. (2002). Evolution and determinants of host
408 specificity in the genus Lamellodiscus (Monogenea). Biol. J. Linnean Soc., 77, 431-443.
409 Edgar, R.C. (2004). MUSCLE: multiple sequence alignment with high accuracy and high
410 throughput. Nucleic Acids Res. 32, 1792-1797.
411 Fallon, S.M., Bermingham, E. & Ricklefs, R.E. (2005). Host specialization and geographic
412 localization of avian malaria parasites: a regional analysis in the Lesser Antilles. Am.
413 Nat., 165(4), 466-480.
414 Faurby, S., Davis, M., Pedersen, R.Ø., Schowanek, S.D., Antonelli, A. & Svenning, J-C. (2018).
415 PHYLACINE 1.2: The Phylogenetic Atlas of Mammal Macroecology. Ecology, 99,
416 2626.
417 Faurby, S. & Svenning, J.C. (2015). A species-level phylogeny of all extant and late Quaternary
418 extinct mammals using a novel heuristic-hierarchical Bayesian approach. Mol.
419 Phylogenet. Evol., 84, 14-26.
420 Felsenstein, J. (1973). Maximum-likelihood estimation of evolutionary trees from continuous
421 characters. Am. J. Hum. Genet., 25, 471-492.
422 Felsenstein, J. (1985). Phylogenies and the comparative method. Am. Nat., 125, 1-15.
423 García-Varela. M, Pérez-Ponce de León, G., Aznar, F.J. & Nadler, S.A. (2013). Phylogenetic
424 relationship among genera of Polymorphidae (Acanthocephala), inferred from nuclear
425 and mitochondrial gene sequences. Mol. Phylo. Evo., 68, 176-184.
20 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
426 Han, B.A., Kramer, A.M. & Drake, J.M. (2016). Global patterns of zoonotic diseases in
427 mammals. Trends in Parasitol., 32, 565-577.
428 Han, B.A., Schmidt, J.P., Bowden, S.E. & Drake, J.M. (2015). Rodent reservoirs of future
429 zoonotic diseases. PNAS, 112, 7039–7044.
430 Hansen, T.F. (1997). Stabilizing selection and the comparative analysis of adaption. Evolution,
431 51(5), 1341-1351.
432 Harnos, A., Lang, Z., Petrás, D., Bush, S.E., Szabó, K. & Rózsa, L. (2017). Size matters for lice
433 on birds: coevolutionary allometry of host and parasite body size. Evolution, 71, 421-431.
434 Hayunga, E.G. (1991). Morphological adaptations of intestinal helminths. J. Parasitol., 77, 865-
435 873.
436 Hellgren, O., Pérez-Tris, J. & Bensch, S. (2009). A jack‐of‐all‐trades and still a master of some:
437 prevalence and host range in avian malaria and related blood parasites. Ecology, 90(10),
438 2840-2849.
439 Huang, X., Ellis, V.A., Jönsson, J. & Bensch, S. (2018). Generalist haemosporidian parasites are
440 better adapted to a subset of host species in a multiple host community. Mol.
441 Ecol., 27(21), 4336-4346.
442 Johnson, C.K., Hitchens, P.L., Evans, T.S., Goldstein, T., Thomas, K., Clements, A., et al.
443 (2015). Spillover and pandemic properties of zoonotic viruses with high host plasticity.
444 Sci. Rep., 5, 14830.
445 Kearn, G.C. (1994). Evolutionary expansion of the Monogenea. Int. J. Parasitol., 24, 1227-1271.
446 Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, M., Sturrock, S., et al. (2012).
447 Geneious Basic: An integrated and extendable desktop software platform for the
448 organization and analysis of sequence data. Bioinformatics, 28, 1647–1649.
21 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
449 Kembel, S.W., Cowan, P.D., Helmus, M.R., Cornwell, W.K., Morlon, H., Ackerly, D.D., et al.
450 (2010). Picante: R tools for integrating phylogenies and ecology. Bioinformatics, 26,
451 1463–1464.
