Research Article: Mining Herbaria for Plant Pathogen Genomes: Back to the Future

Date Published: April 24, 2014

Publisher: Public Library of Science

Author(s): Kentaro Yoshida, Hernán A. Burbano, Johannes Krause, Marco Thines, Detlef Weigel, Sophien Kamoun, Joseph Heitman.


Partial Text

Since the dawn of agriculture, plant pathogens and pests have been a scourge of humanity. Yet we have come a long way since the Romans attempted to mitigate the effects of plant disease by worshipping and honoring the god Robigus [1]. Books in the Middle Ages by Islamic and European scholars described various plant diseases and even proposed particular disease management strategies [1]. Surprisingly, the causes of plant diseases remained a matter of debate over a long period. It took Henri-Louis Duhamel du Monceau’s elegant demonstration in his 1728 “Explication Physique” paper that a “contagious” fungus was responsible for a saffron crocus disease to usher in an era of documented scientific inquiry [2]. Confusion and debate about the exact nature of the causal agents of plant diseases continued until the 19th century, which not only saw the first detailed analyses of plant pathogens but also provided much-needed insight into the mechanisms of plant disease. An example of this is Anton de Bary’s demonstration that a “fungus” is a cause, not a consequence, of plant disease [3]. This coming of age of plant pathology was timely. In the 19th century, severe plant disease epidemics hit Europe and caused economic and social upheaval. These epidemics were not only widely covered in the press but also recognized as serious political issues by governments [1], [4]–[6]. Many of the diseases, including late blight of potato, powdery and downy mildew of grapevine, as well as phylloxera, were due to exotic introductions from the Americas and elsewhere. These and subsequent epidemics motivated scientific investigations into crop breeding and plant disease management that developed into modern plant pathology science over the 20th century.

DNA retrieved from historic and prehistoric sources such as museum specimens, archaeological finds, and fossil remains is collectively known as ancient DNA (aDNA) [12]. The aDNA field goes back to the 1980s. With the invention of PCR, aDNA research came into its own, and PCR-based methods were its mainstay for 20 years. Early on, mycologists recognized the value of herbaria in storage of pathogen aDNA and used PCR to decode fragments of aDNA from dried herbarium specimens [7]–[9], [13]. In contrast to DNA extracted from fresh tissue, aDNA comes, in general, in tiny amounts, is highly fragmented, and contains chemical modifications [12]. Many of the difficulties resulting from these characteristics have been recently overcome, thanks to the advent of high-throughput sequencing and the development of new library-based retrieval of aDNA fragments without relying on direct amplification by PCR [14], [15]. Nowadays, it is possible to sequence complete genomes from organisms that went extinct tens of thousands of years ago, providing unique insight into their history and evolution. For example, the sequencing of the complete genomes of two archaic hominins (Neanderthals and Denisovans) has opened a window to the past and profoundly changed our views on human origins [16]–[18].

To understand disease dynamics, it is of particular importance to accurately date major events, such as epidemics’ emergence and reemergence, and sudden changes in genetic diversity. These dated events can then be correlated with a timeline of historical and socioeconomic information. Fortunately, the genetic information obtained from aDNA sequences provides a unique opportunity to date divergence times in a phylogenetic tree. Genomes from historic samples will accumulate fewer nucleotide substitutions than modern samples, which have continued to accumulate substitutions for many more generations [18], [22]. This can be used to calculate substitution rates and subsequently divergence times using the sample dates as tip calibration points in a Bayesian phylogenetic framework [24], [25]. Herbaria samples usually contain collection date and geographic location, which makes them ideal calibration points in a phylogenetic tree [10]. Historic and modern samples that are more spread out in time yield more accurate calculations of divergence times [24].

