Potential distribution of a new Bacillus species causing anthrax in African rainforests Daniel Romero-Alvarez*1, Lindsay P. Campbell2, A. Townsend Peterson1

1 Department of Ecology and Evolutionary Biology | Biodiversity Institute, University of Kansas, Lawrence, KS, USA. 2 Florida Medical Entomology Laboratory, Entomology and Nematology | IFAS, University of Florida, FL, USA

*Email: daromero88@gmail.com

Bacillus anthracis is a well-known zoonotic bacterium that is the etiological agent of anthrax, also known as ‘Virgil’s plague’ is a disease that causes deadly infections in mammals (including humans), posing an important medical and economic burden worldwide [1]. Anthrax cases are rarely described in rainforest ecosystems, however in 2001 and 2004-2005 a series of carcass from chimpanzees (Pan troglodytes troglodytes) and gorilla (Gorilla gorilla gorilla) were identified as anthrax mortalities in forested natural reserves in Côte d'Ivoire and Cameroon [2]. Thus, a detailed molecular, morphological, and microbiological analysis revealed that the etiological agent was actually a new variety of a different Bacillus species, B. cereus biovar anthracis (Bcbva) [3,4]. Details on its ecology and geographic extend remain unknown; here we are using geographical information systems and ecological niche models to describe its potential distribution and to compare Bcbva’s environmental profile with that of B. anthracis in Africa.

References [1] World Health Organization. Anthrax in Humans and Animals. 4th ed. Geneva: WHO Press; 2008. [2] Leendertz FH, Lankester F, Guislain P, Néel C, Drori O, Dupain J, et al. Anthrax in Western and Central African great apes. 2006;933: 928–933. [3] Klee SR, Brzuszkiewicz EB, Nattermann H, Brüggemann H, Dupke S, Wollherr A, et al. The genome of a Bacillus isolate causing anthrax in chimpanzees combines chromosomal properties of B. cereus with B. anthracis virulence plasmids. PLoS One. 2010;5: e10986. [4] Hoffmann C, Zimmermann F, Biek R, Kuehl H, Nowak K, Mundry R, et al. Persistent anthrax as a major driver of wildlife mortality in a tropical rainforest. Nature. 2017;548: 82–86. [5] Radosavljevic A, Anderson RP. Making better Maxent models of species distributions: complexity, overfitting and evaluation. J Biogeogr. 2014;41: 629–643. [6] Warren DL, Glor RE, Turelli M. Environmental niche equivalency versus conservatism: quantitative approaches to niche evolution. Evolution. 2008;62: 2868–2883.

Methods

Results & Conclusions Results.– -  We developed a total of 576 models for each species; from them, we

depict those models showing the broader prediction with the lowest uncertainty (Occurrences thinned to 30 km with uncorrelated PCs).

-  Suitability for B. anthracis was predicted in an east-west corridor across northern Africa, the majority of east African countries and in Namibia, South Africa, and Bostwana.

-  Suitability for Bcbva was predicted for central Africa in areas neglected by B. anthracis models. Further, small regions of suitability were found in Liberia and southern regions of Ghana, Nigeria and Côte d'Ivoire.

-  Our models suggests that B. anthracis could be using a different set of climates (temperature and humidity) and soils than Bcbva.

Conclusions.– -  To the best of our knowledge this is the first attempt to map the

potential distribution of Bcbva. -  In-field surveys could be developed in areas identified by our models

to collect Bcbva samples in order to develop reliable and easy-to-apply diagnostics for livestock, wildlife, and possibly humans.

-  Although no human cases of anthrax due to Bcbva have been identified, cases reported along ecological transitions from forested to savanna habitats should be investigated thoroughly for correct diagnoses.

Potential distribution of B. anthracis and Bcbva. Panels show the best performing model for each species

Comparison of each environmental profile between the species involved

Confirmed records

Environmental variables (PCs)

Model development [5]

Final models

B. anthracis n = 57

Temperature

Humidity

NDVI

Soils

-10 0 10 20 30 40 50

-40

-30

-20

-10

010

20

Bin MTP

0.0

0.2

0.4

0.6

0.8

1.0

-10 0 10 20 30

-20

-10

010

20

Bin MTP

0.0

0.2

0.4

0.6

0.8

1.0

Algorithm: - Maxent

Evaluation: - pROC - Omission rates - AICc -10 0 10 20 30 40 50

-40

-30

-20

-10

010

20

0

50

100

150

200

-10 0 10 20 30 40 50

-40

-30

-20

-10

010

20

-2000

-1000

0

1000

2000

3000

4000

-10 0 10 20 30 40 50

-40

-30

-20

-10

010

20

-1e+05

-5e+04

0e+00

5e+04

Bac

kgro

und

sim

ilarit

y te

st [6

]

Bcbva n = 14

Confirmed records and accessible areas (M) for model calibration

Acknowledgements Models for this project were developed using the R package ‘ku.enm’ developed by Marlon E. Cobos (https://github.com/marlonecobos/kuenm)

Spatial filtering: 30 – 50 km

+

All PCs and uncorrelated +