452 Kennedy, C.R. (2006). Ecology of the Acanthocephala. Cambridge University Press, Cambridge.
453 Knapp, J., Gottstein, B., Saarma, U. & Millon, L. (2015). Taxonomy, phylogeny and molecular
454 epidemiology of Echinococcus multilocularis: from fundamental knowledge to health
455 ecology. Vet. Parasitol., 213, 85-91.
456 Krasnov, B.R., Poulin, R., Shenbrot, G.I., Mouillot, D. & Khokhlova, I.S. (2005). Host
457 specificity and geographic range in haematophagous ectoparasites. Oikos, 108(3), 449-
458 456.
459 Krasnov, B.R., Shenbrot, G.I., Khokhlova, I.S., Mouillot, D. & Poulin, R. (2008). Latitudinal
460 gradients in niche breadth: empirical evidence from haematophagous ectoparasites. J.
461 Biogeogr., 35(4), 592-601.
462 Lanfear, R., Frandsen, P.B., Wright, A.M., Senfeld, T. & Calcott, B. (2016). PartitionFinder 2:
463 new methods for selecting partitioned models of evolution for molecular and
464 morphological phylogenetic analyses. Mol. Biol. Evol., 34, 772-773.
465 Loiseau, C., Harrigan, R.J., Robert, A., Bowie, R.C., Thomassen, H.A., Smith, T.B., et al.
466 (2012). Host and habitat specialization of avian malaria in Africa. Mol. Ecol., 21, 431-
467 441.
468 Mackiewicz, J.S. (1988). Cestode transmission patterns. J. Parasitol., 74, 60-71.
469 McCoy, K.D., Boulinier, T., Tirard, C. & Michalakis, Y. (2001). Host specificity of a generalist
470 parasite: genetic evidence of sympatric host races in the seabird tick Ixodes uriae. J. Evol.
471 Biol., 14(3), 395-405.
22 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
472 Mouillot, D., Krasnov, B.R., Shenbrot, G.I., Gaston, K.J. & Poulin, R. (2006). Conservatism of
473 host specificity in parasites. Ecography, 29, 596-602.
474 Nadler, S.A. & Hudspeth, D.S.S. (2000). Phylogeny of the Ascaridoidea (Nematoda: Ascaridida)
475 based on three genes and morphology: hypotheses of structural and sequence evolution.
476 J. Parasitol., 86, 380-393.
477 Olival, K.J., Hosseini, P.R., Zambrana-Torrelio, C., Ross, N., Bogich, T.L. & Daszak, P. (2017).
478 Host and viral traits predict zoonotic spillover from mammals. Nature,
479 doi:10.1038/nature22975.
480 Orme, D., Freckleton, R., Thomas, G. & Petzoldt, T. (2013). The caper package: comparative
481 analysis of phylogenetics and evolution in R. R package version, 5, 1-36.
482 Pagel, M. (1999). Inferring the historical patterns of biological evolution. Nature, 401, 877–884.
483 Park, A.W., Farrell, M.J., Schmidt, J.P., Huang, S., Dallas, T.A., Pappalardo, P., et al. (2018).
484 Characterizing the phylogenetic specialism-generalism spectrum of mammal parasites.
485 Proc. R. Soc. B., 285, 20172613.
486 Pedersen, A.B., Altizer, S., Poss, M., Cunningham, A.A. & Nunn, C.L. (2005). Patterns of host
487 specificity and transmission among parasites of wild primates. Int. J. Parasitol., 35, 647-
488 657.
489 Pederson, A.B. & Davies, T.J. (2009). Cross-species pathogen transmission and disease
490 emergence in primates. EcoHealth, 6, 496–508.
491 Poinar Jr, G.O. (2011). The evolutionary history of nematodes—as revealed in stone, amber and
492 mummies. Brill, Leiden, the Netherlands.
493 Poulin, R. (1992). Determinants of host-specificity in parasites of freshwater fishes. Int. J.
494 Parasitol., 22, 753-758.
23 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
495 Poulin, R., Krasnov, B.R. & Mouillot, D. (2011). Host specificity in phylogenetic and
496 geographic space. Trends Parasitol., 27(8), 355-361.
