CLONING AND CHARACTERIZATION OF PWL1 AND PWL2 HOST-SPECIES SPECIFICITY GENES OF THE FINGER MILLET BLAST PATHOGEN, MAGNAPORTHE ORYZAE
Abstract
Blast disease, caused by the filamentous ascomycete fungus, Magnaporthe oryzae, is the most
devastating disease affecting finger millet (Eleusine coracana subsp. coracana) production in
eastern Africa. The fungus also infects many other grass species in the poaceae family. It
produces effector molecules to manipulate host immunity and metabolic processes as they
successfully invade live host, and in this case, finger millet cells. Some of these effectors have
avirulence (AVR) activities and therefore are recognized by the host plant’s R genes, preventing
completion of the infection cycle since blast infection mainly depends on the interaction
between the pathogen’s effector genes and the host’s resistant (R) proteins. This study
investigated the distribution of three members of a rapidly evolving gene family, the PWL
genes, in 226 finger millet blast isolates collected from eastern Africa, to examine the evolution
of these effector genes. This was further narrowed down to understand the roles PWL1 and
PWL2 play in the finger millet blast pathogen towards pathogenicity on weeping lovegrass
(Eragrostis curvula) and other host plants. PWL1 and PWL2 were amplified from a finger millet
blast isolate E22, using high fidelity PCR. It was cloned and transformed individually into U34,
a finger millet blast isolate lacking both PWL1 and PWL2, using Agrobacterium tumefaciens
mediated transformation and followed by infection assays on a panel of grasses. Results from
this study showed that transformed isolate U34 lost its pathogenicity on weeping lovegrass but
remained fully pathogenic on finger millet. Contrary to the current understanding of infections
involving AVR genes, it was established for the first time, that presence of PWL1 and PWL2
prevented the fungus from infecting the grass host Sporobolus phyllotrichus. The study further
revealed that the 226 finger millet blast isolates studied, grouped into two distinct classes,
differentiated by the presence or absence of PWL1 and PWL2 and their pathogenicity towards
weeping lovegrass and Sporobolus phyllotrichus. EC-1 isolates, did not have PWL1 or PWL2
and were highly virulent on both finger millet, weeping lovegrass and Sporobolus phyllotrichus.
EC-2 isolates contained either PWL1 or PWL2 or both and did not infect weeping lovegrass and
Sporobolus phyllotrichus but were highly virulent on finger millet. The avirulence of EC-2vi
isolates on weeping lovegrass and Sporobolus phyllotrichus seemed to be conditioned by the
presence of PWL1 and/or PWL2. Results further indicated that all finger millet blast isolates
with PWL4 were pathogenic on weeping lovegrass and Sporobolous phyllotrichus. It therefore
suggested that PWL4 does not modulate the infectivity of the pathogen. Gene copy number
variations, shared non-synonymous mutations and high levels of nucleotide variation diversity
were observed. Therefore, non-synonymous mutations and/or loss and gain of AVR genes
might be the main evolutionally mechanism underlying the rapid evolution of this fungus in
relation to virulence towards the host plant. Findings from this study improved the
understanding of the function and potential mechanism underlying the evolution and dynamics
of the PWL gene family.