Riverine barriers influenced but were not decisive for the evolution of the Amazon’s huge plant diversity

A study by the University of São Paulo shows that where the Negro River is very wide certain plants found on one bank are not found on the other. Overall, however, the extraordinary variety of species is due more to ecology than geography.

By José Tadeu Arantes  |  Agência FAPESP – Some 50,000 different plant species are estimated to occur in the Amazon. A single hectare (100 meters by 100 meters) contains as many plant species as all of Europe. And this is only a preliminary estimate. For example, Diniza excelsa (in Portuguese, angelim), a tree species that can reach a height of 80 meters, was discovered only a few years ago. If such giant plants remained unknown to science for such a long time, there must surely be many smaller species that have yet to be discovered.

One of the factors that contributed to the Amazon’s extraordinary plant diversity may be the natural barrier represented by wide rivers in the region, which will have prevented some species living on one bank from dispersing to the other side. This is known as the riverine barrier hypothesis and was first proposed in the mid-nineteenth century by Alfred Russel Wallace (1823-1913) to try to explain the distribution of primate species in the Amazon. Might it also explain the pattern of geographic distribution of plant species and their diversification?

“We conducted a study that reinterprets Wallace’s hypothesis in genetic terms. According to this reinterpretation, large rivers like those in the Amazon Basin may prevent or interfere with gene flow between populations of plants on opposite banks, leading to allopatric speciation [differentiation of species due to geographic isolation] and restricting species to specific interfluvial regions,” Alison Nazareno, a professor at the Federal University of Minas Gerais’s Institute of Biological Sciences (ICB-UFMG) in Brazil, told Agência FAPESP.

Lúcia Lohmann, a professor at the University of São Paulo’s Institute of Biosciences (IB-USP), also in Brazil, was the principal investigator for the study conducted by Nazareno and collaborators to test this hypothesis. The results are reported in an article published in Frontiers in Plant Science.

FAPESP supported the study via four projects (13/12633-8, 15/07141-4, 17/02302-5 and 12/50260-6). The study was conducted under the aegis of the FAPESP Research Program on Biodiversity Characterization, Conservation, Restoration and Sustainable Use (BIOTA-FAPESP). 

This latest article is a continuation of four others published by Nazareno and Lohmann in January 2017, April 2017, November 2018, and December 2019. 

“We studied the influence on species differentiation of two rivers: the Negro, which is very ancient and extremely wide in some parts, and the Branco, much more recent and narrower. We found that the Branco isn’t a barrier to gene flow, but the Negro is, at least for some species, such as Amphirrhox longifolia, a member of the family Violaceae,” Lohmann told Agência FAPESP.

A. longifolia is dispersed by fish, yet even so the Negro represents an efficient barrier to gene flow. “The river is tens of kilometers wide in places, so it’s really an extraordinary barrier. The fish that disperse seeds of this plant often live in small niches and don’t go near the opposite bank of the river,” Lohmann said.

“We were surprised to find strong genetic structuring among populations of Buchenavia oxycarpa, a plant species in the family Combretaceae, located on opposite banks of the Negro. This is an unexpected pattern for species not dispersed by primates,” Nazareno said.

According to the researchers, their studies on the Negro were among the first to use plant species to test Wallace’s riverine barrier hypothesis, widely considered proven in the case of vertebrates, with several studies showing the importance of the barrier for birds and primates.

On the Branco, the researchers investigated representatives of four plant families: the bignonias (Bignoniaceae), with more than 800 species of trees and vines, including the jacaranda and trumpet tree; passion flowers (Passifloraceae), with more than 700 species; madders (Rubiaceae), with some 13,000 species, including coffee; and violets (Violaceae), with some 900 species. They considered species with three types of seed dispersal: by wind, water, and animals.

In the article the authors explain that to test whether these distinct seed dispersal modes have a consistent effect on the level of genetic connectivity (gene flow) among populations of riverine plant species, they used unlinked single-nucleotide polymorphisms (SNPs) for eight plant species as markers.

SNPs are DNA sequence variations that occur when a single nucleotide (adenine, thymine, cytosine or guanine) in the genome sequence is altered, and are therefore promising markers in genomic studies of populations.

“Although the wider parts of the Negro did act as a barrier to gene flow in some species, our findings show that the Branco didn’t act as a barrier to gene dispersal for any of the plants analyzed, regardless of their mode of dispersal,” Lohmann said.

This suggests that ecological rather than geographical factors played the key role in the evolutionary history of plants in the Amazon Basin. “These results can help improve conservation and management policies for riparian forests in the Amazon, where degradation and deforestation rates are so high,” the authors conclude.

The article “By animal, water or wind: Can dispersal mode predict genetic connectivity in riverine plant species?” is at: www.frontiersin.org/articles/10.3389/fpls.2021.626405/full#B122.