497 R Core Team. (2018). R: A language and environment for statistical computing.
498 Rambaut, A., Drummond, A.J., Xie, D., Baele, G. & Suchard, M.A. 2018. Posterior
499 summarization in Bayesian phylogenetics using Tracer 1.7. Syst. Biol., 67, 901-904.
500 Revell, L.J. (2012). Phytools: An R package for phylogenetic comparative biology (and other
501 things). Methods Ecol. Evol., 3, 217-223.
502 Ricklefs, R. E. (2010). Evolutionary diversification, coevolution between populations and their
503 antagonists, and the filling of niche space. PNAS, 107(4), 1265-1272.
504 Rohde, K. (1980). Host specificity indices of parasites and their application. Experientia, 36,
505 1369.
506 Sharma, S., Lyngdoh, D., Roy, B. & Tandon, V. (2016). Molecular phylogeny of Cyclophyllidea
507 (Cestoda: Eucestoda): an in-silico analysis based on mtCOI gene. Parasitol. Res., 115,
508 3329-3335.
509 Stephens, P.R., Altizer, S., Smith, K.F., Alonso Aguirre, A., Brown, J.H., Budischak, S.A., et al.
510 (2016). The macroecology of infectious diseases: a new perspective on global drivers of
511 pathogen distributions and impacts. Eco. Lett., 19, 1159-1171.
512 Stephens, P.R., Pappalardo, P., Huang, S., Byers, J.E., Farrell, M.J., Gehman, A., et al. (2017).
513 Global mammal parasite database version 2.0. Ecology, 98, 1476.
514 Stunkard, H.W. (1957). Host-specificity and parallel evolution of parasitic flatworms. Z.
515 Tropenmed. Parasitol., 8, 254-263.
516 Thompson, J.N. (2005). The geographic mosaic of coevolution. University of Chicago Press.
24 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
517 Van der Mescht, L., Matthee, S., & Matthee, C. A. (2015). Comparative phylogeography
518 between two generalist flea species reveal a complex interaction between parasite life
519 history and host vicariance: parasite-host association matters. BMC Evol. Biol., 15(1),
520 105.
521 Vereweyen, L., Klimpel, S. & Palm, H.W. (2011). Molecular phylogeny of the Acanthocephala
522 (class Palaeacanthocephala) with a paraphyletic assemblage of the orders Polymorphida
523 and Echinorhyncida. PLoS One, 6, e28285.
524 Webb, C.O., Ackerly, D.D., McPeek, M.A. & Donoghue, M.J. (2002). Phylogenies and
525 community ecology. Annu. Rev. Ecol. Syst., 33, 475-505.
526 Weinstein, S.B. & Kuris, A.M. (2016). Independent origins of parasitism in Animalia. Biol. Lett.,
527 12, 20160324.
528 Woolhouse, M.E.J. & Gowtage-Sequeria, S. (2005). Host range and emerging and reemerging
529 pathogens. Emerg. Infect. Dis., 11, 1842-1847.
530 Zhang, Z., Strotz, L.C., Topper, T.P., Chen, F., Chen, Y., Liang, Y., et al. (2020). An encrusting
531 kleptoparasite-host interaction from the early Cambrian. Nat. Commun., 11, 1-7.