The oomycete plant pathogen P. infestans causes late blight, the most destructive disease of potatoes and a major disease of tomatoes. Ever since the Irish potato famine in the 1840s, P. infestans has remained one of the more serious threats to food production, resulting in losses that add up to enough crop to feed hundreds of millions of people [26]. Similar to other oomycetes, P. infestans has a complex life cycle that includes both asexual and sexual phases. The ability of P. infestans to generate large amounts of asexual spores is a major determinant of its success as a destructive pathogen [27]. In agricultural systems, asexual reproduction is often dominant, with individual genotypes taking over large portions of the population and occasionally resulting in explosive epidemics (Figure 1) [28], [29]. Sexual reproduction is integral to the life cycle of P. infestans at its center of diversity in Mexico. In some agricultural systems, notably in the Netherlands and the Nordic countries, sexual reproduction is also common, generating an abundance of new pathogen genotypes [30], [31]. P. infestans genotypes with favorable phenotypes, such as increased virulence and spore production, can arise from pockets of sexual reproduction to become invasive when they spread to vulnerable potato production areas [28]. One example, illustrated in Figure 1, is clonal lineage 13_A2, an aggressive genotype that emerged in Europe in the early 21st century, rapidly displaced other genotypes [28], and more recently became pandemic with severe outbreaks in India [32] and China [33].

A widely accepted hypothesis is that P. infestans has originated from Toluca Valley, Mexico, where it naturally infects and coevolves with wild Solanum relatives of the potato [31]. Although the Spanish introduced the potato to Europe in the 16th century, the crop remained free of late blight until the mid-19th century. In 1845, P. infestans finally reached Europe. Potato blight was first detected in Belgium at the end of June 1845. Over the summer, the disease rapidly spread over the European continent to reach the British Isles [4]. In Ireland, socioeconomic and political circumstances coincided to turn the disease into a tragic and deadly affair—the Great Famine triggered by potato late blight killed around one million people and forced even more to emigrate.

An obvious question is whether P. infestans HERB-1 was so devastating because it was extraordinarily virulent. Several different lines of argument suggest that this was not the case. Rather, it was the extreme blight susceptibility of the potato cultivars that were grown in the 19th century, such as the Irish Lumper potato, that was responsible for the unusual severity of the disease [4], [35]. Later on, breeding for potato-blight resistance took off at the beginning of the 20th century based on introgression of disease resistance (R) genes from wild relatives of the potato. Among the first R genes to be bred, the R1, R2, R3a, and R4 genes were introduced into cultivated potatoes from the wild Mexican species Solanum demissum[35], [36]. These R genes encode immune receptors that detect specific avirulence effectors in P. infestans[37]. As expected from the historical records of plant breeding, HERB-1 carries isoforms of several pathogen effectors in their avirulence configuration [4], meaning that they would have triggered a resistance reaction by the R1, R2, R3a, and R4 genes as well as most of the other R genes that are present in modern potato cultivars. This indicates that the HERB-1 genotype of P. infestans would probably be incapable of infecting most modern potato varieties. Indeed, HERB-1 could not be identified among the modern isolates examined to date. However, an exhaustive survey of hundreds of P. infestans genotypes from all corners of the globe is needed before a definitive conclusion about HERB-1 extinction is reached [38]. Nonetheless, a plausible scenario is that the HERB-1 clonal lineage was displaced by US-1 with the emergence of modern potato breeding. Genome analyses of P. infestans population from the first half of the 20th century will help to shed light on this question.

Genome analyses of P. infestans aDNA have demonstrated that the potential of archaeogenomics to solve important questions in history and evolution does not apply to only humans and their pathogens but also to plant pathology. The budding field of archaeogenomics holds particular promise for plant pathogens because of the excellent preservation of genomic DNA in dried plant samples stored in herbaria. Thus, herbaria are hidden treasures, serving as unexploited archives of plant pathogen and plant genomes. Beyond the potato late blight, there are numerous interesting targets for investigation of plant pathogens. European herbaria hold over 800,000 specimens of rust fungi [40], a group of plant pathogens that are a recurring threat to world agriculture [41]. Another, well-sampled plant pathogen is the oomycete Plasmopara viticola, which caused epidemics of downy mildew on grapevine [5]. Mining these and other historic herbarium samples for plant pathogen genomes should inform past population and evolutionary dynamics of important plant pathogens, ultimately helping us to better prepare for future plant epidemics.




0 0 vote
Article Rating
Notify of
Inline Feedbacks
View all comments