25 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
532 Figures 533
A C
e G s o i l a r u n n a r
G a T r a t g o o D r i A i n s r t u h n a c o s h f M T s P r A e p u a c s m
s e P t o a t g o i l a r o n a o i e t m r u u h O a s c n l u s l o h x v t r o l i t g t e i r T u l a t u a e n r a a e a n o o l o i t o o r t i s r a c i o o e p S o n s f t g d h u s s s c t a o M t m u a f i x o T o s i A s l i m k i l r c c c t C t y t c c a l a i i l s o o s i t o m l g t r r o s c p u r a a u i s n u a i r r l i u o r o i i s j r u t d o x n c r o r a n a o o m r i s M a r i m i g e e o l r r O i a n u r n u r h o s s a o r c g i a o s n o a b n g i l n i a v r a e o s u b h a g g a m e l i d p s q u o c a r c e u i g g y c p e d g c u t t n o a y j g s s y a s a a u e y e y v h a l l i s a y y s O r s i a n T d m y n s g e u n i e r x a o t l s b l a m s l n r l l u i s n x s r a l a r u a x a a e u u t d s s o a y u o s o o s o p o s i a o t i a i l a i o t s g s o s m r o O s s o o a s r l s s s s a a s h d r s m r u i t l c n a n n i c i u i s i c i n n r o n l o r o d w r e l i u r l c n r a a g c n n g s f e a i i a i o u a i s h p n c c c s u i r u a o y a b u o a r u y o r y g n r i a r a a i n a a f y l c c y b u o a e x p r l i h N n g e i l a s u i u t s c l a o i e r a p m i r o k i r s n a b a r l c a u a s v o i t s x p r o q f l s m o y t C y h u r c p l u m s a s e s i s i y p e i e o i a r i o i m i r o i B B r m g t u u B y f e S l r T l s u f i i r e i A m t s m r r n i s h i p i a e h u O r n e p t o l t o o a u n T e O s e t f t l l T t w m i t i i a e n P a s r a p T e e o o a h p l t s e a m h v r n s i n t t i i s i L t i a a t b o a c o m n r p i h l p t o l r i L r t l m e b s t o o u e i s i E n o a t w i r e h r e o s e s t i b o p i l a s e m d l i s E r a e l a e j i s r a m u o u E a t c o p a l l t r d o n le s n e l d r n a a u ir s t n s P e r c l a n i o r s o o a k e D p g m u a a u r H e r n e l s n t s u s g r i i S c s l a i i h i r s c u m s e r o n e c e u d i c o a u e o a C n s r a li k t c s u t m C e o n e in y s l o s l r v u a n e c a o t e p e r C s o r a j ip r D r l d a a in s a n c e c v p c o p u a t M s o r a a r a ic a n to m e th t u a M a n c e c r e c t u g n u s a h o r d D y l s b ig m M c h c e c a i o u y p i ta n o n ri a A c c s lu i a e M c h c u a s t a c n tu r O n o t rc la o n O y u c s u u c h a fi i s ill is c o a t l s O n c i e o s n y sl c lu p e a m r c t r e ap rm S lo e a s r n r O n a o n u p o s r u e u is O t h o a e p lif P C tr to u lu e o i e c d ri r to i o C o S o m s s c lu s S n o a a ia c t y ro tr n r k x il ri b ra e a n o g a o fi e s O o f a a i g um rio t o n yl c s la rt o il l rv r lis P s ho c g u a tr r i L ir f a e a a h a a to s y y s n a ia D ro ri c m s lc u T ra m to cl lu e in tu i a a e lo u eq ri p u m us s q u s D t ri on u p a od o m u v ui m e ta l g a b i o te i m co ul n S e ta a m su ow T nt rio m r g u S e zi e tr s a rio op s p pa on a s ip a on ra e u ud d h to ar t a ris D el yl ru p ne a on o m id er tu h g pi ra li oc m to ru u e a s T n s ru fe ir itu p s m nt tu o co pi ra is sp nd C Co ho br m at m G li s u it a o li y ro ru ev e u y co pir m m rec ha ath no s ic tta m C li s im ne a nd o c ni au m Cy co ia ilo m de sto yc pp d i yli lar he ne o P m lus on at C fi c ilo a is ote um la ic a iro tho e em rm ns S ri c b us D n ch lon lifo ge tro ost at iat ca tho ei ni s in Cy C ng om ina us A an ch mo hu tus lico yli ylu um tu c tho is nc ec ste cos s e r m A an rm rhy nsp Cy ph tep de atz Ac ilifo tho co ath anu ha nta ii on an us osto s lo nu tu M rac nch s mu ng s g s ac rhy tulu m t ibu old M tro bo ni us Cy etra rsa i en ollis ma race C lico ca tus C filic alt lind ylic cyc nth ro ollis s cy C ocy lus um P filic chu cens ylic clus ade Pro hyn tus C ocyc elon rsi gior a ob sum ylic lus gat Pla som umo ostep auric us ryno a str ni hanu ulatu Co som gdale Cylico s cal s ryno a ma C cyclu icatu Co som um ylicocy s nas s oryno valid clus le satus C osoma ri C ptosto oryn enhyd ylicocy mum C osoma P clus rad Coryn australe etrovinem iatus osoma a pocul Coryn ispida Cylico atum wellina h cyclus ins South us brevis Cylicost igne thmorhynch ephanus minu Arhy evis Anc tus phorhynchus la ylostoma tubaefor Pom me norhynchus gadi Oesophagostomum bifurc Echi um nthocephalus anguillae Oesophagostomum stephanostomum Aca Spirometra mansonoides Ostertagia dikmansi Pharyngostomum ertagia lyrata cordatum Ost Apophallus ostertagi muehlingi Ostertagia Apophallu udemeri s donicus lopteragia ho Plagiorch Spicu simplex is muris Anisakis Diphyllob cattani T othrium d erranova aenia t endriticu Pseudot asicola T witchelli m gylus n i aenia m krjabin petrov Ta artis S gylus s enia c krjabin cebu Ta rassic S loides ri enia eps ongy weste Tae multic Str ides us nia o eps gylo pillos Tae vis Stron s pa rni nia s oide llebo Taen eria ngyl fue eps T ia m lis Stro ides anic s aen ad gylo s pl oni Ta ia k oqu tron ide ocy s en rab ae S gylo pr rali Ta ia o bei ron ides rco la en mis St ylo ste phi Ta ia r sa ng es ro m en egi Stro oid ae ru Tae ia h s gyl ria uo ni T ni yd ron illa eq rso ae a s ati St ap ris e ca T ni ol ge C ca nd oli i ae a p ium na ras s a ng n T ni is Pa ylu o tia i ae a l ifo ng m av xe H ni at rm ro gia d a s yd a p ico is st lla gia us ru M at o llis ho ha as yl lu s M es ig lya lap rs or ng bo ru e oc er ca re a d ro ro u is M s e a n Pa M ela st p n er e oc st tae th T ho s ym et m V s e oi n a ic ylu g s lu er oc st de ia Tr g es hy a m E s e oi s ef on id p ph su a E ch te st de co or tr ro is e lo at c in ria oi s r m s la k c u t is E h o de li ti is ho fi sa no n nc ll c in c m s ne ric ra ni o ve u o ca E h o oc u c a T a A ig p lic ri ii M c in c c st a tu P tr m a fi t r h o o u e n s u ri s e to a M o in c cc s la is m m e u m u c n o o u o e la u o p ir p ti a S o ie c c s li g m st o d m a ic i p n z o cu m g op to o o o y a p l ri D i ie i c s a o s g C t s c e p D i r z a c u rt d o a a a e to s p o i e u s lt h i n h m a h h ia u lu i D i lo m a x s h il ru s u p e i t a r i p e p g iq o s B o g o ri a p h m r M ip h g b r u c s N ra h a ill a p h o y o tr e a a i u e e c ll i o e l a J o h l n a n n n c la O t n i p r r o s H o s y lo o e s u u r p o p a la o c is y g l b e d a l s is o a il e b s lo H y e lo p r o l A C a i v is o b o o in e s u C p s d u o a A y m u v t r n u c a s u b u r A m x o h u a i s i u s s s n e o t r s c p C le is lo i u i i E n o n ie y h i e le r r s i p v e i u S o g i h l li M c o p o l a r m te ie o c u u r o l i O n la iu c h s h c u t e h p lo o le p tr u u i u i v i e n P m a r u c r c r r r a M p t in l c l p p e la o E i i h t r o s o o e p E r u r c s b v e F i o e i a c n p i i s i s T e s c p s c t T h T s u l t a i a e r p s t c e a r i r a P t c s e i i o c r a a r t u s n q n e r e i T u l m e u s r a P s h t h r l P o o h p u r u t e a d o a u a d o i n n p n o M c o a h e e a i h d s T a C r g r m a t a l h a a O a a s n a c l a e h l i c r l i a n S r a a t i a m a l o a c l a a S c m a l e t i l m u S o i l l n r r a e r p S h b l u p a a l l H m N g w P r e u n M c a l h a h m r c y t i m T e i c t a p i m a t m g i i s c e o i n T e m s a p i c o r c n b t g u s h u g s i l a i p h e a e i l a e i i l u h o s i m l e a r a u h n n c i u f a h s l a c s t u h h g n h r u l o o i a i i s i t r h r i r t a i n r v i i o n t n e s m o l r i o f f g i e T e l s s o a b u i n i e r a c h h r r s s e o n n a o i s m l i a o t t o l m h c c i b t t g e i r t c c n a r n m p t l r i t o a l e m i u a i e o i i m m e r t o i h p i o m o o e o i n i p a A o t o u n T r e a r a m c s p h s i d n h g r s m u m p h b a i r T z s m t a t T o c m s e s t a c m s c r e e e a o u u a m y i a i e o s c i o y i o r c o c u e o i o u h u u r l k r s r a e d i l m a l s w t u m u l r i l r e h e h a m a s l y i m s o m T i a g p a m l s t t p c a k a e o a t t a m h e l l h a h u u o l e u a a o y e c e s a i s A u r i t n e x p r n i m o s o A s i r z m p i r o i c l r t r u h t m y h l o u l a e o i g n a b h h u i p k i C c e l r s a i c s r a o a n n l i r o h p u c p s o o o t y i i g o a m a r m e a t c o e p a r p i y a v g a s h t p a t l h i r m p y h a T r t e m m t a o i o F n m o a s i n i t p h o u u a o e p n u i m w o t a r s h l i o p e p E r e a n i m t m o g a c p e t r o a i m s c o i h t m a l i i o o t a n t c l a o b i a C i a c s m a r e y l h I i t o G a u C a n c i M o h m a E P t C r
O
0 trait value 145.753
length=0.438 534 535
536 Figure 1. GMPD helminth phylogeny colored by MPD (warmer colors indicate higher
537 specialism). Single host helminths are included.
538
26 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
T
m
r
u
o
r
g
e
l s o g M i i
C s a n
n u O N t i g r g r e i e p u P s e o a f s s
l s i n t m l n a n s i r e e n i o i r g n s a a e r a C r n O s i t i y a o n s t l a t u o r h o m a l v o t s u u a g s e m l p d a m o e a l l s t c h i e s u c a b u d i c i l i
r s o r a s s c r u e a w a i o d i l u s i s e o r r a i c D s r n d p v r s i t h p t a a c r l m e u u t i i s s c s i a o t a l c c s i u p o l c i l t o o a c r n p a p s a n y s o v s o a s a n n o g a r a s r D p l i r x i c a v i s a c y t e i l e s o x r A i i h G s s a c a l i D c c n r y i u u T o s n t i i l a u u g r A c y s a T a y s i a lo o l l e i c il n y u t a B a l p i l c r p c o s e r r y s e s y s a n i B o r a lo t u e s a t c u p o l c u B o o o s m u i r a t t s k s h ri a o a e P c i m m f r n la b u O a c il t fi a r s a a i O l i h t tu n r ro a v lc s e b i ia i i r u li rt a n D r e p a a e u ta c a qu gi f m e a e a o S ri m a O d rm a e b st ik t on su er m e e yl s Os ta a S g ra u te g n n iru ine rt ia s o p el St ag ly i G os f ro ia r ic ra ng o at yl ru yl s a C pi s oid te os iti Str es rta lic m is on p gi y im rm gy roc C ria ilifo loid yo fila on es ni iro m ens st s D is ing C erc orm us ap or ilif nch illa alis on rhy ria M tho aer can ctus P oph cra nspe aras ila Ma s co caris chu Pa eq rhyn relap uoru ntro hostro m Ce ngylu otulus s and ollis b erson rofilic i P ceus Tricho cylindra strongyl ynchus us axei Plagiorh O esophagostom rumosum um venulosum Corynosoma st Cooperia punctata Spirometra mansonoides
ollis Plagi Nematodirus filic orchis muris a Ta mmetric enia twi agia asy tchelli iculopter Sp ica Tae hepat nia m llaria artis Capi lica Taen ria p ia cr pilla assi Ca ii Ta ceps tor eni a pu a m llari ulti api us Tae cep C hil nia s rop o ae r T vis us lo ae ole co nia uc dis T s E is is a eri ur ov en ali ich is ia s Tr r s Ta k hu pi e ra ric ul n bb T v T ia e is ra a o i r o en m hu h T i is ic p e a a sa r co a e h T n n n yd o a T ia a i e a t ia rc la e p ig r a e T n is e e t a a ia i n p m s n e fo a o a u H n l r o ri g s a m C m a li y ia ti la M d c is ia m e s p o A r a s m n M e t o ll la e s ig l is a la y A id m g E e o e a o a s c o u e E c r c l i l c n o e a a a E h c o h e n i c s n i i a M t i p u t h M c n t a e c c t M t y s a o h s i i g h i e m a o o s r l r o a n i a c t o
h o d n a n i i c t a i c n a e t n n o o F c l i u n e i p l s o a r s i c l a o i a e s n i u m g c d e o e i e o i e n c s l p z c e f a o c E g k z o c o u y e i u m s o t t s v a r a i i c a i c l i s h a s r m e o o u l d c t p i z e P y u n i c y o i r a b u s s h l s e x i e o l i l e a a s t m p m i p i o g a o i p h n p o l a g t i n p a c n u u e n y e o l n r r b r s e l e s a d a t t g o s o a c i h C e n l r a N n m o a t F r
e r i n u u e y c t n
a i l s o u a e H P m s l t a H y
a u r
H i s s
0 trait value 379.9
length=0.438 539
540 Figure 2. Nearctic helminth phylogeny colored by MPD (single host helminths included).
541
27 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
542 Table 1. Weighted AICc scores for each of the three evolutionary models (a three-regime OU 543 model [OU3], a single-regime OU model [OU1], and a Brownian motion model [BM]) fit to the 544 MPD data for each dataset, with and without specialist helminths. 545 Model Nearctic Nearctic GMPD GMPD (no specialists) (no specialists)
OU3 0.99 0.99 0.99 0.88 OU 9.9x10-3 4.1x10-4 6.5x10-3 0.11
BM 2.2x10-7 5.3x10-5 1.6x10-20 5.9x10-6
546
28 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
547 Table 2. Parameter estimates (confidence intervals) for three-regime OU models (OU3) fit to the 548 MPD data for each dataset, with and without specialist helminths. 549 Dataset Alpha s2 Acanthocephala Nematoda Platyhelminthes
Nearctic 21.2 167000 134 42 95.1 (5.94,1100) (40400, (85.6, 184) (23, 61.7) (73.3, 116) 7930000)
Nearctic 486 2630000 201 52.9 119 (no specialists) (6.9, 1630) (38700, (147, 254) (35.5, 69) (99.3, 139) 9770000)
GMPD 16.8 42500 24.8 21.7 47.6 (11, 33.7) (27700, 76800) (6.19, 43.1) (14.6, 28.6) (39.1, 55.8)
GMPD 6.49 13800 67.7 37.5 68.2 (no specialists) (4.56, 17.1) (10200, 37200) (34.4, 99.8) (24.6, 50.7) (54.4, 81.7) 550
29 bioRxiv preprint doi: https://doi.org/10.1101/2021.02.13.431093; this version posted February 14, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
551 Figure 3. Mean pairwise phylogenetic distance (MPD) of helminths infecting each mammal
552 compared to their phylogenetic distance to humans in the GMPD and Nearctic dataset. Didelphis
553 virginiana and Didelphis marsupialis are most distantly related to humans (PD = 380), thus were
554 excluded from the figure.
555
30